The Internet of Energy

The Internet of Energy
Raj
Sekhar
Krishna
Mahapatra
A Pragmatic Approach Towards Sustainable Development
Providing innovative, efficient, clean, and safe solutions and research for interfacing internet technology with
Apple Academic Press

energy power grids for smart cities and smart transportation, this new volume discusses the use and
automation of electricity infrastructures for energy producers and manufacturers, integrating the
implementation of the Internet of Things (IoT) technology for distributed energy systems in order to optimize
energy efficiency and wastage. The volume offers a wide range of research on using IoT for energy solutions,
such as algorithms for the design and control of energy grids, investigations of thermal efficiency from solar
grids, energy for smart buildings using IoT, deep learning for electrical load forecasting, hybrid
ultracapacitors in solar microgrids, induction motor-driven electric vehicles, power loss reduction and voltage
improvement, and much more.
The
Author Copy
ABOUT THE EDITORS
Sheila Mahapatra, PhD, is Professor and Head, Electrical and Electronics Engineering
Department, and Assistant Dean (Research), Alliance College of Engineering and Design,
Internet of Energy
Alliance University, Bangalore, India. She has a total experience of 18 years in the field of engineering,
academics, and research. Her areas of interest include power systems optimization, FACTS devices,
renewable energy, energy economics, and sustainability. She has many research publications in SCI, SCIE,
and Scopus-indexed journals as well as book chapters and conference proceedings to her credit. She serves
Internet of Energy
as a reviewer for reputed journals.
The
Mohan Krishna S., PhD, is Associate Professor in the Department of Electrical and Electronics Engineering
at Alliance College of Engineering and Design, Alliance University, Bangalore, India. His research interests
include electric vehicles, smart homes and IoT-based building energy management systems, and state
observers for induction motors energy economics and sustainability. He has published his research in
A Pragmatic Approach Towards

journals as well as book chapters and conference proceedings. He is an advisory board member to the energy
section of Heliyon (Elsevier). He serves as Associate Editor of the International Journal of Smart Vehicles and
Smart Transportation and is also editing several books on electric vehicles, smart grids, and energy
sustainability.
B. Chandra Sekhar, PhD, is Specialist–EV in Transportation Business Unit at L & T Technology Services
Sustainable Development
Limited (LTTS), Bangalore, India. Prior to joining LTTS, he worked as a Technical Lead in the Automotive
Business Division at TATA Consultancy Services Limited, and as an R & D Engineer in the Solar Division at
Power One Micro Systems Pvt. Ltd,. Dr. Sekhar has published over 21 papers in peer-reviewed journals and
conferences. His research areas are hybrid electric vehicles, EV chargers, battery management systems,
power electronics converters, DC micro grids, LED drivers, and hybrid renewable energy systems.
Saurav Raj, PhD, is an Assistant Professor with the Electrical Engineering Department at the Institute of
Chemical Technology Mumbai, Marathwada Campus, Jalna, India. He has also worked as an Assistant
Professor at Alliance University, Bangalore, India, and as an Associate Editor of the European Journal of
Electrical Engineering. He has done extensive research work in power system optimization, reactive power
planning, swarm and evolutionary optimization techniques, FACTS devices, computational intelligence, and
renewable energy.
ISBN: 978-1-77491-418-2
90000
Sheila Mahapatra | Mohan Krishna S.

www.appleacademicpress.com 9 781774 914182 B. Chandra Sekhar | Saurav Raj
Editors APPLE ACADEMIC PRESS
Non Commercial Use

Non Commercial Use

Author Copy
THE INTERNET OF ENERGY

Author Copy
Non Commercial Use

Non Commercial Use

Author Copy
THE INTERNET OF ENERGY

Author Copy
Edited by
Sheila Mahapatra, PhD
Mohan Krishna S., PhD
B. Chandra Sekhar, PhD
Saurav Raj, PhD
Non Commercial Use

First edition published 2024
Apple Academic Press Inc. CRC Press
1265 Goldenrod Circle, NE, 2385 NW Executive Center Drive,
Palm Bay, FL 32905 USA Suite 320, Boca Raton FL 33431
760 Laurentian Drive, Unit 19, 4 Park Square, Milton Park,
Burlington, ON L7N 0A4, CANADA Abingdon, Oxon, OX14 4RN UK
© 2024 by Apple Academic Press, Inc.

Apple Academic Press exclusively co-publishes with CRC Press, an imprint of Taylor & Francis Group, LLC
Reasonable efforts have been made to publish reliable data and information, but the authors, editors, and publisher cannot assume
responsibility for the validity of all materials or the consequences of their use. The authors are solely responsible for all the chapter
content, figures, tables, data etc. provided by them. The authors, editors, and publishers have attempted to trace the copyright
holders of all material reproduced in this publication and apologize to copyright holders if permission to publish in this form has
not been obtained. If any copyright material has not been acknowledged, please write and let us know so we may rectify in any
future reprint.
Author Copy
Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or utilized in any
form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying, microfilming, and
recording, or in any information storage or retrieval system, without written permission from the publishers.
For permission to photocopy or use material electronically from this work, access www.copyright.com or contact the Copyright
Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. For works that are not available on CCC
please contact [email protected]
Trademark notice: Product or corporate names may be trademarks or registered trademarks and are used only for identification
and explanation without intent to infringe.
Library and Archives Canada Cataloguing in Publication

Title: The internet of energy : a pragmatic approach towards sustainable development / edited by Sheila Mahapatra, PhD,
Mohan Krishna S., PhD, B. Chandra Sekhar, PhD, Saurav Raj, PhD.
Names: Mahapatra, Sheila, editor. | Krishna, Mohan, 1987- editor. | Sekhar, B. Chandra, editor. | Raj, Saurav, editor.
Description: First edition. | Includes bibliographical references and index.
Identifiers: Canadiana (print) 20230539572 | Canadiana (ebook) 20230539610 | ISBN 9781774914182 (hardcover) |
ISBN 9781774914199 (softcover) | ISBN 9781003399827 (ebook)
Subjects: LCSH: Electric power systems—Automatic control. | LCSH: Electric power distribution—Data processing. |
LCSH: Energy conservation—Data processing. | LCSH: Internet of things.
Classification: LCC TK1007 .I58 2024 | DDC 621.310285/4678—dc23
Library of Congress Cataloging-in-Publication Data
CIP data on file with US Library of C

ongress
ISBN: 978-1-77491-418-2 (hbk)

ISBN: 978-1-77491-419-9 (pbk)
ISBN: 978-1-00339-982-7 (ebk)
Non Commercial Use

About the Editors
Sheila Mahapatra, PhD

Professor and Head, Electrical and Electronics
Engineering Department, Assistant Dean
(Research), Alliance College of Engineering
and Design, Alliance University, Bangalore,
Author Copy
Karnataka, India
Sheila Mahapatra, PhD, is currently working as
a Professor and Head, Electrical and Electronics
Engineering Department, and Assistant Dean
(Research), Alliance College of Engineering and
Design, Alliance University, Bangalore, Karnataka, India. She has a total
experience of 18 years in the field of engineering, academics, and research.
Her areas of interest include power systems optimization, FACTS devices,
renewable energy, energy economics, and sustainability. To her credit, she
has many research publications in SCI/ SCI-E, Scopus-indexed journals,
book chapters, and conference proceedings. CIE, and Scopus-indexed
journals as well as book chapters and conference proceedings to her credit.
She serves as a reviewer for reputed journals including IEEE Transactions
on Power Delivery, IET Generation, Transmission and Distribution, Journal
of Power Technologies, International Journal of Modelling and Simulation,
Journal of Engineering Science and Technology, Journal of Concurrency
and Computation, Practice and Experience, and several others. She is a life
member of reputed professional societies, including IEEE, the Institution of
Engineering and Technology, and the International Society for Technology
in Education. She was awarded a PhD in Electrical Engineering (FACTS
controllers’ implementation in power transmission system) from The
North Cap University, Gurugram, India. She received her BTech degree in
Electrical Engineering from Utkal University, Orissa, and MTech in Power
Systems and Automation from Andhra University, A.P., India, in 2002 and
2008, respectively. She was the batch topper while pursuing her master’s
degree from Andhra University, A.P., India.
Non Commercial Use

vi About the Editors
Mohan Krishna S., PhD

Associate Professor, Department of Electrical
and Electronics Engineering, Alliance College
of Engineering and Design, Alliance University,
Bangalore, India
Mohan Krishna S., PhD, is an Associate
Professor in the Department of Electrical and
Electronics Engineering at Alliance College of
Engineering and Design, Alliance University,
Bangalore, India. His research interests include, electric vehicles, smart
Author Copy
homes and IoT-based building energy management systems, and state
observers for induction motors energy economics and sustainability. He
has published his research in SCI/SCIE and Scopus-indexed journals as
well as book chapters and conference proceedings. He is an advisory board
member to the energy section of Heliyon (Elsevier). He also serves as
Associate Editor of the International Journal of Smart Vehicles and Smart
Transportation. Additionally, he is a member of the editorial review board
of the International Journal of Energy Optimization and Engineering. He
is also editing several books in the domain of electric vehicles, smart grids,
and energy sustainability. He was awarded a PhD in Electrical Engineering
(sensorless control of induction motor drives for EV applications) from
Vellore Institute of Technology (VIT), India, and his BTech and MTech
degrees from Amrita Vishwa Vidyapeetham, Coimbatore, India. He also
acquired a domain-specific MBA (Power Management) from the University
of Petroleum and Energy Studies (UPES), Dehradun, India.
B. Chandra Sekhar, PhD

Specialist–EV in Transportation Business Unit
at L&T Technology Services Limited (LTTS),
Bangalore, India
B. Chandra Sekhar, PhD, is currently working as
a Specialist–EV in Transportation Business Unit
at L&T Technology Services Limited (LTTS),
Bangalore. Prior to joining LTTS, he worked as
a technical lead in Automotive Business Division
at TATA Consultancy Services Limited (TCS),
Bangalore. Subsequently, he worked as a R&D Engineer in Solar Division at
Power One Micro-Systems Pvt. Ltd, Bangalore. He received his PhD degree
Non Commercial Use

About the Editors vii
in Electrical Engineering from the Central Power Research Institute (CPRI)

Research Center, Bangalore, Karnataka. He is a Professional Member of
IEEE (94812129). He is a Board of Studies (BoS) committee member for
New Horizon College of Engineering (NHCE), Bengaluru; SJB Institute of
Technology, Bengaluru; NBKR Institute of Technology, Nellore, Andhra

Pradesh; Global Academy of Technology, Bengaluru; Bapatla Engineering
College, Bapatla, Andhra Pradesh; and Basaveshwar Engineering College
(BEC), Bagalkot, Karnataka. He has published over 21 papers in peer-
reviewed journals and conferences (international journals [4], international
conferences [8], national journal [7], and national conferences [2]). His
research areas are hybrid electric vehicles, EV chargers, battery management
Author Copy
systems (BMS), power electronics converters, DC microgrids, LED drivers,
and hybrid renewable energy systems.
Saurav Raj, PhD

Assistant Professor, Electrical Engineering
Department, Institute of Chemical Technology,
Mumbai, India
Saurav Raj, PhD, is an Assistant Professor in the
Electrical Engineering Department at the Institute
of Chemical Technology Mumbai, Marathwada
Campus, Jalna, India. He has also worked as
Assistant Professor at Alliance University, Banga-
lore, India. He has been an Associate Editor of the
European Journal of Electrical Engineering and Editorial Board Member of
the SCIREA Journal of Electrical Engineering. He has also been serving as a
reviewer for reputed journals and conferences, including IET-GTD, Elsevier,
Springer, Taylor & Francis, IGI Global USA, and several other national/inter-
national conferences of repute. He has done extensive research work in power
system optimization, reactive power planning, swarm, and evolutionary
optimization techniques, FACTS devices, computational intelligence, and
renewable energy. He graduated from R. P. Sharma Institute of Technology,
Patna, India, in Electrical and Electronics Engineering. He received his PhD
from IIT (ISM) Dhanbad, India.
Non Commercial Use

Non Commercial Use

Author Copy
Contents
Contributors..............................................................................................................xi
Abbreviations........................................................................................................... xv
Preface.................................................................................................................... xxi
1. Design and Control Using an ISOGI-Q Algorithm for Grid
Author Copy
Integration of PV-Array..................................................................................1
Alka Singh and Praveen Bansal
2. Thermal Efficiency Investigation of a Solar Parabolic Trough

Collector System Using Two Dissimilar Reflectors.....................................21
S. N. Vijayan, S. Sendhil Kumar, and S. Karthik
3. Smart Buildings Using IoT and Green Engineering...................................37

Amit Kumar Singh, Anshul Gaur, and Pradeep Kumar
4. A Novel Electrical Load Forecasting Model Using a Deep

Learning Approach........................................................................................67
Neelapala Anil Kumar, Ravuri Daniel, and Prudhvi Kiran Pasam
5. Battery Plant Model Development for BMS Application...........................91

G. N. Dhanya and K. V. Abhinand
6. Modeling of Constant Voltage Control in Synchronous

Buck and Boost Converters Using MATLAB/Simulink for
Point-of-Load Application...........................................................................103
Sumukh Surya and Vineeth Patil
7. Determination of Open Loop Responses of Switched DC–DC

Converters Using Various Modeling Techniques......................................125
Sumukh Surya
8. Evolution of Hybrid Ultracapacitors in Solar Microgrids.......................155

J. Pradeep Kumar Rao and H. N. Nagamani
9. Speed Sensorless Model Predictive Current Control of Induction

Motor-Driven Electric Vehicle....................................................................179
Karuna Kiran
Non Commercial Use

x Contents
10. Performance Analysis of Hybrid Off-Grid Two-Axis Photovoltaic

Tracking System/Fuel Cell Energy System Incorporating
High-Efficiency Solar Cell...........................................................................195
Shubhashish Bhakta, Mesfin Megra, Pikaso Pal, and Ashebir Berhanu
11. Efficient Integration of Distributed Generation in Radial

Distribution Network for Voltage Profile Improvement and
Power Loss Minimization via Particle Swarm Optimization..................207
Pragya Guru, Nitin Malik, and Sheila Mahapatra
12. Classical and Predictive Control of Interior Permanent Magnet

Synchronous Motor for Railway Application............................................227
Author Copy
Mannan Hassan, Muhammad Suhail Shaikh, Muhammad Shahid Mastoi,
Rao Atif, Muhammad Farhan, Muhammad Amjad, Muhammad Bilal Shahid, and
Abdul Latif Shah
13. Electric Mobility...........................................................................................251

Ashwini Kumar Sharma, Srimanti Roychoudhury, and Sunam Saha
14. Power Loss Reduction and Voltage Improvement Through

Capacitors and Their Optimization for the Distribution System............267
Shubash Kumar, Chandar Kumar, Muhammad Suhail Shaikh,
Anwar Ali Sahito, and Zahid Ali Arain
15. Solar Photovoltaic Powered Automatic Irrigation System for

the Agriculture Sector..................................................................................295
Himanshu Sharma and Pankaj Kumar
16. Optimal Parameter Estimation of 3-Phase Transmission Line

Using a Grey Wolf Optimization Algorithm..............................................313
Muhammad Suhail Shaikh, Abdul Latif Shah, Shafiq Ur Rehman Massan,
Rabia Ali Khan, Munsif Ali Jatoi, Shubash Kumar, and Mannan Hassan
17. Network Reconfiguration-Based Outage Management for Reliability

Enhancement of Microgrid: A Hardware in Loop Approach..................337
Shruti Prajapati, Sonal, Sourav Kumar Sahu, and Debomita Ghosh
18. A Net Energy Meter-Based Approach for Islanding Detection in

Modern Distribution Systems.....................................................................359
Soham Dutta, Akash Kumar Pandey, Sourav Kumar Sahu, and Pradip Kumar Sadhu
Index......................................................................................................................385
Non Commercial Use

Contributors
K. V. Abhinand
Software Engineer, e-Powertrain KPIT, Bangalore, Karnataka, India
Muhammad Amjad
Department of Electronic Engineering, Faculty of Engineering, The Islamia University of Bahawalpur,
Pakistan
Author Copy
Zahid Ali Arain
Department of Sciences and Technology, Indus University, Karachi, Sindh, Pakistan
Rao Atif
Ministry of Education Key Laboratory of Magnetic Suspension Technology and Maglev Vehicle,
School of Electrical Engineering, Southwest Jiaotong University, Chengdu, China
Praveen Bansal
Department of Electrical Engineering, Delhi Technological University, Delhi, India
Ashebir Berhanu
Department of Electrical Power and Control Engineering, Adama Science and Technology University,
Adama City, Ethiopia
Shubhashish Bhakta
Ravuri Daniel
Department of Computer Science and Engineering, Bapatla Engineering College, Bapatla,
Andhra Pradesh, India
G. N. Dhanya
Soham Dutta
Department of Electrical and Electronics Engineering, Manipal Institute of Technology,
Manipal Academy of Higher Education, Manipal, Karnataka, India
Muhammad Farhan
Department of Electrical Engineering and Technology, Government College University Faisalabad, Pakistan
Anshul Gaur
Electronics and Communication Department, Uttarakhand Technical University, Uttarakhand, India
Debomita Ghosh
Department of Electrical and Electronics Engineering, BIT Mesra, Ranchi, Jharkhand, India
Pragya Guru
School of Engineering and Technology, The NorthCap University, Gurgaon, Haryana, India
Mannan Hassan
School of Electrical Engineering, Southwest Jiaotong University, Chengdu, Republic of China
Non Commercial Use

xii Contributors
Munsif Ali Jatoi

Salim Habib University, Karachi, Pakistan
S. Karthik
Assistant Professor, Department of Mechanical Engineering, Sri Krishna College of Engineering and
Technology, Coimbatore, Tamil Nadu, India

Rabia Ali Khan
Newports Institute of Communications and Economics, Karachi, Pakistan
Karuna Kiran
Indian Institute of Technology (Indian School of Mines), Dhanbad, Jharkhand, India
Chandar Kumar
Department of Sciences and Technology, Indus University, Karachi, Sindh, Pakistan
Author Copy
Neelapala Anil Kumar
Department of Electronics and Communication Engineering (ACED), Alliance University, Bangalore,
Karnataka, India
Pankaj Kumar
Department of EEE, SRMIST, Delhi NCR Campus, Ghaziabad, Uttar Pradesh, India
Pradeep Kumar
Electrical Engineering Department, Ambalika Institute of Technology and Management, Lucknow,
Uttar Pradesh, India
S. Sendhil Kumar
Professor, Department of Aeronautical Engineering, Annasaheb Dange College of Engineering and
Technology, Ashta, Sangli, Maharashtra, India
Shubash Kumar
School of Electrical Engineering, Yanshan University, Qinhuangdao, P. R. China; Department of Sciences
and Technology, Indus University, Karachi, Sindh, Pakistan
Sheila Mahapatra
School of Engineering and Technology, Alliance College of Engineering and Design, Bangalore,
Karnataka, India
Nitin Malik
School of Engineering and Technology, The NorthCap University, Gurgaon, Haryana, India
Shafiq Ur Rehman Massan

Muhammad Shahid Mastoi

School of Electrical Engineering, Southwest Jiaotong University, Chengdu, China
Mesfin Megra
H. N. Nagamani
Addl. Director (Retd.), Central Power Research Institute, Bangalore, Karnataka, India
Pikaso Pal
Department of Electrical Engineering, Indian Institute of Technology (Indian School of Mines),
Dhanbad, Jharkhand, India
Non Commercial Use

Contributors xiii
Akash Kumar Pandey

Design Engineer, Larsen and Toubro, Chennai, India
Prudhvi Kiran Pasam

Department of Information Technology, SRKR Engineering College, Bhimavaram, Andhra Pradesh,
India
Vineeth Patil
Department of Electrical and Electronics Engineering, Manipal Institute of Technology, Manipal Academy
of Higher Education, Manipal, Karnataka, India
Shruti Prajapati
J. Pradeep Kumar Rao

Department of Engineering (R&D), Central Power Research Institute, Bangalore, Karnataka, India
Author Copy
Srimanti Roychoudhury
School of Engineering and Technology, Adamas University, Kolkata, West Bengal, India
Pradip Kumar Sadhu

Department of Electrical Engineering, IIT(ISM) Dhanbad, Jharkhand, India
Sunam Saha
School of Engineering and Technology, Adamas University, Kolkata, West Bengal, India
Anwar Ali Sahito

Department of Electrical Engineering, Mehran University of Engineering and Technology, Jamshoro,
Sindh, Pakistan
Sourav Kumar Sahu
Abdul Latif Shah

Muhammad Bilal Shahid

School of Electrical Engineering, Southwest Jiaotong University, Chengdu, China; Department of
Electronic Engineering, Faculty of Engineering, The Islamia University of Bahawalpur, Pakistan
Muhammad Suhail Shaikh

School of Physics and Electronic Engineering, Hanshan Normal University, Guangdong, China
Ashwini Kumar Sharma

Graphic Era, Deemed to be University, Dehradun, India
Himanshu Sharma
Department of EEE, SRMIST, Delhi NCR Campus, Ghaziabad, Uttar Pradesh, India
Alka Singh
Department of Electrical Engineering, Delhi Technological University, Delhi, India
Amit Kumar Singh

Biomedical Engineering Department, VSB Engineering College, Karur, Tamil Nadu, India
Sonal
Non Commercial Use

xiv The Internet of Energy
Sumukh Surya
Senior Engineer, Bosch Global Software Technologies Private Limited, Bangalore, Karnataka, India
S. N. Vijayan
Assistant Professor, Department of Mechanical Engineering, Karpagam Institute of Technology,
Coimbatore, Tamil Nadu, India
Author Copy
Non Commercial Use

Abbreviations
ABC artificial bee colony

AC alternating current
ACO ant colony optimization
AEN average energy not supplied
AHHO ameliorated Harris Hawks optimization
Author Copy
AI artificial intelligence
ALO ant lion optimizer
ANN artificial neural network
BA bat algorithm
BA bat-inspired
BBO biogeography-based optimization
BEMS building energy management system
BEVs battery electric vehicles
BIBC bus injection to branch current
BioCI bioinformatics CI
BIPV building-integrated photovoltaic
BIPV/L building-integrated PV/light
BIPV/T building-integrated PV/thermal
BLE Bluetooth low energy
BMIS building management and information system
BMS battery management system
BPSO binary particle swarm optimization
C capacitor
CA circuit averaging
CAIDI consumer average interruption duration index
CAS compare and swap algorithm
CC constant current
CCCV constant current constant voltage
CCM continuous conduction mode
CdS cadmium sulfide
CH4 methane
CHB-MLI cascaded H-bridge inverter
ChmCI chemistry CI
CI computational intelligence
Non Commercial Use

xvi Abbreviations
CO carbon mono-oxide
CO2 carbon dioxide
COD chemical oxygen demand
CPRI Central Power Research Institute
CS cuckoo search
CSA cuckoo search algorithm
CV constant voltage
DA dragonfly algorithm
DC direct current
DCM discontinuous conduction mode
DCMLI Diode clamped multilevel inverter
Author Copy
DE differential evolution
DE dolphin echolocation
DERs distributed energy resources
DFIG doubly-fed induction generator
DG distributed generation
DLF direct load flow
DLF distribution load flow
DMF dimethylformamide
DNI direct normal irradiance
DoD depth of discharge
DSTATCOM distributed static compensator
ECM equivalent circuit model
EDLC electrical-double-layer-capacitor
EMSOGI enhanced multilayer second-order generalized integrator
EP evolution programming
ES energy sources
ESR equivalent series resistance
EV electric vehicle
FAME faster adoption and manufacturing of electric
FCMLI flying capacitor inverter
FOA fruit fly optimization algorithm
FPA flower pollination algorithm
GA genetic algorithm
GHI global horizontal irradiance
G-IoT green internet of things
GM gain margin
GP genetic programming
GSA gravitational search algorithm
Non Commercial Use

Abbreviations xvii
GWO grey wolf optimizer

GWOSCACSA grey wolf optimizer sine cosine and crow search algorithm
H.T. high tension
HESCO Hyderabad electric supply company
HEVs hybrid electric vehicles

HHO Harris Hawks optimization
HIL hardware in loop
HOMER hybrid optimization of multiple energy resources
HS Hirschberg–Sinclair algorithm
HVAC heating, ventilation, and air-conditioning
ICA imperialistic competitive algorithm
Author Copy
IEEE Institute of Electrical and Electronics Engineers
IEQ indoor environment quality
IGBT insulated gate bipolar transistor
IM induction motor
IoE internet of energy
IoT Internet of Things
IPMSM interior permanent-magnet synchronous machines
IR infrared
ISOGI-Q Improved Second-Order Generalized Integrator
KH krill herd
kV kilo volt
KVA kilo volt-ampere
KVAR kilo volt ampere reactive
KW kilowatt
LaRC Langley research center
LF load flow
LoRa long-range
LQR linear–quadratic regulator
LSTM long short-term memory
LT Network long term network
LTE long-term evolution
LTE-M long term evolution for machines
MAPE mean absolute percentage
MathCI mathematics CI
MCA modified cultural algorithm
MCS Monte-Carlo simulation
MFO moth–flame optimization
MGs microgrids
Non Commercial Use

xviii Abbreviations
ML machine learning
MLP multilayer perceptron
MPC model predictive control
MPCC model-predictive-current control
MTLBO modified teaching-learning-based optimization

MTPA maximum torque per ampere
MTPV maximum torque per voltage
NASA National Aeronautics and Space Administration
NB-IoT narrow band-internet of things
NDZ non-detection zone
NEM net energy meter
Author Copy
NTC negative temperature coefficient
NTLs non-technical losses
OCP open circuit potential
OCV open circuit voltage
OFDM orthogonal frequency-division multiplexing
OGWO oppositional grey wolf optimization
OHHO oppositional Harris Hawk optimization
OHTL overhead transmission line system
P photovoltaic
P.F. power factor
Pb-C HUC lead carbon hybrid ultracapacitor
PBIL probabilistic incremental learning
PbO2 lead oxide
PCC point of common coupling
PFCC power factor correction capacitor
PHEVs plugin hybrid electric vehicles
PhyCI physicist CI
PI proportional–integral
PIV peak inverse voltage
PLL phase lock loops
PM permanent magnet
PM phase margin
PMU phasor measuring unit
POL point of load
PQ power quality
PRV pressure regulating valve
PSO particle swarm optimization
PSSSINCAL power system simulator siemens calculation
Non Commercial Use

Abbreviations xix
PTC parabolic trough collector

PV photovoltaic
PV photovoltaic array
PVDF polyvinylidene fluoride
PWM pulse width modulation

R feeder resistance
R resistor
RDN radial distribution network
RESs renewable energy resources
RF radiofrequency
RHP right half-plane
Author Copy
RMSE root mean squared
RNN recurrent neural networks
RTD resistance temperature detectors
S apparent power
S&H sample and hold circuit
SA simulated annealing
SAIDI system average interruption duration index
SAIFI system average interruption frequency index
SAW surface acoustic wave
SCA sine–cosine algorithm
SCADA supervisory control and data acquisition
SLFA shuffled frog leap algorithms
SMPS switch mode power supplies
SoC state of charge
SOC state of charge
SOGI second-order generalized integrator
SPMSM surface permanent magnet synchronous machine
SPV solar photovoltaic
SRFT synchronous reference frame theory
SSA state space averaging
SSO shark smell optimization
STATCOM static compensator
STLF short-term load forecasting
STSs solar thermal systems
SVM space vector modulation
THD total harmonic distortion
TLBO teaching-learning-based optimization
TS tabu search
Non Commercial Use

xx Abbreviations
TSC thyristor-switched capacitors

UPS uninterruptible power supplies
USB universal serial bus
UWB ultra-wideband
VCRPP voltage-constrained reactive power planning

Vnl voltage at no load
VR Voltage Regulation
Vrated rated voltage
VSC voltage source converter
WCA water cycle algorithm
WHT Walsh Hadamard transform
Author Copy
Wi-Fi wireless fidelity
WOA whale optimization algorithm
WUSB wireless USB
XL feeder reactance
Z feeder impedance
ZCD zero current detector circuit
Non Commercial Use

Preface
As a result of deregulation in the electrical sector, energy is being treated as a

commodity. Further, with the integration of the renewable sector, smart grid
technology and electric vehicle (EV) technology are a force to reckon with
as a futuristic goal. So, it has now become imperative to collect, organize
and analyze the data from interconnected smart grid infrastructure elements
Author Copy
spread across generation, loads, storage units, and self-regulating apparatus
in distribution centers. Internet of Energy (IoE) is poised to become a neces-
sity with the substantial amplification in data and time required to analyze
information where the SCADA system has its limitations. It paves the way
toward a smarter world that drives energy management to be more resilient,
responsive, and efficient.
Internet of Energy is a technical adage primarily referring to ameliorating
and automating the electrical infrastructure for energy producers and
manufacturers. In simple words, IoE is an intelligent interface between
people, processes, data, and things around us. IoE focuses on the realization
of the technicality of the Internet of Things in distributed energy sectors for
optimizing energy efficacy and ruination in electrical infrastructure.
In the decentralized electrical system, innovative networking technologies
will become increasingly important. Some vital reasons why we need to
transform our old grids into an Internet of Energy include: enhanced global
demand for clean energy, a radical shift in the energy sector, which presents
tremendous challenges that require innovative means in energy management,
centralized grids are no longer adequate and smart grids ensure more reliability.
IoE will act as an enabler for coordinating between consumers and
producers to adequately meet demand and supply in an automated environ-
ment facilitated by smart, intelligent forecasting systems in a pursuit to
envisage future energy demand. This stage is crucial for the operational
optimization of the grid and enhancing system management for adequate
handling of contingencies, storage monitoring, and load shedding.
IoE would play a pivotal role in the cost-effective integration of
renewables to the grid, precise planning for energy markets, and grid exten-
sion, paving the way toward higher profitability and trading, but the most
important challenge would be enhancing grid cybersecurity standards. As
the future of the smart grid is digitalization, and so the call of the hour is
Non Commercial Use

xxii Preface
to prepare ourselves as IoE would possibly be the answer to the integration

of all these and develop skills for the next generation workforce where AI
and quantum computing is transforming the energy sector, which requires
innovative networking technologies.
This book is an attempt by authors to showcase contemporary work in

IoE. We endeavor to project the book as a valuable tool for an extensive set
of audiences to provide immense opportunities in hardware and software
integration as well as data communication for future electric grids. It attempts
to provide innovative solutions for interfacing the internet with the power grid
for smart cities and helping transport to be more efficient, clean, and safe. The
Internet of Energy is more relevant now as decarbonization, decentralization,
Author Copy
and digitization are the three new global trends are transforming the energy
sector and revisiting the way we produce, distribute, and consume electric
power.
Non Commercial Use

CHAPTER 1
Design and Control Using an ISOGI-Q

Algorithm for Grid Integration of PV-Array
ALKA SINGH and PRAVEEN BANSAL
Department of Electrical Engineering, Delhi Technological University,
Author Copy
Delhi, India
ABSTRACT
This chapter proposes the grid integration of a single-stage photovoltaic

array (PV) via a voltage source converter (VSC). The VSC is controlled
for active power injection as well as a distributed static compensator
(DSTATCOM). The control of DSTATCOM is designed using an improved
second-order generalized integrator (ISOGI-Q) based algorithm. In the last
two decades, extensive use of nonlinear loads has polluted the grid severely.
Various power quality (PQ) issues like harmonics injection, overheating of
conductors, power factor, voltage regulation, voltage sag, swell, etc., are
quite common problems.
Mitigation of PQ issues can be easily achieved using DSTATCOM, which
can be designed to inject a current to cancel the harmonics generated by the
nonlinear loads. This chapter discusses a five-level cascaded H-bridge multilevel
inverter (CHB-MLI) realized as DSTATCOM Such a configuration shows
several advantages over conventional two-level inverters. The CHB-MLI is
always preferred due to its modular structure, ease of a developing prototype,
compact size, and no requirement of clamping diodes and flying capacitors as
in conventional multilevel inverters.
The controller for the CHB-MLI is designed on a widely used generalized
integrator (SOGI) algorithm whose performance is further enhanced to
provide good performance in the case of DC offset and high harmonics.
The Internet of Energy: A Pragmatic Approach Towards Sustainable Development. Sheila Mahapatra,
Mohan Krishna S., B. Chandra Sekhar, & Saurav Raj (Eds.)
© 2024 Apple Academic Press, Inc. Co-published with CRC Press (Taylor & Francis)
Non Commercial Use

2 The Internet of Energy
The chapter discusses the ISOGI-Q control scheme implemented to extract

the fundamental load current and eradicate all harmonics generated by
nonlinear loads. Further, to utilize the full potential of PV-DSTATCOM, the
proposed system is simulated to work in two modes viz daytime (Mode-I)
and nighttime (Mode-II). Mode-I refers to the daytime operation when the
active power injection is a priority, while in mode-II, reactive power is fed
to the load during the night for power quality enhancement.
The proposed system is modeled and simulated in MATLAB/Simulink
software. Phase shift pulse width modulation scheme is used to generate
the firing pulses for DSTATCOM. Simulation and experimental results are
discussed in detail, which depicts the satisfactory performance of the system.
Author Copy
Results are highlighted, which show the harmonic content in source current
brought within IEEE stipulated limits.
1.1 INTRODUCTION
The use of renewable energy is always preferred by power engineers and

researchers. Renewable energy sources (ES) are becoming increasingly
important all around the world. The challenges connected with conven-
tional energy resources, such as energy shortages, pollution from fossil
fuel combustion, and resource scarcity in the near future, have prompted
a massive hunt for innovative alternatives in many countries. Although
there are other non-conventional ES available today, solar and wind energy
now dominate the market [1]. A fine example of a viable and independent
energy generation system is a grid-connected PV system [2]. Energy insuf-
ficiency and significant import dependence on oil and coal imply serious
problems with energy security. Energy production and consumption should
be balanced. To minimize environmental changes, local energy generation
must be established in the distribution network through the use of rooftop
photovoltaic (PV) arrays. As a result, rooftop PV array utilization in small-
scale applications is on the upswing [3]. However, it is also noteworthy
that a higher proportion of nonlinear loads in the distribution network is the
primary cause of power quality (PQ) problems.
Nonlinear loads in the distribution network result in poor power quality.
Extensive use of power electronics controlled variable speed drives,
Uninterruptible Power Supplies (UPS), arc furnaces, and residential loads
like computers and printers directly inject harmonics in the grid and pollute
it [4]. In Ref. [5], authors have proposed a different design configuration
Non Commercial Use

Design and Control Using an ISOGI-Q Algorithm 3
of PV integrated voltage source inverters. Wu et al. in Ref. [6] consider

losses of single-stage and double-stage grid-connected PV-tied systems. The
single-stage grid connection shows various advantages over the double stage
in terms of the component cost of capacitors, diodes, switches, etc., and also,
the single stage shows the advantage of higher efficiency, lower losses, and
enhanced utilization of solar PV array. The improvement of power quality is
primarily based on the selection of an inverter.
Conventional 2-level inverters suffer from higher switching losses, espe-
cially in medium and high voltage systems, and also have high peak inverse
voltage (PIV) ratings of switches; therefore, the power switches suffer from
high dv/dt stress. Therefore, nowadays, MLI is widely used for medium and
Author Copy
high-power distribution systems [7], although it has been introduced lately.
It possesses the capability to handle more power with reduced PIV rating of
switches and lower stress, reduced filter, and reduced THD in output voltage
and current.
MLI is broadly classified into three categories viz. (i) diode clamped
inverter (DCMLI); (ii) flying capacitor inverter (FCMLI); and (iii) cascaded
H-bridge inverter (CHB-MLI). In this chapter, CHB-MLI is used as a
DSTATCOM unit [8] because of its modular structure, easy assembly, avail-
ability of redundant states, and no requirement for any clamping diodes and
flying capacitors.
The synchronization and control of the proposed system as a compensator
are important. Synchronization requires the use of commonly used circuits,
viz. Phase Lock Loops (PLL). The amplitude, phase angle, and frequency of
the input signal are all computed by PLLs. The most popular synchroniza-
tion technique is Synchronous Reference Frame Theory (SRFT), proposed
by Golestan et al. [9]; and Panda, Pathak, & Srivastava [10]. The SRF-PLL
works satisfactorily in normal grid conditions, but its performance deteriorates
under abnormal grid conditions. Hence, there is a need to modify or redesign
the synchronizing circuits to handle abnormalities in the grid. In this chapter
[11], the authors have implemented the SRF-PLL for a grid-tied PV array.
Many new and adaptive PLL schemes are now reported in the literature, and
generalized integrators such as Second Order Generalized Integrator (SOGI-
PLL) based PLL have been proposed by Rodriguez et al. [12]. Further authors
in Ref. [13] have presented detailed comparisons of various PLL circuits like
Enhanced PLL, delay, and SOGI-PLL-based harmonic compensation for
single-phase PV integrated grid-connected systems. The two-level inverter
faces several issues, as discussed above; hence an alternative solution is
to use MLIs, especially CHB-MLI. In Ref. [14], authors have discussed a
Non Commercial Use

comprehensive review of 88 papers on PV array integration-based CHB-MLI.

In Ref. [15], authors have experimentally investigated the integration of two
PV arrays with 5-level MLI for a large-scale grid system. In Ref. [16], the
authors have implemented an enhanced multilayer second-order general-
ized integrator (EMSOGI) double-stage single-phase system, which gives

better performance than conventional SOGI in terms DClink settlement and
harmonics mitigation.
The conventional SOGI-PLL suffers from poor harmonic elimination,
and DC offset issues. To overcome these problems, this chapter realizes an
ISOGI-Q adaptive PLL algorithm to mitigate the harmonics generated by the
load current under normal grid conditions. The major contributions of the
Author Copy
chapter are as follows:
• The extradition of the fundamental component of load current using
the ISOGI-Q algorithm.
• Enhanced utilization of PV arrays due to two DC link voltages on the
DSTATCOM side.
• Self-balancing of DC link voltages without the need for complex DC
voltage balancing circuits.
• Under dynamic conditions such as solar irradiance, a feed-forward
PV component is used in the control system to make proper balance
in the system.
• Proposed system is operated in two modes, viz. (i) To provide active
power to the grid and (ii) to provide the reactive compensation and
achieve unity power factor operation.
• The designed ISOGI-Q algorithm ensures better filtering capabilities
as compared to conventional SRFT and SOGI algorithms.
• The proposed system is tested via MATLAB/Simulink under dynamic
load conditions.
1.2 DESIGN AND CONFIGURATION OF PROPOSED SYSTEM
The proposed system considers a single-phase, single-stage grid-connected

system with two PV arrays of 1 kW; each capacity interfaced at the DC
link side of 5-level CHB-MLI. The CHB-MLI is realized as a DSTATCOM
unit. It is controlled to suppress the current harmonics generated by the
nonlinear load modeled in the form of a diode rectifier feeding a resistive-
inductive load. The complete system layout of the proposed system is
shown in Figure 1.1.
Non Commercial Use

Author Copy
FIGURE 1.1 Complete block diagram of a system with ISOGI-Q algorithm.
1.2.1 DESIGN OF DC LINK VOLTAGE REFERENCE
A single phase, 110 V (AC rms) supply is connected at the PCC. The refer-
ence DC link voltage required is calculated by using the given relation [17].
N 2 ×Vg
EDC
= ∑=
E
j =1
DC , j
mi
(1)
where; mi is the modulation index = 0.9, the grid voltage is Vg is considered

to be 110 V (rms), then the calculated DC link reference voltage is 172.82
V, and in the simulation, it is approximated as 200 V, for both the H-bridges
connected in a cascaded fashion.
1.2.2 DESIGN OF INTERFACING INDUCTOR
The value of the interfacing inductor is calculated [17] as follows:

Eo , rms 155.54
=Linf = = 2.52mH (2)
8 × g × f r × ∆I g 8*1.2* 2500* 25.70*0.1
Non Commercial Use

The estimated value of the inductor is 2.52 mH, and in the simulation, it
is considered as 3 mH. Here, Eo,rms is the rms output voltage of DSTATCOM,
g is the overloading factor taken as 1.2, fr is the switching frequency taken
as 2.5 KHz, and the ripple current is considered to be 10% of the peak grid
current. The grid current is calculated as follows:

PPV (max) 2000
Ig
= = = 18.18 A (3)
Vg 110
The peak value of grid current is calculated as:

I peak =2 ×18.18 =25.70 A (4)
Author Copy
1.2.3 DESIGN OF DC LINK CAPACITORS
In the proposed system, there are two DC link capacitors are used with 100
V each; their value can be calculated as
PDCj / EDCj 1000 / 100
=CDC , j = = 3184.7133µ F (5)
2 × ωr × EDC − ripple 2 × 314 × 0.05 × 100
The estimated value of the inductor is 3184.7133 μF, and in the simulation,
it is considered as 3 μF each here; for the jth PV array, the DC power is
PDCj = 1000, and DC link voltage is EDCj = 100 V each, ωr is the angular
frequency taken as 314 rad/sec and the DC voltage ripple is EDC-ripple and it
is considered as 5% of each DC link voltage.
1.3 CONTROL ALGORITHM
The overall implementation of the ISOGI-Q control algorithm is depicted in

Figure 1.2. As shown in Figure 1.2, the fundamental estimated component
of load current (If) is extradited using the ISOGI-Q control algorithm. The
overall control structure involves various calculations to achieve stable closed
loop operations such as unit vector generation, DC link voltage control under
varying load conditions, extraction of the real component of load current,
determining reference current, and then finally, generation of PWM pulses for
the firing of Insulated Gate bipolar junction transistors (IGBT). The proposed
system also comprises a sample and hold circuit (S&H) and a zero current
detector circuit (ZCD). The extracted fundamental load current is fed to the
S&H circuit, and it is synchronized with the ZCD; when the unit synchronizing
Non Commercial Use

template crosses the zero, the ZCD generates the triggering signal, which is
further fed to the S&H circuit. The S&H logic circuit captures the samples of
the sensed load current once it receives the signal from the ZCD circuit. As a
result, an accurate and fast estimation of the signal is achieved.
Author Copy
FIGURE 1.2 Overall closed-loop implementation of ISOGI-Q algorithm.
For a nonlinear load distributed network, the fundamental load compo-

nent of current is extracted from the nonlinear load by using the ISOGI-Q
algorithm. As shown in Figure 1.3, the in-phase and quadrature components
AQ: Eqn 6
are iLα and iLβrespectively. The transfer function can be represented as is missing,
please
iLα ksβ × ω × s 2 check
= 3 (7)
iL s + (k sβ + k DC )ω s 2 + ω 2 s + k DC ω 3
iDC k DC × ω × ( s 2 + ω 2 )
= 3 (8)
iL s + (k sβ + k DC )ω s 2 + ω 2 s + k DC ω 3
Here from Eqns. (7) and (8), to obtain the system stability, the value of ksβ
and kDC must be tuned using the Routh-Hurwitz stability criterion. All roots
must have real parts and be equal.
1.3.1 GENERATION OF UNIT VECTOR TEMPLATE
The unit vector template or synchronizing template is generated from grid

voltage Vg. The grid voltage is passed through a delay of 90°, as shown in
Non Commercial Use

Figure 1.3. The in-phase component is considered as Vgp, and the quadrature
component is Vgq, these voltage vectors are further used to generate Vt. Now
the synchronizing template (up) and quadrature synchronizing template (uq)
are calculated as
Vgp Vgq
up
= ; =
uq ;=
Vt Vgp2 + Vgq2 (9)
Vt Vt
Author Copy
FIGURE 1.3 Generalized structure of ISOGI-Q.
1.3.2 DC LOSS CALCULATION AND GENERATION OF REFERENCE

CURRENT
The proposed structure serves two purposes: (i) providing active power to
the AC grid; (ii) compensating the harmonics generated by the nonlinear
load and correcting the power factor of the supply side to unity. For the
effective operation of the proposed system under both conditions, there is a
necessity to control the fluctuations in DC link voltages obtained across PV
arrays. Therefore, a proportional–integral (PI) control is used to control the
DC link voltages. The DC link error can be estimated as
EDCe EDC − ref − EDC
= (10)
The error signal is fed to the PI controller, and Iloss is calculated, as shown in
Figure 1.2. Mathematically it can be represented as
I loss (n +=
1) I loss (n) + k p {EDCe (n + 1) − EDCe (n)} + ki EDCe (n + 1) (11)
Non Commercial Use

where; kp and ki denote the proportional and integral gains, respectively, and
the system dynamic performance of current and voltage can be improved by
the feed-forward term PPV can be estimated as:
2( PPV 1 + PPV 2 )
I PV = (12)
Vt
The reference current is generated by multiplying the unit synchronizing
template with the estimated load current, represented as:
i*gr = u p I est (13)
I est = I loss + I f − I PV (14)
Author Copy
The fundamental estimated load current is Iest, the fundamental load current
is If, and PV feed-forward current is IPV. The generated reference current is
subtracted from the actual grid current, and further, the signal is compared
with phase-shifted PWM techniques to generate firing pulses for 5-Level
CHB-MLI.
1.4 SIMULATION RESULTS
The proposed system is simulated using MATLAB 2020R with two different
modes under steady state and dynamic load and solar irradiation variations.
The parameters used for simulation are given in Table 1.1.
TABLE 1.1 Parameter Used in Simulation

SL. No. Parameters Symbol Value Used in Simulation
1. Single-phase AC power supply Vg(V) 110 V, RMS
2. DC link capacitors CDC1 and CDC2 3,000 μF each
3. Interfacing Inductor Lf 3 mH
4. Reference DC link voltage Eref 200 V
5. PI gains Kp and Ki 1.22, 0.6
6. Open circuit voltage VOC1 and VOC2 70.6 V each
7. Short circuit current ISC1 and ISC2 18.4 A
1.4.1 STEADY STATE PERFORMANCE OF PV-DSTATCOM IN MODE-1
During Mode-1 operation, the system is working with PV-DSTATCOM and

providing active power to the grid. Figure 1.4 depicts the steady state results
Non Commercial Use

of grid voltage Vg(V), grid current Ig(A), load current IL(A), and total DC
link voltage VDC(V) with fixed solar irradiance of 1,000 W/m2.
Author Copy
FIGURE 1.4 Steady-state results of grid voltage Vg(V), grid current Ig(A), load current
IL(A), and total DC link voltage VDC(V).
Figure 1.6 shows the waveforms of PV array parameters such as currents

(IPV1 & IPV2), voltages (VPV1 & VPV2), and power (PPV1 & PPV2). The active
and reactive power of grid Pg(W), reactive power Qg(VAR), and the load
active and reactive power, i.e., PL(W) and QL(W), are represented in Figure
1.7. It is observed the grid current is out of phase with grid voltage; it shows
the PV-DSTATCOM supplying active power to the grid. Therefore, the grid
power is negative, as shown in Figure 1.6. The THD obtained in the grid
current is 3.70% in Figure 1.5, and it is within the stipulated IEEE-519 limits,
i.e., less than 5%, and THD obtained in load current is 24.51%.
1.4.2 DYNAMIC STATE PERFORMANCE OF PV-DSTATCOM IN MODE-1
During dynamic conditions, the proposed system is tested with sudden load
variation during t = 0.8 s to t = 1.2s and variation in solar irradiance also.
Solar irradiancies varied from a minimum value of 200 W/m2 to a maximum
of 700 W/m2. In mode-I, i.e., during the daytime, real-time irradiation has
been taken into consideration for testing. The real-time data of solar irra-
diation taken [18] from morning 06:00 am to evening 06:00 pm is shown
in Figure 1.8. It is clearly observed that the obtained data have dynamic
Non Commercial Use

and nonlinear characteristics. Therefore, the average reading of irradiation

obtained is taken from the morning to evening and is divided into four equal
intervals of three-hour duration each in the simulation. Table 1.2 shows the
average irradiance reading. The real-time solar irradiance data is shown in
Figure 1.8, and the average variation taken is shown in Figure 1.9.
Author Copy
FIGURE 1.5 Harmonic spectrum of grid current and load current under steady state.
FIGURE 1.6 Steady state waveforms of currents (IPV1 & IPV2), voltages (VPV1 & VPV2) and
power (PPV1 & PPV2).
Figure 1.10 shows the waveforms of grid voltage Vg(V), grid current
Ig(A), load current IL(A), and total DC link voltage VDC(V) and the PV array
parameters such as currents (IPV1 & IPV2), voltages (VPV1 & VPV2) and power
Non Commercial Use

(PPV1 & PPV2) are presented in Figure 1.11. Furthermore, the active and reac-
tive power of grid Pg(W), reactive power Qg(var), and the load active and
reactive power, i.e., PL(W) and QL(W), are represented in Figure 1.12.
Author Copy
FIGURE 1.7 Steady-state waveforms of active power Pg(W), reactive power Qg(VAR) of
the grid, and the load active and reactive power, i.e., PL(W) and QL(W).
TABLE 1.2 Average Value of Real-Time Data of Solar Irradiation

SL. No. Duration of Interval Average Reading Obtained Used in Simulation
1. 06:00 am to 09:00 am 196.15 W/m 2
200 W/m2
2. 09:00 am to 12:00 noon 584.77 W/m 2
600 W/m2
3. 12:00 noon to 03:00 pm 711.55 W/m2 700 W/m2
4. 03:00 pm to 06:00 pm 274.22 W/m 2
300 W/m2
The irradiance variation used in the simulation is shown in Figure 1.8.

The performance is satisfactory during dynamic load and solar irradiance
conditions. The total DC link voltage VDC rapidly achieves to reference value
of 200 V due to the fast action of the PI controller and the proposed ISOGI-Q
algorithm able to mitigate the harmonics generated by the load current and
respond quickly under these sudden dynamic variations in load and solar
irradiance. The grid current is almost sinusoidal but out of phase with the
grid voltage. During daytime operation, the proposed system is able to feed
the real power to the grid, and another advantage is the enhanced utiliza-
tion of two PV arrays due to cascaded connections of inverters. However, in
the literature, most of the authors have implemented single PV array-based
2-Level inverters.
Non Commercial Use

Author Copy
FIGURE 1.8 Real-time solar radiation obtained in Delhi /NCR region.
FIGURE 1.9 Average irradiance is taken for the simulation study.
FIGURE 1.10 Dynamic results of grid voltage Vg(V), grid current Ig(A), load current IL(A),
and total DC link voltage VDC(V) under solar irradiance at t = 0.5 s, 1 s, and 1.5 s and load
variation during t = 2.2 s to 2.4 s.
Non Commercial Use

Author Copy
FIGURE 1.11 Dynamic results of currents (IPV1 & IPV2), voltages (VPV1 & VPV2), and power
(PPV1 & PPV2) under solar irradiance variation at t = 0.5 s, 1 s, and 1.5 s and load variation
during t = 2.2 s to 2.4 s.
FIGURE 1.12 Dynamic results of active power Pg(W), reactive power Qg(VAR) of the grid,
and the load active and reactive power, i.e., PL(W) and QL(W) under solar irradiance variation
at t = 0.5 s, 1 s, and 1.5 s and load variation during t = 2.2 s to 2.4 s.
Non Commercial Use

1.4.3 STEADY STATE PERFORMANCE OF PV-DSTATCOM IN MODE-2
In Mode-2, i.e., during nighttime, the PV array is disconnected, and the proposed
system is acting as a DSTATCOM. Figure 1.13 shows the steady state wave-
forms of grid voltage Vg(V), grid current Ig(A), load current IL(A), and total DC
link voltage VDC(V). The active and reactive power of grid Pg(w), reactive power
Qg(VAR), and the load active and reactive power, i.e., PL(W) and QL(W), are
presented in Figure 1.14. In steady-state waveforms, the grid current and grid
voltage is in a phase, which means the grid is supplying reactive power to the load.
Author Copy
FIGURE 1.13 Steady-state results of grid voltage Vg(V), grid current Ig(A), load current
IL(A), and total DC link voltage VDC(V).
1.4.4 DYNAMIC STATE PERFORMANCE OF PV-DSTATCOM IN MODE-2
The dynamic waveforms of the proposed system under load variations are
shown in Figure 1.15. The grid is not feeding any reactive power to the load,
i.e., the entire reactive power is supplied from DSTATCOM, making the
source current in phase with grid voltage and unity power factor operation.
The effective compensation has been achieved under varying load conditions.
It is observed from Figures 1.15 and 1.16 that during sudden load
variation at t=0.5, there is a small variation in load reactive power (QL ↓).
However, the load active power requirement is increased from 1,500 W to
2,600 W. The active power of the grid (Pg↑), but there is a slight variation in
the reactive power requirement of the grid that means almost the unity power
factor is maintained on the grid side.
Non Commercial Use

Author Copy
FIGURE 1.14 Steady-state waveforms of active power Pg(W), reactive power Qg(VAR) of
the grid, and the load active and reactive power, i.e., PL(W) and QL(W).
FIGURE 1.15 Dynamic state results of grid voltage Vg(V), grid current Ig(A), load current
IL(A), and total DC link voltage VDC(V) during t = 0.5 to 1.1 s.
The performance of the proposed system is compared with conventional

SOGI and SRFT algorithms in terms of fundamental weight extraction and error.
Figures 1.17 and 1.18 shows the fundamental estimated component of weight
and error, i.e., the difference between the actual load current and fundamental
estimated load current, respectively. The load is varied at t = 0.3 s and t = 0.5
Non Commercial Use

s, and it was observed that the SRFT algorithm takes more time to converge as
compared to SOGI. ISOGI-Q and the conventional SOGI converge in almost
the same duration, but sustained oscillations are observed in SOGI.
Author Copy
FIGURE 1.16 Dynamic results of active power Pg(W), reactive power Qg(VAR) of the grid,
and the load active and reactive power, i.e., PL(W) and QL(W) under sudden load variation
during t = 0.5 s to 1.1 s.
FIGURE 1.17 Weight estimation under dynamic load conditions.
FIGURE 1.18 Error estimation.
Non Commercial Use

1.5 CONCLUSION
PV-DSTATCOM-based single-stage grid is implemented in the proposed

system, and the system is tested to operate in two separate modes, viz.
daytime operation, and nighttime operation. The ISOGI-Q algorithm has

been modeled in MATLAB/Simulink environment and tested extensively
under a wide variation of solar irradiation and load disturbances. During
MODE-I operation, i.e., during the day, the PV-DSTATCOM feeds real
power to the grid and reactive power to the load, and during MODE-II
operation, i.e., at night, the PV-DSTATCOM feeds reactive power to the
load and improves power quality. The effective control makes the system
Author Copy
free of harmonics on the grid side, allowing the system to achieve a power
factor of approximately unity. In addition, the ISOGI-Q technique is used to
extract the fundamental current from the nonlinear load current. Because of
the CHB-MLI, the designed system operates at a low switching frequency
and results in a low THD in the grid, currently meeting IEEE-519 norms.
The performance of the proposed system under variable solar irradiation and
load variations is satisfactory. Additionally, the performance of the proposed
algorithm is better as compared to the conventional SOGI and SRFT algo-
rithms in terms of fundamental weight estimation and reduced error under
dynamic load variations.
KEYWORDS
• capacitors
• control algorithm
• DC link voltage reference
• multilevel inverter
• nonlinear load
• PV array
• PV-DSTATCOM
• steady-state performance
• unit vector template
Non Commercial Use

REFERENCES
1. Koraki, D., & Strunz, K., (2018). Wind and solar power integration in electricity
markets and distribution networks through service-centric virtual power plants. In: IEEE
Transactions on Power Systems (Vol. 33, No. 1, pp. 473–485).

2. Yan, G., Cai, Y., Jia, Q., & Liang, S., (2019). Stability analysis of grid-connected PV
generation with an adapted reactive power control strategy. In: The Journal of Engineering
(pp. 2980–2985).
3. Shugar, D. S., & Ramon, S., (1990). Photovoltaic in the utility distribution system: The
evaluation of system and distributed benefits. In: Proc. IEEE Photovoltaic Specialist
Conference (Vol. 2, pp. 836–843).
4. Singh, B., Al-Haddad, K., & Chandra, A., (1999). A review of active filters for power
Author Copy
quality improvement. IEEE Trans. Ind. Electron., 46(5), 960–971.
5. Ogunjuyigbe, A. S. O., Ayodele, T. R., Idika, V. E., & Ojo, O., (2017). Effect of lamp
technologies on the power quality of electrical distribution network. In: Proc. IEEE
Power Energy System (pp. 159–163).
6. Wu, T. F., Chang, C. H., Lin, L. C., & Kuo, C. L., (2011). Power loss comparison of
single–and two-stage grid-connected photovoltaic systems. IEEE Transactions Energy
Conversion, 26(2), 707–715.
7. Leon, J. I., Vazquez, S., & Franquelo, L. G. (2017). Multilevel Converters: Control and
Modulation Techniques for Their Operation and Industrial Applications. Proceedings of
the IEEE, 105(11), 2066–2081.
8. Chitra, S., & Valluvan, K. R., (2020). Design and implementation of cascaded H-bridge
multilevel inverter using FPGA with multiple carrier phase disposition modulation
scheme. Microprocessors and Microsystems, 76, 103–108.
9. Golestan, S., Monfared, M., Freijedo, F. D., & Guerrero, J. M., (2013). Dynamics
assessment of advanced single-phase PLL structures. IEEE Transactions on Industrial
Electronics, 60(6), 2167–2177.
10. Panda, A., Pathak, M. K., & Srivastava, S. P., (2016). Enhanced power quality based
single phase photovoltaic distributed generation system. International Journal of
Electronics, 103(8), 1262–1278.
11. Tripathi, R. N., Singh A., (2013). SRF theory-based interconnected solar photovoltaic
(SPV) system with improved power quality. In: Proc. IEEE Emerging Trends in
Communication, Control, Signal Processing & Computing Application (pp. 1–6).
12. Rodriguez, P., Luna, A., Candela, I., Mujal, R., Teodorescu, R., & Blaabjerg, F., (2016).
Multiresonant frequency-locked loop for grid synchronization of power converters
under distorted grid conditions. IEEE Transactions on Industrial Electronics, 58(1),
127–138.
13. Hemant, S., Alka, S., & Rai, J. N., (2016). Design and analysis of different PLLs as
load compensation techniques in 1-∅ grid-tied PV system. International Journal of
Electronics, 106(11), 1632–1659,
14. Soumyadeep, R., Nitin, G., & Ram, A. G., (2016). A comprehensive review on cascaded
H-bridge inverter-based large-scale grid-connected photovoltaic. IETE Technical
Review, 34(5), 463–477.
Non Commercial Use

15. Ray, S., Gupta, N., & Gupta, R. A., (2017). Improved single-phase SRF algorithm for CHB
inverter-based shunt active power filter under non-ideal supply conditions. IEEE PES
Asia-Pacific Power and Energy Engineering Conference, 2017 APPEEC ‘09. Bangalore.
16. Saxena V., Kumar N., Singh, B., & Panigrahi, B. K., (2020). An enhanced multilayer
GI-based control for grid-integrated solar PV system. IEEE International Conference on

Power Electronics, Drives and Energy Systems (PEDES), 1–6.
17. Bansal, P., & Singh, A., (2021). Nonlinear adaptive normalized Huber control algorithm
for 5-level distribution static compensator. Electrical Engineering.
18. Saurabh, K. R., & Dharmendra, K. D., (2021). Mathematical modeling and experimental
validation for the impact of high solar cell temperature on transformer loading and life.
Renewable Energy Focus, 39, 27–36. ISSN 1755-0084.
Author Copy
Non Commercial Use

CHAPTER 2
Thermal Efficiency Investigation of a

Solar Parabolic Trough Collector System
Using Two Dissimilar Reflectors
S. N. VIJAYAN,1 S. SENDHIL KUMAR,2 and S. KARTHIK3
Author Copy
1
Assistant Professor, Department of Mechanical Engineering, Karpagam
Institute of Technology, Coimbatore, Tamil Nadu, India
2
Professor, Department of Aeronautical Engineering, Annasaheb Dange
College of Engineering and Technology, Ashta, Sangli, Maharashtra, India
3
Assistant Professor, Department of Mechanical Engineering, Sri Krishna
College of Engineering and Technology, Coimbatore, Tamil Nadu, India
ABSTRACT
Solar radiation consists of energy in the form of electromagnetic radiation

with wavelengths ranging from infrared to ultraviolet. Solar thermal collectors
gather heat energy from the sun and transfer it to the working medium to
generate energy. Solar collectors are devices that capture energy from the sun
and are utilized as alternative energy sources (ES) in industrial and domestic
applications. A parabolic trough is a type of solar thermal collector with a
polished metal mirror that is straight in one dimension and has a parabola
curve in the other. In this research, two types of reflective materials, such as
stainless steel 303 and aluminum A011, are employed to conduct experiments
in the western zone of Tamil Nadu with two different flow rates (1.139 kg/s
and 2.174 kg/s). The results are validated to select suitable reflector material
with highest efficiency for the application of industrial and domestic usage.
A maximum temperature of 65°C was achieved using stainless steel 303 as a
Non Commercial Use

reflector, whereas 70.1°C was achieved when aluminum A011 was employed
as a reflector, which yields 5% of higher output temperature and 28.67%
of higher thermal efficiency than stainless steel 303 reflective materials.
Finally, it is investigated that the highest thermal efficiency can be obtained
by employing aluminum A011 reflective material with a minimum flow rate

(1.139 kg/s) of working medium.
2.1 INTRODUCTION
Inappropriate usage of available fossil fuels leads to a shortage of fossil fuels,
Author Copy
which creates a negative imbalance in the environment. Hence need arises
to search for new renewable energy resources to fulfill the requirements
and save the earth from toxic gases. Solar energy is the primary renewable
source of energy on the planet. Emitted energy from the sun is utilized to
generate power by the different types of solar energy collector systems
[1–3]. Heat energy is transferred from the sun to any working medium,
which captures the heat and circulates it [4, 5]. Figure 2.1 represents the
various stages of the heat energy transfer process. High-intensity rays fall
on the reflector, and it reflects the heat energy contained in the intensity rays
to the receiver tube mounted on the focal line. Heat transfer fluid is passed
through the receiver pipe, which receives the heat energy from the solar
rays through the pipe.
FIGURE 2.1 Heat energy transfer process.
Atmospheric temperature working medium (fluid) is given as an inlet

and converted into high-temperature fluid by absorbing heat energy from
the reflector, which can be used directly for the applications or for later
usage saved in another mode. The performance of the system depends on
the irradiation level of the sun. Irradiance is the flux received by a surface
per unit space. Direct normal and global horizontal irradiance (GHI) are the
two levels of the irradiance. Direct normal irradiance (DNI) is the quantity
of radiation received per unit space by a surface that is constantly held
perpendicular (or normal) to the rays from the direction of the sun. The
average value of DNI is 5.35 kWh/m2 for a day. GHI is the total quantity
of shortwave radiation received from the source to the ground; the annual
Non Commercial Use

Thermal Efficiency Investigation 23
average GHI for a day is 5.86 kWh/m2. The irradiation is measured at the
latitude of 11.05° and longitude of 76.95°.
2.2 LITERATURE REVIEW
Stainless steel reflectors are used for the analysis of parabolic trough
collectors with GI absorber tubes [1], and the collector thermal efficiency
depends on the flow rate and temperature of the working medium. Obtained
thermal efficiency of the collector is 65% for the optimum values of input
parameters [4]. Concentrated collectors are used in solar power production
Author Copy
for producing high-temperature output with high thermal efficiency using
a minimum collecting area [6, 7]. A Parabolic Trough Collector (PTC)
consists of a solar collector (reflector) to receive the intensity rays from the
sun and reflect it to the receiver tube. The receiver comprises an absorber
tube made of stainless steel or copper with a selective coating to absorb the
solar radiation. Circulated the working medium through the absorber tubes
to collect the heat energy from solar radiation and transfer the heat to the
steam generator or to the heat storage system [8, 9].
A modified absorber has been used to enhance the efficiency of solar
collector by 42.1%, while in conventional absorber, it is only 26.7% [10].
Performed the investigation by changing the parameters (aperture area,
diameter of the receiver, and working medium) to optimize the design,
thereby increasing the efficiency. To maintain the equilibrium between the
aperture area and optical losses for minimizing the thermal losses [11, 12].
Different modes of heat transfer occur while transferring heat from the
radiation to the working medium through the pipe. During the process,
heat losses occur due to the conduction and convection modes of transfer
on the surface of the receiver pipe and the fluctuation of wind speed [13,
14]. Assessment of year-round performance considering aluminum sheet as
a reflector in a parabolic trough collector for different climatic conditions is
conducted. And found that maximum water temperature was achieved during
the month of April with the mass flow rate of 0.010 kg/s for the aperture area
of 1.34 m2 [15]. The impact of the failure of the absorber tube was analyzed
through experimental and numerical approaches for three different types of
tubes such as vacuum tube, lost vacuum (air), and broken glass (bare) tube.
Heat loss is significantly reduced in a vacuum tube when compared with a
broken tube and a lost vacuum tube [16].
PTC is used for cooking applications due to its ability to produce a high-
temperature outlet with 36 m2 of reflective area, which can produce 84,566
Non Commercial Use

Kcal/day from 6 parabolic troughs with 60% collector efficiency. This energy
is enough to prepare meals three times for 250 people in a day [17]. Thermal
performance analysis of the parabolic trough collector is analyzed for the
application of heating water by using copper and aluminum as absorber
tubes with covered glass to arrest the emitted rays. A copper tube with a glass
cover gives a maximum outlet temperature of 68.7°C, whereas an aluminum
tube with a glass cover produces a maximum temperature of 62.4°C [18].
The solid plug is inserted in the focal line of the absorber tube to reduce the
volume of the flow rate and increase the heat transfer rate from the sun by
increasing flow velocity. Evacuated space prevents heat loss by arresting the
emitted radiation from the tube [19, 20].
Author Copy
In this work, two types of reflective materials were used to inspect the
thermal performance of solar parabolic trough collectors for the applications
of industrial and domestic purposes.
2.3 CONSTRUCTION OF PTC
The main components of the solar collector are:

• Collector (receives the heat energy from the sun).
• Receiver (glass tubes which allow the working medium to circulate
inside and gain the heat from the collector and transfer it to the heat
storage device).
• Working medium or heat transfer medium (organic fluids, water, or
any other fluids which transfer heat effectively).
• Tracking devices (E-W or N-S tracking or both) to increase the efficiency
of the parabolic trough collector.
• Storage tank (used to store the heat energy for later use).
• Temperature sensor (contact type LCD digital thermometer sensor).
2.3.1 DESIGN CALCULATION
The calculation for designing the parabola curve and focal length for the
collector is decided based on the following equations:
1 2
y= x (1)
5
y = x2/4f, where f is its focal length. (2)
Non Commercial Use

x2
f= 125 mm (3)
4y
Figure 2.2 illustrates a parabola curve drawn using coordinates generated
from the Eqns. (1) and (2) [21] and Eqn. (3) is used to calculate the focal
length of the curve to fix the absorber tube for better absorption of radia-
tion. The obtained value of the focal length of the system is 125 mm for the
reflector size of 400 mm in the X direction and 320 mm in the Y direction.
Three-dimensional modeling of the parabolic trough collector is made using
solid works software; it is illustrated in Figure 2.3. The mass flow rate of the
working medium is calculated as 2.17 kg/s using Eqn. (3) and the volume of
Author Copy
the absorber tube is calculated (0.2393 m3) using Eqn. (4). The performance
of the solar parabolic trough collector is based on the heat transfer rate of
the absorber. The thermal efficiency of the parabolic trough collector was
calculated using Eqn. (5) for both reflective materials. In the above equations,
m is mass flow rate, Ao means the length of the parabolic curve (0.33 m3), Gb
is a constant value of solar beam radiation (1,360 w/m2), cp is the specific heat
of water (4.184), and Qu is useful to heat energy delivered.
FIGURE 2.2 Parabola curve.
Volume of the tube

Flow rate, M = (3)
Time flow of water inside the tube
Non Commercial Use

Volume of the absorber tube = π r2h = π×(0.25)2 × 1.219 (4)

V = 0.2393 m3
Qu
Thermal Efficiency = (5)

Qa
Qu = mcp (Tout – Tin) (6)
Qa = A0 × Gb (7)
From the above equations, the thermal efficiency of stainless steel 303
and aluminum was calculated.
Author Copy
FIGURE 2.3 Three-dimensional view of parabolic trough collector.
For stainless steel,

Qu = mcp (Tout-Tin) = 2.175×4.184 × (55.2–29.1) = 0.2375 Kw
Qa = A0 × Gb,
where; A0 = 0.33 m3
Gb = 1360 w/m2
Qa = 0.33×1360=448.8 W=0.448 Kw
Non Commercial Use

Efficiency of Stainless Steel 303 sheet

Qu 0.2375
= × 100 = × 100 = 0.529 × 100 = 52.9%
Qa 0.4488
For Aluminum 011 Sheet,

Qu = mcp(Tout – Tin) = 2.175 × 4.184 × (62.1 – 28.5)
= 305.54 w = 0.3055 Kw
Efficiency for aluminum A011 sheet
Qu 0.3055
Author Copy
= ×100 = ×100 = 0.6807 ×100 = 68.07%
Qa 0.4488
The calculated efficiency of a parabolic trough collector with stainless
steel as reflective material is 52.9%, and for the aluminum sheet is 68.07%.
The maximum efficiency is obtained when the aluminum sheet is employed as
reflective material when compared to stainless steel. The detailed specification
of components used in the parabolic trough collector is tabulated in Table 2.1.
TABLE 2.1 Specifications of PTC

SL. No. Description Unit Dimensions
1. Focal length mm 125
2. Reflector sheet length mm 1,219
3. Reflector sheet width mm 1,075
4. Frame width mm 800
5. Receiver tube diameter (Do) mm 25
6. Receiver tube diameter (Di) mm 22
7. Receiver tube length mm 1,219
8. Volume of tube m3
0.239
9. Length of parabolic curve m3 0.33
10. Solar beam radiation w/m 2
1,360
11. Specific heat of water joules 4.184
2.4 EXPERIMENTAL INVESTIGATION
Two types of reflective materials, such as stainless steel 303 and aluminum
A011, were selected for the investigation of the thermal performance of solar
parabolic trough collectors. Water is used as a heat transfer medium, and the
Non Commercial Use

copper tube is used as the absorber tube. The frame and supporting structures
are fabricated using mild steel material. Experimental analysis was carried
out in the latitude of 11.05°N and longitude of 76.95°E during the month of
March 2017 for both selected reflective materials under similar atmospheric
conditions. The operating parameters of minimum and maximum flow rates

are calculated as 1.139 kg/s and 2.174 kg/s, and the experimental setup of
the parabolic trough collector is illustrated in Figure 2.4. The reflective
capacity of stainless steel 303 sheets and aluminum A011 sheets is shown in
Figures 2.5 and 2.6. Stainless steel material possesses a minimum amount of
reflectivity. However, aluminum A011 gives a massive amount of radiation
to the absorber tube. During the test, temperature measurement error is
Author Copy
assumed to be of negligible quantity.
FIGURE 2.4 Experimental setup of parabolic trough collector.
At first, the stainless steel 303 sheet was fixed in the collector as a
reflector to perform the investigation at a minimum flow rate of the working
medium with constant time intervals (30 minutes) from 10.00 am to 3.00
pm. Initially, water is allowed to flow from the tank through the absorber
tube. During the process, inlet and outlet temperatures were noted using a
temperature sensor. The same process is continued for a maximum flow rate
of the working medium and calculated the thermal efficiency of the collector.
Instead of stainless steel 303 sheets, aluminum A011 reflective material was
Non Commercial Use

replaced, and the same procedure was followed to determine the reflective
capacity and thermal efficiency of the solar parabolic trough collector.
Author Copy
FIGURE 2.5 The focal point of stainless steel 303 sheets.
FIGURE 2.6 The focal point of aluminum A011 sheet.
Non Commercial Use

2.5 RESULTS AND DISCUSSION
Experiments were conducted for both reflector materials for 10 days with
the same operating parameters and conditions, and the output results were
tabulated in Tables 2.2 and 2.3. Output temperatures for stainless steel and
aluminum A011 reflective material for minimum and maximum flow rates
are observed from 10.00 am to 3.00 pm for the respective time interval of 30
minutes. Ambient temperature is increased till noon and decreased gradually.
It decides the temperature of the inlet and outlet.
Author Copy
2.6 STAINLESS STEEL 303 REFLECTIVE MATERIAL
Experimental results are noted from the time of 10.00 to 15.00 for minimum
and maximum flow rates of 1.139 kg/s and 2.174 kg/s with ambient climatic
conditions. The output temperature is varied with respect to ambient
temperature; the highest level of ambient temperature is reached at noon
time; therefore, maximum efficiency can be reached at the time of noon for
an optimum flow rate.
TABLE 2.2 Experimental Results for Stainless Steel 303 Reflective Materials
SL. Time Temperature (°C)
No. Minimum Flow (1.139 kg/s) Maximum Flow (2.174 kg/s)
Ambient Inlet (T1) Outlet (T2) Ambient Inlet (T1) Outlet (T2)
1. 10.00 34 28.1 60.2 34 29.4 48.4
2. 10.30 35.7 29.2 62.4 35.7 28.2 49.5
3. 11.00 36.2 28.9 64.5 36.2 28.5 50.2
4. 11.30 38.5 30.1 63.8 38.5 29.6 51.3
5. 12.00 40.3 29.4 65 40.3 28.9 55.2
6. 12.30 39.4 28.5 62.1 39.4 29.1 50.1
7. 13.00 38 29.2 63.4 38 28.5 51.5
8. 13.30 37.6 30.2 64.1 37.6 28.8 52.3
9. 14.00 36.1 27.5 62.5 36.1 30.1 49.6
10. 14.30 36.2 28.8 59.2 36.2 29.4 48.4
11. 15.00 35 29.3 55.4 35 28.2 43.5
The output temperature gradually increases up to noon time, and further,

it reduces due to the variation of ambient temperature. For the minimum
Non Commercial Use

flow rate of the working medium highest output temperature was obtained
(65°C) at noon time (12.00), while for the maximum flow rate of the working
medium highest output temperature occurred (55.2°C) at noon time (12.00).
This means that the highest heat transfer rate takes place for a minimum flow
rate of the working medium.
2.7 ALUMINUM A011 REFLECTIVE MATERIAL
Reflective material is replaced with Aluminum A011 to conduct an investi-

gation with similar ambient climatic conditions. Experiments were carried
Author Copy
out by maintaining a constant minimum and maximum flow rate of 1.139
kg/s and 2.174 kg/s. Output temperature is gradually increased up to noon
time and further decreased due to the variation of ambient temperature. The
highest output temperature obtained for the minimum and maximum flow
rates is 70.1°C and 62.1°C at noon time (12.00). The lowest heat transfer rate
takes place during low ambient temperatures, and the highest heat transfer
rate takes place during the highest ambient temperature.
TABLE 2.3 Experimental Results for Aluminum A011 Reflective Material

SL. Time Temperature (°C)
No. Minimum Flow (1.139 kg/s) Maximum Flow (2.174 kg/s)
Ambient Inlet (T1) Outlet (T2) Ambient Inlet (T1) Outlet (T2)
1. 10.00 34 28.8 65.1 34 27.3 50.1
2. 10.30 35.7 28.4 66.5 35.7 27.5 52.4
3. 11.00 36.2 29.1 67.5 36.2 28.1 52.5
4. 11.30 38.5 29.5 68.3 38.5 28.5 55.8
5. 12.00 40.3 28.9 70.1 40.3 29.3 62.1
6. 12.30 39.4 28.8 68.8 39.4 28.2 56.1
7. 13.00 38 29.1 68.1 38 27.9 53.1
8. 13.30 37.6 28.8 67.9 37.6 28.3 54.5
9. 14.00 36.1 29.2 65.4 36.1 28.8 51.9
10. 14.30 36.2 29.3 65.2 36.2 29.5 49.8
11. 15.00 35 28.4 61.4 35 28.9 49.5
The graphical representation of output temperature variation is illustrated

in Figures 2.7 and 2.8 for minimum and maximum flow rate of working
medium for stainless steel 303 and aluminum A011 reflective material. The
Non Commercial Use

output temperature is gradually increased from 10.00 am and decreased after

noon time (12.00) due to lack of intensity in radiation. At noon time (12.00),
it can reach the highest temperature because of higher radiation emission.
The highest temperature obtained for the minimum and maximum flow rate
is taken into consideration for the discussion of parabolic trough collector

performance.
Author Copy
FIGURE 2.7 Temperature variation for a minimum flow rate of the working medium.
The highest temperatures of 55.2°C and 62.1°C were obtained at noon

time (12.00) when stainless steel 303 and aluminum A011 were employed
as reflective materials for a maximum flow rate of the working medium.
Similarly, the highest temperatures of 65°C and 70.1°C were observed at
noon time (12.00) for a minimum flow rate of the working medium. The
thermal efficiency of the parabolic trough collector system is calculated using
the analytical methods for stainless steel 303 and aluminum A011 reflective
material as 52.9% and 68.07%, respectively. The highest thermal efficiency
Non Commercial Use

was obtained for aluminum A011 reflective material in the climatic condition
of the western area of Tamil Nadu, India. The highest output temperature and
thermal efficiency of parabolic trough collector aluminum A011 reflective
material indicate that the reflective material has the highest reflectivity and
transfers more heat energy to the absorber tube when compared to stainless
steel 303 reflective materials. The highest thermal efficiency of a parabolic
trough collector can be achieved while giving an optimum flow rate of the
working medium as an inlet with aluminum A011 reflective material. Higher
output temperature is produced by the stainless-steel reflector when compared
with the previous work performed by Syed Mohd. Yasir Arsalan et al. [22]
and using aluminum reflector material by Macedo-Valencia et al. [23].
Author Copy
FIGURE 2.8 Temperature variation for a maximum flow rate of the working medium.
2.8 CONCLUSION
The parabolic trough collector was investigated in the western zone of

Tamil Nadu. Thermal efficiency and maximum reflectivity of two dissimilar
reflective materials (stainless steel 303 and aluminum A011) were analyzed.
Experiments were conducted for minimum and maximum flow rates of
Non Commercial Use

heat transfer medium with constant time intervals of 30 minutes from 10.00
to 15.00. When stainless steel 303 is employed as reflective material, the
highest output temperature is obtained as 65°C at 12.00 with minimum flow
rate, and it was 70.1°C for aluminum A011. To gain the maximum utilization
of reflective heat energy, the parabolic trough collector should be run with
an optimal flow rate of the heat transfer medium. From the result, it is clearly
evident that maximum output temperature and highest thermal efficiency can
be obtained while using aluminum A011 as reflective material. Aluminum
A011 reflective material gives 5% of higher output temperature and 28.67%
of higher thermal efficiency than stainless steel 303 reflective materials.
Due to the highest reflectivity of aluminum A011 reflective material, it is
Author Copy
best suited for parabolic trough collector, which is used for the purpose of
industrial and domestic applications.
KEYWORDS
• domestic applications
• heat transfer
• parabolic trough collector
• reflector
• solar energy
• stainless steel reflector
• thermal efficiency
• thermal performance
REFERENCES
1. Syed, M. et al., (2016). Performance evaluation of solar parabolic trough collector with
stainless steel sheet as a reflector. International Journal of Engineering Research &
Science, 2(6), 22–28.
2. Vijayan, S. N., & Sendhil, K. S., (2020). Numerical and analytical investigation of heat
transfer enhancement in flat plate solar collector using internal fins in absorber tube and
dissimilar working medium. Journal of Xidian University, 14(5), 35–45.
3. Vijayan, S. N., & Sendhil, K. S., (2017). Theoretical review on influencing factors in
the design of parabolic trough collector. World Academy of Science, Engineering and
Technology International Journal of Mechanical and Materials Engineering 11(11).
Non Commercial Use

4. Roy, B. et al., (2016). Parametric study of parabolic trough collector – a case study for the
climatic conditions of Silchar, Assam, India. ISESCO Journal of Science and Technology
12(21), 24–29.
5. Vijayan, S. N., et al., (2017). Performance analysis of non-concentrating solar collector:
A review. International Journal of Advanced Chemical Science and Applications 5(2),

39–43.
6. Christos, T., & Evangelos, B., (2016). The use of parabolic trough collectors for solar
cooling: A case study for Athens climate. Case Studies in Thermal Engineering, 8,
403–413.
7. Avadhesh, Y., et al., (2013). Experimental study and analysis of parabolic trough
collector with various reflectors. International Journal of Mathematical, Computational,
Physical, Electrical and Computer Engineering, 7(12), 1659–1663.
8. Amit, K. P., et al., (2014). Design, fabrication, and performance evaluation of a low-cost
Author Copy
parabolic trough collector with copper receiver. International Journal of Mechanical
Engineering and Technology, 5(5), 122–129.
9. Akdeniz, C., (2016). Design of a parabolic trough solar collector using a concentrator
with high reflectivity. Proceedings of the 2nd World Congress on Mechanical, Chemical
and Material Engineering (pp. 1–5). Budapest, Hungary.
10. Kajavali, A., et al., (2014). Investigation of heat transfer enhancement in a parabolic
trough collector with a modified absorber. International Energy Journal, 14, 177–188.
11. Fauziah, S., et al., (2012). A simulated design and analysis of a solar thermal parabolic
trough concentrator. International Journal of Environmental, Chemical, Ecological,
Geological, and Geophysical Engineering, 6(12), 739–743.
12. Sendhilkumar, S., & Vijyan, S. N., (2016). Solar tracking system using a refrigerant as
a working medium for solar energy conversion. International Journal of Mechanical
Aerospace Industrial Mechatronic and Manufacturing Engineering, 10(8), 1548–1553.
13. Donald, J. G., et al., (2012). Design, Construction, and Test of a Miniature Parabolic
Trough Solar Concentrator (pp. 1–73). Project Report Fall Quarter.
14. Vijayan, S. N., et al., (2013). Performance analysis of solar-assisted drying system.
Elixir International Journal, 63, 18107–18109.
15. Devander, K., & Sudhir, K., (2015). Year-round performance assessment of a solar
parabolic trough collector under the climatic condition of Bhiwani, India: A case study.
Energy Conversion and Management, 106, 224–234.
16. Yaghoubi, M., et al., (2013). Analysis of heat losses of absorber tubes of parabolic through
the collector of Shiraz (Iran) solar power plant. Journal of Clean Energy Technologies,
1(1), 33–37.
17. Shubham, G., & Mishra, R. S., (2016). Design and performance analysis of solar
parabolic trough for cooking application. International Journal of Research and Scientific
Innovation, 3(5), 136–139.
18. Mayank, V., et al., (2014). Thermal performance analysis of water heating system for
a parabolic solar concentrator: An experimental model-based design. International
Journal of Current Engineering and Technology, 4(5), 3649–3654.
19. Kaloudis, E., et al., (2016). Numerical simulations of a parabolic trough solar collector
with nanofluid using a two-phase model. Renewable Energy, 97, 218–229.
20. Yu, Q., et al., (2017). Thermal performance analysis of a parabolic trough solar collector
using supercritical CO2 as heat transfer fluid under non-uniform solar flux. Applied
Thermal Engineering, 115(25), 1255–1265.
Non Commercial Use

21. Valan, A. A., & Sornakumar, T., (2007). Design, manufacture, and testing of fiberglass
reinforced parabola trough for parabolic trough solar collectors. Solar Energy, 81(10),
1273–1279.
22. Syed, M. et al., (2016). To analyze the performance of solar parabolic trough concentrators
with two different reflector materials. International Journal of Scientific Engineering and
Research, 4(5), 65–69.
23. Macedo-Valencia, J., et al., (2014). Design, construction, and evaluation of parabolic
trough collector as a demonstrative prototype. ScienceDirect, Energy Procedia, 57,
989–998.
Author Copy
Non Commercial Use

CHAPTER 3
Smart Buildings Using IoT and Green

Engineering
AMIT KUMAR SINGH,1 ANSHUL GAUR,2 and PRADEEP KUMAR3
1
Biomedical Engineering Department, VSB Engineering College, Karur,
Author Copy
Tamil Nadu, India
Electronics and Communication Department, Uttarakhand Technical
2
University, Uttarakhand, India

3
Electrical Engineering Department, Ambalika Institute of Technology
and Management, Lucknow, Uttar Pradesh, India
ABSTRACT
Residential and commercial buildings are rapidly evolving, and Internet of

Things (IoT) technologies are helping to shape their future. Researchers
have recently exploited the Internet of Things (IoT) in various applications
and situations to change traditional buildings into smart, efficient, and
secure structures. While feasible IoT techniques have been created, there
is still a need to fully develop IoT applications and operations to realize
the technology’s promise. This can be accomplished by filling gaps in
current research and laying a foundation for future studies. This chapter
thoroughly assesses current technologies and IoT applications in residential
and commercial buildings. First, we systematically investigated the BEMS
using an IoT approach and how different segments in smart buildings are
connected. The role of sensors and actuators in smart building systems
was then discussed. Then, we demonstrated the role of IoT in green energy
development for smart buildings and how to use it in smart building grid
management. Solar and biomass energy have generated green energy for
Non Commercial Use

smart buildings. Finally, the trends, existing advantages, hazards, and future
obstacles of IoT deployment inbuilt settings are recognized and analyzed
based on an analysis of each category. The most significant barriers to
deployment have been identified as integrating multiple IoT technologies
with varying capabilities, data storage and processing, and privacy and
security concerns. We have also compared different wireless protocols used
in the smart buildings and showed which protocol could be used for the
smart building IoT infrastructure. This chapter assists IoT developers and
researchers in defining their work boundaries and contributions by proposing
future research directions.
Author Copy
3.1 INTRODUCTION
The growing human population is increasing the energy demand of the

world. Uncontrolled energy consumption will undoubtedly exacerbate
environmental damage, worsening global warming. International Energy
Agency report data suggests that the world’s energy demand will double
its current level by 2035. Buildings consume the most energy, accounting
for a rough 33% of global energy usage [1]. According to Shah et al.,
researchers were more concerned with predicting and optimizing electricity
use in residential buildings to develop IoT-based smart building systems.
Traditional energy managing methods used statistical analysis and machine
learning techniques to gather data from electricity meters and were utilized
for monitoring purposes [2]. Two approaches can be used to deal with
the ongoing increase in demand. The first method is to increase energy
production. The other way to reduce the need for power is to increase energy
consumption efficiency, particularly in the building sector. As a result,
renewable energy increases production capacity while decreasing demand
by managing energy consumption. Energy management is the best practice
for maximizing the energy efficacy parameter of smart buildings. Different
authors’ literature and research work show that the IoT can play a vital role
in increasing the efficiency and monitoring of the smart building parameters
to integrate them with the smart grid.
Because of technological advancements, the IoT has a significant role
in the energy management of buildings, enabling them to become smart.
Security, network architecture, sensors, health services, safety, and overall
smart buildings are the features that make a building smart. The Internet of
Things has a flexible architecture and necessary components to accomplish
this. Sensors and actuators were used to monitor and regulate many smart
Non Commercial Use

Smart Buildings Using IoT and Green Engineering 39
building’s energy consumption parameters. With the advancements in

internet technology, smartphones can easily interface with sensor networks
and actuators to control and monitor them and transfer the real-time data to
the user using IoT. As a result, the Internet of Things (IoT) can significantly
improve the energy efficiency of smart buildings. IoT was already being
used to increase the production capacity of the solar-powered plant, which
improves the efficiency of the existing infrastructure by monitoring and
controlling the different parameters of the solar plant [3]. By properly
monitoring and managing the biogas plant, the study reveals that biomass
has a remarkable capacity to boost the electricity generation of rural/urban
areas utilizing IoT [4].
Author Copy
Different researchers have used different algorithms to improve energy
efficacy in building energy management system (BEMS), which uses the
comfort-index and energy-consumption optimization by the fuzzy logic
and considers the thermal, air quality comfort, and visual with the user’s
satisfaction at the top of the priority list. They also avoid overshooting and
energy wastage while monitoring overall energy consumption to achieve
the comfort index [2]. The IoT-based structure uses different methods to
communicate with the various components of the smart buildings. Verma et
al. demonstrated various gateways for transmitting data between different
parts of the sensor and actuators networks, as well as other smart building
components. Different protocols used for data transmission were Long-range
(LoRa), Narrow Band-Internet of Things (NB-IoT), Long Term Evolution
for Machines (LTE-M), Sigfox, Zig-Bee, wireless fidelity (Wi-Fi), Blue-
tooth, Ultra-wideband (UWB), and Wireless universal serial bus (USB), and
comparing those technologies based on different parameters like the speed
of the network, range, network topology, security, power consumption,
complexity, cost and operating frequency of the network used [5, 53].
Figure 3.1 depicts the major components of the IoT-based smart building
system. The smart building (residential/commercial) is at the heart of the
system, with an IoT system and BEMS support. It gathers all the building
information and sends it to a local server/web application via the internet
gateway. The local server saves all of the needed data generated by the smart
building and makes it available in real time to the building’s end users. Solar
and biomass green energy sources (ES) could be used to meet the smart
building’s energy requirements. When no other energy source is available,
battery-based energy storage provides a vital point for power backup. There
is also a waste disposal system, which produces biomass energy. Smart
buildings use various sensors to monitor all environmental characteristics,
Non Commercial Use

such as temperature sensors for the building’s ambient temperature and

many different sensors for monitoring different parameters. The sensors
and actuator networks established within the building premise control
and monitored the electrical appliances installed at the smart facility. The
installed security system improves the living standards of the persons living
on the building premises. The space occupation system provides real-time
data of human presence in a specific building area and, using various sensor
and actuator networks, can turn on/off specific electrical appliances. This
technology eliminates energy waste and boosts the energy management
efficiency of smart buildings. Several studies suggest that net-zero energy
buildings (nZEBs) can reduce building energy usage and make them self-
Author Copy
sufficient by using efficient and sustainable methods.
FIGURE 3.1 The block diagram of the smart building system (IoT-based).
3.2 ENERGY MANAGEMENT FOR SMART BUILDINGS
According to Nguyen et al., buildings consume a large amount of energy.

The survey also found that buildings consume almost 40% of the total
energy required on the planet, of which 60% of that energy was used solely
for heating and cooling the atmosphere of the buildings [1]. Therefore, for
the sustainable development of the world and controlling air pollution, there
Non Commercial Use

was the requirement of the proper monitoring and handling of the buildings’
electrical equipment and meet the growing demand for the buildings. Ida et
al. have demonstrated a two-way solution to the energy and environmental
problems. To deal with the shortfall of energy, the two options are: (i) more
energy production; and (ii) minimizing available energy resource consump-

tion and reducing wastage. Energy production is costly for human society
that takes more time and money, but it is possible to reduce energy usage by
implementing straightforward preventative actions. The ISO 50001 energy
management standard was developed by the industry sector to provide a
framework for businesses to integrate energy efficiency into their operations
and put it into practice in the commercial building sector [6].
Author Copy
So, Nguyen et al. have shown that a BEMS is an efficient technology
to monitor and control building energy requirements. To solve the present
problems of the current BEMS, such as the large energy data of BEMS,
energy data loss, and overload problems, a BEMS based on the Internet of
Energy (IoE) is currently being used. The reference energy model can assess
the building’s energy efficiency and establish critical setpoints. With the
methodological framework given, the BEMS can improve smart buildings’
energy efficiency and savings when considering energy consumption and
cost-benefit analysis [1]. According to a study conducted by Doukas et al.,
numerous metrics were used to determine the user’s comfort level by ensuring
the desired living quality standards in all of the building’s rooms and the need
for energy savings [7].
Furthermore, the demands for essential thermal comfort, indoor air quality,
and visual comfort are increasing, particularly in the present environment of
price changes, the rapid growth of population, and technological advancement.
The BEMS were typically used to control active systems, such as heating,
ventilation, and air-conditioning (HVAC) systems, determining their operating
times in addition. The major recent advances in BEMS have followed the
developments achieved in computer technology, telecommunications, and
information technology. Major BEMS technologies were implemented on
residential and commercial buildings, transforming them into “intelligent
buildings” or smart buildings [7].
Figure 3.2 shows the block diagram of the BEMS. The block diagram
depicts the flow and the control of the energy common in residential and
commercial buildings. It shows the power source of the building that can be
coming from the external source or the green energy generated within the
building premises itself. The load gets its power through the smart meter,
which stores the energy profile of the building by monitoring the peak loads
Non Commercial Use

and other factors of the energy demand. The building monitoring and control
system keeps the data and uses an IoT framework for data transmission
among the different components of the smart buildings.
Author Copy
FIGURE 3.2 Block diagram of the BEMS.
The BEMS analyzes the gathered information per the user require-
ment, which has entered its energy demand as per user comfort. At the
appropriate space and time, the user can enter input factors such as their
room’s temperature, lighting on and off timings, intensity control, and air
quality. So, the feedback provided by the consumer and the energy demand
manage the actuators controlling different electrical appliances to maintain
the environment of the smart building and by on/off the air conditioner,
heater, lights, and other electrical types of equipment as per the demand.
The BEMS uses advanced artificial intelligence (AI) with different fuzzy
logic techniques to optimize the energy requirement and control the actua-
tors accordingly. Various sensors were deployed at different locations and
for different purposes of the smart buildings, providing real-time data for
the different ambient parameters of the indoor and outdoor environmental
conditions. So, the complete system also uses the local server or cloud-based
servers to store the enormous data generated by the facilities and helps
further optimize energy efficiency and real-time monitoring of the smart
buildings for its better utilization of the resources in terms of money, space,
and time. The system makes the building smart, and IoT gets converted into
IoE. The central building monitoring system identifies the deviations of the
peak load demand. It takes corrective action to address the demand response
Non Commercial Use

and provides the consumer’s security by monitoring the building in real-

time with the help of an intelligent security system installed at the building
premises.
3.2.1 ROLE OF SENSORS AND ACTUATORS USED IN SMART

BUILDING MANAGEMENT
The sensors and actuators are one of the prime blocks of a smart building. The
IoT-enabled sensors increase the efficiency of building management manifold.
Buildings are becoming smarter as sensors and actuators, information and
Author Copy
communication, and Internet of Things (IoT) technologies progress. The
building management and information system (BMIS), together with the
IoT, is called a “Building Internet of Things” [5]. Without accurate data, IoT
would be useless. As a result, sensors and actuators are critical components of
a smart building. According to different parameters inside a smart building,
there are different sensors. Some essential types of sensors are environmental,
optical, touch, level, leak, magnetic, electricity, acoustic, object presence,
motion, gyroscope, and chemical.
Figure 3.3 shows different sensors and actuators used in smart buildings
with IoT infrastructure support. The environmental sensors used in smart
buildings are temperature, humidity, gas, smoke, and others. In addition, the
temperature sensors provide thermal conditions of air inside the building,
working environment (thermal comfort), early fire detection, and others
[8]. The temperature sensors used in smart buildings are of the following
types: Resistance temperature detectors (RTD), thermocouples, thermistors,
thermopiles, infrared (IR) temperature sensors, semiconductor junctions, and
others. These primarily used temperature sensors are negative temperature
coefficient (NTC) thermistors and IR sensors in smart buildings. The
NTC thermistor was a low-cost, nonlinear sensor with good temperature
sensitivity. Through proper signal conditioning, they interface with IoT
hardware devices such as ESP8266. IR temperature sensors are becoming
popular due to their non-contact temperature-sensing feature. These are
useful for providing occupant body temperature conditions and electrical
wire heating status for safe operation. Temperature sensors are also used in
HVAC and occupancy detection systems.
Another critical environmental parameter inside buildings is humidity.
The vaporized water content in the air is humidity, generally expressed in
relative humidity. Therefore, humidity provides indoor environment quality
(IEQ) information in a smart building. Gulnizkij et al. have presented a
Non Commercial Use

hydrogel-based bistable sensor for relative humidity measurement inside a

building [9]. Moisture is a broad term measured by different technologies like
hair tension moisture sensors and psychrometers. The humidity sensors are used
in manufacturing plants to ensure long shelf life and optimum performance.
For example, in museums, hospitals create a healthy environment to prevent

articles from corrosion. In greenhouses, the humidity is maintained at the
optimum value for better plant growth. These were used in weather stations
to transmit humidity-related information to the public.
FIGURE 3.3 Block diagram of the sensors and actuators used in the smart building.
Author Copy
The next environmental parameters monitored inside buildings are the
smoke and gases. The smoke could be due to a fire hazard inside a building.
Most of the time, it occurs earlier as compared to flames. The broad
classification of smoke sensors is nonvisual and visual smoke sensors. The
nonvisual sensors are particle-based, and they suffer particle delay effects.
Hence their response time was poor, whereas the accuracy was good. The
visual sensors are volume sensors. They have a good response time but
suffer false outputs. Machine learning techniques, particularly deep learning
techniques, are used to improve performance by reducing false positives
of visual smoke sensors [10]. Gas sensors used in smart buildings provide
IEQ monitoring, occupancy detection, combustible or air pollution from
Non Commercial Use

gas leakage detection, and others. The significant gases measured inside a
smart building are carbon dioxide (CO2), carbon monoxide (CO), oxygen,
hydrogen, nitrogen oxide, and air pollution gases. Bad air quality inside a
building develops health problems such as tiredness, headaches, and carbon
monoxide could cause the deaths of building occupants [11].

Another category of sensors used in smart buildings is optical sensors. A
camera (visible and infrared range), proximity, photoconductive, photovoltaic,
and other sensors fall under this group. The chief application of these sensors
is to monitor the building space.
Cameras are used to surveillance the building, having the night vision
feature. Different image processing and machine learning algorithms detect
Author Copy
any suspicious activity near and inside the building space, and the fire was
also detected. Optical sensors help adjust the buildings’ surrounding light
and make them smart. In smart buildings, touch (tactile) sensors sense touch
or proximity without physical contact. The main challenge of these sensors
was the lack of moving parts. Table 3.1 shows the types of sensors and
their usage in a smart building. Therefore, they can replace the mechanical
buttons used in several electronic devices inside a smart building. According
to different regulations, touch sensors use Surface Acoustic Wave (SAW),
IR, wire resistive, and capacitive techniques to detect objects. In addition,
level sensors are used to monitor the water level in the tanks of the buildings,
including float, ultrasonic, resistive, capacitive, optical, and other sensors.
Another type of sensor used in the smart building is the leak sensor. These
sensors are of the types such as rope style, spot type, and others are used to
detect leakages in smart buildings. Magnetic sensors were used to provide
a magnetic map of corridors of buildings. These maps are helpful under
low-lighting conditions. These are mainly of the Hall Effect type. Electricity
types of sensors are used in buildings to monitor the electrical load conditions,
which help provide a cost estimate of electricity usage. These are also useful
to measure the electric current flowing through the wires or cables. This
information helped find the possibility of a fire hazard. Because when the
electrical current becomes more prominent, the wire or cable heats up and
may lead to a fire hazard. Acoustic, object presence, motion, gyroscope, and
chemical sensors were also used in buildings.
3.2.2 NETWORK PROTOCOLS USED IN SMART BUILDINGS
The core of smart buildings is smart BEMS. Different communication networks

aid in exchanging data between sensors and control devices. Commercial and
Non Commercial Use

TABLE 3.1
Sensors Used in Smart Buildings 46
SL. Sensor Type Sensors Used in Smart Buildings Usage References
No.
1. Environmental Temperature (RTD, thermistor, thermocouple, HVAC, fire detection, health parameters, occupancy [5, 10]
thermopile, IR temperature sensor, silicon detection
diode, change of state sensors, and others.
2. Environment Humidity (resistive, thermal subtypes, Moisture content in a smart building environment for [12]
capacitive), psychrometer, and Hair tension comfort and extending appliance’s life.
moisture sensor.
3. Environment Smoke, gas (CO, CO2, nitrogen oxides), Gas and smoke sensing for a cleaner environment, fire [13]
photoelectric, ionization, and aspirating. detection, gas leakage detection for safety
4. Optical Camera (IR, visible), proximity sensor, Surveillance, fire detection, occupancy detection, [14]
charge-coupled devices, photoconductive, adjustment of screen brightness of televisions, mobile
and photovoltaic. phones, personal computers, and others depending on
ambient light.
5. Touch (tactile) SAW sensor, IR sensor, wire resistive sensor, Remote control devices, as an alternative to mechanical [15, 16]
and capacitive sensor. buttons, detect touch or proximity.
6. Level Float, ultrasonic, resistive, capacitive, and Flood alarms in buildings, water level detection in [17]
optical. building tanks, and others.
7. Leak Spot leak detectors, rope style, under carpet Detection of leakages in buildings. [18]
leak detectors, and hydroscopic tape-based
Non Commercial Use

sensors.
8. Magnetic Hall effect, and others. Magnetic maps inside buildings are suitable in poor [19]
lighting conditions.
9. Electricity Current sensors and voltage sensors. Monitoring the electricity parameters inside a building. [20]
10. Acoustic Hydrophone and geophone. For picking sound in liquids, to pick up ground [21]
vibrations (seismography).
Author Copy
TABLE 3.1
(Continued)
SL. Sensor Type Sensors Used in Smart Buildings Usage References

No.
11. Object presence Doppler radar and occupancy sensors. Manufacturing machines, robotics, smart parking. [22]
12. Motion Ultrasonic, IR, PIR (passive IR), and radar Surveillance, alarm systems, automatic lighting, and [23]
(active and passive). ventilation systems.
13. Gyroscope Accelerometer (mechanical and optical) Measures rotation and angular velocity, which is beneficial [24]
for athletes for measuring accurate body movements.
14. Chemical E-Nose and others. Industrial security systems, environment protection. [25]
Smart Buildings Using IoT and Green Engineering
Non Commercial Use

47
Author Copy
residential buildings use different protocols based on their energy, distance,

and data load requirements. Wired and wireless are two open protocols. Both
protocols have their benefits and shortcomings. The wireless protocols used
for data transfer in existing buildings and the comparison of different wireless
protocols used in the smart buildings are shown in Table 3.2. The protocols
compared in Table 3.2 are described below:
1. 802.15.3 UWB: 802.15.3 is the Institute of Electrical and Electronics
Engineers (IEEE) standard for a high-data-rate wireless personal
area network. The protocol design provides proper quality of service
for the real-time distribution of video and music content. Therefore,
Author Copy
it ideally suits a home multimedia wireless network.
2. Bluetooth: It is a communication protocol used for short-range
communication. The protocol was commenced by Bluetooth’s special
interest group to exchange the data over short distances between
fixed and mobile devices. Numerous products are compatible with
Bluetooth automation. A low-energy variant of the Bluetooth network
protocol is Bluetooth low energy (BLE) is a part of the Bluetooth
v4.0 and the recent v4.2 stack. BLE is a global wireless communica-
tion protocol for transferring small data pieces. Bluetooth network
protocol has a limited range.
3. Long-Term Evolution (LTE): This wireless network protocol uses
fast wireless data communication. It is based on GSM/EDGE and
UMTS/HSPA network technologies. It is commonly called 4G LTE.
Nevertheless, LTE has been going into service in mobile phone
communication as it is capable of providing multicasting and broad-
casting service [26].
4. LoRaWAN: It refers to a Long-Range Wide Area Network that
senses signals under the noise level over a long distance and low
strength. LoRaWAN uses the internet-based connection of battery-
powered items. The use of LoRaWAN includes smart cities having
private or global networks.
5. NB-IoT: It is a low-power, wide-area network radio technology
standard developed by 3GPP for cellular devices and services. The
protocol utilizes a subset of the LTE standard and limits the band-
width to a narrow band of 200 kHz. In addition, it uses orthogonal
frequency-division multiplexing (OFDM) modulation and SC-FDMA
for downlink and uplink communication, respectively.
6. Sigfox: It was established in 2021 by a French global network
operator that initially built wireless networks to interconnect
Non Commercial Use

TABLE 3.2
Comparison between Different Wireless Technologies Used in IoT-based Smart Buildings
Parameters LoRa Bluetooth Zigbee NB-IoT 802.11 LTE-M 802.15.3 Sigfox Wireless
(Wi-Fi) UWB USB
Networking Star or mesh, Ad Hoc, Ad Hoc, peer Star Point to hub Star Point to point Star Point to
topology point to point small NWs to peer, star, or point
mesh
Security Low Low Low Low Good Moderate Low Moderate Moderate
Complexity Moderate High Low Low High Moderate Low Moderate Moderate
Range 5 km (Urban) 10 m 10 m 15 km 100 m 11 km Less than 50 km 10 m
and 15 km 10 m
(Rural)
Data rate 0.3–38.4 kbps 1 Mbps 250 kbps 158.5 kbps 6.93 Mbps 1 Mbps 100–500 100 bps (UL) 62.5 kbps
(UL) 106 Mbps 600 bps (DL)
kbps (DL)
Power usage Very low Low Very low High High Low Low Low Low
Operating 779–787 2.4 GHz 2.4 GHz Licensed 2.4 and 5 GHz Licensed 3.1–10.6 GHz Sub-GHz 2.4 GHz
Smart Buildings Using IoT and Green Engineering
frequency MHz (China), (worldwide), (700–900 (700–900 ISM: EU 868

863–870 868 MHz MHz) MHz) MHz, US 902
MHz (EU), (Europe), and MHz
902–928 (US) 900–928 MHz
(NA)
Non Commercial Use

Applications Smart Mobile Home/industrial Smart grid Thermostats, Smart meters Localization, Low power Game
buildings phones automation, communica- intelligent in buildings, identification applications, controllers,
(smart Computers, health care, tion devices, Agricultural Smart digital
lighting) digital smart metering and access Monitoring buildings cameras.
devices in renewable in smart (electric
energies buildings plugs)
49
Author Copy
low-power objects. Example of these is energy meters and smart

watches, which continuously send small amounts of data for
communication.
7. Wi-Fi: It stands for Wireless Fidelity. This wireless network protocol
refers to IEEE 802.11 standard for Wireless Local Networks. Wi-Fi

is the registered trademark of the non-profit Wi-Fi Allianz. Wi-Fi
is the cost-effective and easily accessible wireless technology in
commercial buildings that connects IoT devices. Data transferred
faster when Wi-Fi protocol was used in smaller connected devices.
Furthermore, Wi-Fi may be a great choice in smart buildings powered
by cloud-based software applications, i.e., the Chariot platform.
Author Copy
Wi-Fi application areas include thermostats, lighting, intelligent
devices, and broadband internet access in smart buildings.
8. Wireless USB: WUSB stands for Wireless USB (WUSB), a Universal
Serial Bus (USB) protocol. It is a high-bandwidth and short-range
wireless radio communication protocol. The protocol was originally
developed by the Wireless USB Promoter Group and preserved by
the WiMedia Alliance. It uses radiofrequency (RF) links rather than
cables to provide the interfaces between a computer and peripherals,
for example, external drives, monitors, printers, headsets, MP3
players and digital cameras.
9. Zigbee: It is a shorter-range wireless communication standard
specifically for commercial use. It is mostly used in home and
building automation. It uses a mesh network to create long ranges
and fast communications through radiofrequency. The main benefit
of the mesh network is that for any link breakage, devices can search
through the mesh network to find a new route. Thus, the addition
of a new device is simple and less costly. More than 1,200 products
were compatible with the Zigbee protocol. This wireless communi-
cation protocol is best suited for large buildings and campuses. The
Zigbee technology has widespread applications in home automation,
consumer electronics, healthcare, industrial automation, material
tracking, and many others. The latest version of Zigbee is v3.0 is a
single unified ZigBee standard.
3.2.3 BENEFITS AND ROLE OF IoT IN SMART BUILDINGS
IoT has a major role in transforming residential and commercial buildings

into smart residential and commercial buildings. Smart buildings have all the
Non Commercial Use

systems connected through IoT and network protocols. With the help of IoT
technology, smart buildings, regardless of their usage, started to upgrade,
innovate, be efficient in terms of energy, and provide a healthy, dynamic, and
tech-savvy smart society. The multiple profits of smart buildings originate
from the data-generating systems that empower them. For example, it is

much easier to recognize facilities when there is data to inform how people
use them. Here we have discussed some of the benefits of smart buildings:
1. Reduced Energy Consumption: By using IoT in connecting and
controlling different devices of smart buildings, various researchers
try to reduce energy consumption. This controlled system using
Author Copy
lesser energy costs benefits to commerce and the environment.
2. Automation Opportunities: The IoT initiates a robust automation
system for many applications, including links between physical and
digital management systems. For example, it helps implement floor
sensors for occupancy, motion-sensitive lights, beacons to gauge
workspace utilization, and many others.
3. Improved Performance Management: Remote monitoring
permits administrations to integrate information about many
assets and developments, regardless of time and the location of the
resources. BEMS can instantly merge information about the body’s
performance while covering key performance indicators similar to
overall equipment efficiency and total carbon (CO2) emissions.
4. Quantifiable Building Insights: The energy data points provided by
the workplace, which provide information on the energy consumption
patterns of people living and working in and around buildings, aid
in quantifying the building’s energy consumption pattern. Therefore,
each data point created by the IoT is a calculable part of the tangible
workplace which adds the data points to find trends and actionable
insights.
5. Predictive Maintenance: Technology has a more insightful over-
view of the building’s operations through sensors attached to the IoT.
The sensor data aids in the timely upkeep of the equipment installed
for the building functionality at the right time, thereby increasing the
devices’ shelf life. The collected data also helps reduce the expenses
of replacing the costly equipment used in the buildings and making
the devices smart.
6. Better Resource Utilization: IoT helps better manage the workplace
resources, space, and human resources of the smart buildings,
especially in commercial buildings. Smart buildings make and
Non Commercial Use

measure them within broader facilities. The result shows how people
use those resources and information more efficiently.
7. Reduced Operational Costs: With better insights, one can quickly
reduce the buildings’ operating costs, especially for the commercial
domains, by reducing energy wastage. There the IoT helps by control-

ling the wasteful use and thereby reducing the operational cost of
the commercial buildings specially and saving the paramount money
for the human. Moreover, the connection of sanitation devices with
sensors also reduces the costs of smart buildings.
8. New Workplace Opportunities: With the growing technology,
there is a new age of growing work styles and changes in the work-
Author Copy
place. Due to the pandemic, flexible work and agile workspaces have
become the new standard and the demand for systems to accomplish
them. The intelligent IoT networks and strategies that produce data
to support the new workplace can easily handle this oversight.
9. Increase Productivity: Smart buildings make humans more produc-
tive by continually monitoring and controlling the building use and
ensuring the better facility of the workspace or residence. In addi-
tion, smart buildings provide healthier, productive, satisfied workers
and a comfortable workspace for the people staying in with advanced
climatic and lighting controls using IoT technology. Furthermore, a
better ventilation system in the buildings also improves the indoor
environment’s air quality and reduces the sick building syndrome
symptoms.
10. Physical Security: Due to the financial gap in society, there is an
upswing in the community crime rate. So, IoT can provide better
security for the persons residing in the buildings. Moreover, the
advanced sensors and cameras with the IoT interface help provide
safety and security in any building, and the automatic systems of the
camera also reduce the security expenses of the buildings.
11. Smart Buildings Give Smart Data: Smart buildings provide real-
time electricity, water consumption, hourly space utilization, and
other data. It helps the residents to take corrective action about their
constructive utilization.
12. Smart Buildings Result in Increasing the Asset Value: Buildings
are becoming smart buildings when IoT and smart data are used,
increasing the asset value of the system. In addition, smart energy
efficiency measures ensure that the building is well maintained and
hence does not depreciate over some time.
Non Commercial Use

13. Space Allocation Management: With the help of IoT, smart building
managers can take advantage of remote occupancy monitoring
to allocate shared spaces tenants allocations management, thus
improving the efficiency of the space use and increasing the person’s
income.
14. Reduction of Manual Labor: Remote monitoring of the systems
enables remote inspections and reduces the need for hands-on work.
Furthermore, it also helps the employees to utilize the remaining
time to pursue more creative work.
Table 3.2 shows Sigfox protocol has got the highest range for data transfer
Author Copy
while the 802.15.3 UWB protocol has the shortest range. 802.15.3 UWB
has the highest data transfer rate, while the LoRa protocol was the slowest
one. NB-IoT and 802.11 (Wi-Fi) have the highest power consumption, while
the LoRa protocol has the lowest. 802.15.3 UWB protocol uses the highest
bandwidth, and LoRa uses the lowest bandwidth frequency. LoRa has the
lowest cost, while the 802.15.3 UWB protocol was the costliest. The best
networking topology used by LoRa, and ZigBee is Peer to Peer, Star, or
Mesh, which helps in the fastest and most reliable data transfer. Regarding
data security over network communication, the LoRa was the worst, while
the 802.11 has the best security. In terms of the complexity of the hardware,
Bluetooth and 802.11 are highly complex, while the NbIoT is the least
complex one. Hence, in terms of the features like security, good speed,
cost-effectiveness, range, and lesser energy consumption, LoRa, and 802.11
protocols are found to be most suitable for data communication between
different segments of BEMS.
3.3 ROLE OF IoT IN GRID MANAGEMENT
The traditional power grid connects generators, transformers, transmission

lines, distribution systems, and various loads. Moreover, the traditional grid
system is a one-way communication network. Power generation dependency
is huge on fossil fuels in this system, and there is a burden on these resources
due to an increase in electricity demand. Therefore, renewable energy
resources have attracted power utilities in the last few decades. The major
challenge in the power industry is to provide dependable and decent-quality
power to consumers. With new technological advancements, IoT applications
offer boundless solutions to power utility by giving two-way communication,
which helps in transforming traditional power grids into Smart Grid [SGs].
Non Commercial Use

A smart grid is an energy network consisting of control, automation, and

computers, enabling two-way communication between power utilities and
consumers. Figure 3.4 shows the block diagram of the layout of the smart
grid. The application of IoT in smart grids allows the smart grid to share
information between all components in the grid. As a result, the smart grid
provides many great solutions to the challenges in the power industry. In
addition, IoT can support technologies in SG. The applications of IoT in
different parts of the smart grid are as follows:
• IoT is capable of monitoring electricity generation of different kinds
of power plants (such as coal, solar, wind, biomass), energy storage,
Author Copy
gas emissions, energy consumption, and forecasting necessary power
to supply consumers.
FIGURE 3.4 A layout of the smart grid used in the smart building systems.
• IoT uses to gain electricity consumption, monitor, dispatch, and protect

transmission lines, substations, and towers, along with management
of control equipment.
• IoT can be used on the customer side in smart meters to measure various
parameters. These are intelligent power consumption, interoperability
between various networks, charging and discharging electric vehicles
(EVs), and managing energy efficiency for power demand.
Non Commercial Use

3.4 ROLE OF IoT IN GREEN ENERGY
A smart building has continuous energy demand in various forms, such

as heat energy, mechanical energy, light energy, and electrical energy. A
significant share of this requirement is electrical power, which the electricity

department fulfills. However, electricity generation mostly comes from fossil
fuels, decreasing faster. So, the answer lies in the use of non-conventional
sources based on solar energy, hydraulic energy, tidal energy, wind energy,
geothermal energy, biogas, and others.
World bank report shows that around 93% of the universal population
lives in countries with an average daily solar PV capacity between 3.0 and
Author Copy
5.0 kWh/kWp [27]. Where there are humans, there is biomass, and hence
it covers 100% population of the world for the power source. Animals and
plants also provide biomass for ES. In the ARENA 2017 report, geothermal
power plants generated nearly 80.9 TWh in 2015 or approximately 0.3% of
global electricity generation [28], which means that the percentage avail-
ability of geothermal was still not available in all the places in the world and
hence not easy to harvest the potential it has. According to the reports, wind
power’s stake in worldwide electricity usage at the end of 2018 was 4.8%
[53]. Still, it was less as it was unavailable at all the world’s places. Hydro
and tidal energy contributed a higher percentage but were still not available
at all places. Therefore, we have chosen only solar and biomass energy as the
source for the power of IoT-based smart buildings and discussed their role.
Most existing technologies and literature surveys show solar as the green
energy source for smart buildings. Some literature surveys also show solar
and wind using IoT for smart buildings.
However, as we know, only solar and biomass covers more than 90% of
the world’s share as green energy source. Hence, in this chapter, we have
discussed the role of IoT with solar and biomass as the energy source in
smart buildings. Also, we have discussed the BEMS to make the IoT-based
system for buildings smart with the help of sensors and actuators. We have
shown in Table 3.1 the use of different sensors and their functions in smart
buildings. As the IoT requires enormous amounts of data to be transferred
to different segments of the smart buildings, we have also compared the
technical features of the different wireless protocols used in the BEMS and
their pros and cons. However, we have also discussed how the IoT can make
utility electricity accessible for smart buildings through the usage of the
smart grid if and when necessary and supply more power back to the utility
if the building has excess power of its own.
Non Commercial Use

The current literature shows that the development of the latest technolo-
gies, such as IoT, improves the performance of these green ES. Batcha et
al. [29] surveyed IoT-based utilization of renewable energy. They focused
on solar power. According to them, AI-based techniques also improve
the performance of current renewable energy-based systems. Adhya et

al. discussed and proposed IoT-based remote monitoring and control of a
solar power plant [30]. This technique helps provide a solar power plant’s
maintenance, real-time monitoring, and fault detection. Patil et al. proposed
a display system for the real-time power usage of solar energy [31]. They
have used a flask framework based on Raspberry Pi. Hanumanthaiah et al.
presented an IoT-based remote monitoring system. It will disconnect the
Author Copy
load when a short circuit or illegal tapping occurs [32]. Khan et al. focused
on IoT-based real-time solar energy monitoring using voltage and current
sensors [33]. It helps find out the anomalies arising in a solar power plant.
They have used ThingspeakTM, an open-source IoT cloud platform, to monitor
the parameters. Phoolwani et al. have used the thermal camera, solar power
meter, and photovoltaic (PV) analyzer to detect the reduction in efficiency
of the solar panel due to partial shading, dust, delamination, and cracks [34].
Chieochan et al. have developed an IoT-based off-grid solar energy
system [35]. Kumar et al. discussed the role of IoT in monitoring the solar
energy system [36]. According to them, the benefits of IoT are remote
monitoring of the plant site, a reduction in human-to-human and human-to-
computer interaction, and better identification of faults. Choi et al. explored
the LoRa network in a solar-powered home to create an IoT-based renewable
energy monitoring system [37]. Figure 3.5 shows the use of green energy
usage in smart buildings. The energy provided by solar panels or biogas
generated within the smart building’s premises makes the building energy
self-sufficient, and when there is a power deficit, it can draw power from
the utility grid. So, the smart grid system installed at the building site with
the help of IoT helps fulfill all the needs of the smart building and meets the
concept of nZEB.
3.4.1 SOLAR ENERGY
The electricity derived from the sun is known as solar energy. The various
life activities on earth occur due to this massive energy source. Szabo et al.
discuss the milestones achieved by humankind for the conversion and use of
solar energy [38]. Solar energy has the most significant percentage of all ES.
The conversion of solar power into electrical power is one of the significant
Non Commercial Use

achievements of humankind. This section focuses on various aspects of this

energy and its use in the building system. The solar system helps reduce
building’s greenhouse gas generation and improve public health. The main
advantage of solar-based renewable energy is that it requires little or no water
to operate. It works better under heat wave or drought conditions. If the

energy becomes surplus, it can also be shared through grids. According to
one study, solar PV systems having storage cause a 14% reduction in smart
buildings energy demand, further reduced by using intelligent inverters by
12% [39]. This energy was latitude dependent. Countries with plenty of
average sunlight throughout the year have great potential for this energy.
Author Copy
FIGURE 3.5 Block diagram of the green energy usage in the smart building.
The cloudy days produce 10% to 20% of energy compared to energy

produced on a bright sunny day. The national renewable energy lab used
light detection and ranging techniques, geospatial mapping, and large-scale
simulations to detect America’s (US) total rooftop PV energy potential.
According to their analysis, 40% of the whole U.S. electricity generation
was possible through it. Therefore, there is a need for a smart grid to tap
this energy [40]. In addition, Teofilo et al. investigated the potential of large
rooftop solar panels placed in significant space areas such as airports. These
will help overcome the energy crisis in Australia [41].
Non Commercial Use

This form of clean energy can reduce unemployment by creating jobs. It

is possible to save money by reducing or eliminating energy bills, earning
tax credits and rebates, and improving our environment with this clean
energy source [42]. The other advantage of these renewable ES is the local
heat and electricity generation and minimization of transmission losses. The

other benefits of solar renewable energy systems are (i) the subsidies on the
installation of such systems; (ii) an increase in overall property value; (iii)
higher rental income; and (iv) income through selling the electricity through
the grid, and hence non-requirement of batteries. Some drawbacks are also
there, and these are (a) higher initial cost of installation; (b) requirement
of ample space; (c) maintenance cost; and (d) architectural challenges of
Author Copy
installations. The architecture should consider practical issues such as
rainwater sealing protection from overheating. Two types of systems can
tap solar energy. These are the solar thermal systems (STSs) and PVs.
Residential, commercial, industrial, institutional, and agricultural buildings
use PV power. The two PV system types are building-applied photovoltaic
and building-integrated photovoltaic (BIPV) [43]. Kalogirov et al. also
surveyed possible photovoltaic and solar thermal system integration solutions
on building roofs and facades [44]. The solar components on the rooftop are
often seen as foreign elements and need integration.
There are different aspects to be considered before installing STSs and
PVs. These are mechanical firmness and structural integration; protection
from rain, wind, snow, and hail; protection from fire and noise; cooling
arrangement for excessive heat; calculation of the energy demand of the
building.; availability of space; proper orientation of the solar panel to
receive the maximum amount of solar energy; cost; size and type; off-grid
connection with batteries or grid connection using inverters; count of meet-
ings [45]. In addition, grid-connected solar PVs face the challenge of grid
stability when considerable energy goes to the grid when the requirement is
low [46].
The PV panels also provide shading to the rooftop of the building. It
makes the roof relatively cooler. The additional functions of the PV panels
are integrating them into the building envelope (Building-integrated PV
(BIPV)), integrating heat collection purposes into the PV panel (building-
integrated PV/thermal (BIPV/T), integrating light transmission functions
into PV panels (Building-integrated PV/light (BIPV/L)). The PV can act as
a roof or as a façade. Then there are the various sorts of solar collection
technologies in use.
The IoT technology integration with solar systems increases efficiency
and improves overall performance. Kumar et al. focused on the integration of
Non Commercial Use

IoT technology along with solar energy in a smart building environment [36].
They performed experiments in a 12-story building. Moreover, in case of
hardware disruptions or communication-related issues, they ensured reliable
operation by IoT-enabled control techniques. Furthermore, they focused on
solar tracking parameters. Terzio showed that solar energy is preferable to

wind energy [47]. They proposed that the solar panel installation should be
flexible. Some panels should fulfill the instant needs, whereas others are used
to charge the batteries to achieve cost-effectiveness.
3.4.2 BIOGAS ENERGY
Author Copy
Biogas energy comes under the category of clean, renewable sources of
energy and improves the quality of human life. Different countries are
running programs related to biogas. For example, China and India installed
larger plants for electricity and heat applications. The Indian Government
was running programs such as “National Biogas and Manure Management
Program,” “Off-grid Biogas Power Generation Program,” “Biogas based
Distributed/Grid Power Generation Program,” “Recovery of Energy from
Industrial Wastes,” and “Recovery of Energy from Urban Wastes.” China
runs the “Biogas Institute of the Ministry of Agriculture,” part of the Chinese
Academy of Agriculture Sciences. Other countries such as Italy, the US, and
the United Kingdom run several programs. At the same time, some countries
are not utilizing the full potential of biogas. One such country is South
Africa, where commercial biogas plants are underutilized. Kemausuor et al.
reviewed commercial biogas systems that treat organic waste from different
sources [48]. According to them, the serious blockades to commercial biogas
development are (i) the initial capital costs; (ii) weak environmental policies;
(iii) poor institutional framework; (iv) poor infrastructure; and (v) a general
lack of willpower to implement policies and targets.
Biogas is a mixture of methane (CH4) (50%–70%), CO2 (30%–40%),
or other gases (hydrogen (5% to 10%), nitrogen (1% to 2%), water vapor
(0.3%), hydrogen sulfide (traces) produced by process of anaerobic digestion
of organic matter in an oxygen-free atmosphere [49]. Hence, an airtight
container is the main requirement to ensure an oxygen-free environment.
The airtight system where organic material (diluted in water) breaks by
natural microorganisms is called a bio-digester. The whole process of biogas
energy relies on generating a particular kind of bacteria in the proper amount.
The carbon-nitrogen ratio for anaerobic bio-gasification plays an important
role. For that purpose, sensing and IoT technology can play an important
Non Commercial Use

role. Some sources suggest it be between 20:1 and 25:1 ratio, whereas some
suggest it be approximately 30:1 ratio [50]. The generation of the biogas
depends on different chemical reactions. So, chemical sensors and IoT have
an essential role in continuously monitoring the proper amount of these
chemicals. The primary sensors used are the MQ4 sensor to measure the CH4
level, the MG811 sensor to measure the CO2 level, the temperature sensor,
the pH sensor, the load sensor, and others. The Fourier transform infrared
spectroscopy-based spectrometer helps detect chemical oxygen demand
(COD), total organic carbon, and volatile fatty acids of the reactor. Some other
feasible techniques are micro-gas chromatography (suitable to determine
hydrogen, CH4, hydrogen sulfide, nitrogen, and oxygen), membrane inlet
Author Copy
mass spectroscopy (suitable to determine CH4, CO2, hydrogen sulfide,
reduced organic sulfur compounds, and p-cresol) in the gas phase and with
near IR spectroscopy, pH in the liquid phase [51].
The pH level of the digester slurry plays an important role. It has to be in
the neutral or slightly base range. Different microorganisms require different
pH values for better performance. According to three categories of the
anaerobic digestion process, there was a requirement to maintain different
temperature ranges. The temperature affects the enzyme activity, efficiency
of the process, and microbial dynamics. It was also required to monitor
inhibitors or toxins in the digester slurry that may lead to lesser production
of CH4. These are ammonia, sulfides, light metal ions such as chromium,
iron, and other organic compounds.
Biogas energy not only helps in cooking, but the converted energy also
helps generate the electric supply [52]. This kind of generated electricity
comes under the category of green electricity. Power is generated in gas
turbine power generation by using CH4. Other biogas applications include
lighting, food preparation, two/four-wheeler fuel, power generation, and
running motors for pumping water. In addition, biological waste such as cow
dung waste, nourishment waste, farming waste (crops and animal manure),
and municipal waste generate biogas. Therefore, biogas generation leads to
a decrease in waste transportation to the landfills and less emission of green-
house gases. Biogas energy has another advantage: it is more demand-based
energy than photovoltaic and wind power.
The sensor and actuator, IoT, and AI technologies help to improve the
efficacy of a biogas power plant. The anaerobic digestion process needs
improved monitoring. Conventional methods require an up-gradation.
Correlation between different operational parameters is needed, and AI can
fulfill this need. IoT will also help improve performance. Success monitoring
Non Commercial Use

of the biogas generation process is necessary to avoid any kind of crash.

There should be a common control of feeding frequency. The technologies
mentioned above will be helpful. These are also useful for detecting fire,
leakage, explosions, poisoning at biogas plants, and alerting occupants. The
control strategies used at the biogas plant are ON-OFF, PI/PID, adaptive,
fuzzy, artificial neural networks (ANNs), and others.
3.5 CONCLUSION, CHALLENGES, AND FUTURE TRENDS
As the human population grows, so requires more energy. Most of the energy
Author Copy
generated is presently being covered by fossil fuel burning, leading to green-
house gas emissions that further change the earth’s climate. This climate
change has had an unprecedented effect on the human race. So, there should
be sustainable growth in the energy sector. For sustainable development,
energy production should have more percentage of generation from green
energy, and the building sector should also reduce the wastage of energy,
which consumes a significant portion of the generated power. In this chapter,
we have shown the role of IoT and the BEMS in increasing the energy effi-
ciency of the building and making them smart buildings. Several research
studies show the approach to reaching nonzero energy buildings to optimize
building energy consumption and self-dependability through efficient and
sustainable ways. We have also shown how to implement IoT with solar
and biogas-based energy integration with the BEMS to make smart build-
ings self-sustainable and reliable in terms of power, thus becoming smart.
Different types of IoE-based BEMS, such as storage systems and materials,
energy routers, renewable sources, and plug-and-play interfaces, are also
discussed in the chapter. The use of enhanced public safety and surveillance,
street lighting controls, infrastructure monitoring, meter reading, physical
security, and optimization systems to make the city smart was also shown in
the chapter. Different wireless technologies used for the considerable data
transfer of the IoT-based BEMS system and different sensors and actuators
used were also discussed. Hence in terms of security, fastest, cost-effective,
range, and consumes less energy, LoRa and 802.11 protocols are suitable and
can be used for the smart building IoT infrastructure for data communication.
There are still some challenges to deploying IoT in the energy sector, and it
includes privacy and security. Although engineers and researchers are coming
up with good solutions to these challenges, one is blockchain technology.
Survey shows that energy policymakers, economists, and managers with an
overview of IoT’s role in optimizing energy systems can help the sustainable
Non Commercial Use

energy development of the world requirement. The cloud and fog computing
platforms can ease the path for blockchain services in IoT. The energy
ingesting of IoT devices is another critical challenge, especially in the large-
scale deployment of these technologies. The IoT will require a significant
amount of energy to run billions of devices. Many IoT devices will also
yield a boundless deal of electronic waste. Hence, low-carbon and efficient
communication networks are in need to tackle the challenges. Fortunately,
these requirements have led to the appearance of the green Internet of Things
(G-IoT). The key component of G-IoT is its power-efficient characteristics
throughout the life cycle, that is, the design, deployment, production, and
finally, clearance. As a result, smart applications in the energy sector convert
Author Copy
buildings to smart buildings, converting towns to smart cities, which then
proceed to smart states, leading to smart countries, and finally to a smart
world. When human society reaches that level, the sustainable energy dream
will come true.
KEYWORDS
• biogas
• building energy management system
• energy management
• green energy
• grid management
• IoT
• sensors and network protocols
• smart building
• solar energy
REFERENCES
1. Nguyen, V. T., Thanh, L. V., Nam, T. L., & Yeong, M. J., (2018). An overview of Internet
of Energy (IoE) based building energy management system. In: 9th International
Conference on Information and Communication Technology Convergence: ICT
Convergence Powered by Smart Intelligence, ICTC 2018 (pp. 852–855). https://doi.org/
10.1109/ICTC.2018.8539513.
Non Commercial Use

2. Shah, A. S., Haidawati, N., Muhammad, F., Adidah, L., & Asadullah, S., (2019). A
review on energy consumption optimization techniques in IoT-based smart building
environments. Information (Switzerland), 10(3). https://doi.org/10.3390/info10030108.
3. Deshmukh, N. S., Bhuyar, D. L., & Jadhav, A T., (2018). Review on IoT Based Smart
Solar Photovoltaic Plant Remote Monitoring and Control Unit, 3(3), 6–10.
4. Acharya, V., Vinay, V. H., Anjan, K., & Manoj, K. M., (2018). IoT (internet of things)
based efficiency monitoring system for biogas plants. In: 2nd International Conference
on Computational Systems and Information Technology for Sustainable Solutions,
CSITSS 2017, (pp. 113–117). https://doi.org/10.1109/CSITSS.2017.8447567.
5. Verma, A., Surya, P., Vishal, S., Anuj, K., & Subhas, C. M., (2019). Sensing, controlling,
and IoT infrastructure in smart building: A review. IEEE Sensors Journal, 19(20),
9036–9046. https://doi.org/10.1109/JSEN.2019.2922409.
6. Ida, B. G. P., Nyoman, S. I., & Made, S., (2020). Application of IoT-based system for
Author Copy
monitoring energy consumption. International Journal of Engineering and Emerging
Technology, 5(2), 81–93. https://doi.org/10.24843/IJEET.2020.v05.i02.p014.
7. Doukas, H., Konstantinos, D. P., Konstantinos, I., & John, P., (2007). Intelligent Building
Energy Management System Using Rule Sets, 42, 3562–3569. https://doi.org/10.1016/j.
buildenv.2006.10.024.
8. Paniagua, E., Macazana, J., Lopez, J., & Tarrillo, J., (2019). IoT-based temperature
monitoring for buildings thermal comfort analysis. In: 2019 IEEE XXVI International
Conference on Electronics, Electrical Engineering and Computing (INTERCON) (pp.
1–4). https://doi.org/10.1109/INTERCON.2019.8853608.
9. Gulnizkij, N., & Gerlach, G., (2019). Bistable hydrogel-based sensor switch for
monitoring relative humidity. In: 2019 IEEE Sensors (pp. 1–4). https://doi.org/10.1109/
SENSORS43011.2019.8956498.
10. Gaur, A., Singh, A., Kumar, A., Kulkarni, K. S., Lala, S., Kapoor, K., Srivastava, V., et al.,
(2019). Fire sensing technologies: A review. IEEE Sensors Journal, 19(9), 3191–3202.
https://doi.org/10.1109/JSEN.2019.2894665.
11. Liu, X., Sitian, C., Hong, L., Sha, H., Daqiang, Z., Huansheng, N., & Information
Engineering, (2012). A Survey on Gas Sensing Technology, 9635–9665. https://doi.org/
10.3390/s120709635.
12. Roveti, D. K., (2001). Choosing a Humidity Sensor: A Review of Three Technologies.
Sensors Online. https://www.fierceelectronics.com/components/choosing-a-humidity-
sensor-a-review-three-technologies (accessed on 12 June 2023).
13. Prajapati (Shekhar), C., Rohith, S., Rudraswamy, S. B., Nayak, M. M., & Navakanta, B.,
(2017). Single-chip gas sensor array for air quality monitoring. Journal of Microelectro-
mechanical Systems. https://doi.org/10.1109/JMEMS.2017.2657788.
14. Singh, A., Balam, N. B., Kumar, A., & Kumar, A. (2016). An Intelligent Color Sensing
system for building wall. International Conference on Emerging Trends in Communication
Technologies (ETCT), Dehradun, India, 1–4, doi: 10.1109/ETCT.2016.7882930.
15. Kwon, O., An, J., & Hong, S., (2018). Capacitive touch systems with styli for touch
sensors: A review. IEEE Sensors Journal, 18(12), 4832–4846. https://doi.org/10.1109/
JSEN.2018.2830660.
16. Hub, Admini of Electronics, (2019). Introduction to Touch Sensors-Working, Capacitive
and Resistive. Electronics Hub. https://www.electronicshub.org/touch-sensors/ (accessed
on 12 June 2023).
Non Commercial Use

17. Sood, S., Rajinder, S., Karan, S., & Victor, C., (2017). IoT, big data, and HPC-based
smart flood management framework. Sustainable Computing: Informatics and Systems,
20. https://doi.org/10.1016/j.suscom.2017.12.001.
18. Expert, NetworkTech., (2013). Water Detection Sensors: Types and Applications. Network
Technologies Incorporated. https://www.networktechinc.com/blog/water-detection-sensors-

types-and-applications/303/ (accessed on 12 June 2023).
19. Ashraf, I., Yousaf, B. Z., Soojung, H., & Yongwan, P. (2020). A comprehensive analysis of
magnetic field based indoor positioning with smartphones: Opportunities, challenges and
practical limitations. IEEE Access, 4. https://doi.org/10.1109/ACCESS.2020.3046288.
20. Bedi, G., (2018). Internet of Things and Intelligent Technologies for Efficient Energy
Management in a Smart Building Environment. ProQuest Dissertations and Theses. https://
search.proquest.com/docview/2239986856?accountid=14598 (accessed on 12 June 2023).
21. Fernandes, H., Alam, M., Ferreira, J., & Fonseca, J., (2017). Acoustic smart sensors
Author Copy
based integrated system for smart homes. In: 2017 International Smart Cities Conference
(ISC2), (pp. 1–5). https://doi.org/10.1109/ISC2.2017.8090869.
22. Ehsan, Y., Victor, M. L., & Boric-Lubecke, O., (2013). True human presence detection
with radar technology. IEEE Life Sciences. https://lifesciences.ieee.org/lifesciences-
newsletter/2013/february-2013/true-human-presence-detection-with-radar-technology/
(accessed on 12 June 2023).
23. Luppe, C., & Amir, S., (2017). Towards Reliable Intelligent Occupancy Detection for
Smart Building Applications. https://doi.org/10.1109/CCECE.2017.7946831.
24. Marcello, F., Virginia, P., & Daniele, G., (2019). Sensor-based early activity recognition
inside buildings to support energy and comfort management systems. Energies, 12(13).
https://doi.org/10.3390/en12132631.
25. Sánchez-Garrido, C., Javier, M., & González-Jiménez, J., (2014). A configurable smart
e-nose for spatio-temporal olfactory analysis. Proceedings of IEEE Sensors, 2014.
https://doi.org/10.1109/ICSENS.2014.6985418.
26. Jia, M., Ali, K., Yueren, W., & Ravi, S. S., (2019). Adopting Internet of Things for the
development of smart buildings: A review of enabling technologies and applications.
Automation in Construction, 101, 111–126. https://doi.org/https://doi.org/10.1016/j.
autcon.2019.01.023.
27. Report, World Bank, (2020). Solar Photovoltaic Power Potential by Country. World
Bank. https://www.worldbank.org/en/topic/energy/publication/solar-photovoltaic-power-
potential-by-country (accessed on 12 June 2023).
28. Team ARENA, (2017). Annual Report 2017–2018.
29. Batcha, R. R., & Geetha, M. K., (2020). A survey on IoT based on renewable energy
for efficient energy conservation using machine learning approaches. In: 2020 3rd
International Conference on Emerging Technologies in Computer Engineering: Machine
Learning and Internet of Things (ICETCE) (pp. 123–128). https://doi.org/10.1109/
ICETCE48199.2020.9091737.
30. Adhya, S., Dipak, C. S., Abhijit, D., Joydip, J., & Hiranmay, S., (2016). An IoT-based
smart solar photovoltaic remote monitoring and control unit. In: 2016 2nd International
Conference on Control, Instrumentation, Energy & Communication (CIEC) (pp. 432–436).
31. Patil, S. M., Vijayalashmi, M., & Tapaskar, R., (2017). IoT-based solar energy monitoring
system. In: 2017 International Conference on Energy, Communication, Data Analytics
and Soft Computing (ICECDS) (pp. 1574–1579). https://doi.org/10.1109/ICECDS.2017.
8389711.
Non Commercial Use

32. Hanumanthaiah, A., Liya, M. L., Arun, C., & Aswathy, M. (2019). IoT-based solar
power monitor & controller for village electrification. In: 2019 9th International
Symposium on Embedded Computing and System Design (ISED) (pp. 1–5). https://doi.
org/10.1109/ISED48680.2019.9096221.
33. Khan, M. S., Sharma, H., & Haque, A., (2019). IoT enabled real-time energy monitoring
for photovoltaic systems. In: 2019 International Conference on Machine Learning,
Big Data, Cloud and Parallel Computing (COMITCon) (pp. 323–327). https://doi.
org/10.1109/COMITCon.2019.8862246.
34. Phoolwani, U. K., Sharma, T., Singh, A., & Gawre, S. K. (2020). IoT-based solar panel
analysis using thermal imaging. In: 2020 IEEE International Students’ Conference on
Electrical, Electronics and Computer Science (SCEECS) (pp. 1–5). https://doi.org/
10.1109/SCEECS48394.2020.114.
35. Chieochan, O., Anukit, S., & Ekkarat, B., (2017). Internet of Things (IoT) for smart
Author Copy
solar energy: A case study of the smart farm at Maejo University. In: 2017 International
Conference on Control, Automation, and Information Sciences, ICCAIS 2017 (pp.
262–267). https://doi.org/10.1109/ICCAIS.2017.8217588.
36. Kumar, N. M., Karthik, A., & Sriteja, P., (2018). Internet of Things (IoT) in photovoltaic
systems. In: 2018 National Power Engineering Conference, NPEC 2018. https://doi.
org/10.1109/NPEC.2018.8476807.
37. Choi, C., Jeong, J., Lee, I., & Park, W. (2018). LoRa-based renewable energy monitoring
system with open IoT platform. In: 2018 International Conference on Electronics,
Information, and Communication (ICEIC) (pp. 1, 2). https://doi.org/10.23919/
ELINFOCOM.2018.8330550.
38. Szabó, L., (2017). The history of using solar energy. Proceedings – 2017 International
Conference on Modern Power Systems, MPS 2017. https://doi.org/10.1109/MPS.2017.
7974451.
39. King, J., & Christopher, P., (2017). Smart Buildings: Using Smart Technology to Save
Energy in Existing Buildings. American Council for an Energy-Efficient Economy.
https://www.aceee.org/sites/default/files/publications/researchreports/a1701.pdf
(accessed on 26 July 2023).
40. Trinastic, J., (2016). The Potential of Rooftop Solar Energy: 40% of Total U.S. Electricity
Generation is Possible. Blog. https://www.nature.com/scitable/blog/eyes-on-environment/
the_power_of_rooftop_solar (accessed on 12 June 2023).
41. Teofilo, A., Qian (Chayn), S., Nenad, R., Yaguang, T., Jerome, I., & Chengyang, L.,
(2021). Investigating potential rooftop solar energy generated by leased federal airports
in Australia: Framework and implications. Journal of Building Engineering, 41, 102390.
https://doi.org/https://doi.org/10.1016/j.jobe.2021.102390.
42. Holowka, T., (2017). Top Four Benefits of Installing Solar Panels on Your Home. US
Green Building Council. https://www.usgbc.org/articles/top-four-benefits-installing-
solar-panels-your-home (accessed on 12 June 2023).
43. Canada, Govt. of. (2020). Solar Photovoltaic Energy in Buildings. Govt. of Canada.
https://www.nrcan.gc.ca/energy-efficiency/data-research-insights-energy-ef/buildings-
innovation/solar-photovoltaic-energy-buildings/3907 (accessed on 12 June 2023).
44. Kalogirou, S., (2015). Building Integration of Solar Renewable Energy Systems Towards
Zero or Nearly Zero Energy Buildings. 10(4), 379-385. https://doi.org/10.1093/ijlct/ctt071.
45. Ganguli, B., (2019). Home Solar System: Renewable Energy Solutions for Residential
Users. The Economic Times.
Non Commercial Use

46. Phung, M. D., (2019). IoT-Enabled Dependable Control for Solar Energy Harvesting in
Smart Buildings. https://doi.org/10.1049/iet-smc.2019.0052.
47. Terzio, H., (2015). A New Approach to the Installation of Solar Panels. https://doi.
org/10.1109/ICISCE.2015.133.
48. Kemausuor, F., & Muyiwa, S. A., (2018). A Review of Commercial Biogas Systems and
Lessons for Africa, 1–21. https://doi.org/10.3390/en11112984.
49. Singh, A. K., Jegan, R., & Kumar, P., (2022). Review on the usage of green energy
and IoT in smart buildings. In: 2022 6th International Conference on Computing
Methodologies and Communication (ICCMC) (pp. 452–458). https://doi.org/10.1109/
ICCMC53470.2022.9754130.
50. Satyanarayan, S., Kaul, S. N., Badrinath, S. D., & Gadkari, S. K., (1987). Biogas from
Human Waste. Consortium on Rural Technology. https://www.ircwash.org/sites/default/
files/352.1-4961.pdf (accessed on 12 June 2023).
Author Copy
51. Cinar, S., Cinar, S. O., Wieczorek, N., Sohoo, I., & Kuchta, K., (2021). Integration
of Artificial Intelligence into Biogas Plant Operation. Processes. 9(1), 85. https://doi.
org/10.3390/pr9010085.
52. Anbarasu, V., Karthikeyan, P., & Anandaraj, S. P., (2020). Turning human and food
waste into reusable energy in a multilevel apartment using IoT. In: 2020 6th International
Conference on Advanced Computing and Communication Systems, ICACCS 2020 (pp.
440–444). https://doi.org/10.1109/ICACCS48705.2020.9074170.
53. Wind Power by Country – Wikipedia [Internet]. [cited 2023 Jul 20]. Available from:
https://en.wikipedia.org/wiki/Wind_power_by_country#cite_note-9 (accessed on 28 July
2023).
Non Commercial Use

CHAPTER 4
A Novel Electrical Load Forecasting

Model Using a Deep Learning Approach
NEELAPALA ANIL KUMAR,1 RAVURI DANIEL,2 and
PRUDHVI KIRAN PASAM3
Author Copy
1
Department of Electronics and Communication Engineering (ACED),
Alliance University, Bangalore, Karnataka, India
2
Department of Computer Science and Engineering, Prasad V. Potluri
Siddhartha Institute of Technology, Vijayawada, Andhra Pradesh, India
3
Department of Information Technology, SRKR Engineering College,
Bhimavaram, Andhra Pradesh, India
ABSTRACT
The estimate of electricity appeal in modernistic years is becoming

progressively relevant thanks to market-free trade and, thus, the initiation
of sustainable assets. To satisfy the demands, leading intelligent models
are built to form sure explicit power forecasts for multi-time prospects.
The load forecasting of electric Power is a crucial process in devising
the electric industry and operating electric power systems. Short-term
forecasts are adopted to program the power generation and transmission of
electricity. Medium-term forecasts are meant to line up the fuel purchases.
This necessitates the implementation of the productive determination of
algorithms could be a fundamental feature of smart grids and an efficient
tool for determining ambiguity for better cost and energy ability decisions
like slate the origination, authenticity, power escalation of the system,
and monetary smart grid activities. This work introduces a model for the
evaluation of the utilization of electricity, which can accurately forecast
Non Commercial Use

subsequently estimated from minimum to maximum duration with significant

improvement in the accuracy of forecasting through advanced deep learning
techniques. The analyzes or findings also can provide interesting results for
energy consumption with parameters like forecasting efficiency and error
with duration of data monitoring algorithms namely (LSTM)–long short-

term memory (RNN) – recurrent neural networks and multi-layer perceptron
algorithms (MLP). These algorithms furnish the most interesting results
with respective to the duration of data. Mainly, MLP and RNN proved to
produce favorable results for 24-hour data. Similarly, LSTM has proved
better for 15-day data and monthly data with consistency in terms of errors,
squared, and mean square. To anticipate data ranging from day to month, the
Author Copy
minimal Forecasting error was attained by adopting MLP with R2 (0.91). On
hour-based data, R2 of LSTM holds effective for half-monthly and monthly
data with (0.88 and 0.93), RMSE (89.54 and 84.98), MAPE (3.51 and 2.47).
RNN has been proven to attain the moderate outputs comparatively. MLP
for half-monthly and monthly in terms of R2 (0.81 and 0.92), RMSE (90.72
and 85.78) and MAPE (4.25 and 4.01). The result of LSTM acknowledges
the enhanced attainment and substantial achievements of electrical load
forecasting.
4.1 INTRODUCTION
Major aspects of advanced technology are moving around the consideration

of power. Traditional electrical power generation and distribution systems
getting obsolete due to energy loss and less effective affairs. The major
reason for the claim could be outdated structure. On the other hand,
reduction of natural resources and pollution which are used in the power
generation process. Utmost of these problems were addressed by replacing
the traditional grids with smart grids. Smart grids give the installation of
monitoring, optimization, distribution, and consumption of electricity. In the
present payload existence service, at most important and attainable aspects
to satisfy the forthcoming payload appeals mainly by gratuitous loading,
knockouts. This highly requires a measure of precautions and instantaneous
monitoring of power stations with respect to the payload, starting from
minimal level observation from hours to the maximal observations of the
month. From the practical observations, flutters arrive in a short span of
intervals, which demands hourly base payload monitoring [1]. Forecasting of
the payload is a problem based on probability rather than being inconsistent.
This demand is for forecasting stochastic data rather than static data analysis
Non Commercial Use

A Novel Electrical Load Forecasting Model 69
for safeguarding the power plant [2]. In this aspect, most data scientists
followed the statistical approach-based data analysis algorithms to forecast
the payload of power plants [3] and hierarchical forecast [4]. The load for
the casting of power plants demands the dynamic duration of data retrieval
and analysis. This has a common requirement that the extensive approaches
have been monitored for a short time to long time data series intervals in the
process of forecasting data for identifying the failures to safeguard the power
plant and environment. The short-term payload forecasting turn into a severe
problem for the operating and delivering of power networks to intrude pivotal
issues associated with failures in power systems. Taking the considerations
of maintenance cost of the power plant in the present scenario, it is made
Author Copy
important to automate the process of monitoring load. This necessitates the
typical deep learning intelligent systems for effective and economically
safeguarding monitoring process of payloads in a power plant. It is also an
accessory for optimizing the energy conservation while reducing emigration.
Effective electric payload forecasting can protect electricity while using
minimal energy coffers and allowing for easy distribution. Contrarily, if
payload forecasting is done duly and straight, helps in managing the unborn
electric payload demands with efficient consignment planning [5].
4.1.1 ASPECTS OF ELECTRICAL LOAD
Traditional methods incorporated in electrical systems for the operation

might be definite from others, and a single electrical device is used most
frequently. This type of adaption in the existing systems leads to the problem
of variation in separate loads by casting up with major installation weights.
The various aspects affect the electric load involves climate, time aspects,
and arbitrary aspects.
1. Climate: It is one of the considerable impacting aspects of power
forecasting. The climate is an extruded important rudiment among
the capacity cast script. Generally, load pay models are erected
and tested, with the considerations of factual rainfall reading. The
rainfall cast obviously companion to the declination and instill in
the achievement of the model. Forecasting of Weather, speed of
wind, and climatic conditions attain a major part in short-term load
forecasting (STLF) by varying the chart. The weather has major
betters on ST electrical distribution forecasting of power systems [6].
Sensible climatic consignments have a vast effect on scaled-down
Non Commercial Use

artificial energy architectures. Climate rudiments impact hourly

power payload forecasting. In times of maximum humidity climates,
the cooling outfit can be utilized for lengthy duty patterns for
removal of non-significant condensation of conditioned air and holds
the capability of temperature reduction in the air, which impacts the

minimization of bracing loads.
2. Time Aspects: Power load fluctuations are highly related to the
working timings for the non-working day power load cast is the same
as on weekends. During non-working days, the power loads are lower
compared to working days due to hugeness. Power load cast schemes
for day-to-day daily cycle is general all through day and week. While
Author Copy
seasonal issues describe the modifications in long-term improvement
and pattern of rainfall.
3. Random Aspects: Arbitrary conditions which can agitate energy
load definition include required loads that had not been planned
and the absenteeism of workers making the forecast delicate. Load
forecast has been given more importance in terms of recent times
smart energy operation systems. Load forecasting has been added
with users on an enduring base. Substantially, STLF is incorporated
to maintain payload forwarding the transfer energy control schedule
to 30 glitters for a single day. Thus, any improvement gained along
the operational perfection of STLF tends to reduce the reduction of
electrical charges of OS along with the effective energy network
enhancement [7]. STLF facilitates the rearmost forecast information
of rainfall, new payload forecasting strategy, and power system
arbitrary behavior [8]. The load cast is getting more significant
because of its adding importance towards renewable sources of
energy trends, smart grids, and micro-grids. Several strategies
have been imposed for operation and consignment, similarly as
seasonal bus-accumulative, bus-accumulative integrated moving
average models, retrogression, precipitously integrated moving
average. In recent days, alternative ways were found to improve
the effectiveness of load cast operation of power systems and the
elegancy similar to artificial intelligence (AI) and fuzzy sense.
These results were effectively calculated on the technologies of AI
to clarify forecasting load issues.
Ranaweera et al. [9] presented acquainted rules of fuzzy developed by
applying a literacy algorithm type to introduce load operation and data of
literal rainfall. He et al. [8] suggested an advanced system for the quantifying
Non Commercial Use

the stringed unstable power load to gain information subsequently for

awaited load with neural structure quintile regression construction frame in
the development of statistical ways for castings. Later different styles of AI
were implemented for different electrical payloads, including structures of
smart grids [10].

In mere future payload demands, distributed system load forecasting
and artificial neural network (ANN) exogenous auto-regressive vector
inputs. In addition to these AI-based fuzzy logistics specialized network
structures, neural systems based on Bayesian neural networks and SVM are
extensively used for handling the issues of load forecasting of power loads.
STLF advance in getting a claim because of its non-stationary load data in
Author Copy
addition to durable dependencies, carrying related data for forecasting. This
is a reason for thinking about the application of long-short-term memory,
which is an intermittent distinct type of (RNN) structure [11], to overcome
the issue of STLF. LSTM operates adequately about horizon anticipation of
long duration comparatively. For the remaining AI methods derived with
diversified payload for series time outgrowth connections.
4.2 MATERIALS AND METHODOLOGY
In this section, the detailed materials and methodology are presented. Neural
network development is an extended process that requires a lot of thought
behind the architecture and a whole group of nuances that make up the
system.
4.2.1 MATERIALS
In this section, the details of the data set and tools used to develop the
proposed work are presented. Some of the tools used in this work are Scikit
Learn, TensorFlow, and Kera's. The tools are proven very efficient to build
the neural network models.
1. Dataset: The data was chosen from a confluent to fulfill the
maximum payload conditions for analysis. The data was supported
with a Kaggle database from one of its electricity systems based on
the hourly load demands of one grid as utilized in Ref. [12]. The
load time series was generated for min duration starting from hours
to month as short, medium, and maximum durational data intervals.
Non Commercial Use

The data set of the electricity load collected from Kaggle is shown in
Figure 4.1.
Author Copy
FIGURE 4.1 The data set of the electricity load.
2. Tools: The major tools are used in this work Scikit Learn, Tensor-
Flow, and Keras. The Scikit Learn is a simple and efficient tool for
foretelling data analysis. The TensorFlow is a deep learning and
open-source library used for the smooth running of deep learning
applications. The Keras is a high-level neural network API and is
developed mainly to enable faster experimentation.
4.2.2 METHODOLOGY
Figure 4.2 depicts the schematic block diagram of the methodology, which
interprets the forecasting system for electric load for different durations
ranging from day to month data with several innovative forecasting models.
This model is incorporated with STLF [13] and MTLF used for the planning
of the power systems ranging from a single day to a single month. Once
after introducing the data pre-processing, we progress with robust deep
neural network proficiency models, namely MPL, RNN, and LSTM. The
performance was determined based on the ground test set standard error
performance criteria as RMSE, MAPE, and R-squared. Eventually, load
forecasting for short to medium durations was performed in terms of
durational data for forecasting load demands.
1. Multilayer Perceptron (MLP): It can be trained with local learning
procedures. The training process is performed with a few specimens
selected from the neighborhood of the area of interest. From the data
Non Commercial Use

Author Copy
FIGURE 4.2 Electric load forecasting framework schematic diagram for short and medium-
term load forecasting.
observations complication of the local target function is less than the

global complexity target function. This feature facilitates the MLP
design with minimum neurons to train and learn the system faster.
In Ref. [14], the authors addressed that the utility of a one-neuron
sigmoid acceleration function delivers productive output compara-
tively to the other system, functioning across several neurons within
the concealed surface. MLP can be worked with a backpropagation
algorithm which is claimed to be acquisition monitored algorithm
for handling the problems of large-scale prediction implementation.
Few significantly, in annual forecasting of gross electricity demand
Indicator established on climatic and socio-economic parameters.
Closest neighbors’ identification is mainly performed with hyper-
parameter by utilizing the LLP method. Levenberg-Marquardt
algorithm with Bayesian regularization for MLP absorption [15]
minimizes the squared miscalculations consolidation of weights to
avoid overfitting. The effective utility of MLP is represented in the
Non Commercial Use

algorithm shown in Figure 4.3, which depicts the MLP algorithm

architecture and working, which consists of several normalized time
series input loads appeals as well as hidden neurons and actual time
series activation functions. The final error metrics are calculated to
forecast the difference between the actual load and the predicted load
requirements.
Author Copy
FIGURE 4.3 Architecture of multi-layer perceptron (MPL).
Deep learning is a subdivision of AI, works on par with ANNs

and machine learning (ML) performance. AI functions as a
mimicking pattern of the human brain. ANN utilizes informational
data to process the required information using neuron groups which
originate as layers [16]. These neurons deliver information from one
neuron source to another. Where a small quantity of output infor-
mation is fed back to the preceding layer until it has retrieved the
expected output. Information is represented in the form of regression
at the output layer or classification. DL proved to be effective for
forecasting the complex problems [17] with feature extraction and
automatic data method improvements.
2. Recurrent Neural Network (RNN): The most promising and
powerful robust type of neural network, with its single internal
memory of use, can be claimed as RNN. This algorithm is mainly
preferred for its sequence data modeling. The behavior of this algo-
rithm mimics the brain functionality; it keeps generating forecasting
responses. This algorithm is relatively ancient, identified in the 1980s
but came into existence recently to forecast and process the time
series data [18]. RNN is also one of the deep learning algorithms
Non Commercial Use

used to solve the huge computational data with a short series of

time intervals. In the 1990s, the massive utility of RNN came into
existence because of its internal memory considerations. As this
algorithm can only focus on the significant things about the received
input. Which facilitates them to be most particular in forecasting the

next outcome. This feature placed RNN in higher demand, specifi-
cally for its time series sequential data widely utilized for several
applications. RNN placed the highest priority for its most in-depth
understanding of the sequence of data compared to any other algo-
rithms. The understanding of RNN rest in a feed-forward neural
network with time series data [19]. It does not have any notation
Author Copy
of time order. If a decision must make, it investigates the current
input and previously received inputs from what it learned. From the
above details, it is observed that the limitations of RNN regarding the
computation speed are observed to be low, and it is difficult to access
the long-time go information as it cannot consider any future input
for the current state.
Figure 4.4 shows the RNN model's construction. Forecasting
1-dimensional load time sequence as input for RNN model, which
was refined in various RNN model layers, and the output is the error
or variance between actual and forecast load values. RNN holds
better for the constant model size and input, and computation takes
consideration of historical information shared across the time.
FIGURE 4.4 Load forecasting RNN architecture.
3. Long Short-Term Memory (LSTM): It is designed to recognize

sequences of time series data patterns of deep learning applications
[20]. It is represented as artificial recurrent neural network (RNN)
architecture. LSTM is recognized as a famous model expert [21] in
forecasting time sequences that can handle both long and short-term
data dependents effectively. RNN architecture issues for motivation
Non Commercial Use

are mostly designed with LSTM to control disappearing gradients.

Especially for long-term dependent data, acceptable to form the
short- and long-term neural network. Mostly, Recursive Neural
Network architecture is a framework that introduces the forgetting
gate, input barrier, and output gateway neurons. This new approach
to development can efficiently guide the issue of vanishing gradient
[22]. This addons make LSTM architecture most preferred for long-
term data problem dependencies, which makes LSTM to be most
widely used for forecasting the time series. Most of the applications
pertaining to neuron-cognitive performance use RNN's methods
from LSTM. The LSTM solved the well-known problems of machine
Author Copy
learning, like gradient learning with the back-flow error problems. It
also proves to be most appropriate for compressible input sequence,
noise, and efficient adaptive learning, in the absence of short time lag
capabilities losses. It is of utmost importance to solve lag tasks for
complicated problems, which are difficult in conventional machine
learning. The LSTM has linear hidden layers, but the gradient flow
through long sequences is permitted by self-loop memory blocks.
The recurrent blocks of LSTM, called memory blocks, comprise
multiplicative units of three, majorly forget gates, input, output [23],
and recurrent memory cells. Memory sections saving and connect
information for a maximum duration for solving vanishing problems
can be granted by this cell. The LSTM is additionally admitted for
improving the functionality by memory cell resetting [24]. Figure
4.5 represents the general LSTM model architecture.
FIGURE 4.5 Architecture of LSTM model.
Non Commercial Use

As a major innovation, the LSTM memory cell is used to limit the

data-storing state. The forget gate used in the first step will eliminate the
unnecessary information. The application is then followed by a sigmoid
function to measure the accelerated state of forget fg.
=f g σ ( w [h ,x ] + b ) (1)
i t −1 g i
The second step involves locating the necessary data that must be saved
within the cell state. The “input gate layer,” another sigmoid layer, is used to
update information. Then, a vector c̃ t is created using the tanh function, and
new values that must be updated based on the current state.
Author Copy
=ig ( w [ h ,x ] + b ) (2)
j t −1 g i
=C g tanh ( w [h ,x ] + b ) (3)
j t −1 g i
The old cell state cg–1 of the second step is changed to the new cell state
cg. Multiply cg–1 by fg to remove the information content from the ancient
cell. Then add it *c̃ g. The updated data provided is represented by the new
candidate values.
=C f g *C + i *C (4)
g g −1 g t
Finally, the output must be analyzed in several steps, with the sigmoid
function serving as an output barrier for emphasizing the cell state. In addi-
tion, the achieved cell state is transferred across tanh (.) to the output, which
is the product used to calculate the required information.
=og ( w [ h ,x ] + b ) (5)
1 t −1 g 1
hg = o * tanh(C ) (6)
g g
In Eqns. (1)–(5), wi, wj, wk, and w1 represent the weight matrices along
with representation of bias vectors bi, bj, bk, and bl.
4.2.3 METRICS FOR EVALUATING PERFORMANCE
The quantitative casting accuracy was measured in terms of determining

coefficient errors by squared error (R2), root mean squared (RMSE), and
mean absolute percentage (MAPE) errors. The usage of the detailed metrics
for the following renowned error forecasting is summarized [25].
Non Commercial Use

1. Co-Efficient of Determination (R2): The following function shows

the computed R2.
n
∑ (x − y )
2
i i
R2 = i = 1 (7)
n
∑ (x − y )
2
i i
i =1
Here y denotes the average values of the entire samples.
2. Root Mean Squared Error (RMSE): The evaluation parameter
to test the forecasted quality requires a measure for quantitative
Author Copy
accuracy metric. In the prevalent study, a volume termed RMSE is
denoted with the given formula.
n
RMSE
= ∑ (x − y ) (8)
i i
i =1
The measured and forecast values of the ith sample are denoted by
xi and yi, and ‘n’ represents the total number of the training dataset
samples. The RMSE smaller value specifies the better-selected set
descriptors.
3. Mean Absolute Percentage Error (MAPE): It is a measure of
accuracy for forecasting systems where the accuracy is measured
in terms of percentage and also can calculate the average absolute
percentage error for every instant of time period minus actual values
divided by actual values The MAPE is computed using the following
formula.
100 n ( yi − xi )
MAPE = ∑ (9)
n i =1 n
4.3 RESULTS
As a process of forecasting of electricity load at regular instants of time

ranging from day load to monthly load had been performed by various deep
learning algorithms, equipped, and analyzed, with parameters namely MLP,
LSTM, and RNN applied for load forecasting time series data with a load of
day (24-hours) to a 1 month. The determination of performance in terms of
R2, MAPE, and RMSE for forecasting load demands of electricity. The input
data of the different timing data has been collected from the Kaggle database.
The computed difference between obtained and forecast values suggests that
Non Commercial Use

it is smaller and unbiased, indicating that it is the best fit for the given data.
Statistically, the residual plots which measure the better fit reveal undesir-
able residual patterns indicating that misleading results are more effective
than numbers. This metric is known as the coefficient of determination, and
it indicates the degree to which data is correlated.
4.3.1 24 HOURS ELECTRIC LOAD FORECASTING
Table 4.1 represents the 24-hour electrical load forecasting based on the error
parameters R2, RMSE, and MAPE. The following tabular content depicts the
Author Copy
R2 maximum for the MLP with a value of 0.91, moderate for RNN as 0.75,
and comparatively least for LSTM with a value of 0.37. The parameters
RMSE and MAPE holds better for MLP, with a comparative minimum value
of 62.46 and 1.74. This result reveals that MLP is more effective than other
algorithms for the short-term duration (24 hours).
Figure 4.6(a)–(c) shows a pictorial depiction of the results of three
different procedures for obtaining 24-hour electrical load forecasts (MLP,
LSTM, and RNN). The MLP is the closest to the real load, according to
extracted graphs, followed by LSTM and RNN.
TABLE 4.1 Forecasting of One-Day Electricity Load Forecasting

Method R2 RMSE MAPE
MLP 0.91 62.46 1.74
RNN 0.75 91.45 2.53
LSTM 0.37 286.59 5.04
(a) One-day electricity load forecasting using MLP.
Non Commercial Use

Author Copy
(b) One-day electricity load forecasting using RNN.
(c) One-day electricity load forecasting using LSTM.

FIGURE 4.6 One-day load forecasting using (a) MLP; (b) RNN; and (c) LSTM.
4.3.2 15-DAYS ELECTRICITY LOAD FORECASTING
Table 4.2 reveals the forecasting of 15 days electrical load forecasting.

Focused on R2, RMSE, and MAPE with the input database of the electricity
load. The following tabular content depicts the maximum as 0.88 for the
LSTM, moderate for RNN as 0.81, and comparatively least for MLP as 0.78.
The parameters RMSE and MAPE hold better for LSTM with a comparative
minimum value of 89.54 and 3.51. This result reveals that LSTM outper-
forms comparatively for the mid-term duration (15 days).
Non Commercial Use

TABLE 4.2 Forecasting of the 15 Days Electricity Load Forecasting

Method R2 RMSE MAPE
MLP 0.78 120.422 5.44
RNN 0.81 90.72 4.25

LSTM 0.88 89.54 3.51
Figure 4.7(a)–(c) presents the results of 15 days electrical forecasting

payload for the 3 different methods (MLP, LSTM, and RNN) with graphical
representation. From the extracted waveforms, it is evident that LSTM data
values are nearest to the real load curve accompanied by RNN and MLP.
Author Copy
4.3.3 30-DAYS ELECTRICITY LOAD FORECASTING
Table 4.3 represents the forecasting of load forecasting for one month. In
comparison to the R2-squared error method, the LSTM provides the best
forecasting with R2 0.93, accompanied by RNN as R2 0.92 and MLP as R2
0.65. According to MAPE, the longest duration of one month of forecasting
was obtained as 2.47 for LSTM, 4.01 for RNN, and 5.89 for MLP. Likewise,
with the consideration of RMSE, effective values have been stated as 84.98
for LSTM, 85.78 for RNN, and 155.65 for MLP. From the data forecast, it is
evident that LSTM performance will be more effective for the max duration
time series data in providing maximum squared error and minimal quantity
of RMSE and MAPE.
(a) 15-Day’s electricity load forecasting using MLP.
Non Commercial Use

Author Copy
(b) 15-Day’s electricity load forecasting using RNN.
(c) 15-Day’s electricity load forecasting using LSTM.

FIGURE 4.7 15-day electricity load forecasting using (a) MLP; (b) RNN; and (c) LSTM.
TABLE 4.3 Forecasting of 30 Days Electricity Load Forecasting

Method R2 RMSE MAPE
MLP 0.65 155.65 5.89
RNN 0.92 85.78 4.01
LSTM 0.93 84.98 2.47
Figure 4.8(a)–(c) presents the results of 30 days (one month) payload

forecasts acquired utilizing three different techniques (MLP, LSTM, and
RNN). According to the extracted waveforms, the LSTM is the nearest to
the real load profile, accompanied by RNN and MLP.
Non Commercial Use

Author Copy
(a) 30-days electricity load forecasting using MLP.
(b) 30 days electricity load forecasting using RNN.
(c) 30 days electricity load forecasting using LSTM.

FIGURE 4.8 30-day load forecasting using (a) MLP; (b) RNN; and (c) LSTM.
Non Commercial Use

4.3.4 PERFORMANCE OF ALGORITHMS USING R2, RMSE, AND MAPE
Figure 4.9 depicts the comparative performance among the parameter used
for load forecasting of electrical payloads with different time duration of the
input data with the parameters, namely Squared error R2, root mean square
error, MSE, Moving average absolute percentage error MAPE. Figure 4.9(a)
reveals the performance of algorithms respective to R2 and reveals the fact
that for the day data, MLP yields maximum output, while LSTM proves its
effective capability for 15 days data and month data. Figure 4.9(b) marks the
performance of comparative algorithms with the parameter RMSE. RMSE
is proven more effective with LSTM for the day data and MLP for 15 days
Author Copy
data and finally well proved for both CNN and LSTM. Figure 4.9(c) exhibits
the effective performance with the MAPE parameter. Where MLP proved to
be more effective for both the Medium and maximum data, i.e., for 15 days
and one month, and LSTM holds better for small duration data for day data.
As a common view in all the parameter performance, RNN is provided with
the most stable results maintaining the substantial results for only the short
time series data. This result shows the scope for mixed algorithm require-
ments, which proved the LSTM more effective with desired results for all the
parameters with diversified durations.
4.3.5 DISCUSSIONS
Authentic load forecasting can manually mitigate the effect of sustainable

power admittance to the network, expedite the electricity flora to organize
unit protection, and motivates the electricity provider organizations to
expand a possible strength invoice deal. Many electricity gadgets consisting
of the protection of generator scheduling, renewable power integration, and
electricity grids rely on the burden forecasting LSTM is substantially utilized
in forecasting. LSTM can apprehend sample traits in conjunction with
scale-invariant traits whilst the nearby facts have stable relationships with
one some other. The layout of the regionally set route of load informational
records in nearby hours will be extricated via way of means of LSTM. In
some other load forecasting layout makes use of LSTM infrastructure and
make an evaluation of the relaxation of the neural systems. The examinations
exhibit the truth approximately; LSTM infrastructure is an important concern
in load forecasting. Considering the above description, LSTM has nicely
proven to offer excessive exactness forecast in medium and most due to their
wonderful function to seize hid traits.
Non Commercial Use

Author Copy
(a) R2 performance in comparison with MLP, RNN, and LSTM for one day, half-month, and
one month.
(b) RMSE performance in comparison with MLP, RNN, and LSTM for one day, half-month,
and one month.
(c) MAPE performance in comparison with MLP, RNN, and LSTM for one day, half-month,
and one month.
FIGURE 4.9 Parameter performance of algorithms (a) R2; (b) RMSE; and (c) MAPE.
Non Commercial Use

Along those lines, a hybrid neural logical shape is far essential to

capture and contain such unique, unseen developments in order to provide
powerful execution. The LSTM, RNN, and featured fusion modules are
all included in this package. The forget gate in the LSTM module, as well
as the memory mobility in the RNN module, have been used to eliminate
designs of nearby styles and similar layouts that show up in a variety of
places. The included fusion aggregate module is used to keep track of those
unanticipated requirements and generate the most up-to-date forecast. To
predict a real phase electric-powered load collection of time, the suggested
RNN-LSTM framework was constructed and used. In addition, some tech-
niques have been updated to compare to our recommended version. The
Author Copy
RNN and LSTM modules have been tracked individually to demonstrate
the validity of the counseled version. Furthermore, the records file was
divided into segments to test the effectiveness of the advised version. In
summary, this chapter presents a deep mastering shape that can capture
and contain the hidden qualities of RNN and LSTM to execute some
distance with greater accuracy and power. We used MLP, LSTM, and RNN
to compute forecasts for 1 day, 1/2 month, and one month in this study.
R2, MAPE, and RMSE were used to calculate the computational overall
effectiveness. We enhanced the metrics of those methodologies in order
to improve overall effectiveness and reap the previously mentioned STLF
and MTLF. The results of our comparison with previous findings in terms
of RMSE and MAPE for 15 days and one month were identical. This well-
known demonstrates that our recommended fashions with specifications
enhancement provide the best beforehand detection overall performance
when combined with the combined set of rules layout with LSTM and
RNN for STLF and MTLF.
4.4 CONCLUSION
Electrical load forecasting for different duration of time ranging from min
of 24 hours (day) to one month has been verified with different machine
learning and deep learning algorithms. For this study, the database has
been utilized from the Kaggle database feeder, and load forecasting for 24
hours (one day), 15 days, and one month has been computed. To estimate
and improve the forecasting performance of all types of load demands,
various AI algorithms (MLP, LSTM, RNN) have been utilized. Forecasting
performance is analyzed R2 error, MAPE, and RMSE are examples of
Non Commercial Use

robust error metrics. The smallest error value observed for the 24 hours
was 84.98, 2.47. For 15 days and one month, load forecast data indicated
the best fit. The smallest the value of the R2 statistical, the more closely the
data is fitted. From the practical observation with sample input data, the
following conclusions were made w.r.t to parameters and several machine

and deep learning algorithms. The data states that the R2 yields the lowest
error at 85.78 and 84.98 in all these cases except the LSTM, observed by
MAPE and RMSE. A better forecast is obtained with STLF using MLP. In
return, a similar forecast is obtained for MTLF using LSTM and RNN. It
indicates that, for higher data demands, the deep learning models provide
better forecasting with a greater neuron in the hidden layer and improved
Author Copy
activation functions. We can deduce from this experimental validation
that power systems with highly complicated expansion, power production
consumption, and power planning can be better forecasted using this
approach.
4.4.1 FUTURE DIRECTIONS
There are various factors which affect the load demand growth. This must be
enhanced with real-time data, and comparison must be drawn with a practical
grid along with each parameter for understanding the apparent capability and
range. Hence this analysis helps in the practical estimation of load demands
for suitable areas.
KEYWORDS
• deep learning
• electricity load forecasting
• load demand growth
• Long short-term load forecasting (LSTM)
• multi-layer perceptron (MLP)
• real-time data
• recursive neural networks (RNN)
• smart grids
Non Commercial Use

REFERENCES
1. Singla, M. K., & Hans, S., (2018). Load forecasting using fuzzy logic toolbox. Global
Research and Development Journal for Engineering, 38, 12–19.
2. Hong, T., Pinson, P., Fan, S., Zareipour, H., Troccoli, A., & Hyndman, R. J. (2016).
Probabilistic energy forecasting: Global energy forecasting competition 2014 and
beyond. Int. J. Forecast., 32, 896–913.
3. Hong, T., Pinson, P., & Fan, S., (2014). Global energy forecasting competition 2012. Int.
J. Forecast., 30, 357–363.
4. Hussain, L., Nadeem, M. S., & Shah, S. A. A., (2014). Short-term load forecasting system
based on support vector kernel methods. Int. J. Comput. Sci. Inf. Technol., 6, 93–102.
5. Jamaaluddin, J., Hadidjaja, D., Sulistiyowati, I., Suprayitno, E. A., Anshory, I., &
Author Copy
Syahrorini, S., (2018). Very short-term load forecasting peak load time using fuzzy
logic. IOP Conference Series: Materials Science and Engineering, 403, 012070.
6. Hossain, E., Khan, I., Un-Noor, F., Sikander, S. S., & Sunny, M. S. H., (2019). Application
of big data and machine learning in smart grid, and associated security concerns: A
review. IEEE Access, 7, 13960–13988.
7. Raza, M. Q., & Khosravi, A., (2015). A review on artificial intelligence-based load
demand forecasting techniques for smart grid and buildings. Renew. Sustain. Energy Rev.,
50, 1352–1372.
8. He, Y., Xu, Q., Wan, J., & Yang, S., (2016). Short-term power load probability density
forecasting based on quantile regression neural network and triangle kernel function.
Energy, 114, 498–512.
9. Laouafi, J., Mordjaoui, M., & Boukelia, T. E., (2018). An adaptive neuro-fuzzy
inference system-based approach for daily load curve prediction. Journal of Energy
Systems, 115–126.
10. Velasco, L. C. P., Villezas, C. R., Palahang, P. N. C., & Dagaang, J. A. A., (2015).
Next-day electric load forecasting using artificial neural networks. In: 2015 Int. Conf.
Humanoid, Nanotechnology, Inf. Technol. Control. Environ. Manag. (pp. 1–6). IEEE.
11. Williams, R. J., & Zipser, D., (1989). A learning algorithm for continually running fully
recurrent neural networks. Neural Comput., 1, 270–280.
12. Buitrago, J., & Asfour, S., (2017). Short-term forecasting of electric loads using nonlinear
autoregressive artificial neural networks with exogenous vector inputs. Energies, 10, 40.
13. Kim, K. H., Park, J. K., Hwang, K. J., & Kim, S. H., (1995). Implementation of a hybrid
short-term load forecasting system using artificial neural networks and fuzzy expert
systems. IEEE Trans. Power Syst., 10, 1534–1539.
14. Lee, W. J., & Hong, J., (2015). A hybrid dynamic and fuzzy time series model for
mid-term power load forecasting. Int. J. Electr. Power Energy Syst., 64, 1057–1062.
15. Niu, D. X., Shi, H. F., & Wu, D. D., (2012). Short-term load forecasting using Bayesian
neural networks learned by hybrid Monte Carlo algorithm. Appl. Soft Comput., 12,
1822–1827.
16. Hernández, L., Baladrón, C., Aguiar, J. M., Calavia, L., Carro, B., Sánchez-Esguevillas,
A., et al., (2014). Artificial neural network for short-term load forecasting in distribution
systems. Energies, 7, 1576–1598.
17. Li, X., & Wu, X., (2015). Constructing long short-term memory-based deep recurrent
neural networks for large vocabulary speech recognition. In: 2015 IEEE Int. Conf.
Acoust. Speech Signal Process (pp. 4520–4524). IEEE.
Non Commercial Use

18. Zaremba, W., Sutskever, I., & Vinyals, O., (2014). Recurrent Neural Network Regulariza-
tion. arXiv preprint arXiv: 1409.2329.
19. James, P. E., Kit, M. H., Vaithilingam, C. A., & Chiat, A. T. W., (2020). Recurrent neural
network-based speech recognition using MATLAB. International Journal of Intelligent
Enterprise (Vol. 7, pp. 56–66).

20. Jozefowicz, R., Zaremba, W., & Sutskever, I., (2015). An empirical exploration of
recurrent network architectures. International Conference on Machine Learning (ICML),
2342–2350.
21. Wang, F. F., Zhang, X., Liu, Y., Wei, L., & Shi, Y. (2019). LSTM-based short-term load
forecasting for building electricity consumption. In: 2019 IEEE 28th Int. Symp. Ind.
Electron (pp. 1418–1423). IEEE.
22. Glorot, X., & Bengio, Y., (2010). Understanding the difficulty of training deep feedforward
neural networks. In: Proc. Thirteenth Int. Conf. Artif. Intell. Stat. (pp. 249–256).
Author Copy
23. Zheng, H., Yuan, J., & Chen, L., (2017). Short-term load forecasting using EMD-LSTM
neural networks with an Xgboost algorithm for feature importance evaluation. Energies,
10, 1168.
24. Rafi, S. H., & Al-Masood, N., (2020). Highly efficient short-term load forecasting
scheme using long short-term memory network, In 2020 8th Int. Electr. Eng. Congr.
(pp. 1–4). IEE.
25. Dudek, G., (2020). Multilayer perceptron for short-term load forecasting: From global
to local approach. Neural Comput. Appl., 32, 3695–3707.
Non Commercial Use

Non Commercial Use

Author Copy
CHAPTER 5
Battery Plant Model Development for

BMS Application
G. N. DHANYA and K. V. ABHINAND
Author Copy
ABSTRACT
In order to develop an efficient BMS and serve as a key element of vehicle

level simulation-studies for battery pack size and range estimation calcula-
tions, circuit-based models are preferred. Equivalent Circuit Models (ECM)
have gained importance among EV designers for SOC and SOH estimation
mainly due to the simpler mathematical and numerical methods, reducing the
need for computationally intensive techniques. These models use passive and
active electrical components, such as resistors and capacitors, to resemble the
behavior of a real battery. In the proposed work, a simplified ECM for 48V
battery system having cells with two RC pair network is modeled. Parameter
estimation method is employed to generate the required parameters for the
two RC pair ECM. The dependency of the parameters on the cell SOC and
temperature is also considered.
5.1 INTRODUCTION
At present, with the development of new energy vehicles, batteries have

become one of the main power sources of EVs. Among the various types
of batteries, lithium-ion batteries are used in a diverse range of applica-
tions as they possess more advantages. But due to the fragile nature of
these batteries, advanced monitoring is essential for safe operation when
Non Commercial Use

compared to lead-acid or nickel-cadmium batteries. Thus, BMS becomes

a major requirement in order to monitor and maintain the battery state,
measure the secondary data, record the data, and control its environment and
has functions to detect and notify the user in case of irregular conditions such
as over-charging or over-discharging, over-heating and charge balancing

between cells [1]. In particular, a battery management system (BMS) is used
to ensure the safe and efficient use of batteries in electric vehicles.
In the past, a variety of models have been built to describe and simulate
lithium cells. Lumped-parameter equivalent circuit designs have acquired a
lot of popularity among EV designers for real-time battery state estimation
and power management applications mainly due to the simpler mathematical
Author Copy
and numerical methods, which reduce the need for computationally intensive
techniques [3, 5]. These models use passive and active electrical components,
such as resistors and capacitors, to resemble the behavior of a real battery
[6]. It is essential to model the battery accurately in order to predict the
battery performance more closely. One such approach is to estimate the
battery parameters using the parameterization technique [5, 7].
Most of the existing models are done using a fixed-parameter approach
[2]. But, in practice, these parameters vary with respect to state of charge,
temperature, etc. [3, 8]. The accuracy of the fixed parameter model is not
adequate, especially during the transient conditions. Hence the proposed
work aims at developing a more accurate model which takes into account
the dependency of parameters on SOC and temperature.
The objective of the proposed work is to develop a high-fidelity ECM
that will closely describe the behavior of a real battery. A 48 V battery
system having cells with two RC pair networks is modeled by considering
the dependency of the parameters on the cell SOC and temperature, for
which the parameter estimation method is employed to generate the required
parameters for the two RC pair ECM.
5.2 METHODOLOGY
5.2.1 LI-ION CELL MODELING
A study of the battery model can be carried out by building up behavioral/

phenomenological analogs using common circuit elements. The end result of
such a process will be an ‘equivalent circuit model (ECM),’ which will give
a response similar to that of a real-life cell or battery [2].
Non Commercial Use

Battery Plant Model Development for BMS Application 93
The battery is a source of voltage. So, a voltage source is the closest

component that can be used to represent a battery. This is a very oversimplified
model, as the voltage remains constant irrespective of time or current drawn
or any other parameters. It is evident from experiments carried out that when
the battery is charged, it shows a higher OCV, and when it’s discharged, it
shows a lower SOC value. Figure 5.1 shows the change of OCV with respect
to SOC. So, replacing the independent voltage source with a voltage source
that depends on SOC value will increase the fidelity of the model.
Author Copy
FIGURE 5.1 Relationship between SOC and OCV.
If we define the capacity as the total amount of charge removed from the
cell when its SOC is removed from 100% to 0%, SOC can be represented as:
SOC
= ( t ) SOC ( 0 ) + ∫η * I ( t ) (1)
C *3600
where; I(t) is the current flow into or out from the battery; SOC(t)
is determined by the previous SOC(to); C is the capacity in Ah and
is the coulombic efficiency. The terminal voltage of a battery under
a pulsed load is given in Figure 5.2. The voltage source depending
on SOC is inadequate to describe this type of behavior under similar
Non Commercial Use

load conditions. It can be observed that when under load, the battery
terminal voltage drops to a lower value. This voltage dip can be
assumed due to the presence of a resistor within the cell, called the
‘Internal resistance.’ It can be modeled as,
V0 = i (t). R0 (2)
Author Copy
FIGURE 5.2 Terminal voltage waveform of a cell for pulsed discharge current.
The phenomenon is known as linear polarization. So, in the cell model, in

addition to the voltage source, a series resistor is also connected. This model
is good enough for many elementary electronic circuit systems but cannot
be employed for sophisticated consumer electronics or EV applications. As
seen in the pulse response waveform (Figure 5.2), we also observe a dynamic
(non-instantaneous) response to a current pulse. When the cell is allowed to
rest, the terminal voltage is not jumping back to OCV in an instant, but the
process is gradual. This is due to the slow diffusion process within the cell,
and hence it’s called ‘diffusion voltages.’ This behavior can be approximated
by connecting parallel connected resistor-capacitor pairs in series. As the
number of RC pairs increases, the accuracy of the model will also increase,
trading simplicity. Considering two RC pairs are sufficient enough for most
of the applications. The voltage across the two RC pairs can be modeled as
in the equation (Figure 5.3).
Let the current flowing through the resistor and capacitor be, respectively.
Applying KCL to the circuit:
i ( t ) iR ( t ) + iC ( t )
= (3)
where; iR (t) and iC (t) are the currents flowing through the resistor and
capacitor of the RC network. We have voltage across the capacitor:
Non Commercial Use

1
Vc ( t ) = ∫ i ( t ) .dt
c (4)
C
Author Copy
FIGURE 5.3 Parallel R-C branch.
Thus,
dVc ( t )
iC ( t ) = C. (5)
dt
i.e.,
iC1 ( t ) = C1 V̀c1 (6)
Substituting in Eqn. (3) we get,
i ( t ) iR1 ( t ) + C1V̀ c1
= (7)
VR1
i (=
t) + C1 V̀ c1 (8)
R1
Therefore,
−VR1 i ( t )
1( t )
V̀ c= + (9)
R1C1 C1
Since VR1 = VC1, we can re-write the equation as:
−VC1 i ( t )
1( t )
V̀ c= + (10)
R1C1 C1
Similarly,
−VC 2 i ( t )
c 2( t )
V̀ = + (11)
R2 C2 C2
Therefore, the cell terminal voltage is determined by applying KVL to
the equivalent circuit:
Non Commercial Use

Vt = Voc − V0 − Vc1 − Vc 2 (12)

where; Vt is the cell terminal voltage; Voc is the open circuit voltage; V0 is the
drop across the internal resistor R0; Vc1 and Vc2 are assumed as the voltage
across the two RC networks, respectively.

Summing up, all the above inferences are represented in Figure 5.4 as an
equivalent circuit diagram.
Author Copy
FIGURE 5.4 Equivalent circuit model of a cell with two R-C pair branches.
5.2.2 PARAMETER ESTIMATION
For battery systems, the accurate estimation of values representing the

internal states and parameters of the cell is essential. Parameter estimation
is one such technique which is commonly used to fit an ECM to a specific
battery cell. In order to achieve a valid generalized fit between measured
and simulated results, a numerical optimization algorithm using Simulink
Parameter Estimation is used to estimate the values of the parameters [4, 5].
The parameter estimation method is employed to generate the required
parameters for the two RC pair ECM.
In this work, a two RC equivalent cell is modeled to fit the real data
of one RC model. For this, a pulse discharge test is performed. The pulse
data of terminal voltage and the input current of the equivalent cell are
considered for the pulse discharge test. The cell current and terminal voltage
is recorded at three different temperatures (i.e., at 5°C, 25°C, and 45°C) and
Non Commercial Use

loaded to the parameterization algorithm. In the parameterization procedure,

initially, the pulse data considered are loaded into the algorithm. Appropriate
settings are chosen for the optimization algorithm. The algorithm facilitates
pulse discharge sequences from 100% to 0% SOC. Then, the pulses and the
relaxation periods are identified for different SOC values by the algorithm.
Since the algorithm is automated, it estimates the parameters, sets initial
values, optimizes, and generates the parameters for a two RC network which
is then fed to a two RC ECM of a cell.
5.2.3 48 V LI-ION BATTERY PACK MODELING
Author Copy
To realize a specific voltage range on the battery pack level, cells must be
connected in series, while parallel connections increase the capacity. In this
work, the cells modeled by the equivalent cell model are connected in series
and parallel as per the voltage and capacity required. The cell rating is 3.7 V,
2.5 Ah. The battery pack is modeled such that it consists of four modules in
parallel, each module consisting of 13 cells connected in series. Thus, a total
of 52 cells are arranged as a combination of series and parallel connections
appropriately to form a complete battery pack of 48 V with 10 Ah capacity.
5.3.1 EFFECT OF OPERATING PARAMETERS
5.3.1.1 PARAMETER GENERATION
The result of the pulse discharge test is shown in Figure 5.5. The waveform is
of the pulse data of terminal voltage at three different temperatures showing
the variation of terminal voltage with respect to temperature.
FIGURE 5.5 Pulse data of terminal voltage of the equivalent cell at three different temperatures
while discharging.
Non Commercial Use

After the considered pulse data are loaded into the algorithm, the pulses
and the relaxation periods are identified for different SOC values by the
algorithm, as shown in Figure 5.6.
Author Copy
FIGURE 5.6 Identification of pulses shown for voltage response, current, and SOC.
The parameters generated after performing the parameter estimation

technique at three different temperatures (i.e., at 5°C, 25°C, and 45°C) are
as shown in Figure 5.7. It shows the results which provide the generated
parameters, that is, OCV, R0, R1, R2, C1, and C2, which are functions of SOC
and temperature.
FIGURE 5.7 Generated parameters for a two RC network.
Non Commercial Use

5.3.1.2 LI-ION CELL MODEL
The terminal voltage, current pulse, and state of charge during discharging of
a two RC pair equivalent cell are obtained as shown in Figure 5.8. The SOC
dips at every pulse of current, indicating the discharging operation.
Author Copy
FIGURE 5.8 Waveforms of a two RC pair equivalent cell model.
5.3.1.3 BATTERY PACK MODEL
The battery pack voltage for a given current is shown in Figure 5.9. The
battery is modeled for a nominal voltage of 48 V with 10 Ah capacity.
FIGURE 5.9 Battery pack waveforms.
Non Commercial Use

5.4 CONCLUSION AND FUTURE SCOPE
A detailed understanding of the behavior and characteristics of the cell and

equivalent cell modeling approach is described in this chapter. An approach
to generate parameters for a two RC pair equivalent cell model using the
Parameter estimation technique has been illustrated, and the corresponding
equivalent cell is modeled wherein the parameters are a function of both SOC
and temperature. From this, a 48 V, 10 Ah battery pack is being modeled.
Thus, an electrical equivalent battery is modeled, and the results obtained
for the cell and the battery show satisfactory characteristics with the data
researched in the literature survey. Therefore, the battery system modeled
Author Copy
on the Simulink environment can be utilized for virtual validation of BMS
control algorithms in a real-time simulation environment.
In this model, the cell degradation effects are not considered. This is one
of the areas which can be explored further to improve the fidelity of the
battery plant model.
KEYWORDS
• battery
• battery management system
• BMS
• electric vehicles
• equivalent circuit modeling
• EV
• parameter estimation
• SOC
• state of charge
REFERENCES
1. Shen, M., & Qing, G. (2019). A review of the battery management system from the
modeling efforts to its multiapplication and integration. International Journal of Energy
Research, 43(10), 5042–5075.
Non Commercial Use

2. Nemes, R., Sorina, C., Mircea, R., Horia, H., & Claudia, M., (2019). Modeling and
simulation of first-order Li-ion battery cell with experimental validation. In: 2019 8th
International Conference on Modern Power Systems (MPS) (pp. 1–6). IEEE.
3. Lu, Z., Yu, X. L., Wei, L. C., Cao, F., Zhang, L. Y., Meng, X. Z., & Jin, L. W., (2019). A
comprehensive experimental study on temperature-dependent performance of lithium-ion

battery. Applied Thermal Engineering, 158, 113800.
4. Miniguano, H., Andrés, B., Antonio, L., Pablo, Z., & Cristina, F., (2019). General
parameter identification procedure and comparative study of Li-ion battery models.
IEEE Transactions on Vehicular Technology, 69(1), 235–245.
5. Ahmed, R., Javier, G., Simona, O., Saeid, H., Robyn, J., Kevin, R., Jimi, T., & LeSage,
J., (2015). Model-based parameter identification of healthy and aged Li-ion batteries for
electric vehicle applications. SAE International Journal of Alternative Powertrains, 4(2),
233–247.
Author Copy
6. Wang, Y., Jiaqiang, T., Zhendong, S., Li, W., Ruilong, X., Mince, L., & Zonghai, C.,
(2020). A comprehensive review of battery modeling and state estimation approaches
for advanced battery management systems. Renewable and Sustainable Energy Reviews,
131, 110015.
7. Surya, S., Janamejaya, C., Shantanu, D. D., Abhay, S. J., & Ashita, V., (2020). Accurate
battery modeling based on pulse charging using MATLAB/Simulink. In: 2020 IEEE
International Conference on Power Electronics, Drives and Energy Systems (PEDES)
(pp. 1–3). IEEE.
8. Bai, B. (2020). Estimate the parameter and modeling of a battery energy storage system.
In: 2020 Chinese Control and Decision Conference (CCDC) (pp. 5444–5448). IEEE.
Non Commercial Use

Non Commercial Use

Author Copy
CHAPTER 6
Modeling of Constant Voltage Control in

Synchronous Buck and Boost Converters
Using MATLAB/Simulink for Point-of-
Load Application
Author Copy
SUMUKH SURYA1 and VINEETH PATIL2
1
Senior Engineer, Bosch Global Software Technologies Private Limited,
Bangalore, Karnataka, India
2
Department of Electrical and Electronics Engineering, Manipal Institute
of Technology, Manipal Academy of Higher Education, Manipal, Karnataka,
India
ABSTRACT
DC-DC converters play a vital role in the design of Electric Vehicle (EV)
charging systems, LED drivers and power supply units for critical loads like
micro-processors. The converters show different behaviors during ideal and
non-ideal scenarios. In the present work, synchronous buck and synchronous
boost converters operating in Continuous Conduction Mode (CCM)
considering the non-idealities were modeled using MATLAB/Simulink. The
shift in the output voltage was observed when the Equivalent Series Resistance
(ESR) values for the inductor and capacitor were varied. The mathematical
models for the converters were derived using volt-sec and amp-sec balance
equations for the ideal and non-ideal cases. It was observed that the practical
synchronous buck converter showed stability. The practical synchronous boost
converter showed instability due to the right half plane zero in its transfer
function as a result of which phase reversal occurred at high frequencies.
Non Commercial Use

6.1 INTRODUCTION
The primary goal of any converter is to electrically transform power from

one form to another. AC–AC, AC–DC, DC–DC, and DC–AC are the major
converters used in modern power electronics. Switched-mode converters

and linear regulators are more commonly used DC–DC converters. Since
operating the load at the rated load current causes consequent I2R losses in
the latter, the former is preferred. The primary role of DC-DC converters is
to supply the DC power at required regulation to the load at high efficiency.
The converters are designed such that the inductor shall never short-circuit,
and the capacitor shall never be open-circuited. The practical converters
Author Copy
show distinct behavior compared to the ideal converters in terms of duty
cycle and inductor design. Hence, the modeling of practical converters is
close to reality.
In Ref. [1], the modeling of steady state was shown using governing
converter equations. The inductor and capacitor values were derived using
the waveforms for continuous conduction mode (CCM) and discontinuous
conduction mode (DCM). In Ref. [2], the steady-state modeling considering
the inductor ESR was considered, and their effect on the output voltage was
shown. Different methods to obtain the open-loop transfer functions during
CCM and DCM operations were discussed.
In Refs. [3, 4], the mathematical modeling of non-isolated converters was
carried out and analyzed using MATLAB/Simulink. The converter chosen
did not consider any switch drops.
In Ref. [5], the modeling of isolated ideal DC-DC converters, namely
Flyback and Forward converters were performed. The open-loop and closed-
loop analyses were performed using MATLAB/Simulink. It was shown that
in the closed-loop operation, the flyback converter showed a small over-
shoot, and the time taken to reach the steady-state voltage was slower than
the forward converter.
In Ref. [6], the steady-state and the average model for ideal and non-ideal
boost converter operating in CCM and DCM operations were discussed. The
effect of output voltage (V0), a ripple in the inductor current (∆IL), and
maximum and minimum inductor current (ILMax and ILMin) in ideal and
non-ideal cases of CCM and DCM operations were discussed. The choice of
selection of components was also discussed.
In traditional converters, controlled (MOSFET) and uncontrolled switches
(diode) are used. However, in synchronous converters, the uncontrolled switch
is replaced by a controlled switch. One of the many advantages is in regard
to the conduction losses. The voltage drop provided by the diode would be
Non Commercial Use

Modeling of Constant Voltage Control 105
typically 0.65 to 0.7 V. However, the voltage drop across a controlled device
like MOSFET would be typically 0.3 to 0.35 V. In Ref. [7], a small signal
model for a practical synchronous buck converter operating under CCM for
light load conditions was discussed. It was shown that the resonant frequency
drastically increased with the increase in the load resistance, R. Simulation

study using MathCAD software and hardware implementation was performed.
The developed model accurately predicted the AC behavior of the converter.
This generic model can be used for lightly loaded synchronous converters
during the critical conduction mode.
In Ref. [8], non-linear modeling of DC–DC converters using Hammer-
stein’s Approach using black-box identification methods are presented for a
Author Copy
boost converter. Controller design involves two major steps viz: (i) For each
duty cycle value, a controller is designed to operate in a particular opera-
tive situation Gain Scheduling Technique is used; and (ii) to take parameter
variations and unmodeled dynamics into account, a robust controller is
designed assuming that the converter can be described by a set of models
instead of only one model. DC–DC converters play a major role in photo-
voltaic systems. In Ref. [9], computationally efficient DC–DC converters
for PV applications are presented. This approach shown is a model that is
mathematically formed and can be added efficiently in a large simulation
system. This model automatically detects the steady state, and lowered
computational costs are incurred.
In this work, the mathematical modeling for the synchronous buck and
boost converters was performed using governing equations. Synchronous
converters are also referred to as Point of Load (POL) converters, as they are
used as regulators in low-power circuits like a microprocessor. The developed
model was analyzed using MATLAB/Simulink. A small-signal model for
analyzing the control study was derived, and the investigation on their
stability was made. It was noted that instability was observed for constant
voltage (CV) operation in the synchronous buck converter. However, the
synchronous buck converter was stable.
6.2 MATHEMATICAL MODELING OF AN IDEAL SYNCHRONOUS

BUCK CONVERTER
Figure 6.1 shows an ideal converter with two switches S1 and S2. The switches
S1 and S2 are typically MOSFETs. The main advantage of using a MOSFET
at S2 is that the drop across it is small. Hence, the drop at the output voltage
can be minimized.
Non Commercial Use

The operation of the converter is similar to that of the conventional

boost converter except that the diode is replaced by a MOSFET (at S2).
Initially, the switch S1 is closed due to which iL gets charged. The supply
current charges the capacitor. Later, switch S2 is turned ON with S1 in OFF
condition. The inductor reverses its polarity and discharges the capacitor
through the load.
Author Copy
FIGURE 6.1 Circuit diagram of an ideal buck converter.
When S1 ON and S2 OFF,

VL = Vg – V0 (1)
V0
ic= iL − (2)
R
When S2 ON and S1 OFF,
VL = –V0 (3)
V0
ic= iL − (4)
R
Combining Eqns. (1) and (3) with (2) and (4)
diL
VL = L = (Vg − V0 ) s + (1 − s )(−V0 ) (5)
dt
dV0 V V
ic = C = (iL − 0 ) s + (1 − s )(iL − 0 ) (6)
dt R R
Eqns. (5) and (6) were solved using MATLAB/Simulink 2018a with a
proper step size and solver.
Non Commercial Use

6.3 MATHEMATICAL MODELING FOR SYNCHRONOUS BUCK

CONVERTER
Figure 6.2 shows a practical converter. Considering the non-idealities, Eqns.

(5) and (6) would be modified as Eqns. (7) and (8).

diL
VL = L =Vg D − iL D( RL − Ron 2 ) − iL ( Ron1 + Ron 2 ) − V0 (7)
dt
dV0 V V
ic = C = (iL − 0 ) s + (1 − s )(iL − 0 ) (8)
dt R R
Author Copy
where; s is the instantaneous duty cycle.
FIGURE 6.2 Circuit diagram of non-ideal synchronous buck converter.
6.4 SPECIFICATIONS OF THE CONVERTER (Table 6.1)
TABLE 6.1 The Specifications of the Synchronous Buck Converter

SL. No. Specification Value
1. Input voltage (Vg) 5V
2. Output voltage (V0) 1.8 V
3. Output current (I0) 0 to 5 A
4. Inductor (L) 1 µH
5. Inductor ESR (RL) 15 mΩ
6. MOSFET resistances (Ron1 and Ron2) 31 mΩ
7. Duty cycle (D) 0.36
8. Capacitor (C) 200 µH
9. Capacitor ESR (Resr) 0.8 mΩ
Non Commercial Use

6.5 SMALL-SIGNAL MODELING
In the present work, the transfer function for the practical converters is derived and
their characteristics using bode plot are studied using MATLAB/Simulink 2018a.
Applying volt-sec balance to Figure 6.2,

diL
V=
L L = D(Vg − iL ( RL − Ron 2 )) + (1 − D)(−iL ( Ron1 + Ron 2 ) − V0 ) (9)
dt
On Simplification,
VL= Vg D − V0 + iL ( Ron 2 ( D − 1) − DRon1 − RL ) (10)
Author Copy
Perturbing and linearizing Eqn. (10)
VL + vˆL = (Vg + vˆg )( D + dˆ ) − (V0 + vˆ0 ) + (iL + iˆL )
(11)
(− Ron 2 (1 − D − dˆ ) − ( D + dˆ ) Ron1 − RL )
On Simplification and eliminating the product of perturbed quantities,
VL + vˆL= dˆ (Vg + I L ( Ron 2 − Ron1 )) + D(vˆg − iˆL Ron1 + iˆL Ron 2 ) − iˆL ( RL + Ron 2 ) − vˆ0 (12)
where;
V0
IL = (13)
R
Applying amp-sec balance to Figure 6.2,
V0 V
iˆc D( I L −
Ic + = ) + (1 − D)( I L − 0 ) (14)
R R
vˆ
I c + iˆc = iˆc − 0 (15)
R
Equivalent circuit for Eqns. (12) and (15) are shown in Figures 6.3 and
6.4, respectively.
FIGURE 6.3 Equivalent circuit based on Eqn. (12).
Non Commercial Use

where; (1) is ‘D(vˆ − i R + iˆ R )’, ( 2 ) is ‘V + I ( R − R )’ and ( 3) is ‘vˆ ’

g L on1 L on 2 g L on 2 on1 0
Author Copy
FIGURE 6.4 Equivalent circuit based on Eqn. (23).
where; Eqn. (4) is ‘î c,’ Eqn. (5) is ‘î L’ and Eqn. (6) is ‘v̂ 0’
The supply current ig flows in the inductor when the switch S1 is closed.
Hence, it can be averaged as shown in Eqn. (24)
ig = DiL (16)
The small-signal model for Eqn. (24) can be obtained by perturbing and
linearizing it.
I g + iˆg = ( D + dˆ ) ( I L + iˆL ) (17)
= ˆ
iˆg DiˆL + dI (18)
L
The equivalent circuit for Eqn. (18) is shown in Figure 6.5.
FIGURE 6.5 Equivalent circuit.
where; Eqn. (7) is ‘v̂ g ,’ Eqn. (8) is ‘î g,’ Eqn. (9) is d̂ IL and Eqn. (10) is ‘Dî L’
Non Commercial Use

Combining the circuits shown in Figures 6.3–6.6 Canonical model is

obtained.
Author Copy
FIGURE 6.6 The canonical model for a synchronous buck converter.
where; Eqn. (12) is Vg + vˆg

vˆ
For obtaining Gvd = ˆ0 , the terms having v̂ g are set to zero. Hence, the
d
circuit reduces to Figure 6.7.
FIGURE 6.7 Reduced canonical model.
where; Eqn. (13) is v̂ g D and Eqn. (14) is sLD2 with the voltage drop as v̂ L.
When the values shown in Table 6.1,
4000( s + 6.25*106 )
Gvd = (19)
s + 3.08*104 s + 5*109
2
6.6 SMALL-SIGNAL MODELING OF SYNCHRONOUS BOOST

CONVERTER
Applying the volt-sec and amp-sec balance to Eqns. (13) and (14) considering
non-idealities in the circuit
Non Commercial Use

VL = Vg − iL ( DRon 2 + RL + Ron 2 (1 − D)) − V0 (1 − D) (20)

V0
ic =iL − DiL − (21)
R

vˆL = vˆg − dˆ ( I L Ron1 − I L Ron 2 − V0 ) − D
(22)
( Ron1 iˆL − Ron 2 iˆL − vˆ0 ) − iˆL ( RL − Ron 2 ) − vˆ0

vˆ
Author Copy
− I L dˆ + iˆL (1 − D) − 0
iˆc = (23)
R
V0
where; I L =
R(1 − D)
The equivalent circuit for Eqns. (22) and (23) are shown in Figures 6.8
and 6.9, respectively.
where; {1} is v̂ g {16} is d̂ (IL Ron1 – IL Ron2 – V0) and {17} is D(î L Ron1 – î L Ron2 – v̂ 0).
where; {18} is –IL d̂ and {19} is (1 – D)î L
Non Commercial Use

Combining Figures 6.8 and 6.9, the canonical model for the converter is
Author Copy
FIGURE 6.10 Canonical model of a synchronous boost converter.
iL dˆ
where; {20} is vˆg , {21} is , {22} is
1− D
I sL I R I R sL + RL − Ron 2
−dˆ ( I L Ron1 − I L Ron 2 − V0 + L + L L + L on 2 ) and {23} is
1− D 1− D 1− D (1 − D) 2
In order to obtain Gvd, v̂ g sources are zero (shorted) and the circuit reduces
to Figure 6.11.
− dˆ I sL sL + RL − Ron 2
where; {24} is, ( I L Ron1 − I L Ron 2 − V0 + L ), {25} is and {26}
1− D 1− D (1 − D) 2
RResr sC + R
is
RResr sC + 1 + sRC
By plugging in values shown in Table 6.1,

6.6856*105 ( s + 1.327 *105 )
Gvd = (24)
( s + 1.714*104 )( s + 7313)
Non Commercial Use

6.7 RESULTS AND CONCLUSION
Figure 6.12 shows the mathematical model for a practical synchronous buck
converter operating in CCM. The inductor current (iL) and output voltage
(V0) variations with time are observed in Figure 6.13.
Author Copy
FIGURE 6.12 Mathematical model using Simulink.
It is observed that iL initially showed a transient of about 20A and later

settled at 0.526 A. Hence, the inductor must be designed to handle a peak
current of 20 A. Similarly, V0 showed a transient of about 2.8 V and later
settled at 1.8 V. Hence, the voltage withstanding capacity of the capacitor at
the output must be close to 3 V and not 1.8 V.
The variation of V0 as a function of RL is seen in Figure 6.14. A step-change
in RL from 15 mΩ to 17 mΩ in steps of 1 mΩ was applied. It is found that
higher values of RL produced higher V0. From Figure 6.14, it was observed
that RL = 17 mΩ showed higher transients than others. It was observed that
Non Commercial Use

the output voltage and RL are directly proportional when the DC transfer
function was extracted [5].
V0 DVg + Vd (1 − D) Rsw D + Rd (1 − D) + RL + R(1 − D) 2
= * (25)
Vg Vg R(1 − D)
FIGURE 6.13 Inductor current (A) and output voltage (V) vs. time.
The variation of V0 as a function of the Resr is seen in Figure 6.15. A step-

Author Copy
change in Resr from 0.8 mΩ to 1 mΩ in steps of 0.1 m was applied. It was noted
that the higher values of Resr produced higher transient V0. From Figure 6.15,
it was observed that Resr =1 mΩ showed higher transients than others.
Figure 6.16 shows the variation of V0 with respect to control voltage (Vc)
proportional to D, varied from 0.2 to 0.9 in steps of 0.01. It is observed
that V0 showed linear variation with respect to the control voltage. As D
increased, V0 also increased.
Frequency response of Gvd as described in Eqn. (27) is shown in Figure 6.17.
The plot initially had a DC offset of about 14 dB, showed a resonance at
around 105 rad/s and later a decrease in magnitude due to the effect of second-
order poles which were complex in nature. The bode plot shows the converter
is stable in the open loop operation. However, for a DC–DC to be highly
stable the Phase Margin (PM) should be greater than 42° [2].
Non Commercial Use

Author Copy
FIGURE 6.14 V0 as a function of RL.
FIGURE 6.15 V0 as a function of Resr.
Non Commercial Use

Author Copy
FIGURE 6.16 Variation of output voltage vs. duty cycle.
FIGURE 6.17 Bode plot of Gvd in open-loop.
Non Commercial Use

Figure 6.18 shows the root locus of Gvd. It shows a real zero and complex
poles.
Author Copy
FIGURE 6.18 Root locus of Gvd in open-loop.
Figure 6.19 shows the mathematical model of a synchronous boost

converter operating in CCM. Figure 6.20 shows iL and V0 variations. It is
observed that V0 showed an initial transient of about 16A and later settled
around 2.5A at the steady state. Similarly, V0 showed an initial transient of
18 V and a steady state voltage of around 12 V.
The variation of V0 as a function of RL is seen in Figure 6.21. The latter is
varied from 71 mΩ to 91 mΩ in steps of 1 mΩ. In addition, D is varied from
0.62 to 0.65. It was found that the lowest RL showed the highest V0. As seen
from Figure 6.21, 71 mΩ showed the largest V0. This is due to the fact that D
α V0 as shown in Eqn. (33).
The variation of V0 as a function of Resr is seen in Figure 6.22. Resr varied
from 4 mΩ to 6 mΩ in steps of 1 mΩ. D is varied from 0.62 to 0.65. It was
found that the Resr showed no effect on V0. As seen from Figure 6.22, all the
values of Resr converged to yield V0 of around 14.65 V.
Non Commercial Use

Author Copy
FIGURE 6.19 Mathematical modeling of synchronous boost converter.
FIGURE 6.20 iL and V0 vs. time.
Non Commercial Use

Author Copy
FIGURE 6.21 Variation of RL as a function of V0.
FIGURE 6.22 Variation of Resr as a function of V0.
Non Commercial Use

The variation of V0 as a function of Vc is seen in Figure 6.23. Vc propor-

tional to D varied from 0.2 to 0.9 in steps of 0.01. In addition, RL is varied
from 71 mΩ to 91 mΩ. As seen from Figure 6.23, lower RL provided
higher V0.
Author Copy
FIGURE 6.23 Variation of D on V0.
Figure 6.24 shows the bode plot of Gvd. It can be observed that the system
has a negative gain margin (GM) and phase margin. Gvd initially showed a
DC offset of around 60 dB. Later, due to the effect of two poles, the system
showed a – 40 dB/dec fall at high frequency. Root locus for Gvd is shown
in Figure 6.25. It possesses a right-hand side zero, thus making an unstable
system.
The variation of load resistance, R as a function of V0 is shown in
Figure 6.26. R is varied from 6 Ω to 8 Ω in steps of 2 Ω. It was observed that
higher value of resistance results in larger value of resonance.
Non Commercial Use

Author Copy
FIGURE 6.24 Bode plot of Gvd for synchronous boost converter.
FIGURE 6.25 Root locus of Gvd.
Non Commercial Use

FIGURE 6.26 Effect of R on Gvd.
Figure 6.27 shows the physical origin of right half plane zero in the case
Author Copy
of synchronous boost converter. D was varied from 0.628 to 0.828 in steps
of 0.1, and iL, and V0 were plotted.
It is observed that V0 decreased when the duty cycle was changed due to
the load discharging the capacitor. As iL builds up to a larger value (at steady
V0
state, iL = ( ), V0 regains its value.
R(1 − D) 2
The Right Half Plane (RHP) zero causes significant issues during the
feedback design. The general form of a right half plane zero is shown in
Eqn. (34).
s
G (s) = 1 − (26)
ω0
Non Commercial Use

At the low frequency, the transfer function for Eqn. (34) would be one
or zero dB. However, at high frequency (ω0 >> ω), there would be a phase
reversal due to the negative sign.
Author Copy
FIGURE 6.27 Variation of iL and V0 as a function of Vc.
6.8 CONCLUSION
In the present work, synchronous buck and synchronous boost converters

operating in CCM considering the non-idealities are modeled using
MATLAB/Simulink. The shift in the output voltage is observed when the
ESR values for the inductor and capacitor were varied. The mathematical
models for the converters are derived using governing equations for the
ideal and non-ideal cases. It is observed that the practical synchronous buck
converter showed high stability. The practical synchronous boost converter
showed instability due to the RHP zero in its transfer function due to which
phase reversal occurred at high frequencies.
The authors declare no conflict of interest and no data/information were
taken from Bosch Global Software Technologies (BGSW).
Non Commercial Use

KEYWORDS
• CCM
• duty cycle
• instability
• mathematical modeling
• MATLAB/Simulink
• non-ideality
• small signal modeling
Author Copy
• synchronous converters
REFERENCES
1. Hart, D. W., (2011). Power Electronics. Tata McGraw-Hill Education.

2. Erickson, R. W., & Dragan, M., (2007). Fundamentals of Power Electronics. Springer
Science & Business Media.
3. Surya, S., & Arjun, M. N., (2021). Mathematical modeling of power electronic converters.
SN Computer Science, 2(4), 1–9.
4. Hinov, N. L., (2018). Mathematical modeling of transformerless DC-DC converters. In:
2018 IEEE XXVII International Scientific Conference Electronics-ET. IEEE.
5. Surya, S., & Srividya, R., (2021). Isolated converters as LED drivers. Cognitive Infor-
matics and Soft Computing (pp. 167–179). Springer, Singapore.
6. Arjun, M., & Vineeth, P., (2015). Steady state and averaged state space modelling of
non-ideal boost converter. International Journal of Power Electronics, 7(1, 2), 109–133.
7. Zhu, J. Y., (2005). Interpreting small signal behavior of the synchronous buck converter
at light load. IEEE Power Electronics Letters, 3(4), 144–147.
8. Alonge, F., et al., (2007). Nonlinear modeling of DC/DC converters using the
Hammerstein’s approach. IEEE Transactions on Power Electronics, 22(4), 1210–1221.
9. Corti, F., et al., (2020). Computationally efficient modeling of DC-DC converters for PV
applications. Energies, 13(19), 5100.
Non Commercial Use

CHAPTER 7
Determination of Open Loop Responses

of Switched DC–DC Converters Using
Various Modeling Techniques
SUMUKH SURYA
Author Copy
Senior Engineer, Bosch Global Software Technologies (BGSW), Bangalore,
Karnataka, India
ABSTRACT
In this chapter, transfer functions for open loop ideal second order converters
like Buck, Boost and Buck – Boost converters operating in Continuous
Conduction Model (CCM) are derived using different methods viz., (a) Small
signal model and (b) State Space Averaging. Using State space averaging
technique, transfer functions for duty cycle (Gvd) to output voltage and output
to input voltage (Gvg) were derived. Circuit averaging technique provided
the frequency response of Gvd. MATLAB/Simulink were used to study the
open loop behavior using the principles of volt-sec and amp-sec balance
equations. It was observed that the boost and buck-boost converters showed
a right hand side zero, leading to instability in the open loop configuration.
However, the buck converter showed stability with two poles on the left
hand side of the s-plane.
7.1 INTRODUCTION
Switch mode power supplies (SMPS) have become fashionable due to the
advent of Electric Vehicles (EVs). SMPS are considered superior to voltage
regulators in terms of noise-free output, high efficiency, and less weight.
Non Commercial Use

Some of the major applications of SMPS are (a) LED driver; and (b) battery
charging and discharging. The main components of a SMPS are: (a) switches
like MOSFET’s and diodes; (b) inductor (L); (c) capacitor (C); and (d) a
resistor (R). L and C behave as a low-pass filter, filtering out current and
voltage harmonics which are caused due to fast switching.

It is important to maintain constant load voltages and currents in terms
of critical loads. Hence, controlling the output voltage, duty cycle, and the
inductor current irrespective of the Vg changes is extremely important.
In Ref. [1], a small signal model for an ideal buck-boost converter
operating in Continuous Conduction Mode (CCM) was derived using a
canonical model. The concept of State Space Averaging (SSA) was discussed.
Author Copy
Circuit averaging (CA) for second-order converters operating in CCM and
DCM (Discontinuous Conduction Mode) was shown. However, the CA for
higher-order converters, including all the non-idealities, was not proposed.
In Ref. [2], the DC-DC converter is considered as a black box and is
modeled using generalized Power Conservative POPI networks. It is one
of the most effective methods to model the first-order power converter
properties.
DC–DC converters and their behavior have been studied for many years.
A large signal-averaged model for converters was proposed in Ref. [3].
An approach which is generalized is derived independent of the operation
(CCM/DCM). The modeling and experimental results were presented for
buck, boost, and SEPIC converters, µDCM > µCCM
In Ref. [4], a generalized procedure involving input voltage, output
voltage, and duty cycle operating in the DCM for a buck-boost converter
modeling was presented. In reality, increased conduction losses are observed
due to equivalent series resistance (ESR).
In Ref. [5], a switch and an average model were modeled and analyzed
using MATLAB/Simulink and validated using experiments. The transfer
functions Gvd (constant output voltage) and Gvg (constant output voltage)
were derived. Lower input current harmonics for SEPIC was the major
conclusion drawn.
In Ref. [6], DCM operation for a SEPIC was demonstrated using State
Space averaging technique. For frequencies below 10 kHz, a strong match
was observed. For higher frequencies, differences were observed in LTspice
when compared with MATLAB software. It was concluded that the deviation
was due to approximation.
In Ref. [7], for power factor correction fourth order converters operating
under DCM used. A single-phase rectifier was used to feed the converter. In
Non Commercial Use

Determination of Open Loop Responses 127
Ref. [8], an LTSpice-based averaged model of the converters operating in

CCM was designed using a block named CCM1.
In Ref. [9], a generalized signal flow graph model of the fourth-order
DC–DC converter topology is proposed. This technique is useful for gener-
ating unified models. Using these models, the steady-state models were
derived and verified experimentally.
In Ref. [10], the CA for the Cuk converter was carried out using a Saber
circuit simulator, and Gvd and Gvg were found out theoretically and verified
using simulation. Gvd showed complex pole conjugates, and Gvg showed a
Right Half Plane (RHP) zero. This chapter did not account for the diode drop
(Vd) and Dynamic resistance (Rd).
Author Copy
In Ref. [8], CA using LTSpice for basic converters during ideal condi-
tions was carried out. However, the simulation and analysis for non-ideal
higher-order converters was not carried out. In Ref. [12], the modeling
of switching DC-DC converters was shown using state space modeling,
including parasitic values considering switching and conduction losses.
In the present work, the open loop behavior of second-order converters
like Buck, Boost, and Buck Boost converters are studied in CCM opera-
tion using MATLAB/Simulink. Large signal and small signal models for
the converters were derived using the ‘SSA Technique.’ Later, values of R,
L, C, and duty cycle (D) were designed based on the switching frequency
(fs). These values were used to study the open-loop behavior of the ideal
converters using Root Locus and Bode techniques. Most of the papers
concentrate on a specific control tool for the analysis of transfer function.
In this work, three different control tools, namely, small signal model, SSA,
and Generalized CA technique, are applied on the same converter operating
in CCM. The software tools used are MATLAB and LTspice. It is shown
that the bode plots of Gvd obtained for the converters using two different
software tools closely match in terms of low-frequency gain, gain, and phase
margins. Hence, extensive software validation is performed, which plays a
vital role before the hardware development.
7.2 STATE SPACE MODELING
7.2.1 BUCK CONVERTER
Figures 7.1–7.3 shows an ideal buck converter operating in CCM with no

lossy elements.
Non Commercial Use

FIGURE 7.1 Schematic of the converter.
Author Copy
FIGURE 7.2 Switch closed.
FIGURE 7.3 Switch opened.
State space averaging technique was used to find the large signal and
small signal models for the converters.
The basic form of the state space equations is shown in Eqns. (1) and (2)
X̊ = AX + BU (1)
Y = CX + EU (2)
When the switch is closed,
Vg – V0 = VL (3)
V0
ic= iL − (4)
R
Non Commercial Use

When the switch is opened,

VL = –V0 (5)
V0
ic= iL − (6)
R
Reducing Eqns. (3), (4) and (5), (6) in the form of Eqns. (1) and (2)
Assuming [ diL / dt dV0 / dt ]
T
= g, X
u V= and
= Y V0
 −1 
0 L 
A1 =   (7)
Author Copy
1 1 
 C RC 
1
B1 =  L  (8)
 
0
 −1 
0 L 
A2 =   (9)
1 1 
 C RC 
0 
B2 =   (10)
0 
C1 = [ 0 1] (11)
C2 = [ 0 1] (12)
Combining Eqns. (7), (8), …, (12)
 LdiL   −1 
 dt  0 1
L   iL    
=      + L Vg 0  (13)
 CdV0  1 −1  V0   
 dt  0
 C RC 
 LdiL   −1 
 dt  0 L   iL  0  
=    + Vg 0  (14)
 CdV0  1 −1  V0  0  
 dt   C RC 
To obtain the large signal model,

A = A1D + A2 (1 – D) (15)
Non Commercial Use

B = B1D + B2 (1 – D) (16)
C = C1D + C2 (1 – D) (17)
E = E1D + E2 (1 – D) (18)
 −1 
0 D
L 
A= , B =  L 
1 −1   
 C  0
RC 
=C [=
0 1] and E [ 0]
Since, the term X̂ = 0 under steady state conditions,
Author Copy
X = –A–1BU (19)
 −1 1 
D
i L     RC
1 L  
V  = −    −1  L Vg 0 
 0  ∆  −1   
0
 C RC 
Hence, upon simplification
iL D
= (20)
Vg R
V0
=D (21)
Vg
where;
1
∆= (22)
LC
Eqns. (22) and (23) describe the small signal model for the converter.
X̂ ( s ) = [ sI − A]−1{BÛ ( s ) + dˆ[[ A1 − A2 ] X + [ B1 − B2 ]U ]} (23)
Xˆ
To find ˆ , Û(s) should be made zero, as per superposition theorem.
d
1
A1 − A2 [0]
= = & B1 − B2  L 
 
0
 Vg 1 
 ( ) * (s + )
ˆ
X 1 L RC 
=   
dˆ  ∆   Vg 
 LC 
Non Commercial Use

where;
s 1
∆= s 2 + + (24)
RC LC
Vˆ0 1 Vg
= *( ) (25)
dˆ ∆ LC
Vg 1
( ) *( s + )
iˆL (26)
= L RC
dˆ ∆
Author Copy
7.2.2 BOOST CONVERTER
Figure 7.4 depicts a schematic of boost converter operating in CCM.
FIGURE 7.4 Schematic of a circuit.
When the switch is closed (Figure 7.5):
Vg = VL (27)
V0
ic = − (28)
R
When the switch is opened (Figure 7.6).
Non Commercial Use

VL = Vg – V0 (29)
V0
Author Copy
ic= iL − (30)
R
Assuming u = Vg, X = [diL / dtdV0 / dt]T and Y = V0 & reducing it to the form
shown in Eqns. (1) and (2),
0 0 
A1 =  −1  (31)
0
 RC 
1
B1 =  L  (32)
 
0
 −1 
0 L 
A2 =   (33)
1 −1 
 C RC 
1
B2 =  L  (34)
 
0
C1 = [ 0 1] (35)
C2 = [ 0 1] (36)
To obtain the large signal model, Eqns. (15)–(18) are used.
 −(1 − D) 
 0 L 
A= 
1 − D −1 
 C RC 
Non Commercial Use

1
B = L
 
0
C = [ 0 1] &
E = [0]
Since, the term X̂ = 0 under steady state conditions, X = –A–1 BU
 −1 −D ' 
 iL   −1   RC L 
V  =    − D '  Vg 0 
Author Copy
 0  ∆  0 
 C 
Upon simplification,
(1 − D) 2
∆= (37)
LC
iL 1
= (38)
Vg RLC ∆
V0 1
= (39)
Vg 1 − D
To find the small signal model for the converter, Eqn. (23) is considered.
 1 −D ' 
s+
 iL   −1   RC L 
V  =    − D '  Vg 0  (40)
 0  ∆  −1  
 C RC 
V i 1− D
(( s + (1/ RC )) 0 ) − ( L ) *( )
iˆL L C C (41)
=
dˆ ∆
−V 1 − D siL
( 0 )( )−
Vˆ0 (42)
= L C C
ˆ
d ∆
where;
V0
iL = (43)
R(1 − D)
s (1 − D) 2
∆= s 2 + + (44)
RC LC
Non Commercial Use

7.2.3 BUCK BOOST CONVERTER
A Buck Boost converter is a type of SMPS obtained by combining boost and

buck converter. A schematic of Buck-Boost converter is shown in Figure 7.7.
Author Copy
FIGURE 7.7 Schematic of buck boost converter.
When the switch is closed (Figure 7.8).
VL = Vg – V0
V0
ic= iL −
R
When the switch is opened (Figure 7.9).
Non Commercial Use

VL – V0
V0
ic= iL −
R
Under steady state conditions, X̂ = 0

X = –A–1BU
0 0 
A1 =  −1  (45)
0
 RC 
Author Copy
 1 
0 L 
A2 =   (46)
1 −1 
 C RC 
1
B1 =  L  (47)
 
0
0 
B2  
0  (48)
 1− D 
 0 L 
A=  (49)
1 − D −1 
 C RC 
D
B=L (50)
 
0
C = [ 0 1] (51)
 − DVg 
 iL   LC   RLC 
V  =    (52)
  D(1 − D)Vg 
2
 0   (1 − D)
 
 L 
The small signal can be obtained using Eqn. (23) as:
Vg − V0 D −1 sV0
( ) *( + )
Vˆ0 L C RD ' C (53)
=
dˆ ∆
Non Commercial Use

where;
s (1 − D) 2
∆= s 2 + +
RC LC
7.3 SMALL SIGNAL MODELING
In this approach, the steps shown below are followed:

1. Governing equations for inductor and capacitor are determined
2. The quantities are perturbed viz., Vg, V0, D and iL
3. The terms are linearized and the product of terms having double hat
Author Copy
quantities is eliminated.
When the switch is closed,

VL = Vg – V0
V0
iC= iL −
R
When the switch is opened,

VL = –V0
V0
iC= iL −
R
 Step 1: Volt-sec balance equation for inductor and Amp-second
balance for capacitor.
VL = (Vg − V0 ) D − (1 − D)V0 (54)
iC = (iL − V0 / R) D + (iL − V0 / D)(1 − D) (55)
 Step 2: Perturbation:
VL + VˆL = (Vg + Vˆg − V0 − sVˆ0 )( D + dˆ ) − (V0 + Vˆ0 )((1 − D) − dˆ ) (56)
Simplifying (see Figure 7.10),

VˆL = Vg dˆ + DVˆg − Vˆ0 (57)
Non Commercial Use

FIGURE 7.10 Equivalent circuit of Eqn. (57).
Author Copy
where; {1} is V̂ gD, {2} is Vgd̂ and the current in the circuit is î L.
Similarly (see Figure 7.11),
( I L + iˆL ) − (V0 + Vˆ0 )
I c + iˆc = (58)
R
Vˆ
iˆc= iˆL − 0 (59)
R
FIGURE 7.11 Equivalent circuit of Eqn. (59).
where; {3} is î c and {4} is î L.

 Step 3: Linearizing:
V̂ L = Vgd̂ + DV̂ g – V̂ 0 (60)
Vˆ
iˆc= iˆL − 0 (61)
R
The supply current ig can be expressed as:
ig = DiL (62)
Non Commercial Use

Perturbing the above equation:

I g + iˆg = ( I L + iˆL )( D + dˆ ) (63)
Linearizing (see Figure 7.12),
iˆg iL dˆ + iˆL D
=
(64)
Author Copy
FIGURE 7.12 Equivalent circuit for Eqns. (57), (59), and (64).
where; {1} is V̂ g{3} is IL d̂ {4} î L d {X} is V̂ g d and {b} is î L.
FIGURE 7.13 Combining the three circuits.
where; {5} is Vg + d̂ .
FIGURE 7.14 Canonical model.
FIGURE 7.15 Simplifying the network.
Non Commercial Use

where; {a} is V̂ g (see Figures 7.14 and 7.15).

On simplifying the circuit,
Vˆ0 DR
= (65)
ˆ
Vg sL *(1 + RsC ) + R
7.3.2 BOOST CONVERTER
A similar analysis considering the ESR of the inductor is carried out. This
ESR tends to decrease V0 and causes increase in D to meet the desired V0. A
Author Copy
schematic of a boost converter with inductor ESR is shown in Figure 7.16.
From the circuit shown in Figure 7.17, when the switch is closed.
Closed condition (see Figure 7.18),

VL = Vg – IL RL (66)
V0
ic = − (67)
R
Non Commercial Use

Open condition,
VL = Vg – IL RL –V0 (68)
Author Copy
V0
ic= iL − (69)
R
Applying volt sec balance and perturbing,
VL = Vg – IL RL –V0 (1 – D) (70)
VL + VˆL = (Vg + Vˆg ) − ( I L + iˆL ) RL − (V0 + Vˆ0 )( D '− dˆ )
VL + VˆL = Vˆg − iˆL RL + V0 dˆ − Vˆ0 D ' (71)

Upon linearizing,
VˆL =Vˆg − iˆL RL + V0 dˆ − Vˆ0 D '
(72)
The equivalent circuit is shown in Figure 7.19.
FIGURE 7.19 Equivalent circuit for Eqn. (71).
where; {a} is V̂ g{b} is V0 d̂ and {c} is V̂ 0 D'.

Applying amp sec balance and perturbing,
−V0 V
ic
= D + (1 − D)(iL − 0 ) (73)
R R
Non Commercial Use

V0
ic = iL (1 − D) − (74)
R
Linearizing,
Vˆ
− I L dˆ + iˆL D '− 0
iˆc = (75)
R
The equivalent circuit is shown in Figure 7.20.
Author Copy
FIGURE 7.20 Equivalent circuit for Eqn. (71).
Vˆ
where; {e}– IL d̂ {f} iL D', {g} î c and current in R is 0
R
Combining the equations (Figure 7.21).
FIGURE 7.21 Combining the circuits shown in Figures 7.19 and 7.20.
Upon simplification,
Vˆ0 / ((V0 dˆ /D ')((− sL − RL ) / ( RD '2 ) + 1)= 1/ (1 + (2)(3)) (76)
where; (1) RD'2, (2) (R + sL) and (3) is (1 + RsC)
Vˆ0
The equation for ˆ can be derived from the above expression. The
Vg
design of classical controllers for achieving constant voltage (CV)/current
is essential in DC-DC converters. Design of such converters for a Pressure
Regulating Valve (PRV) is shown in Ref. [13].
Non Commercial Use

7.4 CIRCUIT AVERAGING (CA) TECHNIQUE
It is one of the simple techniques used to analyze the stability of the

converter using the switch voltages currents. This method is a generalized
approach and can be applies to any converter having two switches (Diode
and MOSFET), operating in CCM or DCM mode. The following steps are
used to obtain the switch voltages and currents. It was shown for a Cuk
converter that Gvd obtained from small signal model and CA matched [11].
The simulation can be easily performed on LTSpice simulation tool which is
open-source software.
Author Copy
7.5 CONVERTER SPECIFICATIONS (Tables 7.1–7.3)
TABLE 7.1 Specifications of Buck Converter

Specification Value
Input voltage (Vg) 18 V
Output voltage (V0) 5V
Resistance (R) 5Ω
Capacitance (C) 2.5 µF
Inductance (L) 1.8 mH
Duty cycle (D) 0.27
TABLE 7.2 Specifications of Boost Converter

Specification Value
Input voltage (Vg) 15 V
Output voltage (V0) 25 V
Resistance (R) 100 Ω
Capacitance (C) 10 µF
Inductance (L) 2 mH
Duty cycle (D) 0.4
7.5 RESULTS OF LTSPICE SIMULATION
For the simulation, CCM1 module under average .lib was used. Figure 7.22
shows the simulation model using CCM1 module as the modeling and simu-
lation is performed for CCM operation.
Non Commercial Use

TABLE 7.3 Specifications of Buck-Boost Converter

Specifications Value
Input voltage (Vg) 5V
Output voltage (V0) –18 V

Resistance (R) 9Ω
Capacitance (C) 87 µF
Inductance (L) 212.08 µH
Duty cycle (D) 0.78
FIGURE 7.22 LTSpice simulation.
Initially, a transient simulation for 1s is analyzed. The PWM is changed

Author Copy
from 0.27 to 0.5 from 0 to 3 ms. A plot of output voltage and inductor current
is shown in Figure 7.23. Figure 7.24 shows the variation of V0 under varying
D (0.2 to 0.9 in steps of 0.01). Figure 7.25 shows Gvd for a buck converter
inclusive of Inductor ESR.
Similar analysis for the Boost converter was carried out and the plots are
In Figure 7.27, variation of iL and V0 for change in the duty cycle is
shown. Figure 7.28 shows the variation of iL and V0 for change in the duty
cycle from 0.2 to 0.9 in steps of 0.01. It is observed that as D increases, iL
and V0 increase. In Figure 7.29, frequency response of Gvd for fixed R is
shown. It is to be noted that gain margin (GM) and the phase margin (PM)
are negative due to which the system is unstable.
Non Commercial Use

Author Copy
FIGURE 7.23 Variation of V0 and iL vs. time.
FIGURE 7.24 Variation of V0 and iL vs. time.
In Figure 7.30, bode plot of Gvd for varying load is shown. It was observed
that as the resistance increases, the value of resonance also increases.
Non Commercial Use

Author Copy
FIGURE 7.25 Bode plot of Gvd.
FIGURE 7.26 Circuit averaging for a boost converter.
7.6 RESULTS OF MATLAB SIMULATION
To analyze the open loop response of the converter, MATLAB was used to
Vˆ0
obtain the frequency response of ˆ
Vg
Non Commercial Use

Author Copy
FIGURE 7.27 iL and V0 vs. time.
FIGURE 7.28 iL and V0 vs. time for varying D.
5000
(1.25*105 s 2 + s + 2778) (77)
Non Commercial Use

Author Copy
FIGURE 7.29 Gvd for boost converter.
FIGURE 7.30 Bode plot for varying R.
Figures 7.31 and 7.32 show root locus and bode plot for a stable system
with two poles.
Non Commercial Use

Author Copy
Non Commercial Use

7.6.2 BOOST CONVERTER INCLUSIVE OF INDUCTOR ESR
−41667 *( s − 1.799*104 )
(78)
s 2 + 1005s + 1.8*107
As observed, the transfer function possesses a Right Half Plane (RHP) zero
due to which the system is unstable. The origin of right half plane zero is
explained in Ref. [1].
Figures 7.33 and 7.34 show the root locus and bode plot for a two pole
and one zero system. The negative GM and PM indicate the instability in the
system.
Author Copy
7.6.3 BUCK–BOOST CONVERTER INCLUSIVE OF INDUCTOR ESR
−1.0449*105 ( s − 2624)
(79)
( s 2 + 1277 s + 2.623*106 )
As observed, the transfer function possesses a RHP zero due to which the
system is unstable. Figures 7.35 and 7.36 show the root locus and bode plot
for a two pole and one zero system.
Non Commercial Use

Author Copy
7.7 FUTURE SCOPE
Modeling the dynamics of Gvd using different control techniques for non-
isolated topology is presented in this chapter. Similar analyzes for isolated
Non Commercial Use

topologies are recommended. Design of suitable controllers like P, PI, PD

and PID to stabilize the open loop system and achieve sufficient Gain and
Phase Margins is recommended. Analyzes for average current control, Gid
(ratio of perturbed duty cycle to inductor current) can be modeled, and its
characteristics can be studied for isolated and non-isolated topologies.

Author Copy
7.8 CONCLUSION
In this chapter, open loop transfer functions for ideal second order converters
operating in CCM are derived using different control techniques. Using
these techniques, transfer functions like Control to Output (Gvd) and output
to input voltage (Gvg) were derived and analyzed using MATLAB 2018a.
A simpler technique like CA provided the frequency response for Gvd
using built-in libraries in LTSpice software tool. It was observed that the
boost and buck-boost converters showed a right-hand side zero, leading to
instability in the open loop configuration. However, buck converter showed
stability as it had two poles on the LHS of the s-plane. In this work, the
open loop transfer functions for the DC-DC converters have been derived
using different methods. Amongst various methods, CA is the simplest and
Non Commercial Use

provides the frequency response with minimal mathematical analysis. Ideal

Buck converter showed high stability whereas the ideal Boost and ideal
Buck-Boost converters showed instability due to the presence of RHP zero
(Table 7.4).
TABLE 7.4 Nomenclature

SL. No. Symbol Parameter
1. Vg Input voltage (V)
2. V0 Output voltage (V)
3. iL Inductor current (A)
Author Copy
4. ic Capacitor current (A)
5. R Resistance (Ω)
6. L Inductance (H)
7. C Capacitance (F)
8. D Duty cycle
9. RL Inductor ESR (Ω)
Note: All *quantities are perturbed quantities.
KEYWORDS
• buck–boost converter
• CCM
• circuit averaging
• instability
• MATLAB/Simulink
• small signal modeling
• state space modeling
CONFLICT OF INTEREST
The author declares no conflict of interest. No data or information has been

taken from BGSW.
Non Commercial Use

REFERENCES
1. Erickson, R. W., & Dragan, M., (2007). Fundamentals of Power Electronics. Springer
Science & Business Media.
2. Singer, S., & Robert, W. E., (1992). Canonical modeling of power processing circuits
based on the POPI concept. IEEE Transactions on Power Electronics, 7(1), 37–43.
3. Canalli, V. M., Cobos, J. A., Oliver, J. A., & Uceda, J., (1996). Behavioral large signal
averaged model for DC/DC switching power converters. In: 27th Annual IEEE Power
Electronics Specialists Conference (Vol. 2, No. 5, pp. 1675–1681).
4. Cuk, S., & Middlebrook, R. D., (1979). A general uniﬁed approach to modeling switching
DC- to-DC converters in discontinuous conduction mode. Proc. IEEE PESC (pp. 36–57).
5. Bertoldi, B., et al., (2018). A non-ideal SEPIC DCM modeling for LED lighting
Author Copy
applications. In: 2018 IEEE 4th Southern Power Electronics Conference (SPEC). IEEE.
6. Eng, V., & Chanin, B., (2009). Modeling of a SEPIC converter operating in discontinuous
conduction mode. In: 2009 6th International Conference on Electrical Engineering/
Electronics, Computer, Telecommunications and Information Technology (Vol. 1). IEEE.
7. Simonetti, D. S. L., Javier, S., & Javier, U., (1997). The discontinuous conduction mode
SEPIC and Cuk power factor preregulators: Analysis and design. IEEE Transactions on
Industrial Electronics, 44(5), 630–637.
8. Chien-Min, L., & Yen-Shin, L., (2007). Averaged switch modeling of dc/dc converters
using new switch network. In: 2007 7th International Conference on Power Electronics
and Drive Systems. IEEE.
9. Veerachary, M., (2008). Analysis of fourth-order DC-DC converters: A ﬂow graph
approach. IEEE Transactions on Industrial Electronics (Vol. 55, No. 1, pp. 133–141).
10. Kathi, L., Ayachit, A., Saini, D. K., Chadha, A., & Kazimierczuk, M. K., (2018). Open-
loop small-signal modeling of Cuk DC-DC converter in CCM by circuit-averaging
technique. In: 2018 IEEE Texas Power and Energy Conference (TPEC) (pp. 1–6). IEEE.
11. Surya, S., & Sheldon, W., (2021). Modeling of average current in ideal and non-ideal
boost and synchronous boost converters. Energies, 14(16), 5158.
12. Middlebrook, R. D., & Cuk, S., (1976). A general unified approach to modeling switching
converter stages. IEEE Power Electronics Specialists Conf., 18–34.
13. Surya, S., & Singh, D. B., (2019). Comparative study of P, PI, PD and PID controllers
for operation of a pressure regulating valve in a blow-down wind tunnel. In: 2019 IEEE
International Conference on Distributed Computing, VLSI, Electrical Circuits and
Robotics (DISCOVER). IEEE.
Non Commercial Use

Non Commercial Use

Author Copy
CHAPTER 8
Evolution of Hybrid Ultracapacitors in

Solar Microgrids
J. PRADEEP KUMAR RAO1 and H. N. NAGAMANI2
1
Department of Engineering (R&D), Central Power Research Institute,
Author Copy
Bangalore, Karnataka, India
2
Additional Director (Retd.), Central Power Research Institute, Bangalore,
Karnataka, India
ABSTRACT
A lead carbon hybrid ultracapacitor (Pb–C HUC) is a hybrid energy storage

device comprising a battery-type electrode (PbO2: lead oxide electrode)
and an ultracapacitor-type electrode (AC: activated carbon electrode). The
chapter discusses the standardization of test protocol for optimizing the
performance of 12 V 2500F range Pb–C HUCs for solar microgrid applica-
tions. For solar power applications, the charge duration of the energy storage
system would be less than 5.5 hours, as the solar power availability is nearly
5.5 hours on a bright sunny day. No standard test protocol is available to date
to estimate the charge/discharge protocol of HUCs for solar power applica-
tions. A 12 V 2500F HUC is expected to have a charge capacity of 4.167
AH and energy storage capacity of 37.5 Wh. HUCs have been subjected to
12 different protocols of charge/discharge cycles at constant current (CC)
(Ich = Idis = 1 A, 2 A, and 3 A) followed by constant voltage (CV) of 13.8 V
for tcv charging time durations (tcv = 0 h, 1 h, 2 h, and 3 h) on 150 number
of commercially available HUCs. The performance optimization of HUCs
is measured in terms of certain key parameters of HUC, namely, Charge
input/output, Energy input/output, Charge efficiency, Energy efficiency, and
Capacitance offered. It is observed that 96% of the total energy delivered is
Non Commercial Use

due to the capacitance of HUC, and only 4% of energy is lost due to internal
resistance. The analysis establishes that 2A-3h charge/discharge protocol is
the standard test protocol for charging 12 V/2500F HUCs for solar microgrid
applications.
8.1 INTRODUCTION
Grid connected solar photovoltaic power generation is expected to contribute

to combating problems associated with global warming [1]. However, direct
integration of photovoltaic energy sources (ES) with the main grid has certain
Author Copy
practical issues mainly due to intermittent availability of solar power. The
development of microgrids with its own energy resources require an energy
storage system that can be used to store the electric power generated from
solar PV system and then supply it to the main grid as a regulated power
using appropriate power controlling units. Typically, the size of the energy
storage system required for developing a microgrid ranges between 1 kWh
and a few 100 kWh [2–6].
Among the various energy storage systems, electrochemical energy
storage systems, like rechargeable batteries, are considered to be the best
choice due to their flexibility in size and shape [7, 8]. However, most of
the rechargeable batteries have low cycle life, require long periods for
charging and slow discharge schedules. For example, lead-acid batteries
need constant current (CC) charging at C/10 rate followed by constant
voltage (CV) charging for a time duration of 6 h and are discharged at C/5
rate. Lithium batteries can be charged at C/3 to C/5 rate and happen to be
expensive and unsafe. These characteristics make them unsuitable for energy
storage in microgrids. By contrast, hybrid energy storage devices comprising
of a battery type electrode and an ultracapacitor type electrode have higher
cycle life and energy density intermediate between batteries and ultracapaci-
tors [9, 10]. Different types of hybrids ultracapacitor systems such as Pb–C,
Ni–C, MnO2–C, etc., are being developed [11–16]. Among them, Pb–C is an
attractive system due to its lower cost, abundance, higher recyclability, and
simpler manufacturing process.
8.2 LEAD CARBON HYBRID ULTRACAPACITOR
A Pb–C HUC, consists of a lead oxide (PbO2) cathode and an activated

carbon anode with sulfuric acid as an electrolyte. PbO2 electrode acts as the
Non Commercial Use

Evolution of Hybrid Ultracapacitors in Solar Microgrids 157
battery type electrode with the charge/discharge reactions akin to the posi-
tive plate in a typical lead-acid battery. The charge and discharge reactions
for PbO2 electrode and its thermodynamic, reversible potential in HUC can
be expressed as shown in Eqn. (1).
At cathode:
1 charge 1 1

PbSO 4 + H 2 O  
 PbO 2 + H 2SO 4 + H + e
+ -
(1)
2 discharge
2 2
A standard potential of 1.69 V is attained at cathode electrode
( E = 1.67V ) due to chemical reaction. This potential changes with the
o
cathode
concentration of the electrolyte.
Author Copy
Activated carbon anode is an electrical-double-layer-capacitor (EDLC)
type electrode. EDLCs are governed by the same physics as parallel-plate
electrolytic capacitors. As EDLCs use much thinner dielectric medium and
higher surface-area electrodes, EDLCs tend to store relatively larger charge.
The total charge stored in the double layer is proportional to the potential of
the electrode-electrolyte interface with its capacitance as the proportionality
constant, as expressed in Eqn. (2)
Qanode ∝ Canode V (2)
The charge and discharge reactions on the activated carbon anode and its
potential can be expressed as in Eqn. (3). The potential at anode is represented
by Eqn. (4)
At anode:
charge

Cs + H + + e- 
discharge
 +
 (C s / /H )
-
(3)
Qanode
E anode – E (discharge-anode) = (4)
Canode
In Eqn. (3), C –s represents the carbon atoms at the electrode surface, “//”
represents the double layer where charges are accumulated on either side.
Eanode and E(discharge-anode) refer to the potential of the electrode in its charged and
discharged states, respectively. Qanode is the charge on the carbon anode and
Canode is its capacitance in sulfuric acid electrolyte. The nature of interaction
between C –s and H+ has not yet been established and remains a subject of
further study. However, the double layer behavior of C –s //H+ can be inferred
from ac-impedance and cyclic voltammetry studies.
The net cell reaction for Pb–C HUC is expressed in Eqn. (5).
1 1 1
 PbO 2 + H 2SO 4 + (Cs / /H )
charge

PbSO 4 + H 2 O + Cs   - +
(5)
2 discharge
2 2
Non Commercial Use

The cell voltage is expressed as Eqn. (6):

Qanode
E cell = E cathode – (6)
Canode
Accordingly, the cell voltage (Ecell) for Pb–C HUC depends on the anode
capacitance (Canode) and total charge (Qanode) on the carbon anode in sulfuric
acid electrolyte. Typically, the open-circuit voltage (OCV) of the Pb–C cell
is about 2 V. When six of Pb–C cells are connected in series, a 12 V Pb–C
HUC device is realized.
Pb–C HUCs have peculiar charge and discharge characteristics due to
the asymmetric nature of energy storage mechanism associated with the
Author Copy
PbO2 cathode and activated carbon anode. PbO2 electrode involves faradaic
reaction while the carbon electrode involves non-faradaic reactions during
their charge and discharge cycles. Hence, the electrochemical characteristics
of these two electrodes are different and are governed by different energy
storage principles. It is reported that [17] Pb–C hybrid device requires
more than one step charging to achieve State of Charge (SoC) of 100%.
SoC is estimated from depth of discharge (DoD) of energy storage system as
expressed in Eqn. (7)
%SoC = (100 – DoD)% (7)
Conway et al. reported that the Pb–C asymmetric cell is charged at CC to
cut-off voltage of 2.3 V and subsequently the charging current is gradually
decreased stepwise to achieve the required SoC [17]. Similarly, Andrew
Burke et al. have reported that a PbO2–C HUC charged in two steps with CC
to a cut-off voltage 2.2 V and later at CV at 2.2 V to fully charge the PbO2–C
cell [18]. These studies establish that Pb–C HUCs require more than one step
charging.
In Refs. [19, 20], the characteristics of a 12 V/kF range Pb–C HUCs have
been reported, which are charged by single step CC to a cut-off voltage of
13.8 V. Although these Pb–C HUCs with substrate-integrated PbO2 electrodes
and activated carbon double layer electrodes can be charged quickly by a
single CC step, their capacitance is found to decline during repeated charge
and discharge cycles suggesting that these HUCs need two-step charging,
first with CC followed by a CV charging.
Performance optimization of 12V 2500F range Pb – C HUCs with stan-
dard test procedure is not established for solar microgrid application till date;
hence this work is undertaken where HUCs are optimized in terms of charge
input/output, Energy input/output, charge efficiency, Energy efficiency
and Capacitance offered for various constant current and constant voltage
Non Commercial Use

protocols. These parameters are determined from the voltage current (v-i)
charge/discharge curves of HUC. Typical v-i characteristics of HUC for one
cycle is shown in Figure 8.1.
Author Copy
FIGURE 8.1 A single charge/discharge cycle of HUC module with constant current (CC),
constant voltage (CV) charging and constant current (Cdis) discharging.
From the variables denoted in Figure 8.1, the mathematical representa-

tion of key parameters from the v-i curves is explained as follows:
During charging, charge taken by HUC (AHinput) in constant current
constant voltage (CCCV) mode is mathematically represented by Eqn. (8)
t3 t3
AH input = ∫ i HUC (t)dt = Ich *(t 2 - t1 ) + ∫ i (t)dt (8)
t = t1 t = t 2 HUC
Non Commercial Use

During discharging, charge delivered by HUC (AHoutput) in CC mode is

mathematically represented by Eqn. (9)
t4
AH output = ∫ i (t)dt = Idis * (t 4 – t 3 ) (9)
t = t 3 HUC
During charging, input energy (Einput) taken by HUC is mathematically

expressed by Eqn. (10)
t3
E input = ∫ v HUC (t)i HUC (t)dt (10)
t = t1
Similarly, during discharging, energy output (Eoutput) given by HUC is

mathematically expressed by Eqn. (11)
Author Copy
t4
E output = ∫ v HUC (t)i HUC (t)dt (11)
t = t3
where; iHUC is the HUC current.

HUC current will be constant (Ich) in CC mode and during CV mode it
decays exponentially to leakage current.
Energy efficiency and charge efficiencies are calculated from Eqns. (12)
and (13).
Idis *(t 4 – t 3 )
ηc =
Ich *(t 2 – t1 ) + ∫
t3
i (t)dt (12)
t = t 2 HUC
t4
ηe =
∫ t = t3
v HUC (t)i HUC (t)dt
t3 (13)
∫ t = t1
v HUC (t)i HUC (t)dt
Voltage discharge curves are drawn separately in OriginTM software to

determine the slope of the linear region of voltage discharge curve using
least square linear curve fitting technique. Capacitance CHUC offered by
HUC is estimated from the slope of voltage discharge curve of HUC at Idis as
calculated by Eqn. (14)
Idis
CHUC = – (14)
( slope )
In order to estimate the capacitance offered by the HUC, the following
process is adopted during discharging the HUC. A HUC is generally
considered as a healthy/conditioned one, if the slope of the voltage
discharge curve is linear and the slope doesn’t change with number of
charge/discharge cycles. For the present study, the tolerance allowed in
slope is 10%.
Non Commercial Use

A fully charged HUC retains an open circuit voltage of 1.15 times the
rated voltage and it could be discharged at various load currents. If a healthy
HUC is loaded at load current Iload, the voltage of HUC linearly drops from
Vo (output voltage) at t1 to Vuv (undervoltage) at t2, as shown in Figure. 8.2.
FIGURE 8.2 Theoretical representation of voltage discharge curve of HUC.

Author Copy
Capacitance of HUC is calculated by Eqn. (15)
∆t
CHUC = I load * (15)
∆v
where; Δv = Vo – Vuv and Δt = t2 – t1.
Energy output plays an important role in energy storage systems for power
applications. In view of this, an analysis has been carried out to estimate the
percentage of pure capacitance and internal resistance in HUC in terms of
energy output using the voltage discharge curve.
A typical voltage discharge curve of HUC as shown in Figure 8.2 is
redrawn as in Figure 8.3. As it is seen from Figure 8.3, the complete voltage
discharge curve can be divided into two regions, a nonlinear region from t0
to t1 and a linear region from t1 to t2.
Non Commercial Use

Applying a CC load onto the HUC, a sudden dip in the voltage is restricted
by the internal capacitance of the HUC from its basic physics.
Hence this sudden nonlinear drop of voltage discharge curve from 1.15
times the rated voltage (1.15*Vrated) to V2, could be considered as energy
consumed by internal resistance of HUC. This part of voltage discharge

curve is non-linear represents the ESR of HUCs. Linear voltage drop from
V2 to V1 represents the energy delivered to CC load from pure capacitance
of the HUC.
Author Copy
FIGURE 8.3 A typical voltage discharge curve of HUC representing linear and non-linear
region.
The total energy output delivered by HUC is represented in Eqn. (11).

Applying Eqn. (14) between voltages V1 and V2, capacitance of HUC is
Non Commercial Use

determined by rewriting Eqn. (15) as Eqn. (16) and the corresponding energy
output by Eqn. (17)
I dis
CHUC =
 V −V  (16)
− 1 2 
 t1 − t2 
Elinear _ output 0.5* CHUC * (V12 − V22 )
= (17)
The difference between total energy output (under the entire discharge
curve) and energy output due to capacitance (under linear region) represents
the energy loss due to internal resistance (Eres) of HUC, as given by Eqn. (18).
Author Copy
Eres Eoutput − Elinear _ output
= (18)
Implementing constant current constant voltage charge/discharge profiles
on HUC, we can calculate and make a lookup table of performance parameters
of HUC as explained in equations (8) – (18). The input parameters of CCCV
are chosen such that an optimal performance of charge/ discharging the
energy storage unit is seen in 5.5 hours.
The optimal charging protocol of HUCs is estimated through v-i
characteristics, for achieving a better tradeoff between capacitance, charge,
and energy outputs with better efficiencies.
8.3 EXPERIMENTAL
8.3.1 DESCRIPTION OF COMMERCIAL GRADE 12 V/2500F PB–C HUCS
The development of Pb–C HUCs with substrate-integrated PbO2 positive

electrode and activated-carbon-based double layer capacitor negative elec-
trode is described elsewhere [21, 23]. The present study has been carried
out on approximately, 150 numbers of commercially available 12 V/2500F
HUCs. These HUC modules consist of six cells connected in series with each
cell comprising 9 positive electrodes and 8 negative electrodes. Lead metal
sheets of 0.3 mm were employed as substrate for making substrate integrated
Pb/PbO2 electrodes. Flexible graphite sheets of 0.75 mm thickness are coated
with activated carbon mixed with Polyvinylidene fluoride (PVDF) binder in
dimethylformamide (DMF) solvent.
A 12 V 2500F HUC is expected to have a charge capacity of 4.167 AH
and energy storage capacity of 37.5 Wh.
Non Commercial Use

8.3.2 PRECONDITIONING OF PB–C HUCS
The fresh HUC needs to be pre-conditioned before subjecting them for

characterization [24, 25]. For preconditioning, HUCs are charged in a CCCV
mode of 0.5 A, 13.8 V for 48 h. At the end of 48 h, HUCs having open circuit
potential (OCP) less than 12 V were considered as faulty and discarded. HUCs
showing OCP as 12 V were considered for further charge/discharge protocols.
8.3.3 CHARGE/DISCHARGE PROTOCOL
Author Copy
HUC charge/discharge protocol is represented as “Ich/IdisA_tcvh” with respect
to Figure 8.1. In the representation, Ich/Idis corresponds to charging/discharging
currents and tcv is time duration of CV mode (CV mode) charging. For the
sake of simplicity, charging current (Ich) and discharging current (Idis) are
taken same.
Each charge/discharge cycle of 12 V, 2500F HUC has three modes of
operation:
• Constant current charging of Ich till HUC reaches 13.8 V.
• Constant voltage charging of 13.8 V for time duration of “tcv;” and
• Constant current discharge of Idis till 6 V.
HUCs have been subjected to 12 different protocols of charge/discharge
cycles at CC (Ich = Idis = 1 A, 2 A, and 3 A) followed by CV of 13.8 V for
tcv charging time durations (tcv = 0 h, 1 h, 2 h, and 3 h). The HUCs were
discharged till the cut-off voltage of 6 V with rest of 5 minutes between each
charge and discharge step. The HUCs were subjected to 5 to 10 cycles of
each charge/discharge protocol.
The key parameters namely charge output, energy output, and capaci-
tance are determined from charge/discharge curves [26–28]. An optimal test
protocol has been determined from comparing the estimated values of key
parameters from different test protocols [29–31].
8.4 RESULTS
Voltage discharge characteristics of two healthy 12 V/2500F HUC samples

(HUC25F15L070046) and (HUC25F15L90034) accepted for standardiza-
tion of test protocol are shown in Figures 8.4 and 8.5, respectively.
Non Commercial Use

From Figure 8.4, it is seen that the voltage discharge curve of HUC
sample, HUC25F15L090034 was a steady voltage discharge curve whose
capacitance is calculated as 2.7 kF on constant current (CC) load of 5 A.
From Figure 8.5, it is seen that the voltage discharge curve of HUC sample,
HUC25F15L070046 was a steady voltage discharge curve whose capaci-

tance is calculated as 2.2 kF on CC of 5 A. With a tolerance band of 12%,
these HUC samples are accepted. Further these HUC samples are used in
standardizing the charge discharge procedure of HUC to deliver optimal
performance.
Author Copy
FIGURE 8.4 Voltage discharge characteristics of HUC25F15L090034 for three cycles of

5A constant current discharge.
Figures 8.6 and 8.7 represents the voltage discharge curves of a faulty 12
V/2500F range HUCs. From Figure 8.6, It is seen that the voltage discharge
curve of HUC sample, HUC25F15L090038 show a sudden dip of voltage
from 13.8 V to 3–5 V on constant load of 5A for seven cycles. This behavior
Non Commercial Use

represents the HUC unable to deliver the energy on load. From Figure 8.7,
it is seen that the voltage discharge curves of an another HUC sample,
HUC25F15L070042 show a droop effect on 2A CC load. This behavior
represents the HUC is unable to deliver the steady energy requirements on
the CC loads. Hence, the HUC which show inconsistent behavior in voltage
discharge characteristics are considered as the faulty capacitors.
Author Copy
FIGURE 8.5 Voltage discharge characteristics of HUC25F15L070046 for three cycles of
5A constant current discharge.
Out of 150 HUC samples the HUCs whose performance curves are
similar to HUC2515FL090038 or HUC25F15L070042 are considered as
faulty HUCs and are discarded in development of standard test proce-
dure, i.e., charge/discharge procedure of HUCs. Manufacturing processes
and purity of electrodes, electrolyte is some of the reasons for failure of
HUCs.
The voltage current (v–i) characteristics of HUC were obtained for
charge/discharge currents of 1 A, 2 A, and 3A and CV of 13.8 V for tcv of 0 h,
1 h, 2 h and 3 h. V-I characteristic curves are shown for 0 h, 1 h, 2 h, and 3 h
tcv of 1 A, 2 A, and 3A currents in Figures 8.8–8.10, respectively.
Non Commercial Use

Author Copy
FIGURE 8.6 Voltage discharge characteristics of HUC25F15L090038 with zero charge
retention for seven cycles of 5A constant current discharge.
FIGURE 8.7 Voltage discharge characteristics of HUC25F15L070042 with interim voltage

dips for nine cycles of 2A constant current discharge.
Non Commercial Use

Author Copy
FIGURE 8.8 V-I characteristics of HUC for 1A_tcvh protocol (a) 1A_0h; (b) 1A_1h; (c)
1A_2h; and (d) 1A_3h.
2A_2h; and (d) 2A_3h.
Non Commercial Use

Author Copy
3A_2h; and (d) 3A_3h.
The key parameters, charge input/output, energy input/output, charge

efficiency, energy efficiency and capacitance offered by HUC are determined
through V-I curves and Eqns. (8)–(14). The estimated values of key param-
eters for 12 charge/discharge protocols are tabulated as shown in Table 8.1.
For the sake of brevity, the results reported are the average of five charge/
discharge cycles in each charge/discharge protocol.
The percentage of energy output delivered to CC load and percentage of
energy lost in the internal resistance of the HUC are calculated and tabulated
in Table 8.2 for the 12 charge/discharge protocols of the 12V/2500F HUCs.
8.5 DISCUSSION
For optimizing HUC charge/discharge protocol, one of the scopes of present

work is to find out, whether HUC is a single step or two step charging devices.
Single step charge/discharge cycle will have only CC charging and CC
discharging. This process does not have a CV charging followed by CC
charging. Hence tcv in single step charging will be zero. Figures 8.8(a)–8.10(a)
show single step charge/discharge protocol of HUC at 1 A, 2 A, and 3 A,
Non Commercial Use

respectively. Table 8.1 shows that the capacitance delivered is only 50% of
the rated capacitance and hence single step charging of HUCs is discarded
from optimal protocol.
TABLE 8.1 Key Parameters of HUC Estimated for Different Charge/Discharge Protocols
SL. Protocol Charge [Ah] Energy [Wh] Efficiency [%] Cap [F]
No. Input Output Input Output Charge Energy
1. 1A_0h 2.308 1.832 27.46 17.565 79.382 63.966 1252.78
2. 1A_1h 4.733 3.576 57.356 32.283 75.555 56.285 2223.65
3. 1A_2h 5.888 3.827 72.562 34.506 64.987 47.554 2305.97
Author Copy
4. 1A_3h 6.346 4.039 85.338 34.322 63.646 40.219 2494.3
5. 2A_0h 2.258 1.993 27.52 16.304 88.264 59.244 1739.13
6. 2A_1h 3.715 3.072 47.52 25.62 82.692 53.914 2638.56
7. 2A_2h 4.383 3.248 56.61 30.56 74.104 53.983 2751.72
8. 2A_3h 4.555 3.961 57.84 35.71 86.959 61.739 2873.09
9. 3A_0h 1.813 1.676 21.03 15.12 92.443 71.897 1040.16
10. 3A_1h 2.539 2.021 33.03 16.52 79.598 50.015 1710.88
11. 3A_2h 3.325 2.318 43.58 19.55 69.714 44.86 2079.65
12. 3A_3h 4.141 2.579 54.6 20.93 62.28 38.333 2320.24
TABLE 8.2 Percentage of Energy Output Delivered and Energy Output Lost for 12
12V/2500F HUC Charge Discharge Protocols
SL. Charge/ CHUC V2 [V] V1 [V] t1 [h] t2 [h] Percentage Percentage of
No. Discharge [kF] of Energy Energy Loss
Protocol Output Due to Due to Internal
Capacitance Resistance
1. 1A_0h 1.53 11.91 8.17 0.031 1.590 95.66 4.34
2. 1A_1h 1.84 11.77 7.51 0.031 2.177 95.22 4.78
3. 1A_2h 2.32 11.81 6.17 0.031 3.635 97.94 2.06
4. 1A_3h 2.50 11.30 5.55 0.031 3.993 97.90 2.10
5. 2A_0h 1.74 10.14 6.10 0.031 0.976 97.26 2.74
6. 2A_1h 2.64 10.32 6.15 0.031 1.529 98.29 1.71
7. 2A_2h 2.77 10.18 6.00 0.031 1.608 93.72 6.28
8. 2A_3h 2.99 11.25 7.22 0.031 1.674 97.07 2.93
9. 3A_0h 1.04 11.62 6.07 0.031 0.534 93.95 6.05
10. 3A_1h 1.71 10.13 6.10 0.031 0.638 93.99 6.01
11. 3A_2h 2.08 10.07 6.12 0.031 0.761 94.51 5.49
12. 3A_3h 2.46 10.10 6.16 0.031 0.897 96.55 3.45
Non Commercial Use

Two step charge/discharge cycle will have CCCV charging followed

by CC discharging. Figures 8.8(b)–(d)–8.10(b)–(d) show two step charge/
discharge protocol with tcv as 1 h, 2 h, and 3 h, respectively. Table 8.1 shows
that the capacitance delivered by HUCs is closer to the rated capacitance
indicating that, HUCs are two-step charging devices.

Based on the availability of solar power for 5.5 h on a bright sunny day,
the following are considered as the deciding criteria for optimizing the stan-
dard protocol:
• Total charging time (tch) should be less than 5.5 h.
• Capacitance deliverance (C) should be as close as possible to the rated
Author Copy
capacitance (i.e., 2500F).
• Charge capacity of 4.167 AH and energy storage capacity of 37.5 Wh
should be achieved.
From Figures 8.8(b)–(d), it could be observed that charging time (tchcc) is
nearly 4 h at 1 A. From Figures 8.9(b)–(d), and 8.10(b)–(d); tchcc is nearly 1.5 h
and 0.75 h at 2 A, and 3 A, respectively. The total charging time (tch = tchcc + tcv)
of HUC at 1A, 2A, and 3A is summarized in Table 8.3.
TABLE 8.3 Total Charging Time of HUC at Constant Current Charge of 1 A, 2 A, and 3A
and Constant Voltage Charge of 13.8 V for tcv of 1 h, 2 h, and 3 h
SL. Ich tch/Capacitance
No. tcv = 1 h tcv = 2 h tcv = 3 h
1. 1A 5.0 h/2.2 kF 6.0 h/2.3 kF 7.0 h/2.5 kF
2. 2A 2.5 h/2.6 kF 3.5 h/2.8 kF 4.5 h/2.9 kF
3. 3A 1.75 h/1.7 kF 2.75 h/2.1 kF 3.75 h/2.3 kF
As the charging times of HUC for 1A_2h and 1A_3h protocols are
greater than 5.5 h, these protocols are discarded as standard protocols of
HUC charge/discharge for solar power applications. Remaining protocols
have been further analyzed for determination of standard test protocol.
From Figures 8.8–8.10 and Tables 8.1 and 8.2, following are the impor-
tant observations:
• Capacitance deliverance of HUC improves with total charging time (tch).
• For each charging current (Ich), the energy output has increased with tch.
• Energy output has marginally decreased with increase in Ich.
• Charge efficiency and energy efficiency has no correlation with tch and
Ich. However, for two step charging, charge efficiency and energy effi-
ciency lie in the range of (62% to 92%) and (38% to 72%), respectively.
Non Commercial Use

Charging at 1A is not meeting the requirement for the following reasons:

• Charging time is more than 5.5 h (except for tcv = 1 h).
• Capacitance deliverance is less than 2.5 kF.
Following are the observations during charge/discharge at 3A:

• HUCs are getting charged faster with tch ranging from 1.75 h to 3.75 h.
• Capacitance deliverance is less than 2.5 kF.
• Charge output and energy output are 2.6 AH and 21 Wh which are only
50% of the rated values.
Charging at 3A has also been discarded as the estimated values of key
Author Copy
parameters are not meeting the rated values of HUCs, in spite of HUCs
getting charged faster.
Regarding 2A charging, following are the observations:
• HUCs charge with tch ranging from 2.5 h to 4.5 h.
• Capacitance deliverance is in the range of 2.6 kF to 2.9 kF range,
which is better than the rated capacitance.
• Charge output and energy output are 3.0 AH, 3.2 AH, 4.0 AH and 26
Wh, 31 Wh, 36 Wh for tcv of 1 h, 2 h, and 3 h, respectively.
• Charge efficiency and energy efficiencies are 54%, 54%, 62% and
83%, 74% and 87%.
• The maximum charge output, energy output and capacitance of 4.0 AH,
36 Wh, and 2.9 kF were delivered in 2A_3h charge/discharge protocol.
As can be seen from Table 8.2, the capacitance deliverance increases with
tcv. In order to avoid, overcharging of HUCs, tcv has been limited to 3 h in the
present study.
12V/2500F HUCs are 4 AH energy storage systems. To operate HUC
safely and ensure battery electrode of HUC is not sulfated, the C-rating of
12V/2500F range HUC is limited to 0.75C, i.e., 3A.
The above analysis establishes that 2A_3h charge/discharge protocol is
the standard test protocol for charging 12V/2500F HUCs for solar microgrid
applications.
8.5.1 STANDARD PROTOCOL FOR OPTIMAL PERFORMANCE OF HUCS
The experimental work reported, and the analysis made establish that
2A_3h charge/discharge protocol is the standard test protocol for charging
12V/2500F HUCs for solar microgrid applications. The standard procedure
Non Commercial Use

for one cycle of charge/discharge of 12V/2500F HUC module with “2A_3h”

charge/discharge protocol consists of following sequence:
 Step 1: Rest the HUC module for 5 min.
 Step 2: Charge HUC at 2A till the voltage across HUC reaches 13.8 V
 Step 3: Charge HUC with constant voltage of 13.8 V for 3 h
 Step 4: Rest the HUC module for 5 min
 Step 5: Discharge HUC at 2A till the voltage across HUC reaches 6 V.
 Step 6: Repeat Step 1 to step 5 for ‘N’ number of cycles.
Number of cycles (N) required to estimate the capacitance deliverance
of HUC is of specific interest. As of now, there is no standard protocol for
Author Copy
deciding the number of cycles. Based on the experimental study carried out
on approximately 150 numbers of 12 V, 2500 kF HUCs, N is chosen between
5 and 10.
8.5.2 ANALYSIS OF ENERGY OUTPUT
Comparing the energy output delivered in the linear and nonlinear regions
from Table 8.2, it could be seen that 96% of total energy delivered is due to
capacitance of HUC and only 4% of energy is lost due to internal resistance.
Therefore, HUC could be considered predominantly as a capacitor having
substantially low internal resistance.
8.6 CONCLUSION
The present research work has been carried out on approximately 150
numbers of 12 V Pb–C HUCs, having rated capacitance of 2500F, rated
charge capacity of 4.167 AH and rated energy storage capacity of 37.5 Wh.
Performance optimization of HUCs has been attempted employing different
charge/discharge protocols at appropriate voltage, current and time dura-
tions. Key parameters, namely, charge output, energy output and capacitance
have been estimated and analyzed to arrive at a standard test protocol for
HUCs as energy storage devices for solar power applications. Availability of
solar power has been considered as 5 ½ h on a bright sunny day.
Following are the major conclusions of the study conducted on 12V
2500F Pb–C HUC modules:
1. Single step charging is not applicable as it yields only 50% of the
rated capacitance. Whereas the capacitance deliverance during two
Non Commercial Use

step charge/discharge cycle is closer to the rated capacitance indi-

cating that HUCs are two-step charging devices.
2. For the nine different two step charge/discharge protocols at CC of
1A, 2A, and 3A followed by CV of 13.8 V, the total charging time is
nearly 5 h, 3.5 h, and 3.75 h, respectively.

3. Capacitance deliverance of HUC marginally improves with total
charging time. For example, an increase from 2.6 kF @ 2A-1h
charging to 2.9 kF @ 2A-3h has been seen. Similarly, energy output
improves from 26 Wh to 36 Wh with charging time of 1 h to 3 h.
4. Charge efficiency and energy efficiency has no correlation with
charging time and charging current. However, charge efficiency and
Author Copy
energy efficiency lie in the range of (62% to 92%) and (38% to 72%),
respectively.
5. To operate HUC safely and to ensure that battery electrode of HUC is
not sulfated, C-rating is limited to 0.75C, i.e., 3A. In order to avoid,
overcharging of HUCs, charging time is limited to 3 h in the present
study.
6. HUC can be treated predominantly as a capacitor, as the energy
output delivered due to capacitance is 96% of total energy and only
4% of energy is lost due to internal resistance.
7. Charge/discharge protocol of 2A_3h has yielded values of capaci-
tance, charge output and energy output of 2.9 kF, 4.0 AH and 36 Wh,
which are close to the rated values of HUC.
8. Also, the total charging time for 2A_3h charge/discharge protocol is
4.5 h. Based on the assumption of availability of 5 ½ h of sunlight in
one single day would charge the HUC fully, 2A_3h charge/discharge
protocol has therefore been considered as the standard test protocol
for charging 12V 2500F HUCs for solar power applications.
ACKNOWLEDGMENT
The authors acknowledge the Central Power Research Institute (CPRI),

Bangalore, India for the encouragement and support extended in forming
this publication. The authors would also like to thank the officials of the
Solid State and Structural Chemistry unit of Indian Institute of Science,
Bangalore, Capacitors Division and R&D Management of CPRI for their
help in carrying out the research work.
Non Commercial Use

KEYWORDS
• discharge protocol
• hybrid ultracapacitors
• lead carbon hybrid ultracapacitor
• optimal performance
• performance characterization
• solar microgrid
• standard test protocol
Author Copy
REFERENCES
1. Chu, S., & Majumdar, A., (2012). Opportunities and challenges for a sustainable energy
future. Nature, 488, 294–303.
2. Hittinger, E., Wiley, T., Kluza, J., & Whitacre, J., (2015). Evaluating the value of batteries
in microgrid electricity systems using an improved energy systems model. Energy
Conversion and Management, 89, 458–472.
3. Abu-Sharkh, S., Arnold, R. J., Kohler, J., Li, R., Markvart, T., Ross, J. N., Steemers, K., et
al., (2006). Can microgrids make a major contribution to UK energy supply? Renewable
and Sustainable Energy Sources, 10(2), 78–127.
4. Tan, X., Li, Q., & Wang, H., (2013). Advances and trends of energy storage technology in
microgrid. International Journal of Electrical Power and Energy Systems, 44(1), 179–191.
5. Pradeep, K. R. J., & Nagamani, H. N., (2016). Modeling of hybrid ultra-capacitors for
power electronic applications using PSPICE/ORCADTM. Power Research–A Journal of
CPRI, 12, 3, 547–554.
6. Ravikumar, M. K., Rathod, S., Jaiswal, N., Patil, S., & Shukla, A. K., (2017). The
renaissance in redox flow batteries. J. Solid State Electrochemistry, 21(9), 2467–2488.
7. Yang, Z., Zhang, J., Meyer, M. C. W. K., Lu, X., Choi, D., Lemmon, J. P., & Liu, J.,
(2011). Electrochemical energy storage for green grid. Chem. Rev., 111, 3577–3613.
8. Dunn, B., Kamath, H., & Tarascon, J. M., (2011). Electrical energy storage for the
grid–A battery of choices. Science, 334, 928–935.
9. Dubal, D. P., Ayyad, O., Ruiz, V., & Gòmez–Romero, P., (2015). Hybrid energy storage–the
merging of battery and supercapacitor chemistries. Chem. Soc. Rev., 44(7), 1777–1790.
10. Bélanger, D., Brousse, T., & Long, J. W., (2008). Manganese oxides–battery materials
make the leap to electrochemical capacitors. The Electrochemical Society Interface,
49–52.
11. Zuo, W., Li, R., Zhou, C., Li, Y., Xia, J., & Jingpin, L., (2017). Battery-supercapacitor
hybrid devices: Recent progress and future prospects. Adv. Sci., 4. Open Access. https://
doi.org/10.1002/advs.201600539.
Non Commercial Use

12. Yu, N., Gao, L., Zhao, S., & Wang, Z., (2009). Electrodeposited PbO2 thin film as positive
electrode in PbO2/AC hybrid capacitor. Electrochimica. Acta, 54(14), 3835–3841.
13. Banerjee, A., Ravikumar, M. K., Jalajakshi, A., Suresh, K. P., Gaffoor, S. A., & Shukla, A.
K., (2012). Substrate integrated lead-carbon hybrid ultracapacitor with flooded, absorbent
glass mat and silica-gel electrolyte configurations. Journal of Chemical Sciences, 124(4),
747–762.
14. Wang, J., Yang, Y., Huang, Z. H., & Kang, F., (2013). A high-performance asymmetric
supercapacitor based on carbon and carbon–MnO2 nanofiber electrodes. Carbon, 61,
190–199.
15. Zhao, C., Ren, F., Xue, X., Zheng, W., Wang, X., & Chang, L., (2016). A high-performance
asymmetric supercapacitor based on Co(OH)2/graphene and activated carbon electrodes.
Journal of Electroanalytical Chemistry, 782, 98–102.
16. Lang, J. W., Kong, L. B., Liu, M., Luo, Y., & Kang, L., (2010). Asymmetric
Author Copy
supercapacitors based on stabilized α-Ni(OH)2 and activated carbon. Journal of Solid
State Electrochemistry, 14(8), 1533–1539.
17. Pell, W. G., & Conway, B. E., (2004). Peculiarities and requirements of asymmetric
capacitor devices based on combination of capacitor and battery-type electrodes. Journal
of Power Sources, 136(2), 334–345.
18. Burke, A., & Miller, M., (2010). Testing of electrochemical capacitors: Capacitance,
resistance, energy density, and power capability. Electrochimica. Acta, 55(25), 7538–7548.
19. Banerjee, A., Ravikumar, M. K., Jalajakshi, A., Gaffoor, S. A., & Shukla, A. K., (2012). A
12 V substrate-integrated PbO2-activated carbon asymmetric hybrid ultracapacitor with
silica-gel-based inorganic-polymer electrolyte. ECS Transactions, 41(13), 101–113.
20. Shukla, A. K., Banerjee, A., Jalajakshi, A., & Ravikumar, M. K., (2013). 12 V / kilo-
farad range lead-carbon hybrid ultracapacitors and their envisaged applications. ECS
Transactions, 50(31), 367–376.
21. Shukla, A. K., Ravikumar, M. K., & Gaffoor, S. A., (2015). Energy Storage Device and
Method Thereof. US Patent US 9,147,529.
22. Shukla, A. K., Banerjee, A., Ravikumar, M. K., & Gaffoor, S. A., (2015). Energy Storage
Device, an Inorganic Gelled Electrolyte and Methods Thereof. US Patent US 9,036,332.
23. Shukla, A. K., Banerjee, A., & Ravikumar, M. K., (2012). Electrochemical capacitors:
Technical challenges and prognosis for future markets. Electrochimica. Acta, 84, 165–173.
24. Pradeep, K. R. J., Nagamani, H. N., Vaidhyanathan, V., & Bhavani, S. T., (2017). A
study on preliminary screening of 12V/ 2500F range lead carbon hybrid ultracapacitors
as energy storage device for solar power applications. Power Research–A Journal of
CPRI, 13(3), 503–510.
25. Pradeep, K. R. J., Nagamani, H. N., Vaidhyanathan, V., & Bhavani, S. T., (2017).
Charge–discharge behavior of lead acid battery and lead carbon hybrid ultracapacitors
as an integrated system for solar power applications. Power Research–A Journal of
CPRI, 13(3), 517–524.
26. Sanjeevi, K. P., Neeraj, P., Mahajan, S. B., Bo Holm–Nielsen, J., & Eklas, H., (2019).
A hybrid photovoltaic–fuel cell for grid integration with Jaya based maximum power
point tracking: Experimental performance evaluation. IEEE Access, 7, 82978–82990.
27. Yong, S., Xu–Wei, G., Ji, X., Xin, W., & Xu, Y., (2019). Large power hybrid soft
switching mode PWM full bridge DC–DC converter with minimized turn–ON and turn
–OFF switching loss. IEEE Transactions on Power Electronics, 34(12), 11629–11644.
Non Commercial Use

28. Ben, S. S., (2020). Intelligent energy management for off–grid renewable hybrid system
using multi–agent approach. IEEE Access, 8, 8681–8696.
29. Chen, Z., He, Y., & Chengbin, M., (2020). Equivalent series resistance based real–time
control of battery–ultracapacitor hybrid energy storage systems. IEEE Transactions on

30. Siyuan, C., Quifan, Y., Jianyu, Z., & Xia, C., (2021). A model predictive control method
for hybrid energy storage systems. CSEE Journal of Power and Energy Systems, 7(2),
329–338.
31. Mince, L., Li, W., Yujie, W., & Zonghai, C., (2021). Sizing optimization and energy
management strategy for hybrid energy storage system using multiobjective optimization
and random forest. IEEE Transactions on Power Electronics, 36(10), 11421–11430.
Author Copy
Non Commercial Use

Non Commercial Use

Author Copy
CHAPTER 9
Speed Sensorless Model Predictive

Current Control of Induction
Motor-Driven Electric Vehicle
KARUNA KIRAN
Author Copy
Indian Institute of Technology (Indian School of Mines), Dhanbad,
Jharkhand, India
ABSTRACT
The chapter deals with an analysis of speed-sensorless model-predictive-

current control (MPCC) technique of induction motor (IM) driven electric
vehicle (EV). MPCC is a direct control approach in which the errors between
the reference stator current and the predicted stator current are converges
to minimize the cost function. The cost function is defined for each of
the predicted stator current corresponding to every voltage vector of the
inverter. The presence of an optimized inner current controller with cost
function and absence of pulse width modulator (PWM) are the key features
of the proposed MPCC scheme. Moreover, stator current (is) based model
reference adaptive system (is-MRAS) is proposed as speed estimator owing
to its accuracy in estimation and simplicity. The absence of PWM as well as
the employment of is-MRAS aid in making the overall drive system modest
and cost effective. The simulation results in MATLAB/Simulink accomplish
satisfactory dynamic behavior and uphold smooth steady state performance.
9.1 INTRODUCTION
In recent years, electric vehicles (EVs) are gaining remarkable research

efforts due to sharp depletion of petroleum resources as well as climate and
Non Commercial Use

environmental awareness. Electrical machine drive is a crucial component

of EV powertrain as it gives the vital power to propel the EV in an efficient
manner [1]. From the industrial applications point of view, induction motor
(IM) shows strong candidature as EV drive for sensorless speed control in
respect of low-cost and high performance [2, 3]. Moreover, it is efficient,

trustworthy and offers an extended range of stability at different speeds [1].
Literature survey shows that manifold analytical research and efficacious
practical implementations for the speed control of IM have been carried out
[3, 4]. Some of the classic control approaches such as direct torque control
(DTC) [6] and vector control (VC)/field-oriented control (FOC) [5] have
been comprehensively employed during recent few decades. FOC scheme
Author Copy
comprises linear proportional integral (PI) controllers as well as pulse-width
modulator (PWM) [5]. FOC scheme is widely applicable for low and medium
power electrical drives. However, DTC appeared as a plausible solution for the
moderate to high-power applications [2]. However, the outcomes of hyster-
esis torque and flux controllers in DTC vitiate the steady state performance.
On contrary, FOC has merit of outstanding performance at steady state [5].
However, inherent slow dynamics of the inner current control loop degrades
its dynamic response [4]. Being a linear control strategy, FOC has drawbacks
pertaining to limited bandwidth and requirement of extra PWM hardware for
pulse generation of voltage source inverter with two-levels (2L-VSI) [7].
To overcome the above-mentioned concerns, the emerging model-
predictive-control (MPC) is deputized as encouraging approach to tackle the
aforementioned FOC drawbacks [8, 9]. With the advent of high-performance
power electronic devices and powerful digital signal processors (DSPs),
MPC emerged as an attractive area of research for nonlinear and direct
control of electric drives and power electronic devices [10]. It utilizes
model of system and a cost/objective function to directly govern the control
variables [11]. Optimal solution is determined by meticulous searching of
voltage vectors in the specified viable margin, according to the principle
of cost function minimization [9]. The scheme can be classified into two
mainstream categories: model predictive current and model predictive
torque control (MPCC [7], MPTC [12]). Alike classical FOC [4], MPCC
makes use of orthogonal current components only in the cost function and
hence weighing factor is not mandatory [7, 13]. However, whereas in case
of MPTC the cost function contains both the torque and the flux terms and
hence a combination of two different units and order of magnitude needs to
be made, thus requiring a weighing factor which is challenging to tune [12].
Therefore, MPCC strategy is recommended in this work over MPTC due to
its simplicity.
Non Commercial Use

Speed Sensorless Model Predictive Current Control 181
Apart from this, the installation of a physical speed sensor (such as

tachometers, optical encoders, etc.) on the rotor shaft of EV motor to get the
instantaneous speed information is troublesome [4]. It not only reduces the
system reliability in hostile environment but also imposes cost burden and
hardware intricacy to the overall drive system [14]. Subsequently, the estima-
tion of speed from electrical measures (i.e., machine terminal phase currents
and phase voltages) is favored than speed measurement [15]. A number of
sensorless techniques have been suggested for IM drive [14]. Recent litera-
ture survey recommends that MRAS-based methods for rotor speed estima-
tion show more promising performance over other model based and signal
injection-based speed estimation techniques because of its simplicity and
Author Copy
physical exposition. It is, however, difficult to improve the performance of
these schemes at very low speeds. Various MRAS configurations have been
analyzed on the basis of simplicity, ease in implementation, stability, and
high performances in all the four quadrants and immunity towards machine
parameter variation. In this context, MRAS speed estimators using rotor flux
[16], back EMF [17], electromagnetic torque [18], active power or reactive
power [15, 19], stator current as control variables are investigated compre-
hensively [20]. Rotor flux MRAS is the classical MRAS strategy, first intro-
duced by Schauder [21]. It suffers from pure integrator related issues. Back
EMF scheme avoids pure integration process but shows low noise immunity
and may become unstable at low or zero supply frequencies. Reactive power
MRAS overcomes parameter sensitivity effects and pure integrator related
issues like drift or saturation [22] but suffers from instability at few operating
points. This instability issue is avoided using active power MRAS.
The present work extends the concept of MPCC as discussed in Ref. [7]
for the sensorless operation. In this context, a stator current based MRAS
algorithm (is-MRAS) in association with MPCC scheme is developed for
the first time. The reference model of the estimator is designed using the
instantaneous terminal stator current. Moreover, the speed dependent instan-
taneous predicted current from the MPCC algorithm is unit delayed and
employed as adaptive model. The developed is-MRAS based MPCC scheme
is investigated in MATLAB/Simulink for wide-ranging speed variations.
The present work is divided into six sections. The fundamentals (2L-VSI)
and IM drive model formulation are discussed in Section 9.2. In Section 9.3,
the overall speed control using MPCC scheme of IM driven EV is described
in detail. To make the developed MPCC scheme speed sensorless, a stator
current MRAS for speed estimation is integrated in Section 9.4. To investi-
gate the efficacy of the acclaimed scheme, some important outcomes of the
Non Commercial Use

simulation in MATLAB/Simulink are discussed in Section 9.5. In the final

Section 9.6, the work has been concluded.
9.2 MATHEMATICAL MODELING OF VOLTAGE SOURCE INVERTER

AND INDUCTION MOTOR
In this section, simple mathematical models of the three-phase 2L-VSI and

IM are presented.
Author Copy
9.2.1 MODELING OF 2L–VSI
The DC-link power is converted to three-phase alternating (AC) power by the

power circuit of three-phase 2L-VSI (see Figure 9.1(a)). To circumvent short-
circuiting of DC source, the complementary switches in each phase operate
at a time [8]. Hence, switching states of the six switches (Sx, S̅ x = 1, 0 and
x = 1,2,3) can be represented by switching signals Sa, Sb, Sc as [7]:
1 if S1 → on, and S1 → off
Sa ⇒  (1)
0 if S1 → off, and S1 → on
1 if S 2 → on, and S 2 → off

Sb ⇒  (2)
0 if S 2 → off, and S 2 → on
1 if S3 → on, and S3 → off

Sc ⇒  (3)
0 if S3 → off, and S3 → on
The phase to neutral (N) voltages of 2L-VSI are defined in terms of

switching signals Eqns. (1)–(3) as:
vaN = S aVdc (4)
vbN = SbVdc (5)
vcN = ScVdc (6)
where; Vdc: voltage across the DC-link.
The voltage vector output is the vector sum of three-phase to neutral
voltages as:
2
v=
3
( vaN + avbN + a 2 vcN ) . (7)
Non Commercial Use

where;
j 2π
a=e 3 (8)
Therefore, eight switching states are obtained using all feasible arrange-
ments of the gating signals (Sa, Sb and Sc). Consequently, eight different
voltage vectors are generated [9]. These voltage vectors are represented in
Figure 9.1(b).
Author Copy
SFIGURE 9.1 Three-phase 2L-VSI topology and inverter voltage vectors in complex plane.
Non Commercial Use

9.2.2 MODELING OF IM
In stationary (superscript ‘s’) reference frame, the classical electrical dynamic

equations are described below as following set of equations:
d s
v ss iss Rs +
= ψs (9)
dt
d s
0 =irs Rr + ψ r − jωψ rs (10)
dt
ψ
= s
s Ls iss + Lm irs (11)
Author Copy
ψ
= s
r Lr irs + Lm iss (12)
3
=Te p Im {ψ sr ⋅ iss } (13)
2
where; p,ω and Te denote number of pole pairs, electrical speed and gener-
ated electromagnetic torque of the motor, respectively.
From IM model described in Eqns. (9)–(12), the current in the stator
winding can be evaluated as:
1   diss 1   
iss =
−   Lσ − kr  − jω ψ rs  − v ss  (14)
Rσ   dt τr   
where;
Lm
k=
r ; L=
σ σ Ls ; and R=
σ Rs + kr2 Rr (15)
Lr
9.3 MODEL PREDICTIVE CURRENT CONTROL
Figure 9.2 represents the comprehensive block diagram of IM driven EV with

rotor flux oriented sensorless MPCC scheme. The rotor speed estimates are
obtained from the is-MRAS block. The direct and quadrature axes reference
current vector components are generated using the rotor flux and load torque
reference, respectively. The rotor flux estimation is done using the stator
current in rotating reference frame to fulfill the rotor flux orientation strategy
so that the decoupled torque and flux control could be possible alike VC.
However, the remaining control mechanism is executed in the stationary
frame. The calculation of unit vector is done to perform conversions
from stationary to synchronously rotating frame and contrariwise. For the
calculation of unit vector, slip speed is determined using the stator current
Non Commercial Use

in stationary reference frame. Hence, the control mechanism becomes very

simple. Consequently, the rotating frame reference currents are transferred
to the static reference frame as a parameter in the MPCC algorithm. Using
the speed estimates from is–MRAS, instantaneous stator current and 2L-VSI
voltage signals, the stator current for the next sampling interval is predicted
corresponding to all feasible inverter voltage vectors. Comparing the
predicted currents to that of the stationary frame reference stator current
components; a cost function is designed corresponding to all the predicted
currents. The optimization of the cost function in the MPCC algorithm is
performed using the algorithm depicted in flow diagram – Figure 9.3. The
voltage vector that corresponds to the optimal value of cost function is
Author Copy
applied for the upcoming control cycle.
FIGURE 9.2 is-MRAS based sensorless predictive current control of IM driven EV.
The entire MPCC scheme can be explained through the algorithm depicted
in Figure 9.3. First of all, voltage and current in the stator is measured at the
terminal. Then, this information is used for the estimation of rotor flux and
speed. These estimated quantities along with the reference values of current
(ir*ds and ir*qs) obtained from flux and torque references are utilized to calculate
of stator current in stationary frame. Thereafter, prediction of current for
Non Commercial Use

the next control cycle is done corresponding to each voltage vectors. On

comparison of these predicted currents to their respective references, the
cost functions are calculated. The predicted value that minimizes the cost
function is selected and its conforming voltage vector is fed to VSI for the
next control cycle. The step-by-step process of the scheme as accomplished

in Figure 9.3 is discussed as in subsections.
Author Copy
FIGURE 9.3 Model predictive current control algorithm.
Non Commercial Use

9.3.1 GENERATION OF REFERENCE CURRENT
The actual stator current is measured at the terminal. For the execution of
the MPCC scheme, the direct and quadrature axes current references need to
be generated. The quadrature-axis current reference is obtained using torque

reference which is generated as an output of the outer speed loop. However,
the direct-axis current reference is obtained by flux reference which is taken
as a constant.
idsr * = ψ drs* / Lm (16)
(K + Ki s ) T *
Author Copy
iqsr *
= p (17)
where; Kp and Ki are the speed PI regulator’s proportional and integral

constants.
The stator current reference signals generated from the above direct and
quadrature components are further converted to stationary reference frame
(iα* and iβ*) for the evaluation of cost function.
9.3.2 PREDICTION OF STATOR CURRENT FOR THE NEXT CONTROL

INTERVAL
The stator current in the succeeding control interval i ss(k + 1) is predicted

for all feasible inverter voltage vectors using system model Eqn. (14) as [7]:
 T  Ts  1  k r  s  
iss ( k + 1) = 1 + s  iss ( k ) +   − kr jω ψ r ( k ) + v s ( k )  
s
(18)
 τσ  τ σ + Ts  Rσ  τ r   
where; τσ = Lσ / Rσ
9.3.3 EVALUATION OF COST FUNCTION
The cost function that is described in terms of the error in predicted currents
with respect to the reference current and is evaluated as:
g y = iα* − iα ( k + 1) y + iβ* − iβ ( k + 1) y (19)
y = 0,1,2..7
where; iα ( k + 1) y , iβ ( k + 1) y are real and imaginary components of predicted

current for yth iteration.
Non Commercial Use

The switching state signals corresponding to that voltage vector is

selected which minimizes the current error. Finally, the generated output
inverter voltage is fed as an input to the stator of the induction motor.
9.4 FORMULATION OF is-MRAS-BASED SENSORLESS MPCC SCHEME
The main purpose of a MRAS is to make the state variable adjust itself
as per the quantity to be observed in the system. For a is-MRAS, the state
variable is the instantaneous stator current. The elementary arrangement
of is-MRAS is shown in Figure 9.4. It comprises of reference model and
Author Copy
adaptive model. The reference model incorporates a set of equations that
is free from the parameter to be estimated (i.e., speed). The instantaneous
current at motor terminal is used as reference model. The adaptive model
computes the same state variables using speed (parameter to be estimated)
dependent equations. In this case, the predicted current signal at instant k
+1 is passed through a unit delay to obtain the instantaneous current for
the adaptive model. An adaptive mechanism is employed to do away with
the error between these two models to accomplish exact estimation of the
desired quantity.
FIGURE 9.4 Basic structure of is–MRAS.
9.5 SIMULATION STUDY AND RESULTS
The simulation study in MATLAB/Simulink environment validates

developed MPCC based speed control scheme of the motor drive. The motor
parameters and PI controller gains for speed and is-MRAS are revealed in
Non Commercial Use

the Appendix A. The performance of drive for steep step speed variation is
represented in this section.
Figure 9.5(a) shows that the speed reference is set in step changing pattern.
The starting and throughout speed tracking performance is very smooth. The
estimated speed ω̂r trails the reference ω*r and actual speed ωr very precisely
at all the operating points even during the speed transitions as well. In Figure
9.5(b), absolute speed inaccuracy is always below than 1.5 rad/s at steady
state and within the permissible limit during start-up and speed transitions
which shows good overall response of the drive system. Figure 9.5(c) shows
that electromagnetic torque produced in motor and the applied constant load
torque (one-fourth of the rated value). The applied load torque is followed
Author Copy
by electromagnetic torque during steady-state and remains in stable range
within the transient period as well. The maximum values of absolute torque
error at steady-state and transient states are always within the reasonably
satisfactory range as shown in Figure 9.5(d). In Figure 9.5(e), the amplitude
of one of the stator phase currents remains constant at constant load torque
despite of change in speed. Figure 9.5(f) shows the rotating frame rotor flux
which validates the rotor flux orientation.
9.6 CONCLUSION
The present research work is dedicated to the speed sensorless MPCC scheme
of IM drive which is suitable for electric vehicle applications. To mitigate
the defects of degraded reliability and increased cost, stator current MRAS
observer-based speed estimation algorithm is integrated with the MPCC
technique. In is-MRAS, the reference model is the instantaneous terminal
stator current, however the predicted stator current for the next control cycle
of model predictive current controller is unit delayed in order to acquire
the estimated current for the adaptive model. The adaptive mechanism
constructed by employing a PI controller provides the rotor speed estimates.
Moreover, the independent control of electromagnetic torque and flux (rotor)
is also attained using FOC method. This inventive incorporation comprises
the merits of both the intuitive sensorless method and MPCC scheme. Conse-
quently, the system is robust, cost-effective, and easy to implement. It shows
good speed tracking and dynamic performances for a wide speed range. Due
to the presence of model predictive controller, the scheme works well even
at closely zero speeds. The simulation results validate the soundness of the
proposed scheme.
Non Commercial Use

(a) (b)
Author Copy
(c) (d)
(e) (f)
FIGURE 9.5 Simulation results for variable speed constant torque operation of IM drive:
(a) Reference, estimated and actual speed; (b) absolute error between estimated and actual
speed; (c) electromagnetic and load torque; (d) torque error; (e) stator phase current; and (f)
rotating frame rotor flux.
KEYWORDS
• electric vehicle
• induction motor
• model predictive current control
• MRAS current estimator
• sensorless
• simulation study
• voltage source inverter
Non Commercial Use

REFERENCES
1. Hadraoui, H., Zegrari, M., Chebak, A., Laayati, O., & Guennouni, N., (2022). A multi-
criteria analysis and trends of electric motors for electric vehicles. World Electric Vehicle
Journal, 13(65), 1–28.

2. Rodríguez, J., et al., (2011). High-performance control strategies for electrical drives:
An electrical assessment. IEEE Trans. Ind. Electron, 812–820.
3. Wallace, A., Spee, R., & Alexander, G., (1993). Adjustable speed drive and variable
speed generation systems with reduced power converter requirements. In: IEEE
International Symposium on Industrial Electronics, Conference Proceedings, (ISIE’93).
Budapest, Hungary.
4. Bose, B. K., (2002). Modern Power Electronics and AC Drives. Prentice–Hall.
Author Copy
5. Karuna Kiran (2018) An improved rotor flux space vector-based MRAS for field
oriented control of induction motor drives. IEEE Transactions on Power Electronics,
33(6), 5131–5141.
6. Das, S., Pal, A., & Manohar, M., (2017). Adaptive quadratic interpolation for loss
minimization of direct torque controlled induction motor driven electric vehicle. In:
IEEE 15th International Conference on Industrial Informatics (INDIN) (pp. 641–646).
Emden, Germany.
7. Wang, F., Mei, X., Tao, P., Kennel, R., & Rodriguez, J., (2017). Predictive field-oriented
control for electric drives. Chinese Journal of Electrical Engineering, 3(1).
8. Rodriguez, J., et al., (2007). Predictive current control of a voltage source inverter. IEEE
Transactions on Industrial Electronics, 54(1), 495–503.
9. Rodriguez, J., & Cortes, P., (2012). Predictive Control of Power Converters and Electrical
Drives. Valparasio, Chile: A John Wiley & Sons Ltd.
10. Cortes, P., Kazmierkowski, M. P., Kennel, R. M., Quevedo, D. E., & Rodriguez, J.,
(2008). Predictive control in power electronics and drives. IEEE Trans. Ind. Electron.,
55(12), 4312–4324.
11. Kennel, R., Rodríguez, J., & Espinoza, J., (2010). High performance speed control
methods for electrical machines: An assessment. In: IEEE International Conference on
Industrial Technology (ICIT). Vina del Mar, Chile.
12. Wang, F., et al., (2015). Model-based predictive direct control strategies for electrical
drives: An experimental evaluation of PTC and PCC methods. IEEE Trans. Ind.
Informat., 11(3), 671–681.
13. Fuentes, E., Rodrigues, J., Silva, C., Diaz, S., & Quevedo, D., (2009). Speed control of
a permanent magnet synchronous motor using predictive current control. In: IEEE 6th
International Power Electronics and Motion Control Conference (IPEMC), 390–395.
14. Holtz, J., (2006). Sensorless control of induction machines-with or without signal
injection? IEEE Trans. Ind. Electron, 53(1), 7–30.
15. Ravi, T. A. V., Verma, V., & Chakraborty, C., (2015). A new formulation of reactive-
power-based model reference adaptive system for sensorless induction motor drive.
IEEE Trans. Ind. Electron., 62(11), 6797–6808.
16. Rehman, H. U., Derdiyok, A., Guven, M. K., & Xu, L., (2001). An MRAS scheme for
on-line rotor resistance adaptation of an induction machine. In: IEEE 32nd Annual Power
Electronics Specialists Conference, 2001; PESC: 2001. Vancouver, BC, Canada.
17. Rashed, M., & Stronach, A. F., (2004). A stable back-EMF MRAS-based sensorless
low-speed induction motor drive insensitive to stator resistance variation. IEE Proceedings–
Electric Power Applications, 151(6), 685–693.
Non Commercial Use

18. Kojabadi, H. M., Abarzadeh, M., & Chang, L., (2015). A comparative study of various
methods of IM’s rotor resistance estimation. In: IEEE Conference on Energy Conversion
Congress and Exposition (ECCE), 2015. Montreal, QC, Canada.
19. Ta, C. M., Uchida, T., & Hori, Y., (2001). MRAS-based speed sensorless control for
induction motor drives using instantaneous reactive power. In: The 27th Annual Conference
of the IEEE Industrial Electronics Society, 2001: IECON ’01. Denver, CO, USA.
20. Kumar, R., Das, S., & Chattopadhyay, A. K., (2016). Comparative assessment of two
different model reference adaptive system schemes for speed-sensorless control of
induction motor drives. IET Electric Power Applications, 10(2), 141–154.
21. Schauder, C., (1992). Adaptive speed identification for vector control of induction motors
without rotational transducers. IEEE Transactions on Industry Applications, 28(5),
1054–1061.
22. Maiti, S., Chakraborty, C., Hori, Y., & Ta, M. C., (2008). Model reference adaptive
Author Copy
controller-based rotor resistance and speed estimation techniques for vector controlled
induction motor drive utilizing reactive power. IEEE Trans. Ind. Electron., 55(2),
594–601.
Non Commercial Use

APPENDIX A
Specifications of IM
Shaft power (rated) 2.2 kilo Watt

Current (A) rating 2/2 Amperes
Voltage (V) rating 415/415 Volts
Frequency of stator supply (Hz) 50 Hertz
Synchronous speed (revolution per minute) 1,441 rpm
p 2
Author Copy
Rs, R '
r
4.125 Ω, 2.486 Ω
Lm, Lls, Rlr' 0.2848 H, 0.01557H, 0.01557H
J 0.0182 kg-m2
Xm 89.477
Speed PI controller gains (Kp, Ki, Slim) 0.3, 3, 6
is-MRAS adjustable mechanism PI controller gains 0.6, 10, 600
(mKp, mKi, mSlim)
Sampling frequency (discrete) for simulation study 20 kHz
Non Commercial Use

Non Commercial Use

Author Copy
CHAPTER 10
Performance Analysis of Hybrid Off-Grid

Two-Axis Photovoltaic Tracking System/
Fuel Cell Energy System Incorporating
High-Efficiency Solar Cell
Author Copy
SHUBHASHISH BHAKTA,1 MESFIN MEGRA,1 PIKASO PAL,2 and
ASHEBIR BERHANU1
1
Department of Electrical Power and Control Engineering,
Adama Science and Technology University, Adama City, Ethiopia
2
Department of Electrical Engineering, Indian Institute of Technology
(Indian School of Mines), Dhanbad, Jharkhand, India
ABSTRACT
The present chapter deals with a hybrid off-grid two-axis photovoltaic (PV)
tracking system with a hydrogen-based fuel cell (FC) system for Adama
City, Ethiopia. The hybrid system is designed and simulated using HOMER
software. The high efficiency (20%) monocrystalline silicon PV panel is
considered to analyze the performance of the PV-FC energy system. From
the simulation results, it is found that with the assumed typical load profile
of 11.7 kW peak and average annual solar radiation of 6.06 kWh/m2/d,
the proposed hybrid system is capable of generating total annual power of
107.39 kW and 19.9498 kW from PV array and FC, respectively, with no
unmet electricity. Moreover, the renewable fraction obtained for the hybrid
energy system obtained is 100%.
Non Commercial Use

10.1 INTRODUCTION
Nowadays, electricity generation using fossil fuel-based energy sources (ES)

is still considered the major energy resource and holds most of the energy
sectors worldwide. These conventional ES are finite and depleting rapidly. Due
to the profound impact on environmental issues as the production of harmful
gases and nuclear waste products and rapidly growing electricity demand
by end-users, renewable energy system-based electricity generation from
renewable energy resources (RESs) has become an essential and urgent need
across the globe [1]. The RES examples are wave, solar, geothermal, wind,
and biomass energy. These RES can be properly utilized for the generation
Author Copy
of electricity to mitigate future electricity generation issues for the end-users.
Among all the RESs, electricity generation using solar photovoltaic (PV)
technologies is considered more popular and adopted by developing and
developed countries. However, PV system output is considerably impacted
by the variation of solar radiation during the daytime, environmental factors
such as cloudiness, and ambient temperature. Hence, to generate a steady
supply of electricity, the PV system connected to electrical loads needs to
incorporate suitable energy storage systems that can act as backup units
during the nighttime, rainy season, and even in the cloudy days. The backup
units in PV systems are optimal-sized battery banks, diesel generators, and
fuel cells run from hydrogen tanks. In the case of the battery, despite the
high efficiency of the most common batteries used in a PV system, such
as a lead-acid battery which has the disadvantage of a short life period and
degradation in capacity [2]. On the other hand, diesel fuel-operated diesel
generators produce harmful gases, which is a concern for environmental
issues. Therefore, PV systems incorporating hydrogen-based FC systems as
backup systems called hybrid energy systems may be the viable option to
generate clean electricity for end-user load demand applications, especially
in developing African countries where the population is expected to increase
to 2.4 billion by 2050 [3]. The electricity consumption in Africa per person
is relatively low; however, in 2019, the net electricity generated in Africa
is only about 804 TWh compared to the United States, which has total net
electricity generated only 20% in the same year. Considering Ethiopia,
where 3% of urban electricity have access to off-grid electricity and in rural
areas have off-grid electricity access up to 49% only [4], yet as of 2018,
approximately 44.98% of the populace have access to electricity [5].
Numerous studies being carried in the research field of RESs technologies
by researchers. Castañeda et al. [6] proposed control strategies and sizing
technique in a standalone PV/hydrogen-FC/battery energy system in order
Non Commercial Use

Performance Analysis of Hybrid Off-Grid Two-Axis 197
to well utilization of ES to satisfy the load demand, maintain hydrogen

energy reserve and state of charge of the battery bank based on Simulink
Design Optimization. Khalid et al. [7] analyzed PV/wind/hydrogen-FC/
battery system for the application of power supply to the residential building
to equalize the disparity between supply and demand utilizing the excess
electricity generation by the considered energy system using Hybrid
Optimization of Multiple Energy Resources (HOMER) software. Tzamalis
et al. [8] examined an autonomous power supply of PV/hydrogen-FC/battery
and PV/diesel generator/battery system for the rural and remote building with
the aid of HOMER simulation software. The simulation results indicated that
the increasing nominal PV capacity from 5.5 to 19 kW may eradicate the
Author Copy
diesel generator in end-user power systems in both optimal-sized PV and
hydrogen technologies. To supply electricity to an inaccessible population in
the Amazon region (Tocantins, Brazil), Silva et al. [9] examined an optimized
PV/FC/battery system based on technical and economic issues. Ghenai et al.
[10] optimized an off-grid PV/FC system to a required residential electrical
load consisting of 150 homes located in the Emirate of Sharjah (United Arab
Emirates). The renewable fraction acquired from the considered system was
40.2% only.
In the present study, the investigation geographical location considered
is Adama City, Ethiopia. The PV system incorporating a two-axis tracker
connected to a hydrogen-based FC system and typical electrical load profiles
is considered to carry out the performance analysis of the hybrid PV/
hydrogen FC system through HOMER simulation software.
10.2 SYSTEM ARCHITECTURE
The hybrid PV/hydrogen-based FC system architecture is depicted in

Figure 10.1. The hybrid off-grid energy system comprises PV modules,
direct current (DC) electrolyzer, hydrogen tank and DC FC system. The DC
busbar is interlinked with PV modules, FC, electrolyzer and electrical load.
The PV modules generate electricity when a sufficient amount of solar irra-
diation falls on the top of solar panels. In contrast, FC generates electricity
by ingesting hydrogen gas as fuel from a hydrogen tank through a chemical
reaction. The hydrogen fuel is produced by electrolyzer through electrolysis
of water and the produced hydrogen gas is accumulated in the hydrogen
tank. The mathematical equations related to hybrid energy components of
the considered system are described in the literature [11].
Non Commercial Use

Author Copy
FIGURE 10.1 Hybrid off-grid PV/hydrogen FC energy system.
The PV technology considered is monocrystalline silicon having average

efficiency equal to 20%. The nominal operating cell temperature, PV panel
lifetime, temperature coefficient and derating factor of power are assumed
as 45°C, 25 yrs., –0.35%/°C and 90%, respectively. The ground reflectance
value considered is 20%. The PV tracking system incorporated in the present
study is a two-axis PV tracking system. The FC lifetime operation hours and
minimum load ratio considered are 40,000 hrs. and 0%, respectively. The
considered values of electrolyzer lifetime, efficiency and minimum load ratio
are 15 yrs., 85% and 0%, respectively. The initial intake level of hydrogen
tank and lifetime assumed is 10% and 25 yrs., respectively. The estimated
electrical load connected to the hybrid energy system having scaled annual
average value obtained is 99.5 kWh/d with average power, peak power and
load factor estimated to be 4.14 kW, 11.7 kW and 0.355, respectively. The
daily load profile is shown in Figure 10.2.
10.3 SOLAR RADIATION AND TEMPERATURE DATA
The geographical site under investigation is Adama city, Ethiopia. From the
National Aeronautics and Space Administration (NASA) Langley Research
Non Commercial Use

Center (LaRC) POWER website [12], the monthly average global horizontal
solar radiation and temperature data incurred using geographical coordinates
of considered location, i.e., latitude: 8.56331°N and longitude: 39.2884°E as
shown in Figure 10.3. It can be seen from Figure 10.3 that the estimated
range of monthly average daily solar radiation is found to be 5.23 (July)–

6.57 (February) kWh/m2/d, while the average monthly temperature range is
estimated to be 16.77°C (December)–20.98°C (June). The average annual
solar radiation and temperature are calculated as 6.06 kWh/m2/d and 18.94°C,
respectively.
Author Copy
FIGURE 10.2 Typical electric load profile.
10.4 RESULTS AND DISCUSSIONS
In the present study, the hybrid off-grid two-axis PV/FC system is designed,
and simulation is carried out using HOMER software. The input parameters
Non Commercial Use

such as the size of the PV system considered to be 0, 10, 20, 25, 30 kW.
The size considered for FC is 5 kW and 10 kW, hydrogen tank capacity
is 0, 20, 25, 30, 35, 40, 45 kg and electrolyzer is 5, 10 kW. The random
variability such as day-to-day and time-step-to-time-step are considered to
be 2.02% and 12.2%, respectively, for the considered load profile. Moreover,
in the simulation, the effect of monthly averaged ambient temperature on
PV panels is considered. HOMER accounts for the maximum power point
tracker efficiency, which is included in the derating factor. Concerning the
connected electrical load, the optimal sizing of the considered energy system
is 10 kW FC, 45 kg hydrogen tank, 30 kWp rated PV array and 20 kW
electrolyzer.
Author Copy
FIGURE 10.3 Monthly average solar radiation and ambient temperature of Adama City,
Ethiopia.
The net annual electrical generated from rated PV array found to be

78,336 kWh/yr. (84%) and 14,597 kWh/yr. (16%) from the FC, while the
total yearly electrical production from the energy system estimated to be
98,933 kWh/yr. The monthly average electric generation of an optimally
sized energy system is presented in Table 10.1. From the rated PV array, the
maximum and minimum monthly mean electric power production obtained
is 10.204 (December) kW and 7.121 (July) kW, respectively, while the
corresponding values for FC were found to be 20.359 (September) kW and
Non Commercial Use

1.3762 (January) kW, in order. The net power generated by rated PV array
obtained is 107.39 kW and 19.9498 kW from the FC. The DC primary load
and the electrolyzer annual electric energy consumption were found to be
36,317 (46%) kWh/yr. and 42,447 (54%) kWh/yr., respectively. The total
annual electric utilization by corresponding electric load estimated as 78,764

kWh/yr. Besides, the energy system is capable of producing excess annual
electricity, which is found to be 14,169 (15.2%) kWh/yr. The excess electric
production profile is presented in Table 10.2. Also, the energy system found
to be zero unmet electric loads and zero capacity shortage throughout the
year, while the renewable fraction obtained is 100%.
Author Copy
TABLE 10.1 Monthly Average Electric Production from the PV/Hydrogen FC Energy System
Month Two-Axis PV Array (kW) Fuel Cell (kW)
Jan 9.887 1.3762
Feb 9.998 1.1599
Mar 9.276 1.5219
Apr 8.711 1.4152
May 9.044 1.9311
Jun 8.071 1.4872
Jul 7.121 1.9372
Aug 7.349 1.8995
Sep 8.226 2.0359
Oct 9.372 1.4883
Nov 10.131 1.8235
Dec 10.204 1.8739
Total 107.39 19.9498
With PV penetration of 216%, the estimated daily average energy output

and the average power output from the rated PV array obtained is 2,015
kWh/d and 8.9 kW over the year, respectively, while the PV capacity factor
obtained is 29.8%. The total PV hour of operation over the year is seen to
be 4,421 hr./yr., with the maximum PV power output value found to be 32.7
kW over the year. The hourly power output from the PV array over the year
is shown in Figure 10.4.
In the case of the FC system, the net run time of FC generator over the
year obtained is 5,249 hr./yr. and the number of times FC generator started
annually is estimated to be 428 start/yr., while the operational life incurred
by the FC before replacement found to be 7.62 yr. The net annual hydrogen
Non Commercial Use

TABLE 10.2
Daily Profile of Excess Electric Production for Considered PV/Hydrogen FC Energy System 202
Hour Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
0 0 0 0 0 0 0 0 0 0 0 0 0
1 0 0 0 0 0 0 0 0 0 0 0 0
2 0 0 0 0 0 0 0 0 0 0 0 0
3 0 0 0 0 0 0 0 0 0 0 0 0
4 0 0 0 0 0 0 0 0 0 0 0 0
5 0 0 0 0 0 0 0 0 0 0 0 0
6 0 0 0 0 0 0 0 0 0 0 0 0
7 0.56 0.47 0.58 0.18 0.26 0.11 0.11 0.05 0.07 0.32 0.72 0.65
8 1.73 0.97 1.84 0.65 0.10 0.18 0.53 0.23 0.51 0.53 0.99 1.51
9 2.64 2.61 1.68 0.60 0.27 0.44 0.75 0.21 0.76 1.22 1.65 2.00
10 3.06 3.02 1.69 1.14 0.15 1.54 0.38 0.19 0.42 0.42 1.12 2.00
11 2.79 12.67 1.37 1.78 0.01 3.29 0.07 0.11 0.34 0.33 0.86 1.44
12 4.87 18.49 9.82 9.03 0.62 4.99 0.32 0.36 0.69 1.27 2.91 2.72
13 5.11 18.23 13.12 11.56 2.85 6.15 0.05 0.27 0.30 1.07 6.23 6.96
14 5.31 19.50 14.00 13.94 5.47 7.11 0.28 0.21 0.26 0.88 8.60 10.51
15 3.52 16.49 12.84 12.84 7.84 7.08 0.21 0.03 0.21 0.74 7.89 10.33
16 3.21 14.86 10.16 9.82 6.06 5.56 0.02 0.01 0.01 0.23 7.46 9.79
17 2.29 11.33 7.56 7.48 4.46 4.29 0.19 0 0 0.10 2.28 3.29
Non Commercial Use

18 0 0.04 0 0 0 0 0 0 0 0 0 0
19 0 0 0 0 0 0 0 0 0 0 0 0
20 0 0 0 0 0 0 0 0 0 0 0 0
21 0 0 0 0 0 0 0 0 0 0 0 0
22 0 0 0 0 0 0 0 0 0 0 0 0
23 0 0 0 0 0 0 0 0 0 0 0 0
Author Copy
consumption by FC is found to be 876 kg/yr., which is equivalent to the total

annual fuel energy input of 29,194 kWh/yr. fed to the FC generator. The
capacity factor, average electrical efficiency maximum electrical output and
mean electrical output of the FC found to be 16.7%, 50%, 8.4 kW and 2.78
kW, respectively. The hourly FC power output over the operational year is
Author Copy
FIGURE 10.4 Hourly PV power output over the year.
FIGURE 10.5 Hourly FC power output over the year.
The overall hydrogen production by the rated electrolyzer (20 kW)

over the year is found to be 915 kg/yr. with a capacity factor value equal
to 0.242. The autonomy of the hydrogen tank obtained is 362 hours.
Considering the electrolyzer, the monthly average hydrogen production
is presented in Table 10.3. The monthly average maximum hydrogen
production by electrolyzer obtained is 3.2556 (January) kg/d and minimum
hydrogen production is 1.6804 (February) kg/d. The annual average value
of hydrogen production by electrolyzer is found to be 2.4983 kg/d. The
hourly hydrogen tank storage level is shown in Figure 10.6, which shows
from February–June and November-December month, the hydrogen level
is above 40 kg.
Non Commercial Use

TABLE 10.3 Hydrogen Production by the Electrolyzer

Month Monthly Average Hydrogen Production (kg/d)
Jan 3.2556
Feb 1.6804
Mar 2.1694
Apr 2.0441
May 2.7722
Jun 2.1559
Jul 2.1806
Aug 2.2964
Author Copy
Sep 2.7241
Oct 3.2809
Nov 2.7233
Dec 2.6973
Average 2.4983
FIGURE 10.6 Hourly hydrogen tank level over the year.
10.5 CONCLUSION
In this chapter, a hybrid off-grid two-axis PV tracking system with a hydrogen-

based FC system is designed for Adama city, Ethiopia. A typical load profile
is assumed to have a peak demand of 12 kW. To obtain and analyze the system
component’s performance, the energy system is simulated in HOMER soft-
ware. From the simulation results, it is found that the energy system is capable
of generating access electricity. The hydrogen storage tank is capable of
storing hydrogen up to 40 kg from February-June and November–December
month.
Furthermore, from the two-axis PV array tracking system and FC system,
the total annual power obtained is 107.39 kW and 19.9498 kW, respectively.
Non Commercial Use

In addition, during the energy system operation it is found that no unmet

electric load and capacity shortage over the year. Hence, implementing such
an energy system may mitigate energy security issues in the nearer future for
the end-user electric load demand applications.
KEYWORDS
• hybrid energy system

• hydrogen fuel cell
Author Copy
• photovoltaic
• photovoltaic tracking system
• renewable fraction
• solar cell
• solar radiation
• temperature data
REFERENCES
1. Salameh, T., Abdelkareem, M. A., Olabi, A. G., Sayed, E. T., Al-Chaderchi, M., &
Rezk, H., 2021. Integrated standalone hybrid solar PV, fuel cell and diesel generator
power system for battery or supercapacitor storage systems in Khorfakkan, United Arab
Emirates. International Journal of Hydrogen Energy, 46(8), 6014–6027.
2. Zhang, Y., & Wei, W., (2020). Decentralized coordination control of PV generators,
storage battery, hydrogen production unit and fuel cell in islanded DC microgrid.
International Journal of Hydrogen Energy, 45(15), 8243–8256.
3. Section Issue Africa. Available from: https://www.eia.gov/outlooks/ieo/section_issue_
Africa.php (accessed on 12 June 2023).
4. Multi-Tier Framework for Energy, (2018). Available from: https://trackingsdg7.esmap.
org/data/files/download-documents/2019-Tracking%20SDG7-Full%20Report.pdf
(accessed on 12 June 2023).
5. Access to electricity-Ethiopia. Available from: https://data.worldbank.org/indicator/
EG.ELC.ACCS.ZS?end=2018&locations=ET&start=2018 (accessed on 12 June 2023).
6. Castañeda, M., Antonio, C., Francisco, J., Higinio, S., & Luis, M. F., (2013). Sizing
optimization, dynamic modeling and energy management strategies of a stand-alone
PV/hydrogen/battery-based hybrid system. International Journal of Hydrogen Energy,
38(10), 3830–3845.
7. Khalid, F., Ibrahim, D., & Marc, A. R., (2016). Analysis and assessment of an integrated
hydrogen energy system. In: International Journal of Hydrogen Energy.
Non Commercial Use

8. Tzamalis, G., Zoulias, E. I., Stamatakis, E., Varkaraki, E., Lois, E., & Zannikos, F.,
(2011). Techno-economic analysis of an autonomous power system integrating hydrogen
technology as energy storage medium. Renewable Energy, 36(1), 118–124.
9. Silva, S. B., Severino, M. M., & De Oliveira, M. A. G., (2013). A stand-alone hybrid
photovoltaic, fuel cell and battery system: A case study of Tocantins, Brazil. Renewable
Energy, 57, 384–389.
10. Ghenai, C., Tareq, S., & Adel, M., (2020). Technico-economic analysis of off grid solar
PV/fuel cell energy system for residential community in desert region. International
Journal of Hydrogen Energy.
11. Singh, A., Prashant, B., & Bhupendra, G., (2017). Techno-economic feasibility analysis
of hydrogen fuel cell and solar photovoltaic hybrid renewable energy system for academic
research building. Energy Conversion and Management.
12. Worldwide Renewable Resources Data Access Viewer. Available from: https://power.
Author Copy
larc.nasa.gov/data-access-viewer/ (accessed on 12 June 2023).
Non Commercial Use

CHAPTER 11
Efficient Integration of Distributed

Generation in Radial Distribution
Network for Voltage Profile Improvement
and Power Loss Minimization via Particle
Swarm Optimization
Author Copy
PRAGYA GURU,1 NITIN MALIK,1 and SHEILA MAHAPATRA2
1
School of Engineering and Technology, The NorthCap University,
Gurgaon, Haryana, India
2
School of Engineering and Technology, Alliance College of Engineering
and Design, Bangalore, Karnataka, India
ABSTRACT
The distribution system is a last and the final link between the generating
stations and the consumers end. The integration of distributed generation
(DG) resources in distribution system is crucially important due to ever-
growing increased energy demand. The ideal allocation (sizing and siting)
of DG could lead to accomplish the various benefits like improved voltage
profile and reduced losses. The load is modeled as constant complex
power. The power flow solution for the type-1 DG integrated test system is
computed using direct load flow approach under normal loading conditions
for a balanced distribution system. The problem is formulated as single-
objective constrained optimization problem. PSO algorithm is used for
optimal allocation of DG in radial distribution system. The performance is
tested and validated on four IEEE standard bus systems viz. 12-bus, 33-bus,
34-bus and 69 bus systems. The percentage of real power loss reduction
Non Commercial Use

using PSO approach is by 48.30%, 47.36%, 57.73% and 63.05% for IEEE
12-bus, 33-bus, 34-bus, and 69-bus system respectively. The appreciable
rise in the minimum voltage using proposed methodology is in the range
of 3.8%–6.5% for all the four test systems respectively. The annual energy
loss savings for type I DG also shows the remarkable benefits compared to
other published results. The proposed methodology is further compared and
tabulated with analytical approach as well as other methods to exemplify the
superiority of the proposed work.
11.1 INTRODUCTION
Author Copy
11.1.1 MOTIVATION
The incorporation of distributed generation (DG) resources in a distribu-

tion system are important due to ever-growing increased energy demand,
complexity and economic concern with traditional power generation
sources, gradual depletion of conventional resources, and environmental
concern [1]. The term DG at distribution level represents small, dispersed
generation units close to load center to fulfill the local energy demand. If
the operation of the DG system is technically efficient, it helps to cleaner
power production, i.e., reduction of greenhouse gas emission, because it
uses non-conventional resources like small hydro, fuel cells, photovoltaic
systems, wind turbines, etc.
With regards to the ideal allocation (location and capacity) of DG, we
can accomplish the various merits, including improving the power quality,
increasing the voltage profile while reducing the contamination of the
environment. But inappropriate allocation may lead to adverse effects on the
system stability and some other parameters like excessive temperature and
power losses [2, 3]. The impacts of DG may manifest them either positively
or negatively according to the DG characteristics or system operating
conditions. For the operation of distribution power utilities, future planning,
and power transfer between utilities, load flow (LF) calculations are required.
11.1.2 LITERATURE REVIEW
The increased load demand and the allocation schemes of the DG are
categorized as either analytical method or a meta-heuristic method.
Analytical methods use mathematical expressions to identify locations. A
Non Commercial Use

Efficient Integration of Distributed Generation 209
new expression is proposed in Refs. [4, 5] to calculate the DG capacity but

without cost-benefit consideration. Backtracking algorithm was used in Ref.
[6] to find out the most effective capacity and place for DG. In Ref. [7],
authors’ presents mixed-integer non-linear programming for the most appro-
priate DG allocation. The impact of ZIP load model has been analyzed in
Ref. [8] but without cost analysis. In Ref. [9], author represents a new voltage
stability index for most advantageous DG penetration for different types of
DGs. The authors presented an index-based multi-objective approach [10]
for finding the ideal size and place of DG but only type I DG was considered.
A methodology [11] established from the real and reactive branch current
components was evolved to minimize the loss in RDN by locating DG.
Author Copy
The sensitivity analysis-based approach [12] to power loss minimization is
applied for allocation.
A variety of meta-heuristic approaches have been utilized for DG
allocation problems, including non-dominated sorting genetic algorithm
(GA) [13] and ant lion optimization algorithm [14]. GA [15] is employed
for DG incorporation at various loading conditions. The allocation problem
can be resolved with bacterial foraging algorithm [16]. To determine the
optimal allocation under varying loading conditions, other approaches like
bat algorithm (BA) [17], artificial bee colony (ABC) algorithm [18] and
hybrid GA-particle swarm optimization (PSO) algorithm [19] are employed.
11.1.3 PAPER TARGET
A massive motivation of the presented work is developing an efficient

methodology to discover the optimal place and capacity of DG for reduced
network losses and achieving a better voltage profile in the RDN using
population based PSO [20] method. The recommended approach has been
examined on various standard IEEE test networks. The final evaluation of
the approach is compared to the existing techniques and is found out to be
superior in terms of minimum losses and refinement in voltage profile.
11.1.4 ORGANIZATION
The systematized paper arrangement is as follows: Section 11.2 explains

the mathematical modeling of RDN. Explanation of the objective function
formation is in Section 11.3. Section 11.4 is dealing with the direct load
flow (DLF) solution. A short explanation of the PSO algorithm is depicted
Non Commercial Use

in Section 11.5. In Sections 11.6 and 11.7, solution methodology and results
of the test network simulations are shown, respectively. Finally, Section 11.8
concluded the overall work carried out.
11.2 MATHEMATICAL MODELING OF RDN
11.2.1 LINE MODELING
Figure 11.1 demonstrates a sample RDN taken for a case study and Figure 11.2
shows an equivalent circuit diagram of single branch section connecting
Author Copy
sending node i and receiving node j, respectively. From Figure 11.2, we find,
Si = Pi + jQi (1)
Si = Vi × Ii* (2)
where; Vi, Qi, and Pi shows the receiving end voltage, reactive power load,
and the real power load at bus i, respectively.
FIGURE 11.1 Sample RDN.
FIGURE 11.2 Equivalent circuit network of single branch.
Non Commercial Use

The corresponding injected branch current at the ith bus (Ii) is given as:
*
 Si  Pi − jQi
=I i =  (3)
V
 i Vi*
where; sending end voltage is Vi. The branch real power loss (Ploss) between
two nodes is given by Eqn. (4)
 P2 + Q 2 
Ploss (i, j ) =  i 2 i R (4)
 Vi 
Author Copy
11.2.2 LOAD MODELING
The load actions can be modeled by analyzing the variation in their reac-
tive and active power demands due to changes in network voltage. The load
model is assumed as constant power load type.
Pi = P0V 0 (5)
Qi = Q0V 0 (6)
where; Q0 and P0 are the nominal reactive and active power corresponding to
the initial operating voltage, respectively.
11.2.3 DISTRIBUTED GENERATION (DG)
With small size, DG supplies are generally served as constant negative PQ

load (constant power mode) with current injecting at a node. If we choose
node i for DG penetration, then Eqn. (7) and (8) can be used to calculate the
equivalent real as well as reactive load at bus i, respectively.
Peq,i = Pi – PDG,i (7)
Qeq,i = Qi – QDG,i (8)
where; PDG,i is generated real and QDG,i is the reactive power of DG at ith node.
11.2.4 COST ANALYSIS FOR ANNUAL ENERGY LOSSES AND DG COST
The total cost of energy losses per year (CL) is given by Eqn. (9).
Non Commercial Use

CL = ((Totalrealpowerloss ) * ( EC * T ) $ (9)
where; EC and T are the energy rate and annual time duration, respectively
with values of 0.06 $/kWh and 8,760 hours.
The cost characteristics of DG for real power are based on information

available from [21] and expressed in terms of system’s real power generation.
C ( PDG ) = a * PDG
2
+ b * PDG + c $/h (10)
Based on Ref. [21], cost coefficients are: a = 0, b = 20 and c = 0.25
Author Copy
11.3 PROBLEM FORMULATION
The key aim for DG placement is to minimize power losses subject to

constraints [22]. The objective function is given by Eqn. (11)
F = minimise ( Ploss (i, j ) (11)
Subjected to the following constraints:
PGi – PDi – PLi = 0 Real power balance (12)
QGi – QDi – QLi = 0 Reactive power balance (13)
0.95 p.u ≤ Vi ≤ 1.05 p.u (14)
ii ≤ iiMax line current at node I (15)
PGiMin ≤ PGi ≤ PGiMax Real power limits (16)
Min Max
Q Gi ≤ QGi ≤ Q Gi Reactive power limits (17)
Max
0 ≤ PDG ,i ≤ PDG , i (18)
Max
0 ≤ QDG ,i ≤ Q DG , i (19)
QGi and PGi are the generated reactive and real power at ith bus, respectively.
Max
QLi and PLi are the reactive and real power loads at ith bus, respectively. PGi ,
PGiMin , QGiMin and QGiMax are the max real, the min real, min reactive and the max
reactive power generation at ith bus, respectively. PDG Max
,i
is the maximum real
power and QDG ,i is the reactive power propagated by DG at the ith node.
Max
11.4 LOAD FLOW ANALYSIS
Conventional LF approaches, for example, Newton-Raphson and fast

decoupled methods are less effective in distribution networks because of its
Non Commercial Use

low X/R ratio. A number of specially designed LF approaches have been

proposed in the literature [23–29] for distribution systems. In this technique
a DLF approach [16] has been used, developed on the basis of topological
structure and widely implemented for distribution system LF analysis.
This LF approach is established from the analog’s current injection. With

the help of the Kirchhoff’s current law, the correlation among the bus current
injections (from Eqn. (3)) and branch currents is accomplished.
B1 = I1 + I2 + I3 + I4 + I5 + I6 + I7 + I8 (20)
B3 = I3 + I4 + I5 + I6 + I7 + I8 (21)
B4 = I4 + I5 + I6 + I7 + I8 (22)
Author Copy
B6 = I6 (23)
B7 = I7 + I8 (24)
B8 = I8 (25)
The matrix formation of above equations can be given by,
 B1  1 1 1 1 1 1 1 1   I1 
 B 2 0 1 1 1 1 1 1 1   I 2 
  
 B3   0 0 1 1 1 1 1 1  I 3
    
 B 4 = 0 0 0 1 1 1 0 0  I 4
 B5   0 (26)
0 0 0 1 1 0 0  I 5
    
 B6  0 0 0 0 0 1 0 0  I 6
 B7  0 0 0 0 0 0 1 1  I 7
    
 B8   0 0 0 0 0 0 0 1   I 8 
By rewriting Eqn. (26),

[B] = [BIBC][I] (27)
where; bus-injections to branch-currents matrix are shown by BIBC. The
voltages at bus 3, bus 4, and bus 5 can be given as:
V3 = V2 – (B3 * Z23) (28a)
V4 = V3 – (B4 * Z34) (28b)
V5 = V4 – (B5 * Z45) (28c)
From the above equations,
V5 = V2 – (B3 * Z23) – (B4 * Z34) – (B5 * Z45) (29)
Non Commercial Use

Similarly, the voltage drops of a sample RDN given in Figure 11.1 can
be obtained by:
V 1 V 2   Z12 0 0 0 0 0 0 0   B1 
V 1 V 3   Z12 0   B 2 
    Z 23 0 0 0 0 0
V 1 V 4   Z12 Z 23 Z 34 0 0 0 0 0   B3 
     
V 1 V 5  =  Z12 Z 23 Z 34 Z 45 0 0 0 0   B4
V 1 V 6   Z12 (30)
Z 23 Z 34 Z 45 Z 56 0 0 0   B5 
     
V 1 V 7   Z12 Z 23 Z 34 Z 45 Z 56 Z 67 0 0   B6 
V 1 V 8   Z12 Z 23 Z 34 0 0 0 Z 78 0   B 7 
     
Author Copy
V 1 V 9   Z12 Z 23 Z 34 0 0 0 Z 78 Z 89   B8 
Eqn. (30) can be written as,

[∆V] = [BCBV][B] (31)
Where branch currents and buses voltages are correlated by BCBV matrix.
From the Eqns. (27) and (31),
[∆V] = [BCBV][BIBC][B] (32)
[∆V] = [DLF][I] (33)
Solutions of the Eqns. (34)–(36) gives LF results.
*
S 
I i k =  ik  (34)
 Vi 
[∆V k] = [DLF][I k] (35)
[V k+1] = [V0][∆V k+1] (36)
where; initial voltage V0 and the iteration count is ‘k.’
11.5 PARTICLE SWARM OPTIMIZATION
PSO is a probabilistic computation inspired by a simplified societal system

like fish schooling, bird flocking, etc. [31]. PSO is a successful optimization
approach having balanced global and local search throughout the run and
has been tested in many application areas. PSO gets faster and better results
with minimum computational burden when compared to the other methods.
In this approach, the current state of each particle changes in a search space.
If Sid ( Si1 , Si 2 ,…, Sin ) and
= d
= vid (vi1 , vi 2 , …, vin ) represent the position and
d
Non Commercial Use

velocity of particle p, respectively; d = 1, 2, … nd and i = 1, 2, … N. Swarm

size represented by N and dimension represented by nd. The velocity (v) and
position (S ) of particles in d-dimensional hyperspace are updated by using
Eqns. (37) and (38), respectively.
( ) (
vid ( q +1) = ωi vid ( q ) + C1rand1* Pbestid − Sid ( q ) + C2 rand 2* Gbestid − Sid ( q ) ) (37)
( q +1)
Sid= Sid ( q ) + vid ( q +1) (38)
The inertia weight given to each velocity is decreased linearly with itera-
tion number as in Eqn. (39).
Author Copy
[(ωmax − ωmin ) * currentiterationnumber ]
ωi ωmax −
= (39)
Maximumiteration
where; Sid(q) is position and vid(q) is particle (p) current velocity at qth iteration.
The random values rand1 and rand2 are generated in the span of [0,1].
Cognitive learning factor is C1 and social coefficient is C2. The term Gbest
and Pbest represent the global best value and the personal best value of the
particle, depending on its personal experience and experience of the overall
swarm, respectively.
11.6 SOLUTION METHODOLOGY
The algorithmic steps in optimal DG allocation are as follows:

 Step 1: Initialize line data, bus data and bus voltage.
 Step 2: Run the direct power flow algorithm for the base case as
explained in Section 11.4 to determine system power losses, branch
current, phase angle and bus voltage magnitude at all the nodes.
 Step 3: Choose the algorithm parameters swarm size, inertia weights
and acceleration coefficients for objective function minimization.
The selected parameters are given in Table 11.1.
 Step 4: Initialize the iteration count as zero k=0.
 Step 5: For each particle, compute the objective function value as
given in Eqn. (11) and compare it with the individual Pbest.
 Step 6: If there is no violation in operational constraints, calculate
the total power loss (TRPL) for each particle. Else particle is rejected,
regarded as infeasible solution.
 Step 7: Make a comparison of the objective function value of every
particle over the individual Pbest value. A lower function value than
Pbest is set to become the new Pbest.
Non Commercial Use

 Step 8: Among all the Pbest values, lowest Pbest is assigned as the
Gbest.
 Step 9: Utilizing Eqns. (31)–(33) for updating velocity, position and
weight, respectively for each particle.
 Step 10: Increase the iteration count by 1 and repeat steps 5 through
9 until convergence criterion determined by maximum number of
iterations is achieved.
 Step 11: Print out the optimal solution for the best possible place
and size for DG. The corresponding fitness values indicate minimal
system loss.
Author Copy
The flowchart for DG placement is shown in Figure 11.3.
The suggested technique is simulated for the TRPL, bus voltages, CL and
cost of DG power. The obtained outcomes are analyzed with the previously
published outcomes for the identical load models and base values. The
simulations are running on Intel Pentium dual-core processor 2 GHz, 3
GB RAM and carried out in commercial software MATLAB R2018a with
Windows 10 operating system. The 12-bus RDN has (0.4350+0.4050i) MVA
power of demand. There are three laterals in 33-bus system [32] consuming
(3.715+2.3i) MVA power demand. The 34-bus system has three laterals
and (4.6365+2.8735i) MVA power demand whereas the large scale 69-bus
system [33] has seven laterals and 48 load points with (3.803+j.22693) MVA
load. The base values for all four systems are chosen as 12.66 kV and 100
MVA. The four test systems are examined for two individual cases.
 Case 1: In the absence of DG. Base case results are shown in Table
11.2 for all four test systems.
 Case 2: With DG. Tables 11.3–11.6 shows the test results intended
for standard IEEE bus systems.
11.7.1 RESULTS FOR 12 BUS SYSTEM
The TRPL for 12 bus system and minimum voltage value (at bus 12) has been
calculated as 0.0207 MW and 0.9434 p.u, respectively. The CL before placing
the DG is 10,887 $. With DG allocation, the power loss is decreased to 0.0107
MW. The favorable size of DG is 0.2370 MW at bus 9. The minimum voltage
Non Commercial Use

at bus 12 has been increased to 0.9835 p.u. The CL after DG allocation is

decreased to 5,662 $. The cost of PDG is 5.006 $/h (Table 11.4).
Author Copy
FIGURE 11.3 Flow chart for DG allocation in RDN using PSO algorithm.
TABLE 11.1 Implementation Parameters for PSO

Population Size Max Iteration C1, C2 ωmax, ωmin
10 100 0.7, 0.7 0.9, 0.1
Non Commercial Use

TABLE 11.2 Base Case Results

Bus Network TRPL (MW) Voltage (p.u.) Cost of Energy Loss ($)
12 0.0207 0.9434 10,887
33 0.2109 0.9038 1,10,890

34 0.2217 0.9417 1,16,550
69 0.2249 0.9417 1,18,160
TABLE 11.3 12-Bus Results After DG Placement

Test Network With DG
Author Copy
Bus no. for DG allocation 9
Size of DG (MW) 0.2370
TRPL (MW) 0.0107
Percentage reduction in loss 48.30
Voltage (p.u.) 0.9835
CL ($) 5,662
Cost of PDG ($/h) 5.006
FIGURE 11.4 Voltage profile improvement for 12-bus system.
Non Commercial Use

The TRPL for the 33-bus network and minimum voltage value (at 18 bus)
have been calculated as 0.2109 MW and 0.9038 p.u. The CL before placing
the DG is 110890 $. The system real power losses after DG allocation are
0.1110 MW. The favorable DG size is 2.5902 MW and placed at 6th bus. The
voltage at 18th bus is enhanced to 0.9424 pu. The CL after DG allocation is
minimized to 58356 $. The cost of PDG is 52.04 $/h.
TABLE 11.4 Results for the 33-Bus Network with DG
Author Copy
TRPL (MW) 0.1110
% Reduction in loss 47.36
CL ($) 58,356
The TRPL for 34-bus network and minimum voltage value (at 27 bus) has
been calculated as 0.2217 MW and 0.9417 p.u. The CL before the DG instal-
lation is 116550 $. The system real power losses with DG allocation are
0.0937 MW. The most favorable size of DG is 2.9666 MW and placed at 21st
bus. The voltage at the bus 27 is increased to 0.9835 p.u. The CL after DG
allocation is decreased to 49,272 $. The cost of PDG is 59.58 $/h (Table 11.5).
TABLE 11.5 Results for 34-Bus System After DG Placement

Test System With DG
Real power loss (MW) 0.0937
CL ($) 49,272
Non Commercial Use

Author Copy
FIGURE 11.5 Improved voltage profile for 33-bus system.
The TRPL for 69 bus network and minimum voltage value (at 65 bus) has been
calculated as 0.2249 MW and 0.9417 p.u. The CL before the DG installation
is 1,18,160 $. The TRPL with DG is 0.0831 MW. The most favorable DG
size is 1.8717 MW and placed at bus 61. The voltage at 65 bus is increased
to 0.9789 p.u. The CL after DG allocation is reduced to 43,710 $. The cost of
PDG is 36.40 $/h (Tables 11.6 and 11.7).
TABLE 11.6 Results for 69-Bus Network After DG Placement

TRPL (MW) 0.0831
CL ($) 43,710
Non Commercial Use

Author Copy
FIGURE 11.6 Improved voltage profile for 34-bus network.
FIGURE 11.7 Voltage profile improvement in 69-bus system.
Non Commercial Use

TABLE 11.7 Comparative Study of Outcomes for Different Test Networks

Bus Total Load Method Optimum Optimum ∑ Ploss Ratio (Reduction
Network (MVA) DG DG Size with DG in Power Loss/DG
Allocation (MW) (MW) Size)
12 0.4350 + Proposed 9 0.2378 0.0107 0.042

0.4050i Acharya [3] 9 0.2272 0.0108 0.043
Gozel [4] 9 0.2272 0.0113 0.041
33 3.8021 + Proposed 6 2.5902 0.1110 0.038
2.6945i Kansal [34] 6 3.150 0.1150 0.030
Lalitha [30] 6 2.4948 0.1114 0.038
Author Copy
34 4.6365 + Proposed 21 2.9666 0.0937 0.043
2.8735i Acharya [3] 21 2.8848 0.0938 0.044
Gozel [4] 21 2.8848 0.0990 0.042
69 3.8021 + Proposed 61 1.8717 0.0831 0.075
2.6945i Acharya [3] 61 1.8078 0.0833 0.078
Gozel [4] 61 1.8078 0.0920 0.073
11.8 CONCLUSIONS
In this chapter, PSO algorithm is used for optimum positioning of DG

in RDN. This analysis is mapped out as a constrained single-objective
optimization problem. The power flow solution incorporated with DG is
computed using DLF approach under normal loading conditions. The CL
and expenditure on DG are also taken into account. The upgraded outcomes
are observed for Type I DG placement. The appreciable rise in minimum
voltage, significant diminution in network power loss, reduced CL, cost of
DG and the ALS is also analyzed. The percentage diminution in real power
loss by employing PSO is 48.30%, 47.36%, 57.73% and 63.05% for IEEE
12-bus, 33-bus, 34-bus, and 69-bus system, respectively. The appreciable
rise in the minimum voltage by using proposed methodology is in the range
of 3.8%–6.5% for all the four test systems, respectively. The ALS for Type
I DG also showed significant advantages over the published results. The
proposed methodology is further compared and tabulated in Table 11.7 with
analytical approach as well as other methods to exemplify the supremacy of
the planned work.
Non Commercial Use

KEYWORDS
• bus system
• cost analysis
• distributed generation
• line modeling
• load flow analysis
• optimal positioning
• particle swarm optimization
Author Copy
• radial distribution network
REFERENCES
1. Dey, B., Raj, S., Mahapatra, S., & Fausto, P. G. M. (2022). Optimal scheduling of
distributed energy resources in microgrid systems based on electricity market pricing
strategies by a novel hybrid optimization technique. International Journal of Electrical
Power & Energy Systems, 134, 107419.
2. Borges, C. L. T., (2012). An overview of reliability models and methods for distribution
systems with renewable energy distributed generation. Renewable and Sustainable
Energy Reviews, 16, 4008–4015.
3. Escalera, A., Hayes, B., & Prodanović, M., (2018). A survey of reliability assessment
techniques for modern distribution networks. Renewable and Sustainable Energy
Reviews, 91, 344–357.
4. Acharya, N., Mahat, P., & Mithulananthan, N., (2006). An analytical approach for DG
allocation in primary distribution network. International Journal of Electrical Power &
Energy Systems, 28, 669–678.
5. Gözel, T., & Hocaoglu, M. H., (2009). An analytical method for the sizing and siting of
distributed generators in radial systems. Electric Power Systems Research, 79(6), 912–918.
6. El-Fergany, A., (2015). Optimal allocation of multi-type distributed generators using
backtracking search optimization algorithm. International Journal of Electrical Power
& Energy Systems, 64, 1197–1205.
7. Rueda-Medina, A. C., Franco, J. F., Rider, M. J., Padilha-Feltrin, A., & Romero, R.,
(2013). A mixed-integer linear programming approach for optimal type, size and
allocation of distributed generation in radial distribution systems. Electric Power Systems
Research, 97, 133–143.
8. Parihar, S. S., & Malik, N., (2021). Optimal allocation of multiple DG in RDS using
PSO and its impact on system reliability. Facta Universitatis, Series: Electronics and
Energetics, 34, 219–237.
Non Commercial Use

9. Parihar, S. S., & Malik, N., (2020). Optimal integration of multi-type DG in RDS based
on novel voltage stability index with future load growth. Evolving Systems, 1–15.
10. Aman, M. M., Jasmon, G. B., Mokhlis, H., & Bakar, A. H. A., (2012). Optimal placement
and sizing of a DG based on a new power stability index and line losses. International
Journal of Electrical Power & Energy Systems, 43, 1296–1304.

11. Viral, R., & Khatod, D. K., (2015). An analytical approach for sizing and siting of DGs
in balanced radial distribution networks for loss minimization. International Journal of
Electrical Power & Energy Systems, 67, 191–201.
12. Kashem, M. A., Le, A. D., Negnevitsky, M., & Ledwich, G., (2006). Distributed
generation for minimization of power losses in distribution systems. In: 2006 IEEE
Power Engineering Society General Meeting (p. 8).
13. Singh, D., Singh, D., & Verma, K. S., (2008). GA based energy loss minimization
approach for optimal sizing & placement of distributed generation. International Journal
Author Copy
of Knowledge-based and Intelligent Engineering Systems, 12, 147–156.
14. Ali, E. S., Abd, E. S. M., & Abdelaziz, A. Y., (2017). Ant lion optimization algorithm
for optimal location and sizing of renewable distributed generations. Renewable Energy,
101, 1311–1324.
15. Shukla, T. N., Singh, S. P., Srinivasarao, V., & Naik, K. B., (2010). Optimal sizing of
distributed generation placed on radial distribution systems. Electric Power Components
and Systems, 38, 260–274.
16. Devi, S., & Geethanjali, M., (2014). Application of modified bacterial foraging
optimization algorithm for optimal placement and sizing of distributed generation.
Expert Systems with Applications, 41, 2772–2781.
17. Candelo-Becerra, J. E., & Hernández-Riaño, H. E., (2015). Distributed generation
placement in radial distribution networks using a bat-inspired algorithm. Dyna, 82,
60–67.
18. Abu-Mouti, F. S., & El-Hawary, M. E., (2011). Optimal distributed generation allocation
and sizing in distribution systems via artificial bee colony algorithm. IEEE Transactions
on Power Delivery, 26, 2090–2101.
19. Moradi, M. H., & Abedini, M., (2012). A combination of genetic algorithm and particle
swarm optimization for optimal DG location and sizing in distribution systems.
International Journal of Electrical Power & Energy Systems, 34, 66–74.
20. Mahapatra, S., Badi, M., & Raj, S., (2019). Implementation of PSO, it’s variants and
hybrid GWO-PSO for improving reactive power planning. In: 2019 Global Conference
for Advancement in Technology (GCAT) (pp. 1–6). IEEE, Bangalore.
21. Murthy, V. V. S. N., & Kumar, A., (2013). Comparison of optimal DG allocation methods
in radial distribution systems based on sensitivity approaches. International Journal of
22. Mahapatra, S., Malik, N., & Jha, A. N., (2020). Cuckoo search algorithm and ant lion
optimizer for optimal allocation of TCSC and voltage stability constrained optimal power
flow. In: International Conference on Intelligent Computing and Smart Communication
(Vol. 2019, pp. 889–905). Springer, Singapore.
23. Zimmerman, R. D., & Chiang, H. D., (1995). Fast decoupled power flow for unbalanced
radial distribution systems. IEEE Transactions on Power Systems, 10, 2045–2052.
24. Zhang, F., & Cheng, C. S., (1997). A modified Newton method for radial distribution
system power flow analysis. IEEE Transactions on Power Systems, 12, 389–397.
Non Commercial Use

25. Expósito, A. G., & Ramos, E. R., (1999). Reliable load flow technique for radial
distribution networks. IEEE Transactions on Power Systems, 14, 1063–1069.
26. Chang, G. W., Chu, S. Y., & Wang, H. L., (2007). An improved backward/forward
sweep load flow algorithm for radial distribution systems. IEEE Transactions on Power
Systems, 22, 882–884.

27. Kresting, W. H., & Mendive, D. L., (1976). An application of ladder network theory
to the solution of three phase radial load flow problem. In: IEEE PES Winter Meeting,
76044–76048.
28. Jabr, R. A., (2006). Radial distribution load flow using conic programming. IEEE
Transactions on Power Systems, 21, 1458–1459.
29. Wang, Z., Chen, F., & Li, J., (2004). Implementing transformer nodal admittance
matrices into backward/forward sweep-based power flow analysis for unbalanced radial
distribution systems. IEEE Transactions on Power Systems, 19, 1831–1836.
Author Copy
30. Lalitha, M. P., Reddy, V. V., Reddy, N. S., & Reddy, V. U., (2011). DG source allocation
by fuzzy and clonal selection algorithm for minimum loss in distribution system.
Distributed Generation & Alternative Energy Journal, 26, 17–35.
31. Guru, P., Malik, N., & Mahapatra, S., (2019). Optimal allocation of distributed generation
for power loss minimization using PSO algorithm. In: 2019 3rd International Conference
on Recent Developments in Control, Automation & Power Engineering 2019 (pp. 22–26).
IEEE, Noida.
32. Baran, M. E., & Wu, F. F., (1989). Network reconfiguration in distribution systems for
loss reduction and load balancing. IEEE Power Engineering Review, 9, 101–102.
33. Ranjan, R., Venkatesh, B., & Das, D., (2003). Load-flow algorithm of radial distribution
networks incorporating composite load model. International Journal of Power & Energy
Systems, 23, 71–76.
34. Kansal, S., Kumar, V., & Tyagi, B., (2013). Optimal placement of different type of DG
sources in distribution networks. International Journal of Electrical Power & Energy
Systems, 53, 752–760.
Non Commercial Use

Non Commercial Use

Author Copy
CHAPTER 12
Classical and Predictive Control of

Interior Permanent Magnet Synchronous
Motor for Railway Application
MANNAN HASSAN,1 MUHAMMAD SUHAIL SHAIKH,2
Author Copy
MUHAMMAD SHAHID MASTOI,1 RAO ATIF,1 MUHAMMAD FARHAN,3
MUHAMMAD AMJAD,4 MUHAMMAD BILAL SHAHID,1,4 and
ABDUL LATIF SHAH5
1
Ministry of Education Key Laboratory of Magnetic Suspension
Technology and Maglev Vehicle, School of Electrical Engineering,
Southwest Jiaotong University, Chengdu, China
2
School of Physics and Electronic Engineering, Hanshan Normal University,
Guangdong, 52104, PR China
Department of Electrical Engineering and Technology, Government College
3
University Faisalabad, Pakistan

4
Department of Electronic Engineering, Faculty of Engineering,
The Islamia University of Bahawalpur, Pakistan
5
ABSTRACT
The classical and predictive controllers for interior permanent-magnet

synchronous machines (IPMSM) railway traction drives were covered in
this chapter. Introduction to traction drive systems, mathematical modeling
of ipmsm in three phases and dq, predictive control, direct torque control,
field-oriented controllers are described in the sections.
Non Commercial Use

12.1 INTRODUCTION
According to DIN 57115 (VDE 0115) part 1, “track-bound and non-track-

bound transport systems for passenger and freight transportation” [1] is the
most complete definition of a “railway.” A strength converter, an everlasting

magnet ac machine (PMAM), sensors, and a managed set of rules are defined
because of the 4 principal additives of an ac force in modern-day transportation
systems. Permanent magnet synchronous automobiles are presently broadly
hired withinside the railway traction business (trains, trams, locomotives,
and so on). Due to their advantages, higher efficiency drive systems, smaller
size, low electrical losses, power density, low torque ripples, flux weakening
Author Copy
performance, and torque-speed cures of PMSM smooth and favorable for
traction drives as compared to other drives [3, 4].
The basic traction drive system is described in Figure 12.1. Traction
drive system owing to their performance capacity, robustness, cost, etc., in a
simplified way, that it is composed of:
• An electric motor, which generates the rotational movement;
• A power electronic converter, which supplies the electric motor taking
the energy from a specific source of energy, enabling the controlled
rotational movement of an electric motor;
• A control algorithm, which is in charge of controlling the power
electronic converter to obtain the desired performance of the electric
motor; and
• An energy source, which in some cases is part of the electric drive and
in other cases is considered an external element.
FIGURE 12.1 Basic electric drive configuration.
Non Commercial Use

Classical and Predictive Control 229
12.2 MATHEMATICAL MODELING OF INTERIOR PERMANENT

MAGNET SYNCHRONOUS MOTOR
12.2.1 INTRODUCTION OF PMSM

Synchronous AC machines are those in which the rotor speed in electrical

units is equal to the frequency of the stator current in steady-state circum-
stances. In general, synchronous machines outperform induction motors in
terms of performance, but at a greater cost and with less resilience, due to
a more complex rotor design. Synchronous machines have been utilized for
decades in high-power applications, and they continue to play an important
Author Copy
role. Synchronous machines (rotating or linear and three-phase or multi-
phase) are different types. Wound field (WFSM), Synchronous reluctance
(SyRM), Permanent magnet (PM) with different structures such as Axial,
Transversal, Radial, Interior, Surface, Trapezoidal (brushless DC, BLDC),
and Sinusoidal.
The design of electrical machines was presented in materials to hold the
magnetic field in the 1950s. Permanent magnets are used in the rotor of the
PMSM design to set it apart from conventional motors. A permanent magnet
for a high-speed application is mounted on an inside magnet, and magnets
are surface mounted for maximum power density. The internal and surface
mount motor rotor designs are shown in Figure 12.2. The salient pole is
the mathematical modeling of the motor with three-phase stator winding,
sinusoidal back emf, and rotor construction. For the following differential
equations, eddy current, iron saturation, and hysteresis losses are considered
to be zero.
FIGURE 12.2 Structure of permanent magnet synchronous machine (a) IPMSM; and (b)
SPMSM.
Non Commercial Use

Magnets embedded in the rotor generate the rotor flux. The magnets
are buried in the rotor in Figure 12.2(a), resulting in an internal permanent
magnet synchronous machine. As a result of the magnet, the d axis has a
huge air gap. The rotor on the q axis, on the other hand, is largely made of
iron. As a result, the permeability along the d axis is much lower than the
permeability along the q axis (i.e., Lds< Lqs). In Figure 12.2(b), magnets
are placed on the rotor surface, resulting in a surface permanent magnet
synchronous machine (SPMSM).
12.2.2 DYNAMIC MODELS OF SYNCHRONOUS MACHINES USING
Author Copy
THREE-PHASE VARIABLES
Major properties of the stator of three-phase synchronous machines are

similar to those of the induction machine covered in the previous chapter. A
three-phase synchronous machine’s stator voltage equations are thus similar
to those of an induction machine. For your convenience, we’ve replicated
them here:
vas  ias  ψ as 
    d  
 vbs  Rs ibs  + dt ψ bs 
= (1)
 vcs   ics  ψ cs 
The fluxes in the stator are as follows:
ψ as   Laa Lab Lac  ias  ψ am 
ψ  L Lbb Lbc  ibs  + ψ bm 
= bs   ba (2)
ψ cs   Lca Lca Lcc   ics  ψ cm 
The basic wave of the fluxes created in the rotor by the magnets or the
rotor circuit may be described as follows:
 
 ψ ⋅ cos θ 
ψ apm   pm m 
    2π 
ψ bpm  =ψ pm ⋅ cos  θ m −
3
 (3)
ψ cpm    
  
 2π 
ψ pm ⋅ cos  θ m + 
  3 
Asymmetric synchronous machine’s stator-voltage equation:
Non Commercial Use

 
 sin θ m 
vas  ias   Ls 0 0 ias   
 v  = R ⋅ i  +  0 d   2π 
Ls 0  ⋅ ibs  − ωm ⋅ψ pm sin  θ m −  (4)
 bs  s  bs   dt 3
 vcs   ics   0 0 Ls   ics    
  2π 
sin  θ m + 
  3 
12.2.3 DYNAMIC MODELS IN A SYNCHRONOUS REFERENCE FRAME
A revolving d–q reference frame with the d axis aligned with the rotor flux
Author Copy
can be used to change the dynamic model of synchronous machines. It
should be noted that the rotor flux angle for synchronous machines is the
same as the rotor angle. This transformation has significant advantages for
both analysis and control.
dψ ds
Vds =Rs ids − ωmψ qs + (5)
dt
dψ qs
Vqs =Rs iqs + ωmψ ds + (6)
dt
dids R L V
− s ids + q ωe iqs + ds
= (7)
dt Ld Ld Ld
diqs L R V ψ
− ds ωe ids − s iqs + ds + m ωe
= (8)
dt Lq Lq Lq Lq
The following are the motor’s electromagnetic torque and speed equations:
= (
Te 1.5 p ψ m iqs + ( Ld − Lq ) ids iqs (9)
d ωm 1
=
dt J
( (
1.5 p ψ m iqs + ( Ld − Lq ) ids iqs − Bωm − Tl ) ) (10)
The d–q axis stator currents and voltages are represented by iq, id, Vq
and Vd, respectively. The permanent magnet’s flux linkage is m, the stator
resistance is Rs, the mechanical rotor angular speed is m, and the d–q axis
inductances are Lq and Ld.
The flow equation for the Stator is:
ψ qs = iqs Lqs
ψ ds ids Lds + λm
=
(11)
ψ s∣ ψ + ψ
= 2
ds
2
qs
Non Commercial Use

As shown in Figure 12.3, a variety of control strategies are employed to

regulate the variables of power converters. The most well-known and easiest
to implement of these are hysteresis and linear controls with PWM [5–7].
Highly configurable digital signal processors and microcontrollers are being
developed. However, the implementation of current and more intelligent

control methods is possible.
Author Copy
FIGURE 12.3 Control techniques for electrical drives.
12.2.4 REGION OF OPERATION
Figure 12.4 shows a typical IPMSM current trajectory on the id iq plane,

which may consist of three or four lines or regions. The maximum torque per
ampere (MTPA) methdology delivers desired torque with the least amount
of phase current in Region 1 shown in Figure 12.4 with the curve OX. Along
the torque curve that represent as hyperbola, Region 2 (hyperbola XY) pulls
the existing trajectory away from the MTPA curve. Optional Region 3 (arc
YB) initiates the flux-weakening process by moving the current trajectory
along the current restriction circle. In Region 4 (curve BE), the maximum
torque per voltage (MTPV) operation provides the greatest torque possible
under the inverter voltage limitation [26].
12.3 FIELD ORIENTED CONTROL
Blachke [8] proposed the first work on FOC in 1971 for induction motors,
and the approach has since evolved to its current state. The use of a rotating
Non Commercial Use

coordinate system enables decoupled control of electrical torque and rotor

flux, which is one of FOC’s main advantages. Eqn. (9) describes the torque
and flux caused by PMSM currents Eqn. (11) There are three steps in the
FOC process: The three-phase currents are first converted to dq currents. The
park inverse transformation is then used to calculate voltage signals. Finally,

the SVPWM modulation may be used to compute the inverter switch signals.
Figure 12.5 depicts the fundamental diagram of PMSM’s FOC.
Author Copy
FIGURE 12.4 Region of operation of IPMSM drives.
FIGURE 12.5 Control block diagram of field-oriented control.
Non Commercial Use

The sensor measured ia, ib, ic three-phase currents and motor speed,
then converts ia and ib into identity notification and iq using the Park and
Clarke transformation. The outside velocity loop provides the velocity sign
mistakes, whereas the inside modern-day loop provides the modern-day sign
blunders. The PI controller receives the velocity sign blunders, which creates
the iq reference modern-day, and the PI controller receives the modern-day
sign blunders, which generates the ud and uq reference voltages. Inverse
park transformation is applied for voltage signals ua and ub generated from ud
and uq. To manage the inverter detailed below, the PWM responsibility cycle
is computed, and the SVPWM modulator delivers pulsating alerts.
Author Copy
12.3.1 PULSE WIDTH MODULATION SPACE VECTOR THEORY
SVM (space vector PWM) is a sophisticated and computationally demanding

PWM algorithm. It might be the finest PWM approach among other inverter
PWM methods. SVM (space vector modulation) is a pulse width modulation
control method (PWM). It is most typically used to drive three-phase AC
motors at varied DC speeds utilizing several Class D amplifiers to create
alternating current (AC) signals. The decrease of total harmonic distortion
(THD) produced by the frequent changes in these algorithms is an important
topic of research.
Ts=1/fs samples the reference signal Vref at a frequency of fs. With
the help of the transform, the reference signal may be created from three
independent phase references. After that, the reference vector is created by
combining the two neighboring energetic switching vectors and one or each
of the zero vectors. The order of the vectors and the zero vector(s) to utilize
can be determined using a variety of ways. The harmonic content material
and switching losses will be affected by the strategy used.
Switching sequence of voltage vector is shown in Figure 12.6. A 60°
angle occurs between two non-zero vectors that are contiguous. The two zero
vectors (V0 and V7) are the origins and deliver zero voltage to the load at the
same time. V0, V1, V2, V3, V4, V5, V6, and V7 are the eight vectors that make
up the spatial basis vectors. In the d-q plane, the required output voltage
may be translated in the same way to provide the required reference voltage
vector Vref. Using eight switching modes, the PWM space vector method
approximates the reference voltage vector Vref. When an investor trades for a
short period of time, a simple approximation is to acquire an average result,
in which case T equals Vref for the same time period.
Non Commercial Use

Author Copy
FIGURE 12.6 Sectors and basic switching vectors.
As a result, space vector PWM can be applied using following methods

below:
 Step 1: Determine Vd, Vq, Vref, and angle (α) in step one.
Vbn Vcn
Vd =Van − Vbn cos(60) − Vcn cos(30) =Van − − (12)
2 2
V V
0 − Vbn cos(30) − Vcn cos(30) =
Vq = 3 bn − 3 cn (13)
2 2
V
=ref Vd2 + Vq2 (14)
V
= arctan ( q=
α ) ω=
t 2π ft (15)
Vd
 Step 2: T1, T2, and T0 are the time intervals to be determined

(Figure 12.7). The following formula may be used to calculate the
change over time duration:
Non Commercial Use

Author Copy
FIGURE 12.7 In sector 1, the reference vector is made up of neighboring vectors.
• Sector 1 switching time period:

Tz T1 T1 + T2 Tz
∫ Vref dt =∫ V1 dt + ∫ V2 dt + ∫ V0 dt (16)
0 0 T1 T1 + T
∴ Tz ⋅ Vref = ( T ⋅V + T
1 1 2 ⋅ V2 )
cos( α ) 2 1  2 cos( 60 ) (17)
⇒ Tz⋅ | Vref | ⋅   = T1 ⋅ ⋅ Vdc ⋅   + T2 ⋅ ⋅ Vdc ⋅  
 sin( α )  3 0 3  sin( 60 ) 
where; (0 ≤ α ≤ 60)
π
sin( −α )
3 1
∴T1 =
Tz.a. and T0 = Tz – (T1 + T2), where Tz = and
π fz
sin( )
3
|Vref |
α=
2
Vdc
3
Non Commercial Use

• Switching time in the sector:

3 .Tz . Vref   π n −1   3 .Tz . Vref   π 
=T1  sin  = −α + π   sin  n − α  
Vdc   3 3  Vdc   3 
3 .Tz .|Vref | π π
= sin( n )cos( α ) − cos( n ) sin( α ) (18)
Vdc 3 3
3.Tz . Vref   n −1  
=∴ T2  sin  α − π 
Vdc   3 
3.Tz . Vref n −1 n −1
Author Copy
sin(α )cos ( π ) − cos (α ) sin ( π) (19)
Vdc 3 3
where; n = 1, 2, 3, 4, 5, 6 (i.e., sector 1 to 6).

 Step 3: Sector determination: Each PWM cycle samples the refer-
ence vector at a specified input sampling frequency (fs). The sector
is selected at this time, and the modulation vector is assigned to two
neighboring sectors. The non-zero vector looks like this:
π
2 j ( m −1)
vm = vdc e 3
(20)
3
Therefore the non-zero vector for v̅ m and v̅ m + 1 become
2  π π
=vm vdc cos (m − 1) + j sin(m − 1)  (21)
3  3 3
2  π π 
=vm +1 vdc cos (m ) + j sin(m )  (22)
3  3 3 
The reference voltage vector v̅ ref can be expressed as a function of v̅ m
and v̅ m + 1 as follows:
Ts
vref= vmTa + vm +1Tb
2
vref= vd + j vq (23)
The active state vector and the time required for each sample period
are represented by Ta and Tb, respectively, while k is the sector
number indicating the reference point. Calculate Ta and Tb, then
apply them to the switch to construct the PWM vector space’s sector
switching pattern (Table 12.1).
Non Commercial Use

TABLE 12.1 Duty Cycle Calculation

Sector Degrees Ta Tb
1 0 ≤ θ ≤ 60° 3vα v vβ
− 3 β 3
4 4 2
2 60° ≤ θ ≤ 120° 3vα v −3vα v

+ 3 β + 3 β
4 4 4 4
3 120° ≤ θ ≤ 180° vβ −3vα v

3 − 3 β
2 4 4
4 180° ≤ θ ≤ 240° −3vα v vβ
Author Copy
+ 3 β − 3
4 4 2
5 240° ≤ θ ≤ 300° −3v á v 3vα v

− 3 â − 3 β
4 4 4 4
6 300° ≤ θ ≤ 360° vβ 3vα v

− 3 + 3 β
2 4 4
 Step 4: Control the switching time of each switch (S1 to S6). The
switching sequence should be configured to minimize the switching
frequency of each inverter branch. You can run SVPWM using
different switching modes. Only two adjacent active vectors and two
zero vectors are used to avoid switching losses. To meet this optimal
condition, each switching cycle begins with a zero vector and ends
with a different zero vector.
12.4 DIRECT TORQUE CONTROL
The inverter’s optimal voltage switching vectors directly regulate the stator
flux linkage and electromagnetic torque, whereas the voltage source inverter
that supplies the PM synchronous motor is controlled by direct torque. The
major aim is to determine the voltage conversion vectors that will result in the
quickest electromagnetic torque response. Both flux error and torque error
are inputs to the flux linkage and torque delay comparator, and VSI’s DTC
PM synchronous motor control diagram has all of the necessary information.
Flux error and torque error are used as inputs for flux linkage and torque-
delay comparison.
Non Commercial Use

The electromagnetic torque comparator is a two-level comparator,

whereas the magnetic flux coupling comparator is a three-level comparator.
The hysteresis comparator’s discretized output is utilized as an entry for
top-of-the-line voltage switching. The stator’s location, on the other hand,
is undetermined. Within the lookup table, the flux linkage area vector (area
number) is also provided. Six voltage switching vectors and six voltage
switching vectors are shown in Figure 12.8 Angles u (l)…. u Eqn. (6).
Author Copy
FIGURE 12.8 Block diagram of DTC.
Using the converter’s active space vectors, the stator flux may be influ-
enced in several ways throughout the time period Ts. Regardless of flux
location, three of these voltage vectors enhance flux magnitude while others
reduce it. In addition, three of the space vectors induce an anticlockwise
rotation, which raises the relative angle, while the others create a clockwise
rotation, which reduces it.
To separately regulate stator flux and torque, it must be assured that the
amount of flux is exclusively regulated by the application of suitable voltage
vectors, while torque may be raised or lowered by modifying the relative
angle (–) within the interval [–/2; /2]. The direct component of the voltage
vector-only affects the magnitude of the stator flux vector, but the indirect
component affects torque by changing the angle of the stator flux vector.
Furthermore, both have indirect components that cause torque to rise,
whereas and may be utilized to decrease torque. As a consequence, both the
amplitude of the flux and the angle of the torque may be increased [9, 10]. The
voltage vector can also be employed to lower stator flux and torque. The link
between voltage space vectors and their effects on flux and torque within each
field is summarized in a look-up table (Table 12.2). The control error sampled
Non Commercial Use

at the time instant t0 = nTs determines the flux controller’s output value. As a
result, a similar relationship exists between the torque controller’s controller
error eT and controller output hT. The torque controller’s output can be one
of three values: –1, 0, or 1. Only the numbers –1 and 1 are accepted by the
flux controller. The corresponding control variable must be lowered if the

controller output is –1, whereas the control variable must be increased if the
controller output is 1. The torque controller, on the other hand, must keep the
control variable constant if the output is zero.
TABLE 12.2 Lookup Table for DTC
Author Copy
hψ hT S1 S2 S3 S4 S5 S6
1 1 V2 V3 V4 V5 V6 V1
1 0 V0 V7 V0 V7 V0 V7
1 –1 V6 V1 V2 V3 V4 V5
–1 1 V3 V4 V5 V6 V1 V2
–1 0 V7 V0 V7 V0 V7 V0
–1 –1 V5 V6 V1 V2 V3 V4
12.5 MODEL PREDICTIVE CONTROL
Predictive control has emerged as a viable alternative to converter control

among all of these current control strategies. It includes a variety of controllers,
including MPC, deadbeat, hysteresis-based, and trajectory-based control. All
predictive control is based on obtaining a mathematical model of the system in
order to anticipate the future response of the quantities to be controlled. As a
result, based on the established optimization condition, select the appropriate
actuation. The goal of the hysteresis-based predictive control approach is to
keep the controlled variable inside the boundaries of an area, whereas the
variables in the trajectory-based control technique must follow a predetermined
course.
Figure 12.9 provides a basic concept of the prediction horizon of MPC.
Furthermore, MPC’s fundamental principle is to employ the system’s model
to predict a few steps forward in time (Prediction horizon) based on present
conditions, which may be observed or approximated. A cost function is used
to decrease the differences between the reference and predicted variables of
interest. This technique results in an optimal control law of a certain dimension
(Control horizon). In the next sample, just the first entry of this dimension
will be used, and the operation will be repeated (Receding Horizon).
Non Commercial Use

Author Copy
FIGURE 12.9 The basic concept of MPC.
The optimization criteria in the deadbeat control approach are to pick actua-
tion that causes the error to be near to zero at future sampling instants. Cost
function The distinction between these two control methods is that continuous
control-set MPC and deadbeat control both need the use of a modulator to
provide the appropriate voltage. As a result, the switching frequency will
remain constant. The other predictive controllers generate switching signals
directly to operate the converter without the need for a modulator. As a result,
the fluctuating switching frequency will result [11]. Predictive control offers a
number of advantages that make it a good choice for power converter control:
minimization or optimization, which is employed in MPC, is a more flexible
optimization criterion.
• Nonlinearities may be accounted for in modeling without the necessity
for model linearization;
• Concepts are natural and easy to grasp;
• Technique is applicable to a wide range of constructions;
• Nonlinearities and limitations may be dealt with easily;
• The controller is simple to set up;
• Avoiding cascaded schemes is simple;
• It is possible to produce fast transient behavior.
The books [12–14] and survey studies [15–18] cited above provide
applications and theoretical conclusions. MPC has the virtue of being able
Non Commercial Use

to govern limited nonlinear MIMO systems collectively and transparently.

According to the control set of the power converter, there are two basic types
of MPC.
• MPC continuous control set (CCS-MPC);

• MPC finite control set (FCS-MPC).
Model predictive control (MPC) is a sophisticated process control
technology that is used to govern a process while meeting a set of criteria.
The main benefit of MPC is that it allows you to optimize the current timeslot
while also considering future timeslots. This is accomplished by optimizing
a finite time horizon while only implementing the current timeslot and then
Author Copy
optimizing again and again, as opposed to a linear–quadratic regulator (LQR).
In addition, MPC can predict future occurrences and execute appropriate
control actions.
12.5.1 MODEL PREDICTIVE CURRENT CONTROL
A shown in Figure 12.10. To simplify the FCS-MPC optimization process,

the modulation stage can be removed by adopting a discrete model of the
power converter. Because there are a finite number of switching states,
the optimization process can be shortened by selecting the optimum state
from among all those that correspond to the lowest cost function value.
Furthermore, if the horizon length is set to N=1, the computation overhead is
minimized and implementation becomes simple.
FIGURE 12.10 Block diagram of FCS-MPC.
Voltage and current sensors are employed in this stage to monitor control
variables including phase current, capacitor voltages, and dc-link at the
nth instant. The continuous-time model in Eqns. (7) and (8) is converted to
Non Commercial Use

discrete-time using the Euler approximation approach to forecast d–q axis

stator currents.
Ld − Ts Rs Ts Lq T V (N)
ids ( N + 1)
= ids ( N ) + ωe ( N ) iqs ( N ) + s ds (24)
Ld Lq Ld
Ld − Ts Rs TL T V ( N ) Tsψ m
iqs ( N + 1) =− iqs ( N ) − s d ωe ( N ) ids ( N ) + s ds + ωe ( N ) (25)
Lq Lq Lq Lq
Block diagram of model predictive current control is shown in

Figure 12.11. In the discrete-time domain, the traditional MPC algorithm
is developed. The MPC control technique necessitates several calculations,
Author Copy
resulting in action delay. If the delay problem is not addressed, the controller
control process will be delayed and will continue, potentially deteriorating
control performance. As a result, time-delay compensation must be
considered. Time delay compensation is a simple and effective method of
computing time, and prediction control will anticipate the amount of delay
compensation in a single value. A PI controller is used to reference the torque-
producing current, which is then used to regulate the speed. This current is
tracked using a predictive current controller. The stator current components
for the seven various voltage vectors generated by the inverter are predicted
using the machine’s discrete-time model in the predictive approach. For the
whole sampling interval, the voltage vector that minimizes a cost function is
chosen and used [19].
FIGURE 12.11 Block diagram of MPCC.
Non Commercial Use

The goals of the predictive current control scheme are as follows:

• Current magnitude limiting;
• Torque current reference tracking;
• Torque by ampere optimization.

Tracking changes in reference and decreasing tracking errors are key
components of model predictive control. As a result, cost functions must
play an important role in control systems depending on the cost functions
selected.
) (ids* ( N ) − ids ( N + 1))2 + (iqs* ( N ) − iqs ( N + 1))2 + Lˆ
g ( N= (26)
Author Copy
A non-linear function is inserted in the cost function for the present
restriction, which may be stated as:
∞ if ids ( N + 1) ≥ imax or iqs ( N + 1) ≥ imax 

ˆ ( N + 1) , i ( N + 1) = 
Lids qs   (27)
 0 if ids ( N + 1) ≤ imax or iqs ( N + 1) ≤ imax 
12.5.2 MODEL PREDICTIVE TORQUE CONTROL
The following steps are included in model predictive torque control measure-
ment and estimate, prediction, and cost function reduction. The suggested
method measures the stator current, voltage, and speed directly from the
inverter terminals and the motor shaft. The PI-speed controller provides the
torque reference. To develop a current prediction equation and forecast the
current under the operation of eight voltage vectors, first construct a model
predictive torque control equation. Second, the following control cycle’s
flux and torque are projected using the expected current, which is scrolled
and compared. Finally, in the following control cycle, the voltage vector
with the effect closest to the goal flux and torque is chosen and applied.
Calculate the difference between the anticipated flux and the torque using the
provided flux and torque Eqn. (28). Compare and choose the voltage vector
that results in the minimum flux and torque differential, which is then used
in the next control cycle. Because the variables had various units and orders
of magnitude in value, the control technique was implemented by including
the control variables in the cost function with weighting factors. The initial
weighting factor λ in the cost function modifies the trade-off between the
torque and flux terms. It’s worth noting that changing the weighting factor
between two primary variables like torque and flux has a direct impact on
how well they compare.
Non Commercial Use

 1 1
g ( k=
)  × Te* (k ) − Te (k + 1) + λ × × ψ s* −ψ s (k + 1) (28)
 Ten ψn
Because MPTC combines the system model directly with the finite
switching states, it is more effective and accurate in voltage vector selection

than direct torque control. The FCS-MPTC has no inner current controller
loop. Therefore, it has a fast dynamic response. The speed of the motor
is controlled by an outer speed-loop with PI controller. According to the
research, in comparison with industry practiced control strategies such as
FOC and DTC, FCS-MPTC’s structure is simple and it can achieve similar
Author Copy
and in some cases, better performance. According to the literature, a detailed
comparison between DTC, FOC, and FCS-MPTC is listed in Table 12.3.
Nevertheless, FCS-MPTC has some limitations such as high computational
time, variable switching frequency, weighting factor tuning, availability of
the finite number of voltage vectors and higher current distortion which are
the new research focuses of the MPC-researchers [20–27].
TABLE 12.3 Comparison of MPTC. MPCC, DTC, and FOC Methods

Index MPTC MPCC FOC DTC
Algorithm Simple Simple Complex Simple
Axis transformation Not required Not required Required Not required
Parameter sensitivity Low Low High Low
Position sensor Not required Not required Required Not required
Switching frequency Variable Variable Constant Variable
Dynamic response Fast Fast Slow Fast
Switching table Can be used Not required Not required Required
System’s nonlinearity Can be included Can be included Hard to include Hard to include
Modulation block No No Yes No
Weighting factor Yes No No No
Computational time High High Low Low
Variations in load and speed are used to test the performance of the proposed
design and other controllers. IPMSM motor specifications used in the
railway traction drive system [27–31] is used in this chapter. For the PI speed
Non Commercial Use

controller, the anti-windup technique clamping is employed. The given

design has been tested for high-speed operation. In the suggested design,
the Maximum Torque per Ampere (MTPA) principle is applied for reference
current.
12.6.1 STEADY-STATE RESPONSE
The speed response findings show significant disparities between the predic-
tive-based flux-weakening algorithm and the conventional controller, even
though both predictive control methods can track the orders well. Analysis of
Author Copy
steady-state response predictive controls and classical controls is performed
at rated speed and rated loaded torque of the system. Time 0-sec reference
speed of 189 rad/sec is applied and load torque is 800 Nm. Speed analysis of
describe controller is shown in Figure 12.12.
FIGURE 12.12 Steady-state response of speed of MPTC, MPCC, DTC, and FOC.
12.6.2 DYNAMIC RESPONSE
At step input of reference speed of 50 rad/sec up to 2 sec and 189 rad/sec

up to 5 sec, a constant load of 800 Nm is utilized to validate the dynamic
performance of the IPMSM. Figure 12.13 depicts the speed performance of
all methods. In Figure 12.14, the simulation results of MPTC, MPCC, FOC,
and MMPC proposed methods are shown.
Non Commercial Use

Author Copy
FIGURE 12.13 Dynamic performance of speed at constant load and step reference speed.
FIGURE 12.14 Comparison of current dynamic performance. (a) The FOC method; (b)
method of the MPCC; (c) the DTC method; and (d) method of MTPC.
12.6.3 DYNAMIC PERFORMANCE OF CONSTANT LOAD AND RAMP

SPEED
The train goes through acceleration, uniform motion, and a slowdown in

the railway traction system. These sorts of procedures will now have to
be certified for dynamic analysis. Table 12.4 shows the operating status
in detail. Figure 12.15 depicts the speed performance of a constant load
and a ramp. The conclusion is that the suggested architecture performs
better than alternative controllers. As a result, during acceleration and
Non Commercial Use

deceleration, the suggested mixed model predictive control is efficient

and resilient.
TABLE 12.4 Reference Ramp Speed Details

Application Time (Sec) Load (Nm) Speed (rad/sec)
0–5 800 100
5–10 20*(t–5) + 100
10–15 200
15–20 –20*(t–5) + 200
20–25 100
Author Copy
FIGURE 12.15 Dynamic performance of speed at constant load and ramp reference speed
of MPTC, MPCC, DTC, and FOC.
12.6.4 DISCUSSION
In this chapter, four control methods MPCC, MPTC FOC, and DTC are
applied to the traction drive system of PMSM. The comparison is carried out
to examine several aspects such as torque performance in stable and dynamic
states, torque, and flux ripple. To guarantee smooth reference tracking under
varying operating circumstances, a PI compensator is employed. It is critical
to consider objectives and restrictions, for example, for an application, before
deciding on a control algorithm.
Non Commercial Use

KEYWORDS
• dynamic models
• dynamic performance
• dynamic response
• field oriented control
• modulation space vector theory
• ramp speed
• steady-state response
Author Copy
REFERENCES
1. Steimel, A., (2008). Electric Traction-Motive Power and Energy Supply: Basics and
Practical Experience. Oldenbourg Industrieverlag.
2. Krause, P. C., Oleg, W., Scott, D. S., & Steven, D. P., (2013). Analysis of Electric
Machinery and Drive Systems (Vol. 75). John Wiley & Sons.
3. Türker, T., Umit, B., & Faruk, B. A., (2016). A robust predictive current controller for
PMSM drives. IEEE Transactions on Industrial Electronics, 63(6), 3906–3914.
4. Jöckel, A., & Knaak, H. J., (2002). INTRA ICE-A novel direct drive system for future
high speed trains. In: Proceedings ICEM.
5. Kazmierkowski, M. P., Ramu, K., & Frede, B., (2002). Control in Power Electronics
(Vol. 17). San Diego: Academic press.
6. Mohan, N., Tore, M. U., & William, P. R., (2003). Power Electronics: Converters,
Applications, and Design. John Wiley & Sons.
7. Linder, A., (2006). Modellbasierte Praediktivregelung in der Antriebstechnik. Logos-Verlag.
8. Blaschke, F., (1972). The principle of field orientation as applied to the new transvektor
closed-loop control system for rotating field machines. Siemens Review, 34(1).
9. Takahashi, I., & Youichi, O., (1989). High-performance direct torque control of an
induction motor. IEEE Transactions on Industry Applications, 25(2), 257–264.
10. Sorchini, Z., & Philip, T. K., (2006). Formal derivation of direct torque control for
induction machines. IEEE Transactions on Power Electronics, 21(5), 1428–1436.
11. Cortés, P., Marian, P. K., Ralph, M. K., Daniel, E. Q., & José, R., (2008). Predictive
control in power electronics and drives. IEEE Transactions on Industrial Electronics,
55(12), 4312–4324.
12. Camacho, E. F., & Carlos, B., (1999). Model Predictive Control (Advanced Textbooks in
Control and Signal Processing).
13. Maciejowski, J. M., (2002). Predictive Control: With Constraints. Pearson education.
14. Goodwin, G., María, M. S., & José, A. De. D., (2006). Constrained Control and
Estimation: An Optimisation Approach. Springer Science & Business Media.
15. Qin, S. J., & Thomas, A. B., (2003). A survey of industrial model predictive control
technology. Control Engineering Practice, 11(7), 733–764.
Non Commercial Use

16. Mayne, D. Q., James, B. R., Christopher, V. R., & Pierre, O. M. S., (2000). Constrained
model predictive control: Stability and optimality. Automatica, 36(6), 789–814.
17. Garcia, C. E., David, M. P., & Manfred, M., (1989). Model predictive control: Theory
and practice—A survey. Automatica, 25(3), 335–348.
18. Cortés, P., José, R., Daniel, E. Q., & Cesar, S., (2008). Predictive current control strategy
with imposed load current spectrum. IEEE Transactions on Power Electronics, 23(2),
612–618.
19. Rodriguez, J., & Patricio, C., (2012). Predictive Control of Power Converters and
Electrical Drives. John Wiley & Sons.
20. Habibullah, M., Dah-Chuan Lu, D., Dan, X., & Muhammed, F. R., (2016). Finite-state
predictive torque control of induction motor supplied from a three-level NPC voltage
source inverter. IEEE Transactions on Power Electronics, 32(1), 479–489.
21. Kennel, R., Rodriguez, J., Espinoza, J., & Trincado, M., (2010). High performance speed
Author Copy
control methods for electrical machines: An assessment. In: 2010 IEEE International
Conference on Industrial Technology (pp. 1793–1799). IEEE.
22. Rodríguez, J., Ralph, M. K., José, R. E., Mauricio, T., César, A. S., & Christian, A.
R., (2011). High-performance control strategies for electrical drives: An experimental
assessment. IEEE Transactions on Industrial Electronics, 59(2), 812–820.
23. Geyer, T., (2011). Computationally efficient model predictive direct torque control. IEEE
Transactions on Power Electronics, 26(10), 2804–2816.
24. Ma, Z., Saeid, S., & Ralph, K., (2014). FPGA implementation of model predictive
control with constant switching frequency for PMSM drives. IEEE Transactions on
Industrial Informatics 10(4), 2055–2063.
25. Rojas, C. A., Jose, R., Felipe, V., José, R. E., César, A. S., & Mauricio, T., (2012).
Predictive torque and flux control without weighting factors. IEEE Transactions on
26. Ming-Shyan, W., Min-Fu, H., & Hsin-Yu, L., (2018). Operational improvement of
interior permanent magnet synchronous motor using fuzzy field-weakening control.
Electronics, 7(12), 452.
27. Zhang, Z., Xinglai, G., Zisi, T., Xiaohua, Z., Qidi, T., & Xiaoyun, F., (2017). A PWM
for minimum current harmonic distortion in metro traction PMSM with saliency ratio
and load angle constrains. IEEE Transactions on Power Electronics, 33(5), 4498–4511.
28. Woldegiorgis, A. T., Xinglai, G., Songtao, L., & Mannan, H., (2019). Extended sliding
mode disturbance observer-based sensorless control of IPMSM for medium and high-
speed range considering railway application. IEEE Access, 7, 175302–175312.
29. Woldegiorgis, A. T., Xinglai, G., Huimin, W., & Mannan, H., (2020). A new frequency
adaptive second-order disturbance observer for sensorless vector control of interior
permanent magnet synchronous motor. IEEE Transactions on Industrial Electronics,
68(12), 11847–11857.
30. Woldegiorgis, A. T., Xinglai, G., Songtao, L., & Yun, Z., (2021). An improved sensorless
control of IPMSM based on pulsating high-frequency signal injection with less filtering for
rail transit applications. IEEE Transactions on Vehicular Technology, 70(6), 5605–5617.
31. Woldegiorgis, A. T., Xinglai, G., Huimin, W., & Yun, Z., (2022). An active flux
estimation in the estimated reference frame for sensorless control of IPMSM. IEEE
Transactions on Power Electronics.
Non Commercial Use

CHAPTER 13
Electric Mobility
ASHWINI KUMAR SHARMA,1 SRIMANTI ROYCHOUDHURY,2 and
SUNAM SAHA3
1
Graphic Era Deemed to be University, Dehradun, India
Author Copy
School of Engineering and Technology, Adamas University, Kolkata,
2
West Bengal, India
ABSTRACT
Climate is considered as one of the pillars for sustainable development. In

order to curb the challenges posed by the air pollution on the environment
now a days, companies have started rolling out the different vehicle with
technologies including electric vehicles in the market. In present chapter
the authors discussed about the scenario of electric vehicles in India their
types and challenges in adopting the new technologies. It has been found that
the three-wheeler market has adopted the change rapidly followed by four
wheelers and two wheelers. The government regulation has also encouraged
the acceptability of electric vehicles in different segments by defining new
policies. Overall, a fundamental change is happening with the automotive
industry for sustainable environment.
13.1 INTRODUCTION
With electric vehicles (EV) the first thought that comes is, why EVs are
becoming popular around the world? There are many strong reasons
behind this fact as in the present scenario many countries are facing huge
pollution problems because of the use of conventional vehicles, especially
Non Commercial Use

in the more populated cities. The increasing cost of fuel and scarcity of the
same has compelled the engineers and industries around the world to think
about a better alternative that can mitigate the issue. Increasing customer
consciousness about the comfort and luxury experience in the vehicle has
led the engineers to think seriously about the option of electric mobility as a
promising alternative.
As shown in Figure 13.1 there is a continuous rise in the electric vehicle
demand since 2019 and China is in the lead with 1.149 million units followed
by Europe. This indicates a new era of development in the mobility segment
with possibilities for the future.
Author Copy
FIGURE 13.1 New EV registration by area for H1 2021.
Source: EV Volumes.
Though the present-day roads are populated by conventional vehicles, in

near future, say within three to four years, markets are going to be captured
by EVs only. For the past five years, Governments in different countries
especially in Southeast Asia have undertaken many initiatives to promote
and force the use of electric vehicles in the market. Based on this many
automotive companies are shifting toward EVs and lined up several models
in the coming years. We can understand these electric vehicles have some
add-on advantages which pass over initial resistance towards accelerated
adaptation of electric vehicles.
Non Commercial Use

Electric Mobility 253
In many countries, electric mobility has already been embraced positively.

For example, this technology has proved its worth in terms of public
transport as e-rickshaw are being used by people everywhere, especially in
India, China, Europe, USA. Now the industries are venturing into the other
domains of transport such as luxury segment as well as heavy-duty transport

with this alternative.
13.2 HISTORY BEHIND ELECTRIC MOBILITY
Electric vehicle, also known as EVs operates with the help of an electric
Author Copy
motor. The internal combustion engine of the current generation automobile,
which operates by burning fuel and gases has been replaced by the electric
motor. The electric motor in an EV is operated with electric power, which
is clean energy. As, a result, electric vehicles are seen as a replacement for
the current automobiles. The outline of the historical timeline of electric
vehicles is shown in Figure 13.2.
FIGURE 13.2 Electric vehicle history chart.
Sibrandus Stratingh traced that the first Electrical vehicle was used in
the Netherlands run by Non-rechargeable batteries. Vehicles running with
Non-rechargeable primary batteries were found to be used till 1859. Using
rechargeable batteries in place of primary batteries was not a viable option
as the cost of rechargeable batteries that were required in huge quantities
was very high, which increased the cost of the e-vehicle. This drawback
was addressed with the introduction of the first lead-acid battery by Gaston
Plante. The introduction of this battery made a revolution in the entire world
and many countries started to produce electric cars. And the U.S.A. was the
first to invent the six-seater car with 22 kilometers per hour speed.
Non Commercial Use

Today we refer the Electric vehicles as the transportation of the future due
to their efficiency and minimal carbon emission. This was the same situation
with electric vehicles in comparison with engine-based automobiles at the
commencement of the 20th century. The first motor for running the electric
vehicles was prototyped in 1828 by Anyos Jedlik while a model for the
vehicles was conceptualized by Homes, Davenport in 1834 [1]. In that era,
three alternatives had the same probabilities for development and use in the
future. Out of three, steam engines were fast and cheap, but they required a
lot of time to prepare before driving, also it is needed to stop to get water.
The gasoline-powered vehicle was fast with a satisfactory covering distance
but their problem was noise, complication, and too many parts. Electric-
Author Copy
powered vehicles were not able to match the speed of combustion engines
but were environment friendly, comparatively silent, and handling was also
easy. It has been traced and surprising that New York, Boston, and Chicago
had one-third of vehicles electric powered in 1904 [2]. But in the 1920s
petrol engine vehicles replaced the e-vehicles because the cost of the electric
vehicle was very high, the travel speed of the e-vehicle was only around
20–30 miles per hour, which was very less, and with the development of the
roads, fast vehicles were the requirement [3]. Also, the development of an
electric self-starter to start the engine and the abundance and low prices of
gasoline led to gain the mass market of gasoline-powered vehicles.
The second popularity period of the electric vehicle was when environ-
mental issues gained interest around the year the 1960s and were shadowed
by the oil emergency in the year 1973. During this time most electric vehicles
were developed for government institutions [4].
The third admiration wave for electric vehicles started around the 1990s
due to environmental awareness, pollution after-effects, and development in
e-vehicle technology which is continued till today. Across the globe, several
governments regulation has been passed on reducing the use of fossil fuels
owing to its environmental impact and scarcity of fossil fuels. The electric
vehicle seems to have become a trend, showing its interest in environmental
problems. Every company is developing and in process of launching more
and more electric vehicles in different segments such as 2 wheeler, 3 wheeler,
and 4-wheelers [3]. As the concern of environmental issues and oil depletion
remains as severe as ever, governments are proactively considering different
other options to old-fashioned transport technologies.
Battery electric vehicles (BEVs) are now being considered an encouraging
option that has the potential for the vehicle independence from oil. The
principle of operation is simple. Here the internal combustion engine and the
Non Commercial Use

tank of a conventional vehicle are replaced by the electric motor powered by

a battery. When not in use the battery of the vehicle is charged by plugging
into a charging spot [5–7]. But the challenges in BEVs are the technology
readiness of the battery and the energy storage systems. Some of the popular
battery technologies have been discussed below.

Lead-acid batteries are well-known batteries and their improvement
potential is low. It has low specific energy (20–40 Wh/kg) compared to
Lithium-ion batteries. Typically to achieve a particular range for example, 200
km the weight of led acid batteries is three times more than the Lithium-ion
batteries. This has rendered the lead-acid batteries not being used for the
electric vehicles, particularly for the long-range [8, 9].
Author Copy
The other popular battery is Nickel Metal Hydride Battery. This is now
being adopted for Hybrid vehicles by Toyota in their Prius model. This is
less costly compared to a lithium-ion battery [10]. Nickel Metal Hydride
Battery is considered an established technology and has achieved its greatest
prospective in reduced cost reduction and operation characteristics. Its energy
density is between 60 and 80 Wh/kg and is insufficient for BEVs [11].
To date, in the ever-changing battery technologies, Lithium-ion batteries
are being adopted as the supreme encouraging option for future use in BEVs.
Therefore, Lithium-ion batteries are researched and studied in detail since it
has the highest electrochemical potential resulting in high energy density and
low equivalent mass [12, 13]. It has high effectiveness and a longer lifecycle,
and its potential to further improve has led it to be used virtually for all the
vehicles [14, 15]. There are certain disadvantages of the lithium-ion battery
as well like it is expensive and it presents safety issues. Overcharging may
lead to fire and destruction. The energy density is still not sufficient as per
the requirement [16–18].
Another popular option for battery in research is the Sodium Nickel
Chloride battery. This battery has many advantages. It is considered to have
a longer life span and a less costly and safer option and also can be drained
entirely without degrading its life prospect. Though the specific energy it has
(less than 150 W/Kg) is satisfying, because of the lower specific power, it is
not a good option to power BEVs. But efforts are being made to associate it
with other sources like supercapacitors which can make it usable to power
electric vehicles [19].
Improvement in battery technology can also come from a battery
management system (BMS) that manages the use of batteries in different
situations also known as BMS. The BMS has two major roles to play.
Firstly, it optimizes the charging and discharging behavior of the battery, and
Non Commercial Use

secondly, it controls the battery with safety and efficiency [20, 21]. Studies
are being done especially in the area of improving battery efficiency through
BMS.
13.3 PRESENT SCENARIO
As a formal definition, we know these EVs are run by rechargeable batteries

which can easily be charged by external sources of electricity. For EVs, we
need separate supply equipment, which is formally known as EV Supply
Equipment or EVSE. This entire EVSE system consists of a separate power
Author Copy
source for an EV for providing an easy charging facility, network connectivity,
smart metering, etc. The charging of EVs is simple in terms of the process
however it requires thorough planning and procedure for installation to be
used by the public as there should not be any hazard due to mishandling of
charging equipment. The charging stations may be supplied by conventional
electricity or suitable renewable energy such as wind, and solar. With the
decresing cost of photovoltaic cells day by day the cost of solar-based
charging is going to reduce shortly and viable which will make the charging
independent of fossil fuels. This is the promising option being studied and
implemented by many countries and the lead taken by the Netherlands.
Nowadays the electric vehicle industry is becoming a center of the new
manufacturing hub, with added advanced technologies in BMS, and its control
mechanism. All over the world Government is giving different subsidies and
technological support for advancements in this new era of industry. India
with the added advantage of being the world’s largest producer of electric
vehicles has proactively taken an initiative to produce the batteries of global
standard which will be the other growth drivers for the Indian EV industry.
Recently, in September 2021, the cabinet approved the production linked
incentive scheme which will aid in supporting the automotive industry in
manufacturing electric vehicles and hydrogen fuel cell vehicles. The world-
wide conventional automotive industry is experiencing an exemplar shift to
switch to alternative smart vehicles with higher efficiency, which is nothing
other than electric vehicles. In the list of all those countries, India is also
serving a prominent role in the electric mobility shift.
Figure 13.3 indicates the continuously rising demand for electric cars
owing to the different reasons, possibly the prominent one is rising pollution
due to conventionally powered vehicles. In the last few decades, the burden
of rising pollution, greenhouse effects, and global climate changes has
perpetually become major factors that motivated the rapid transition from
Non Commercial Use

traditional automotive vehicles to electric mobility. This has created a global

awareness of a sustainable solution that can take care of these factors. A
combination of Electric-powered vehicles with renewable energy-based
charging infrastructure is one of the most sought-after alternatives being
researched by researchers across the globe. There are countries especially in

Europe using this method thereby trying to reduce the carbon emission to a
net-zero level.
Author Copy
FIGURE 13.3 Global electric car fleet.
Sources: EV Volumes and Global EV Outlook 2021.
As per Table 13.1, we can see the promises being offered by electric
vehicles over the conventional ones.
13.4 TYPES OF EVS
Broadly if we divide the EVs based on their construction, they can be of

three types:
1. Battery Electric Vehicles (BEVs): These are powered only by a
battery. This battery is recharged by an external source of electricity
that may include plugging into conventional electricity or through
renewable energy-based charging stations adding a further advantage
to vehicular systems. Most electric vehicles including two-wheelers
are BEVs.
Non Commercial Use

2. Hybrid Electric Vehicles (HEVs): Hybrid cars have a battery and

a combustion engine. In this case, both the IC engine and electric
propulsion system work together to add efficiency to the existing
system. The regenerative braking process used in EVs enhances the
efficiency of the entire vehicular system. So, that energy will be

stored in the battery for future use.
3. Plugin Hybrid Electric Vehicles (PHEVs): Here, both the internal
combustion engine and electric battery are present. The battery plugin
cables with the vehicle can be attached to an outside electric power
source to recharge the battery. Similar to other electric vehicles,
PHEV reduces greenhouse gas emissions. Plugin Hybrid vehicles are
Author Copy
making it possible to reduce emissions to a great extent by tapping
on the advantages of both conventional and electric vehicles.
TABLE 13.1 Comparison of Conventional Vehicle and Electric Vehicle

SL. Conventional Vehicle Electric Vehicle
No.
1. Power is generated by conventional IC In Electric vehicles, electric motors are used
engines. in place of IC engines.
2. The efficiency of the IC engine vehicle With the use of electric motors, the
is approximately 30%, which is very efficiency of EVs has reached up to 85%
low and makes its running cost high. resulting in lower running costs.
3. It emits greenhouse gases, which cause By using electric energy, there is no harmful
global warming, and nowadays it is a emission.
severe issue.
4. One of the main reasons behind lower Using a regenerative braking mechanism, the
efficiency is the loss of braking energy, braking energy can be restored and reused.
which cannot be restored for future use.
5. IC engine vehicle uses complex gear Electric Vehicle employs electric motor.
systems.
6. High maintenance cost. Low maintenance cost as it uses fewer
components.
7. Reuse after its intended life is costly. The reuse of batteries makes it a viable option
and increases the total lifetime of the vehicle.
13.5 CONSTRUCTIONAL DETAILS OF ELECTRIC VEHICLES
The electric vehicle consists of an electric traction motor, charging port,

onboard charger, DC/DC Converter, Thermal system, Traction battery pack,
transmission, BMS, etc.
Non Commercial Use

If we think about, electric traction motors, we use motors of high

efficiency like BLDC, Induction motors, etc., where the efficiency goes up
to 88%. These motors use power from the traction battery pack, which stores
electrical energy. Electric vehicles have an added advantage of a charging
port, with the help of which we can charge the battery using an external
power supply. The onboard charger used in the electric vehicle helps in
charging the traction battery. DC/DC converter helps in power modulation
and here, it takes the power from the traction battery pack and converts it
into the required amount for the accessories of the electric vehicle. A thermal
system helps in maintaining the proper temperature of the electric motor,
battery pack, power electronics devices, and other components of an electric
Author Copy
vehicle. A transmission transfers mechanical power from the motor to the
driving wheels. The block diagram of a typical basic electric vehicle with
different components is shown in Figure 13.4.
FIGURE 13.4 Block diagram of electric car.
As the electric vehicle operates at high voltage there lies a risk of what
may be a fail-safe condition? Therefore, a special management system is
required to control the operation of the battery arrangement in the electric
vehicle which is referred to as a BMS. The primary purpose of the BMS
is to continue the workflow with the battery and also to enhance the safety
limit of the entire system. Another important role of the BMS is to provide
optimization over battery operation, which in turn enhances the life of a
battery. BMS also monitors each battery cell to reduce voltage fluctuations
within the system and reduce the chance of possible system failure.
The rechargeable battery packs in electric vehicles can be made up of
multiple cells, either connected in series or parallel, based on the require-
ment of the vehicular system. These battery cells can produce thousands of
voltage to supply, which is to be controlled by BMS. The key purpose of the
BMS is to make sure that there is no unbalance in supply voltage. Also, the
BMS provides extra support to ensure zero supply failure.
The primary objective of using BMS is to ensure the safe action of the
battery and also to provide its safety. Apart from this BMS checks the condi-
tion of the charge of the battery pack. BMS also manages cell balancing
followed by its battery optimization which indirectly improves the longevity
Non Commercial Use

of the battery. This BMS also monitors the variations in voltage, different
physical constraints like ambient temperature, atmospheric pressure, the flow
of coolant, etc. Lithium-ion batteries possess a high charge density. Although
these batteries are not so large, still they can be extremely unstable, which
is not desirable. Therefore, it is recommended that these batteries should on

no occasion be charged too much or be permitted to reach a circumstance
of complete discharge on any occasion. When the current is supplied to
charge the batteries to store the energy there may be a situation of rising in
the cell temperature this phenomenon is called a Thermal Runaway. This is
a highly undesired situation in electric vehicles because it reduces the life
of the batteries drastically and sometimes results in the catching of fire. To
Author Copy
avoid such hazardous circumstances an automatic arrangement to monitor
its voltage and current is required. As this is a highly difficult process to
control in which each cell needs to be monitored closely and the action needs
to be taken based on the requirement of the vehicle for safe and effective
operation the specially designed control system called BMS is employed. In
addition, this BMS is required to compute various parameters of the system
to such an extent that the vehicle can be monitored remotely with connected
technologies and potentially unsafe situations can be avoided.
13.6 FUTURE SCOPE AND LIMITATION
India comes the fifth number in the car market in the world. It is estimated
that by 2030 around 35 crores of consumers will need a promising solution
for conveyance or transport. But considering the objectives set under the
Paris treaty, the growing number of customers in the mobility domain shall
not demand an increase in the ingestion of conventional fuels.
This indicates that to reduce the carbon footprint by reducing fossil fuel
consumption, electric mobility is one of the promising technology for the
future.
With more and more renewable energy being used for electricity genera-
tion across the globe in terms of wind, solar, etc., it has grasped the attention
of many researchers and practitioners. As shown in Figure 13.5 the advance-
ment in smart grid technology is helping to harness the renewable energy
along with the conventiontional energy in much more effective way as the
former is being the point of attention for all. A lot of effort is being put on
distributed generation (DG) system with major focus on solar generaton in
the countries where solar power is available for maximum time in the year.
Non Commercial Use

Author Copy
FIGURE 13.5 A schematic view of smart grid with distributed generation technology.
Source: Ref. [22].
To ensure achieving India’s Net Zero Emissions by 2070 as per the

government, a huge transport restructuring is required. The present situation
is demanding superior and efficient transport for the public which includes
roads, and railways. The efficient transport is already adopted as electrically
driven by railways and now it’s the time for the electric-powered cars. In
this regard, the government of various countries has come up in support of
growing the use of electric-powered vehicles by reducing duties and taxes.
The changed and more encouraging faster adoption and manufacturing of
electric vehicles (FAME II) scheme in India is an example of the same. With
the promises of electric mobility as above there come some associated chal-
lenges which need to be addressed to make it successful.
The first challenge is the nonexistence of an industrial base for battery
manufacturing in India, therefore to meet the rising demand of the electric
vehicle market it is hugely dependent on the import from outside the country.
The import bill related to electric vehicles, especially for batteries in the
year 2021 was more than 1 billion dollars as per the government sources. As
Non Commercial Use

compared to the demand and future projections there is tiny penetration of

electric vehicles. Moreover, this dependence will continue as the availability
of lithium and cobalt which are the most important constituents for battery
production. In addition, there are no clear guidelines to end of life of the
batteries which will be a quite serious issue in the coming future.

The second challenge is charging infrastructure, which is quite less, than
the immediate counterparts who already had a huge number of charging
stations. The lack of charging stations in different states in India is a big distrust
for the consumers to buy electric vehicles as they are not able to achieve the
range presently. Further, it takes long hours up to 8 hours for a complete
charge of a vehicle through the home socket using a company-supplied slow
Author Copy
charger. And above this, the cost of cars running on conventional fuel is
much less than the comparable electric cars of the dame segment. The cost
factor is comparable for not only the cars but the electric two-wheelers also
which come out to be nearly double the cost of conventional two-wheelers.
Recently renewable energy-based charging systems especially solar are
being installed in different areas to make the charging of electric vehicles
independent of the conventional grid.
The third challenge is Policy Challenges, As EV production requires huge
capital and long-term preparation to realize the profits. This lack of govern-
ment on the policy front with the absence of clear guidelines, processes, and
governance is resulting in the discouragement of new ventures to enter the
market. The policies regarding electric mobility are still in nescent phase and
are being developed with the time. Europian countries are ahead with clear
policies and system for the electric vehicles as they have started adopting
them early and contributing in reduction of carbon emission in a huge way.
The fourth challenge is the availability of Skilled Labor and technology.
India needs self-sufficient production of electronics which still is a far-fetched
dream. Electronics being in the heart to control and monitor the entire function
of any electric vehicle, require huge and proactive initiatives for the in-house
production of semiconductors. In addition, there is a huge demand for the
skilled workforce in the EV domain but still, no dedicated course as of now
is available in this segment. The only availability is conventional engineering
courses which do not touch upon the electric vehicles in detail. Surprisingly
there is a lack of laboratories related to hands-on learning about electric vehicle
technologies even with the ranked and accreditated institutions of repute.
With the above discussion, it is evident that there is a enormous scope
for electric mobility in the coming decades provided the challenges are met
head-on to make it the mobility of the future. Also with such initiatives the
Non Commercial Use

major factor for the carbon emission that is transport sector accounting for
27% of total emission can very wellbe reduced as a result it will contribute
towards the net zero emission goals of different nations.
13.7 CONCLUSION
With the above discussion on electric-powered vehicles being the possible

solution for the new age mobility solution, there are still continuous enhance-
ments being worked upon to make them more attractive to the masses in
terms of features, efficiency safety, and handling. Such enhancements
Author Copy
include ADAS, connected vehicle technology, etc.
Interestingly the electric mobility first was seen with the three-wheelers
being used for the public transport in India which percolated to the remote
areas owing to its advantages for the owners. Most of these three-wheelers
are custom-made and use lead-acid batteries which results in a reduction in
the price. Further, the electric power was seen with the four-wheelers that
are cars and which aligned with the subsidy scheme of the government. Day
by day the improvement is happening in electric cars and many automo-
tive companies are considering switching and producing electric-powered
cars with different capacities. Last but not least all the major two-wheeler
(e-scooters) manufacturers are now developing different models due to
rising demand as per the requirement of the masses which may result in
reducing vehicular pollution to a large extent. The sales of electric two
wheelers accounted for the 64.29% of the total electric vehicles in March
2022 [23]. Electric-powered buses are also on the card to manufacture in
India for which many joint ventures are working.
The efforts are being made by the automotive industry and it will require
some time for the consumer to get confidence in electric-powered vehicles
and surpass the advantages of conventional vehicles in safety and efficiency.
The government of India also taking proactive measures towards encouraging
the transformation of the transport scenario with electric vehicles to reach
70% for commercial cars, 30% for private cars, 40% for buses, and 80% for
two- and three-wheelers by 2030.
A great deal of the achieving the goals for a greener future by transforming
the transport depends largely on the clear policies by the government,
technological breakthroughs in research and development towards battery
technologies, charging infrastructure, reuse of batteries, safety assurance,
and at par vehicle cost with faster adoption by the masses. This can surely
create a better future for all.
Non Commercial Use

KEYWORDS
• connected vehicle technology

• conventional vehicles
• electric mobility
• electric vehicles
• e-scooters
• safety assurance
• vehicular pollution
Author Copy
REFERENCES
1. Jones, W. D., (2003). Hybrids to the rescue [hybrid electric vehicles]. IEEE Spectrum,
40(1), 70, 71.
2. Curtis, D. A., & Judy, A., (2010). Electric and Hybrid Cars: A History (2nd edn., p. 257).
McFarland & Company, North Carolina.
3. Jones, W. D., (2005). Take this car and plug it [plug-in hybrid vehicles]. Spectrum, IEEE,
(Vol. 42, No. 7, pp. 10–13).
4. Wager, G., Whale, J., & Braun, T., (2016). Driving electric vehicles at highway speeds:
The effect of higher driving speeds on energy consumption and driving range for electric
vehicles in Australia. Renew Sustain Energy Rev., 63, 158–165. doi: 10.1016/j.rser.2016.
05.060.
5. Al-Alawi, B. M., & Bradley, T. H., (2013). Review of hybrid, plug-in hybrid, and
electric vehicle market modeling studies. Renew Sustain Energy Rev., 21, 190–203. doi:
10.1016/j.rser.2012.12.048.
6. Hannan, M. A., Azidin, F. A., & Mohamed, A., (2014). Hybrid electric vehicles and
their challenges: A review. Renew Sustain Energy Rev., 29, 135–150. doi: 10.1016/j.
rser.2013.08.097.
7. Arora, S., Shen, W., & Kapoor, A., (2016). Review of mechanical design and strategic
placement technique of a robust battery pack for electric vehicles. Renew Sustain Energy
Rev., 60, 1319–1331. doi: 10.1016/j.rser.2016.03.013.
8. Miller, J. M., (2009). Energy storage system technology challenges facing strong hybrid,
plug-in, and battery electric vehicles. Veh Power Propuls Conference 2009: VPPC ’09
(pp. 4–10). IEEE. doi: 10.1109/VPPC.2009.5289879.
9. Offer, G. J., Howey, D., Contestabile, M., Clague, R., & Brandon, N. P., (2010). Comparative
analysis of battery-electric, hydrogen fuel cell, and hybrid vehicles in a future sustainable
road transport system. Energy Policy, 38, 24–29. doi: 10.1016/ j.enpol.2009.08.040.
10. Burke, A., Jungers, B., Yang, C., & Ogden, J., (2007). Battery electric vehicles: An
assessment of the technology and factors influencing market readiness. Public Interest
Energy Res. Progr. Calif. Energy Comm., 1–24.
Non Commercial Use

11. Miller, J. M., Bohn, T., Dougherty, T. J., & Deshpande, U., (2009). Why hybridization
of energy storage is essential for future hybrid, plug-in, and battery electric vehicles. In:
2009 IEEE Energy Convers. Congr. Expo. ECCE 2009 (pp. 2614–2620). doi: 10.1109/
ECCE.2009.5316096.
12. Li, Y., Yang, J., & Song, J., (2017). Nano energy system model and nanoscale effect of
graphene battery in renewable energy electric vehicle. Renew Sustain. Energy Rev., 69,
652–663. doi: 10.1016/j.rser.2016.11.118.
13. Sabri, M. F. M., Danapalasingam, K. A., & Rahmat, M. F., (2016). A review on hybrid
electric vehicles architecture and energy management strategies. Renew Sustain. Energy
Rev., 53, 1433–1442. doi: 10.1016/j.rser.2015.09.036.
14. Deberitz, J., (2006). Lithium Production and Application of a Fascinating and Versatile
Element (2nd edn.). Verlag Moderne Industrie.
15. Scrosati, B., & Garche, J., (2010). Lithium batteries: Status, prospects, and future. J
Author Copy
Power Sources, 195, 2419–2430. doi: 10.1016/j.jpowsour.2009.11.048.
16. Riba, J. R., López-Torres, C., Romeral, L., & Garcia, A., (2016). Rare-earth-free
propulsion motors for electric vehicles: A technology review. Renew Sustain. Energy
Rev., 57, 367–379. doi: 10.1016/j.rser.2015.12.121.
17. Tahil, W., (2008). The Trouble with Lithium 2 Under the Microscope, Meridian
International Research, Les Legers, 27210 Martainville, France, pp. 39–48. https://
d1wqtxts1xzle7.cloudfront.net/60272860/Lithium_Microscope20190812-6534-
jkxysn-libre.pdf?1565643520=&response-content-disposition=attachment%3B+fi
lename%3DThe_Trouble_with_Lithium_2_Under_the_Mic.pdf&Expires=168976
3676&Signature=K8xWiyOgSpfiCa2io0sdLz3BNSQAHBtHe5XttVbclWr2S7VM
Z81kjbpWlfvByGAn4OAzjvZYTaGABP1~4sADfUoLRpgPR9DUt~VLZLJWV
RARs3ut4-KLvgQox6pVvRAfubLOc~63bz0IPacMq02Ck~JMQIUTKnWpV6mrt
TydApWsHzAv-BpZ8crfw9RHqs~k28kZ6PkVyAmkpJYrBAJ8v5eNF7q3URWH
vxvaFCaMFCR3f23spS1sfO-AIG2ZFwGPya5PdFdHlVfxWRDIbqPIZcO1lcU2Z
AivXT1LAyl2moNL1H~lv8kTSTlcYYMW2RIXD4k40VoZ75rQL5CtC3wrkA__
&Key-Pair-Id=APKAJLOHF5GGSLRBV4ZA.
18. Amiribavandpour, P., Shen, W., Mu, D., & Kapoor, A., (2015). An improved theoretical
electrochemical-thermal modeling of lithium-ion battery packs in electric vehicles. J.
19. Li, Y., Yang, J., & Song, J., (2015). Electromagnetic effects model and design of energy
systems for lithium batteries with gradient structure in sustainable energy electric
vehicles. Renew Sustain. Energy Rev., 52, 842–851. doi: 10.1016/ j.rser.2015.07.155.
20. Bauer, S., Suchaneck, A., & Puente, L. F., (2014). Thermal and energy battery
management optimization in electric vehicles using Pontryagin’s maximum principle. J.
21. Zhang, X., Mi, C. C., & Yin, C., (2014). Active-charging based powertrain control in
series hybrid electric vehicles for efficiency improvement and battery life extension. J.
22. Eolas Magazine, (2018). Smart Grid Evolution. Available online: https://www.eolasmaga-
zine.ie/smart-grid-evolution/ (accessed on 12 June 2023).
23. https://evreporter.com/indias-electric-vehicle-sales-trend-march-2022/ (accessed on 12 June
2023).
Non Commercial Use

Non Commercial Use

Author Copy
CHAPTER 14
Power Loss Reduction and Voltage

Improvement Through Capacitors and
Their Optimization for the Distribution
System
Author Copy
SHUBASH KUMAR,1,2 CHANDAR KUMAR,2 MUHAMMAD SUHAIL SHAIKH,3
ANWAR ALI SAHITO,4 and ZAHID ALI ARAIN2
1
School of Electrical Engineering, Yanshan University, Qinhuangdao,
P. R. China
2
Department of Sciences and Technology, Indus University, Karachi,
Sindh, Pakistan
3
Guangdong, China
4
Department of Electrical Engineering, Mehran University of Engineering
and Technology, Jamshoro, Sindh, Pakistan
ABSTRACT
To offer customers a constant power supply, a robust electrical power

distribution system is necessary to be in place. The fluctuation in voltage
is present in practically every distribution system, resulting in a significant
voltage drop at the consumer’s point of use. It is a prerequisite of a good
distribution system that consumers be given a sufficiently consistent voltage
for varied loads to operate satisfactorily. The modern usage of electronic
ballast and energy conservation measures have modified the load pattern of
business and residential users, which lowers the power factor to their desired
Non Commercial Use

value, because the load on the distribution system is primarily inductive and
requires a lot of reactive power, most power quality issues can be handled by
controlling the reactive power. The installation of capacitors in the distribution
system is widely used to accomplish reactive power compensation. According
to studies, the distribution system wastes around 13% of the generated power
due to Ohm’s losses (I2R) losses. High technical losses are being experienced,
owing to an aging network and poor operating circumstances. The low power
factor is one of the main causes of voltage variation and power loss in the
distribution system. Voltages are given to consumers that are considerably
below the lower permissible limits. In this chapter, the causes of the poor power
factor, power losses, and voltage drop has been discussed. Different techniques
Author Copy
have been used to minimize the above problems but optimal placement of
capacitors in the utility network using the latest optimization technique. In
this chapter, power capacitor compensation has been proposed and discussed
briefly their advantages and optimal location in the distribution system. In the
end comparison of the latest optimization techniques used for losses and voltage
stability by optimal placement of the capacitor in the distribution system is
also discussed. Installing capacitors and applying optimization techniques for
appropriate placement in the distribution system, resulting in enhanced voltage
profiles and reduction in power losses in the distribution system, can improve
the stability and reliability of the power distribution system.
14.1 INTRODUCTION
The distribution system is the last step before electricity can be delivered to
end-users, in any power system. Distribution feeders provide electricity from
a distribution network to various end-users. It lowers the sub-transmission
voltage to the distribution voltage level with this device (11 kV). The voltage
of an 11 kV feeder is stepped down for different applications. Electricity
is supplied to utility customers via the secondary side of the distribution
transformer. Technical and non-technical losses can both occur in a distribu-
tion system. According to research, the distribution network wastes about
13% of electricity generated in the system. Because of the increased voltage
drop along any line or transformer, as the system load grows, voltage drops
at consumers’ terminals [1]. Reduced consumer voltage has an impact on
the performance and lifespan of consumer appliances [2]. It also raises
the current required to operate loads, resulting in greater power losses.
Distributor efficiency has a direct impact on both consumer appliances
and power system economics [3]. A better understanding of voltage drop
Non Commercial Use

Power Loss Reduction and Voltage Improvement 269
and power losses due to a low load power factor will enhance the system’s
overall performance and dependability various methods have been employed
to solve these issues, including
Network reconfiguration, DG placement, and capacitor allocation. In the
distribution system, shunt capacitors are used to reduce power loss, improve
the voltage profile along with feeders, and increase the life of the equipment
used for the distribution system. Power distribution networks often have
shunt capacitors built to compensate for reactive power loss and decrease
it. In distribution networks, however, shunt capacitors must be placed in
the proper position and of the right size. When correctly installing shunt
capacitors, capacitor location is critical to reducing loss as much as possible
Author Copy
while keeping shunt capacitor prices low [4]. Power system architects and
academics have had a difficult time figuring out where and how big to make
shunt capacitors in distribution networks. There were several academic
studies and publications on reactive power planning with optimization can
be found in the literature. In Ref. [5], load flow analysis of the distribution
feeder before and after installation of DG in the power loss reduction. In
Ref. [6], a hybrid shuffling frog jumping and particle swarm optimization
is given for the improvement of voltage profile and loss reduction in the
radial distribution system. In Ref. [7], the author offers binary particle
swarm optimization and shuffled frog leap (BPSO-SLFA) algorithms for
optimal distributed generation (DG) placement in the radial distribution
system to enhance voltage profile and power loss reduction. In Ref. [8], the
proper placement of capacitors and PV systems with geographical location
constraints is necessary for active power loss reduction in the radial distribu-
tion system. In Ref. [9], a hybrid optimization design was used to determine
the best capacitor location and reconfiguration to improve the distribution
system’s performance. Firefly optimization is utilized in a radial distribution
system to discover the ideal capacitor location and size [10].
This challenging combinatorial issue has been solved using a variety
of methodological approaches. The problem of capacitor position and size
is made considerably more complicated by imbalanced operating circum-
stances and the existence of harmonic sources in distribution systems. Many
capacitor optimization studies have neglected to take these considerations
into account. To solve capacitor placement problems, numerous different
strategies have been developed, including analytical methodologies and
numerical programming, as well as heuristic or AI-based solutions [11].
In this chapter, a variety of strategies have been used to tackle the
optimum capacitor placement problem, including heuristics and artificial
Non Commercial Use

intelligence (AI). To deal with the nonlinearity of practical systems, AI is a

potent knowledge-based technique. AI can simplify mathematical problems
and respond quickly, making it useful for transitory analyzes like capacitor
allocation and DG placement.
14.1.1 POWER LOSSES
Transformers, cables, overhead lines, and other infrastructure are used to

transference electrical power from power plants to homes, businesses, and
other consumers of electricity. To put it another way, the amount of electric
Author Copy
energy created does not equal the amount of electric energy provided to
consumers. There is a loss of a particular number of units. Transmission
and distribution losses are defined as the difference between produced and
dispersed units. The following categories can be used to categorize power
system losses.
14.1.1.1 TECHNICAL LOSSES
The energy dissipated in the power system’s conductors and equipment

results in technical losses. The primary and secondary power lines of the
electric grid have suffered significant technical losses. This means that the
primary and secondary distribution networks must be well-designed or func-
tion within acceptable parameters. Technical losses have been higher than
usual due to unanticipated increases in load. It’s the system’s current and
Line characteristics that are to blame for the failure. Ohm’s losses are the
most common technical losses in transformers and conductors. Transformer
losses also include losses from irons. Design maintenance and operation can
help decrease technical losses, but they cannot be eliminated.
Technical failures can be caused by a wide range of circumstances. The
overload, imbalance, and the lower power factor are the most important
effects of a bad distribution system. Other factors to consider are the
length of the feeder, the age of the distribution system, and the size of
the conductor. The problem begins with poor design and planning, as well
as an incorrect load estimate. Then there’s the issue of mishandling and
neglecting the system altogether. Penalties for new installations will be
applied before those for existing ones, according to the current network
load flow. Loads are carefully monitored, but they aren’t stored. There’s no
regular upkeep going on. Due to the lack of a current maintenance schedule,
Non Commercial Use

burning transformers and line losses result in large losses. When a problem
arises, the power supply must be restored and repair methods that are
below par must be employed. It’s the result of poor system planning and
maintenance that the distribution network is in such poor shape, to begin
with. There are a few instances of distribution system components that are
in a terrible condition of repair including the old network, large feeders, an
inappropriate size of conductors, improper earthing or grounds, and poor
quality of the material.
14.1.1.2 NON-TECHNICAL LOSSES (NTLS)
Author Copy
NTLs occur in a power system when the computation of external actions and
the load and circumstances are not taken into consideration. NTL’s customer
management systems and awareness of the program may contain several
fraudulent automobiles. Theft of electricity is a primary cause of NTL.
14.2 VOLTAGE REGULATION
Transmission and distribution systems deliver energy to end-users in an

electrical power system. Almost all distribution circuits have some degree
of voltage fluctuation. Lighting, appliances, motors, and other loads require
a somewhat steady voltage supply to operate properly [3]. As a result, the
system as a whole operates at peak efficiency. Voltage regulation is defined
as the difference between the voltages at the sending and receiving ends of a
circuit. The variations should not exceed ±6% of the time.
14.2.1 CAUSES OF VOLTAGE VARIATIONS
There is a direct correlation between the amount of load placed on the supply
system and voltage at the customers’ terminal as well as the voltage drop
along with the distribution network. When a system’s load drops, the voltage
at the consumer end rises because the voltage loss is reduced.
Voltage regulation requirements are described as: There is a ±6% need
for a 120 V or 230 V system and a –2.5% to a + 6% proposed regulation for
voltages more than 600 V [1].
Short intervals or unusual situations have an authorized upper limit of
–8.3% to +5.8%. These circumstances are not explained in any way. Another
Non Commercial Use

definition of voltage regulation is “the difference between a device’s no-load

and full-load output voltage and the full-load output voltage.”
Vnl −VRated
Voltage Regulation
= (VR ) ×100% (1)
VRated
where; Vnl is the voltage when no load is applied, and VRated is the voltage
when the load is applied to the device. It is preferable if VR has a lower
value. There will be less fluctuation in the voltage at the consumer terminals
as a result of this. To keep the voltage differential between the transmitting
and receiving ends under control, it is necessary to install voltage regulating
equipment.
Author Copy
14.2.2 IMPORTANCE OF VOLTAGE REGULATION
The supply network’s load changes cause the voltage to fluctuate at the
consumer terminal. Voltage fluctuations can be dangerous, hence they must
be controlled for the following reasons:
1. When the voltage increases by 6%, the illuminating power of a lamp
increases by 50%, but when the voltage declines by 6%, the illumi-
nating power of a lamp drops by 20% [12].
2. Low supply voltage might limit the starting torque of an induction
motor. This can lead to inefficient operation and poor power factor as
well as high magnetizing current.
3. Overvoltage produces extra heating and changes rated appliances
like transformers, etc., to differing levels.
Because of this, voltage variations in power systems must be maintained to
a minimum for better supply services. Transmission and distribution systems
in electrical power systems include significant voltage drops, as well as
varied circuits with load characteristics that differ from generation station to
the consumer [13]. Each circuit must be able to be controlled independently.
If the voltage drop exceeds permissible limits, voltage control equipment can
be utilized at the generating station, substations, and feeders.
14.3 POWER FACTOR
In an AC circuit, the cosine angle between current and voltage is called the
power factor (cos∅).
Non Commercial Use

Active Power kW
Power Factor
= ( Cos∅ ) = (2)
Apparent Power kVA
Inductive loads are primarily responsible for poor power factor, and
poor power factor results in high apparent power need for continuous active
power demand. To high apparent power demand, greater capacity trans-
formers and transmission lines are required. Overall, the cost of the system
components rises as a result of this. PF is stated as a percentage or as a unit
of measurement.
Author Copy
14.3.1 REASONS FOR POOR POWER FACTOR
Knowing that a low power factor means that KW is less than KVA, this
makes sense. Equipment having poor power factor includes pure inductive
circuits, some lighting ballast, Induction motors, high-intensity discharge
lights, transformers, and generators. As a result, inductive loads have higher
reactive and apparent power and poor power factor (also known as efficiency)
since the current is needed to create a magnetic field that does the intended
job. In Figure 14.1, the power triangle is used to explain this impact.
FIGURE 14.1 The impact of increasing the power angle.
14.3.2 IMPACTS OF A POOR POWER FACTOR
A low power factor has several detrimental consequences for the entire
power system, which are detailed in subsections.
Non Commercial Use

14.3.2.1 LARGE KVA RATING OF THE EQUIPMENT
kW
KVA = (3)
Cos∅
KVA is the unit of measurement for electrical equipment. A higher KVA

rating signifies that the KVA ratings of the equipment must be higher to
fulfill the system requirements for the same KW demand [1]. It raises the
price of the equipment and its associated accessories by a significant amount.
Author Copy
14.3.2.2 GREATER CONDUCTOR SIZE
As power consumption rises, the conductor size needs to be increased.

Increased current flow in a system or piece of equipment due to a low power
factor is not natural. Conductor size is therefore an additional factor that
contributes to increased costs under these circumstances.
14.3.2.3 I2R LOSSES
Higher I2R losses result from increased current flow, which reduces the
overall efficiency of the power system [14].
14.3.2.4 LOW VOLTAGE REGULATION
The low lagging power factor causes a big voltage drop, which in turn causes
a larger voltage drop. Low lagging power factor To keep voltage drops within
a certain range, voltage regulators must be installed.
14.3.3 TECHNIQUES OF IMPROVING SYSTEM POWER FACTOR
Power capacitors, one approach used to increase system power factor, will
be addressed in subsections.
14.3.3.1 CAPACITORS
Using capacitors prevents inductive loads from causing trailing power.

Figure 14.2 shows how this works quite clearly. Because inductive loads use
Non Commercial Use

lagging current and capacitance consumes leading current, adding capaci-

tance balances out the effects of inductive loads overall.
Capacitance
(KVAR)
Working Power
Author Copy
(KW)
Reactance
(KVAR)
FIGURE 14.2 Impact of capacitors.
14.3.3.2 SHUNT CAPACITORS
The addition of shunt or parallel capacitor banks improves the power factor of
the system [15]. Figure 14.2 illustrates how to shunt capacitors may enhance
the power factor.
Let the feeder’s impedance is, if Z = R + jKL
Then Current IS = IR
From the phasor diagram shown in Figure 14.3(b).
VD = IR + IX XL (4)
here; XL is the feeder reactance; and R is the feeder resistance. IR and IX
represent the real and reactive parts of a current, respectively.
Referring to Figure 14.4(a), when shunt capacitors are installed, the
leading capacitor current balances out the trailing inductive current, resulting
in a reduction in total current. Figure 14.4(b) depicts this. The low voltage
drop from permissible current indicates a higher power factor and better
voltage management. However, a shunt capacitor has limitations and cannot
boost the system power factor any further.
Non Commercial Use

Author Copy
FIGURE 14.3(a) The feeder circuit without capacitor.
FIGURE 14.3(b) The Phasor diagram from Ref. [16].
FIGURE 14.4(a) The circuit of a feeder with a capacitor.
Non Commercial Use

Author Copy
FIGURE 14.4(b) The phasor diagram.
Source: Ref. [16]
Here’s how to figure out the voltage drop now:

VD = IRRR + IX XL – IC XC (5)
here; the reactive component is IC, and the voltage rise caused by the instal-
lation of a shunt capacitor may be expressed b.
VD = IC XL
14.3.3.3 BENEFITS OF SHUNT CAPACITORS
Shunt capacitors have the following benefits in a power system:

• High efficiency because of the little loss of energy during the operation.
• Because of the aging and insulation concerns, the reduced heating
effect helps extend the life of machinery.
• There is a reduction in demand (KVA) for a specific active power
(kW).
• At greater power factors, energy dissipation (i.e., arcing time) is reduced,
which reduces switchgear problems.
• Improved loading results and machine life.
• Improved control of voltage
• Generators with reduced excitation losses
• Generator capacity may be expanded without requiring further improve-
ments to meet rising demand.
Non Commercial Use

14.3.3.4 LOCATION OF SHUNT CAPACITORS
Capacitors must be installed everywhere reactive power needs are high to

meet the system requirement. Capacitors should be installed close to the
load since reactive power demand is higher there. Areas with high levels of
low voltage are given precedence over those with low levels of low voltage.
Transmission lines, distribution feeders, distribution and transmission
substations, and the Long Term Network (LT Network) all benefit from
the use of shunt capacitors. As a result, various compensation percentages
are implemented on systems in different countries. For distribution feeders,
shunt compensation is typically 60%, 30%, and 10% of the transmission
Author Copy
system. A variety of techniques for locating the capacitor in an electrical
system are discussed in the sections that follow.
14.3.3.4.1 Capacitor at Load
The capacitor in this manner is linked in such a way that it is energized when
the load is operated [17]. The power factor correction capacitor (PFCC) is
connected to the motor in the manner seen in Figure 14.5.
FIGURE 14.5 Separate compensation via a capacitor.
The following are the advantages of this method:

• It eliminates the need for a transformer and reduces loss;
• The most technically efficient and most adaptable solution;
• The decrease of voltage drop allows for better performance of the
load;
• It is quite simple to install.
Non Commercial Use

14.3.3.4.2 Fixed Capacitor Group
This method is commonly used when a huge number of powerful motors

are readily available, allowing the business to make a valuable contribu-
tion to society. Controlling motors may be accomplished by adding a fixed

capacitance value to the main bus bar control center. In most cases, the fixed
capacitor bank is located near the service switch entrance, making it easy
for customers to operate. To ensure the capacitor bank is safe from harm, a
switch fuse or circuit breaker must always be positioned in front of it.
There will be an increase in the voltage value, which will lead to the
failure of motors, lights, and other controls and devices when placing a fixed
Author Copy
bank on a light load (for example, on weekends or holidays). Unstable loads
or other related circumstances can cause harmonic distortion. In the study’s
conclusions, no more than 20% of the rated power of the kVA transformer
should be in the fixed capacitance kVAR. Unexpectedly high numbers may
have a harmful effect. Listed below are some advantages of this type of
compensation:
• This method has the potential to enhance the power factor every month;
• Constant-load electrical facilities that are always on and drawing power;
• The correction of transformer reactive power;
• Motor compensation differs depending on the motor type;
• A fixed compensation amount is applicable.
14.3.3.4.3 Automatic Capacitor Bank
When reactive power requirements change dramatically over throughout

operation, an automated system is a viable alternative for improving the
power factor of the whole facility’s inductive loads. The automated power
factor capacitor system keeps track of reactive power needs and switches
to the best switching capacitor whenever a high power factor is required.
The capacitance is automatically altered to keep the right power factor in all
working conditions, including fluctuating loads [14].
14.3.3.4.4 Combined Methodology
System integration may also serve as a compensating mechanism. Designing

such a system necessitates thinking about things like cost and accessibility.
Non Commercial Use

14.4 COMPENSATION FOR DISTRIBUTION FEEDER
When choosing a compensation scheme, it is vital to examine the following

factors:
1. The fixed parallel capacitor’s location about the mean reactive load
should be considered.
2. The use of a big capacitor of the same size can be used to replace
more than one capacitor group at the same time. Because of its
economy, fixed capacitors of the same size may be used in a variety
of locations.
Author Copy
3. Fixed capacitors with a high reactive load and the uniform load can
be used across a distance of 2/3 of the feeder carrier length.
14.5 RESULT AND DISCUSSION
An 11 kV feeder from Hyderabad Electric Supply Company (HESCO),

Sindh Pakistan, was used as a case study in Power System Simulator Siemens
Calculation (PSSSINCAL) software to better understand the effect of voltage
variation and power losses on the distribution system. In the program, three
scenarios are created.
i. Existing system;
ii. After installing capacitors on H.T. network; and
iii. After installing capacitors on L.T. network.
The simulation findings reveal that as the number of nodes in the present
system grows, the voltage loss increases. After installing capacitors on HT
and LT networks, the accompanying Figures 14.6(a) and (b) demonstrate a
significant reduction in voltage dips.
Furthermore, the existing system has the highest possible technical losses,
which must be minimized. Capacitors are used in H.T. and L.T. systems, and
they reduce losses significantly. Table 14.1 presents a summary of losses in
the distribution feeder.
The process for estimating economic advantages for existing network
power losses of 724.555 kW, loss reduction, and yearly energy savings is
shown in Table 14.2.
The present feeder of a distribution system is overloaded, as evidenced
by the aforementioned figures and statistics, and has a maximum voltage
drop in different nodes, as well as larger power losses. The voltage profile
Non Commercial Use

of the feeder is improved and power losses are reduced once capacitors are
installed in the feeder (Figure 14.7).
Author Copy
FIGURE 14.6(a) H.T. node voltages.
FIGURE 14.6(b) L.T. bus voltages.
Non Commercial Use

TABLE 14.1 Summary of Power Losses

Description Existing Power Losses (kW) Power Losses (kW)
Power Losses After Installing After Installing
(kW) Capacitors on H.T. Capacitors on L.T.
Power losses (H.T.) 311.812 190.741 174.892

Power losses (transformer) 74.281 71.061 51.987
Power losses (L.T.) 338.462 301.459 221.962
Power losses (Total) 724.555 563.261 448.481
TABLE 14.2 Economic Benefits of Capacitors
Author Copy
Description After Installing After Installing
Capacitors on H.T. Capacitors on L.T.
Power losses (kW) 563.261 448.481
Power loss reduction (kW) 161.294 276.074
Load factor 0.5 0.5
Loss factor 0.325 0.325
FL = 0.3 * L.F + 0.7 * L.F 2
Annual energy savings (kwh) 459204.02 785982.68

ES (Annual) = PL * 8760 * FL
FIGURE 14.7 Total power losses in the feeder of a distribution system.
Non Commercial Use

14.6 OVERVIEW OF OPTIMIZATION TECHNIQUES
The entire power system is divided into three divisions that are generation,
transmission, and distribution. The generated power from different generating
stations is supplied to the transmission line which is fed to a distribution

system at substations. The fluctuation in voltage is present in practically every
distribution system, resulting in a substantial voltage drop at the point of
consumption. In line with the rising voltage drop and thus the rising reactive
power demand, the amount of reactive power needed to maintain the same
level of voltage at the consumer’s terminal increases, leading to the poor
load power factor and consequently increased reactive power consumption
Author Copy
[19]. Approximately 70% of losses occur at the distribution system level,
which comprises the primary and secondary distribution systems with the
remaining 30% happening at the transmission system level, As a result,
distribution systems are currently of primary concern, with losses targeted
at the distribution level amounting to around 7.5% of total losses to reduce
these losses optimal capacitors and DG are located in distribution system
through optimization [20, 21].
Studies show that I2R losses consume roughly 13% of generated power
in the distribution system. Capacitors are commonly used for reactive power
compensation in distribution systems due to their low cost, low repair, and
maintenance requirements, and superior economic efficiency as compared to
other techniques such as reactive power compensators [2]. The Inappropriate
capacitor installation in any part of the power system increases losses and
costs [23]. The distribution system’s capacitor placement and size must be
adjusted for optimal results.
Micro-grids, which are tiny, decentralized power systems, have emerged
in recent years as a new way of thinking about the power system. Many
dispersed and interconnected energy storage and generating units may
be found in a typical micro-grid system. These systems can be islanding
or grid-connected [24]. Poor capacitor placement can cause problems in
systems that use capacitors. Optimal placement of capacitors in islanding
and grid-connected modes for reactive power compensation is necessary for
microgrids [25]. Based on electricity market pricing techniques, a unique
hybrid modified grey wolf optimizer sine cosine and crow search algorithm
(GWOSCACSA) is employed for optimal scheduling of distributed energy
resources (DERs) in microgrid systems [26]. In Ref. [27], the author
presented an oppositional gray wolf optimization (OGWO) for optimal
reactive power when bus vulnerability was taken into account. In Ref. [28],
author’s suggestion for the solution of voltage constrained reactive power
Non Commercial Use

planning (VCRPP) in the power system, the Ameliorated Harris Hawks

optimization (AHHO) and Harris Hawks optimization (HHO) algorithms
were used. The nature-inspired different optimization algorithms proposed
in Ref. [29] Oppositional Harris Hawk Optimization (OHHO) are found to
deliver higher performance on the proposed test system for reactive power
planning in power systems. In Ref. [30], author proposes a Sine-Cosine
optimization algorithm for power loss reduction of transmission lines of the
power system. In Ref. [31], the author provided a study on the HHOPSO
algorithm’s hybrid technique for voltage constrained reactive power planning.
The simulation results demonstrate that active power loss and operating
expenses are significantly reduced while voltage uniformity is maintained.
Author Copy
In Ref. [32], author suggested an oppositional crow search optimization
for optimal placement of capacitors for the reduction of transmission
losses. In Ref. [33], author presented GWO-PSO hybrid optimization for
reactive power planning in power systems, the presented system generates
a reduction of operating cost in the IEEE 30 bus system and new England
39 bus system. In Ref. [34], the author devised a flux linkage approach for
estimating transmission line parameters using multiple bundled conductors.
The exact calculation of transmission-line properties, such as temperature
correction resistance, using power-flow equations helps improve power-
flow efficiency in the power system. In Ref. [35], the author develops and
modifies a whale optimization technique for transmission line parameters
estimation for various bundled conductors. Because of their high efficiency
in discovering the optimal results, meta-heuristic optimization algorithms
have gained favor in recent years.
To achieve this, many optimization algorithms are investigated, each
with varying degrees of convergence power and precision. Researchers have
studied the optimal placement of capacitors in distribution networks using a
variety of optimization methodologies.
14.6.1 TEACHING-LEARNING-BASED OPTIMIZATION (TLBO)
Teachers have an impact on their students’ performance [36] created a

teaching-learning process-inspired algorithm called TLBO that considers
it. The teacher phase (also known as algorithmic learning) and engaging
with other learners are the two primary modalities of learning described
by the algorithm (known as the learner phase). Using this optimization
process, a population of students is treated as the optimization problem’s
distinct design factors, and the results of the students are compared to the
Non Commercial Use

optimization problem’s ‘fitness’ value. Among all people, the teacher is

thought to be the best answer. It is the objective function’s best value that
determines the optimum solution for a particular optimization issue, not the
design variables [37]. The author used modified TLBO (MTLBO) with a
constant and effective load to place capacitors where they will have the least
influence on system performance.
14.6.2 PARTICLE SWARM OPTIMIZER (PSO)
Particles “fly” around a multidimensional search space modeling social
Author Copy
activity, with each particle representing an intersection of all the search
dimensions. Particles in a small region exchange memories of their “ideal”
places for a brief time and utilize those memories to change their velocities
and subsequent placements [38]. It offers the benefits of parallel processing
and resilience, and it has a higher chance of finding the global optimum
solution than other techniques. PSO is easy to implement, has a fast conver-
gence rate, and has an intuitive design. PSO may be used in a variety of
sectors, including scientific research and engineering. In Ref. [39], to reduce
energy losses, the PSO algorithm creates and acquires capacitors in a radial
distribution network [40]. The author proposes a multi-objective optimiza-
tion technique for DG resources and capacitors in the distribution system. In
Ref. [41], the capacitor bank is used to compensate for total reactive power,
lowering the current and reducing power loss. In this study, for the best
arrangement of a capacitor, energy storage system, and size in a microgrid,
PSO and a generic algebraic modeling approach were applied.
14.6.3 THE GENETIC ALGORITHM (GA)
The genetic algorithm (GA) seeks the best solution utilizing evolutionary
principles based on a string that is assessed and passed on to the next
generation. The technique is meant to ensure that the “fitter” strings live
and reproduce. GA’s main benefit is that the answer is worldwide. GA may
also find the global solution to functions that are not differentiable, linear or
nonlinear, continuous or discrete, analytical or procedural. GAs is part of a
wider family of evolutionary algorithms that solve optimization problems
utilizing natural evolution mechanisms including inheritance, mutation,
selection, and the crossover [42]. To reduce all functions simultaneously, a
multi-objective technique is presented. Have recommended a site and size for
Non Commercial Use

the DG. The method is evaluated on an IEEE 69 bus system and compared
to nonlinear optimization. According to the DG study, improper DG location
and size would reduce system stability and dependability, increasing system
losses. This strategy minimizes power losses and DG investment costs
while optimizing DG placement and size [43]. A microgrid in both grid-

connected and islanded mode is shown with the optimal capacitor allocation.
Non-dispatch-able DGs account for the load. The imperialistic competitive
algorithm (ICA) and GA are utilized in Ref. [44].
14.6.4 CUCKOO SEARCH (CS) ALGORITHM
Author Copy
Structural optimization problems are addressed using the cuckoo search (CS)
approach. An initial benchmark nonlinear restricted optimization problem
demonstrates the efficacy of the unique CS technique with Levy flights.
The majority of structural engineering design optimization problems are
nonlinear, and typically include a large number of diverse design variables
under severe constraints. Material quality ranges, maximum stress, maximum
deflection, minimum load capacity, and geometrical configuration are all
examples of restrictions. Limited to basic limits or nonlinear interactions.
This nonlinearity produces a multimodal response landscape. To improve
the voltage profile while lowering the cost of reactive power, the CSA
(cuckoo search algorithm) for optimal placement of shunt static capacitors
in distribution networks is presented in Ref. [45].
14.6.5 ANT COLONY OPTIMIZATION (ACO)
Ant colonies are another popular evolutionary computing strategy that

examines a large number of populations in parallel before assessing each
population’s competence using a cost function until convergence. The ant
algorithm, which is based on ant behavior, calculates the quickest path from
home to food. Pheromone trails are used by ants to communicate with one
another and provide information about their journeys. Initially, a swarm of ants
creates chaotic trails of constant density. As a result, denser trails for shorter
courses are more relevant for later searches. These routes let the ants travel
more efficiently [46]. The technology uses powerful ant colony optimization
(ACO) to reconfigure feeders and insert capacitors in distribution networks.
This research aimed to provide unique approaches to the best capacitor
location, optimal feeder reconfiguration, and a combination of the two.
Non Commercial Use

14.6.6 GREY WOLF OPTIMIZER (GWO)
The social (leadership) structure and hunting behaviors of wild Grey Wolves
were replicated using a meta-heuristic algorithm. There are four subspecies
of grey wolf optimizer (GWO) to consider (alpha, beta, delta, and omega).
These models approximate the leadership hierarchy, which is based on the
evolutionary struggle for the survival of the human brain. Assaulting your
prey requires three separate tough maneuvers, thus planning a collective hunt
will be necessary [47]. This chapter goes into considerable detail about these
tactics. To reduce voltage loss and fluctuation, reactive power is also taken
into consideration. To reduce network power loss, and voltage variation,
Author Copy
and promote voltage stability, WCA and GWO are utilized to solve optimal
capacitor placement problems in diverse distribution networks [48].
14.6.7 TABU SEARCH (TS)
Tabu exploration involves randomly searching the solution space. Fred

Glover and colleagues created tabu search to find or regulate local optima.
Using Tabu search effectively locates the history of the halt cycle and avoids
being caught in local minima [49]. The higher-level heuristic algorithm is
solved, and the answer is evaluated. Nara et al. [50] proposed a Tabu search
to discover the ideal DG placement with the least amount of distribution
network or system losses. The author of this study evaluated how much the
power system’s loss percentage may be decreased by optimum DG alloca-
tion on the demand side. This method assumes that the total number of DGs
and total DG production capacity are equal to or greater than 1. A hybrid
technique including TS has been used to position optimal capacitors [51]. TS
currently incorporates components of practical heuristics and combinatorial
techniques like GAs and simulated annealing. The method has been fully
tested in a real-world 135-bus network as well as several other networks,
with better quality and cost outcomes.
14.6.8 APPLICATION OF OTHER OPTIMIZATION TECHNIQUES FOR

CAPACITOR ALLOCATION
It is proposed in Ref. [52] to use the artificial bee colony (ABC) optimizer to
find the most optimum position and size possible in the distribution system to
minimize the power losses and stability of voltage profile, besides lowering
Non Commercial Use

the lower of acquisition and installation capacitors. In Ref. [53], the shrimp
straw method is used to determine the best position for the capacitor and the
best way to reconfigure the distribution network to minimize distribution
network losses under various load scenarios. For the optimal size of capaci-
tive banks in the distribution system, the flower pollination algorithm (FPA)
is utilized [54]. The biogeography-based optimization (BBO) is proposed
in Ref. [55], to reduce the costs of power, as well as the quantity of energy
collected from the upstream network. The harmony search method is used
in Ref. [56] to offer the ideal position of capacitors to decrease losses and
expenses (HSA). The ideal placement and size of shunt capacitors in the
distribution network are found in Ref. [57] using a gravitational search
Author Copy
algorithm (GSA) to reduce losses and increase financial advantage owing
to the usage of capacitors (GSA). In Ref. [58], the allocation of capacitors is
investigated using the sine–cosine algorithm (SCA) to maximize net savings
while also improving dependability. In Ref. [59], the researcher proposed
the modified cultural algorithm (MCA) to minimize the power loss of a
distribution system by optimal placement of capacitors. For bus placement
in RDNs, the author has used a combination of LSF and voltage stability
index (VSI) in Ref. [60]. In Ref. [61], the authors show how they used shark
smell optimization (SSO) to enhance OC allocation in small and large RDNs.
Using a modified cultural algorithm (WOA), researchers in Ref. [62] were
able to reduce the operational and energy expenses of RDNs. In Ref. [22],
the authors employed a FPA to reduce the cost of the capacitor as well as
the amount of energy lost. In Ref. [18], the author used a butterfly optimiza-
tion approach to reduce the framework’s generation costs to a minimum.
Observation of four case test frameworks reveals that the proposed tech-
nique provides a superior arrangement regarding the speed of execution and
the capacity to sustain regular costs. The literature review demonstrated that
the meta-heuristic approach is extremely effective in determining the best
location and capacity of reactive resources to obtain the best distribution
network performance feasible. As a result, meta-heuristic techniques with
high optimization power and low computing cost are necessary to achieve
the best network performance.
14.7 CONCLUSION AND FUTURE WORK
In the entire power system approximately 70% of losses occur in distribu-

tion systems, as a result, the distribution system is the primary concern to
Non Commercial Use

minimize the losses through different techniques to provide reliable and

stable power for consumption. In the field of optimization, also known as
mathematical optimization, the process of picking the best element from a
prospective collection of alternatives to meet particular criteria is known as
optimization. This theory, which is concerned with finding the best and most
optimum solution for a certain problem, applies to almost all areas of scien-
tific and social sciences, and it has applications in a wide range of fields.
The optimization domain may be used to maximize or reduce the value of
a real function by using a systematic selection of input values from within
a permissible set and calculation of the value of the function, respectively.
Furthermore, for the challenges of power systems, a variety of optimization
Author Copy
strategies are employed.
Several studies have demonstrated that line loss in low voltage and the
weak distribution system wastes electrical power. However, distribution
networks suffer the worst losses in rural areas of underdeveloped countries
like Pakistan. As a result, reducing network loss in distribution networks
is crucial. Distribution feeders in developing nations have a radial shape
and are long. These issues cause poor voltage control in many feeds. As
a result, it is employed in RDNs to minimize power losses and voltage
variation while simultaneously improving the stability of voltage. However,
the aforesaid advantages are largely dependent on the number, location,
and size, of capacitors. As an optimal capacitor placement problem, it is
a complex combinatorial problem to maximize net savings while reducing
total expenses as an objective function. The classical approaches have been
discovered to be easier, but they have drawbacks such as poor handling of
qualitative constraints, poor convergence, and sluggish computing with
many variables. Furthermore, for future work, AI approaches are quick and
flexible. Designing capacitor banks with automated switching circuits for
distribution networks and analyzing their cost-benefit through optimization
techniques to attain the optimum network performance of the system.
This chapter will assist researchers in locating earlier work done on the
topic of capacitor placement and reactive power planning for distribution
system stability and reliability.
Furthermore, the chapter only discusses the optimal placement of capaci-
tors and their optimization techniques for active and reactive power losses
and voltage profile improvement, whereas the other devices (STATCOM,
TSC, etc.) and distribution generations (PV, wind, fuel cell, etc.) may be
considered for the same problem.
Non Commercial Use

KEYWORDS
• capacitors
• distribution system
• optimization techniques
• power factor
• power loss reduction
• voltage profile improvement
Author Copy
REFERENCES
1. Umer, F., Noor, K. A., Anwar, A. S., & Agha, Z. P., (2013). Critical analysis of capacitors
as a potential solution to achieve optimum level voltage regulation in HESCO network.
In: 2013 The International Conference on Technological Advances in Electrical,
Electronics and Computer Engineering (TAEECE) (pp. 488–497). IEEE.
2. Sedghi, M., Aliakbar-Golkar, M., & Haghifam, M. R., (2013). Distribution network
expansion considering distributed generation and storage units using modified PSO
algorithm. International Journal of Electrical Power & Energy Systems, 52, 221–230.
3. Kumar, S., Chandar, K. F. Ur. R., Shoaib, A. S., & Anwar, A. S., (2018). Voltage
improvement and power loss reduction through capacitors in utility network. In: 2018
International Conference on Computing, Mathematics and Engineering Technologies
(iCoMET) (pp. 1–5). IEEE.
4. Sirjani, R., Azah, M., & Hussain, S., (2012). Heuristic optimization techniques to
determine optimal capacitor placement and sizing in radial distribution networks: A
comprehensive review. Organ, 7, 12.
5. Farooq, U., Shahryar, Q., Sanaullah, A., & Fazal, W. K., (2017). Load flow analysis of 11
kV test feeder with and without the injection of DG. International Journal of Computer
Science and Information Security, 15(3), 1.
6. Lotfi, H., Mahdi, S., & Ali, D., (2016). Optimal capacitor placement and sizing in radial
distribution system using an improved particle swarm optimization algorithm. In: 2016
21st Conference on Electrical Power Distribution Networks Conference (EPDC) (pp.
147–152). IEEE.
7. Hassan, A. S., Yanxia, S., & Zenghui, W., (2020). Multi-objective for optimal placement
and sizing DG units in reducing loss of power and enhancing voltage profile using
BPSO-SLFA. Energy Reports, 6, 1581–1589.
8. Nguyen, T. T., Bach, H. D., Thai, D. P., & Thang, T. N., (2020). Active power loss
reduction for radial distribution systems by placing capacitors and PV systems with
geography location constraints. Sustainability, 12(18), 7806.
9. Hussain, A. N., Al-Jubori, W. K. S., & Haider, F. K., (2019). Hybrid design of optimal
capacitor placement and reconfiguration for performance improvement in a radial
distribution system. Journal of Engineering, 2019.
Non Commercial Use

10. Olabode, E., Ajewole, T., Okakwu, I., & Ade-Ikuesan, O., (2019). Optimal sitting and
sizing of shunt capacitor for real power loss reduction on radial distribution system using
firefly algorithm: A case study of Nigerian system. Energy Sources, Part A: Recovery,
Utilization, and Environmental Effects, 1–13.
11. Eslami, M., Hussain, S., & Azah, M., (2011). Application of artificial intelligent techniques
in PSS design: A survey of the state-of-the-art methods. Przegląd Elektrotechniczny
(Electrical Review), 87(4), 188–197.
12. Mehta, V. K., & Rohit, M., (2005). Principles of Power System: Including Generation,
Transmission, Distribution, Switchgear and Protection: For BE/B. Tech., AMIE and
Other Engineering Examinations. S. Chand Publishing.
13. Sahito, A. A., Jatoi, A. M., Memon, S. A., & Shah, S. A. A., (2015). Transmission system
performance improvement through reactive power compensation. Sindh University
Research Journal-SURJ (Science Series), 47(3).
Author Copy
14. Tsong-Liang, H., Ying-Tung, H., Chih-Han, C., & Joe-Air, J., (2008). Optimal placement
of capacitors in distribution systems using an immune multi-objective algorithm.
International Journal of Electrical Power & Energy Systems, 30(3), 184–192.
15. Ahmadi, H., José, R. M., & Hermann, W. D., (2014). A framework for volt-VAR
optimization in distribution systems. IEEE Transactions on Smart Grid, 6(3), 1473–1483.
16. Gonen, T., (2015). Electric Power Distribution Engineering. CRC press.
17. Almasoud, H., (2009). Shunt capacitance for a practical 380 kV system. International
Journal of Electrical and Computer Science (IJECS), 9(10), 23–27.
18. Ansari, M. M., Chuangxin, G., Muhammad, S., Nitish, C., Bo, Y., Jun, P., Yishun, Z.,
& Xurui, H., (2020). Considering the uncertainty of hydrothermal wind and solar-based
DG. Alexandria Engineering Journal, 59(6), 4211–4236.
19. Reddy, M. D., & Kumar, N. V., (2012). Optimal capacitor placement for loss reduction
in distribution systems using fuzzy and harmony search algorithm. ARPN Journal of
Engineering and Applied Sciences, 7(1), 15–19.
20. Pazouki, S., Amin, M., Mahmoud-Reza, H., & Shahab, A., (2015). Simultaneous allocation
of charging stations and capacitors in distribution networks improving voltage and power
loss. Canadian Journal of Electrical and Computer Engineering, 38(2), 100–105.
21. Dinakara Prasasd Reddy, P., Veera Reddy, V. C., & Gowri Manohar, T. (2018). Ant
lion optimization algorithm for optimal sizing of renewable energy resources for
loss reduction in distribution systems. Journal of Electrical Systems and Information
Technology, 5(3), 663–680.
22. Tamilselvan, V., Jayabarathi, T., Raghunathan, T., & Xin-She, Y., (2018). Optimal
capacitor placement in radial distribution systems using flower pollination algorithm.
Alexandria Engineering Journal, 57(4), 2775–2786.
23. Naderipour, A., Abdul-Malek, Z., Saber, A. N., Hesam, K., Amir, R. R., Saman, S., &
Jiří, J. K., (2021). Comparative evaluation of hybrid photovoltaic, wind, tidal and fuel
cell clean system design for different regions with remote application considering the
cost. Journal of Cleaner Production, 283, 124207.
24. Ramakrishnan, K., Vijeswaran, D., & Manikandan, V., (2019). Stability analysis of
networked micro-grid load frequency control system. The Journal of Analysis, 27(2),
567–581.
25. Baghaee, H. R., Mojtaba, M., Gevork, B. G., & Heidar, A. T., (2017). Nonlinear load sharing
and voltage compensation of microgrids based on harmonic power-flow calculations using
radial basis function neural networks. IEEE Systems Journal, 12(3), 2749–2759.
Non Commercial Use

26. Dey, B., Saurav, R., Sheila, M., & Fausto, P. G. M., (2022). Optimal scheduling of
distributed energy resources in microgrid systems based on electricity market pricing
strategies by a novel hybrid optimization technique. International Journal of Electrical
Power & Energy Systems, 134, 107419.
27. Babu, R., Saurav, R., Bishwajit, D., & Biplab, B., (2021). Optimal reactive power
planning using oppositional grey wolf optimization by considering bus vulnerability
analysis. Energy Conversion and Economics.
28. Swetha, S. G., Sheila, M., & Saurav, R., (2021). Voltage constrained reactive power
planning by ameliorated HHO technique. In: Recent Advances in Power Systems (pp.
435–443). Springer, Singapore.
29. Gudadappanavar, S. S., & Sheila, M., (2021). Metaheuristic nature-based algorithm for
optimal reactive power planning. International Journal of System Assurance Engineering
and Management, 1–14.
Author Copy
30. Babu, R., Vishnu, K., Chandan, K. S., Saurav, R., & Biplab, B., (2022). Application of
sine–cosine optimization algorithm for minimization of transmission loss. Technology
and Economics of Smart Grids and Sustainable Energy, 7(1), 1–12.
31. Shekarappa, G. S., Sheila, M., & Saurav, R., (2021). Voltage constrained reactive power
planning problem for reactive loading variation using hybrid Harris hawk particle
swarm optimizer. Electric Power Components and Systems, 49(4, 5), 421–435.
32. Shiva, C. K., Swetha, S. G., Basetti, V., Rohit, B., Saurav, R., & Biplab, B., (2022).
Fuzzy-based shunt VAR source placement and sizing by oppositional crow search
algorithm. Journal of Control, Automation and Electrical Systems, 1–16.
33. Badi, M., Sheila, M., Bishwajit, D., & Saurav, R., (2022). A hybrid GWO-PSO technique
for the solution of reactive power planning problem. International Journal of Swarm
Intelligence Research (IJSIR), 13(1), 1–30.
34. Shaikh, M. S., Changchun, H., Munsif, A. J., Muhammad, M. A., & Aleem, A. Q., (2021).
Parameter estimation of AC transmission line considering different bundle conductors
using flux linkage technique. IEEE Canadian Journal of Electrical and Computer
Engineering, 44(3), 313–320.
35. Shaikh, M. S., Changchun, H., Saurav, R., Shubash, K., Mannan, H., Muhammad,
M. A., & Munsif, A. J., (2022). Optimal parameter estimation of 1-phase and 3-phase
transmission line for various bundle conductor’s using modified whale optimization
algorithm. International Journal of Electrical Power & Energy Systems, 138, 107893.
36. Rao, R., & Vivek, P., (2012). An elitist teaching-learning-based optimization algorithm
for solving complex constrained optimization problems. International Journal of
Industrial Engineering Computations, 3(4), 535–560.
37. Bhattacharyya, B., & Rohit, B., (2016). Teaching learning-based optimization algorithm
for reactive power planning. International Journal of Electrical Power & Energy
Systems, 81, 248–253.
38. Bai, H., & Bo, Z., (2006). A survey on application of swarm intelligence computation
to electric power system. In: 2006 6th World Congress on Intelligent Control and
Automation, (Vol. 2, pp. 7587–7591). IEEE.
39. Ramadan, H. S., Bendary, A. F., & Nagy, S., (2017). Particle swarm optimization
algorithm for capacitor allocation problem in distribution systems with wind turbine
generators. International Journal of Electrical Power & Energy Systems, 84, 143–152.
40. Zeinalzadeh, A., Younes, M., & Mohammad, H. M., (2015). Optimal multi objective
placement and sizing of multiple DGs and shunt capacitor banks simultaneously
Non Commercial Use

considering load uncertainty via MOPSO approach. International Journal of Electrical

Power & Energy Systems, 67, 336–349.
41. Rajamand, S., (2020). Loss cost reduction and power quality improvement with applying
robust optimization algorithm for optimum energy storage system placement and capacitor
bank allocation. International Journal of Energy Research, 44(14), 11973–11984.

42. Shaikh, M. S., Changchun, H., Mannan, H., Saurav, R., Munsif, A. J., & Muhammad,
M. A., (2021). Optimal parameter estimation of overhead transmission line considering
different bundle conductors with the uncertainty of load modeling. Optimal Control
Applications and Methods.
43. Naderipour, A., Abdul-Malek, Z., Mohammad, H., Zahra, M. S., Mohammad, A. F.,
Saber, A. N., & Iraj, F. D., (2021). Spotted hyena optimizer algorithm for capacitor
allocation in radial distribution system with distributed generation and microgrid
operation considering different load types. Scientific Reports, 11(1), 1–15.
Author Copy
44. Moradi, M. H., Arash, Z., Younes, M., & Mohammad, A., (2014). An efficient hybrid
method for solving the optimal sitting and sizing problem of DG and shunt capacitor
banks simultaneously based on imperialist competitive algorithm and genetic algorithm.
45. El-Fergany, A. A., & Almoataz, Y. A., (2014). Capacitor allocations in radial distribution
networks using cuckoo search algorithm. IET Generation, Transmission & Distribution,
8(2), 223–232.
46. Chung-Fu, C., (2008). Reconfiguration and capacitor placement for loss reduction of
distribution systems by ant colony search algorithm. IEEE Transactions on Power
Systems, 23(4), 1747–1755.
47. Shaikh, M. S., Changchun, H., Munsif, A. J., Muhammad, M. A., & Aleem, A. Q., (2021).
Application of grey wolf optimization algorithm in parameter calculation of overhead
transmission line system. IET Science, Measurement & Technology, 15(2), 218–231.
48. Kola, S. S., & Jayabarathi, T., (2020). Optimal capacitor allocation in distribution
networks for minimization of power loss and overall cost using water cycle algorithm
and grey wolf optimizer. International Transactions on Electrical Energy Systems, 30(5),
e12320.
49. Sirjani, R., Azah, M., & Hussain, S., (2012). Heuristic optimization techniques to
determine optimal capacitor placement and sizing in radial distribution networks: A
comprehensive review. Organ, 7, 12.
50. Nara, K., Hayashi, Y., Yamafuji, Y., Tanaka, H., Hagihara, J., Muto, S., Takaoka, S., &
Sakuraoka, M., (1996). A tabu search algorithm for determining distribution tie lines.
In: Proceedings of International Conference on Intelligent System Application to Power
Systems (pp. 266–270). IEEE.
51. Mori, H., & Shingo, T., (2005). Variable neighborhood tabu search for capacitor
placement in distribution systems. In: 2005 IEEE International Symposium on Circuits
and Systems (pp. 4747–4750). IEEE.
52. El-Fergany, A. A., & Almoataz, Y. A., (2014). Capacitor placement for net saving
maximization and system stability enhancement in distribution networks using artificial
bee colony-based approach. International Journal of Electrical Power & Energy
Systems, 54, 235–243.
53. Sultana, S., & Provas, K. R., (2015). Oppositional krill herd algorithm for optimal
location of distributed generator in radial distribution system. International Journal of
Non Commercial Use

54. Abdelaziz, A. Y., Ehab, S. A., & Abd, E. S. M., (2016). Flower pollination algorithm
and loss sensitivity factors for optimal sizing and placement of capacitors in radial
distribution systems. International Journal of Electrical Power & Energy Systems, 78,
207–214.
55. Ghaffarzadeh, N., & Hassan, S., (2016). A new efficient BBO based method for
simultaneous placement of inverter-based DG units and capacitors considering harmonic
limits. International Journal of Electrical Power & Energy Systems, 80, 37–45.
56. Ali, E. S., Abd, E. S. M., & Abdelaziz, A. Y., (2016). Improved harmony algorithm and
power loss index for optimal locations and sizing of capacitors in radial distribution
systems. International Journal of Electrical Power & Energy Systems, 80, 252–263.
57. Shuaib, Y. M., Surya, K. M., & Christober, A. R. C., (2015). Optimal capacitor placement
in radial distribution system using gravitational search algorithm. International Journal
of Electrical Power & Energy Systems, 64, 384–397.
Author Copy
58. Abdelsalam, A. A., & Hany, S. E. M., (2019). Optimal allocation and hourly scheduling
of capacitor banks using sine cosine algorithm for maximizing technical and economic
benefits. Electric Power Components and Systems, 47(11, 12), 1025–1039.
59. Haldar, V., & Niladri, C., (2015). Power loss minimization by optimal capacitor
placement in radial distribution system using modified cultural algorithm. International
Transactions on Electrical Energy Systems, 25(1), 54–71.
60. Devabalaji, K. R., Ravi, K., & Kothari, D. P., (2015). Optimal location and sizing of
capacitor placement in radial distribution system using bacterial foraging optimization
algorithm. Int. J. Electr. Power Energy Syst., 71, 383–390. https://doi.org/10.1016/j.
ijepes.2015.
61. Gnanasekaran, N., Chandramohan, S., Kumar, P. S., & Mohamed, I. A., (2016). Optimal
placement of capacitors in radial distribution system using shark smell optimization
algorithm. Ain. Shams Eng. J., 7, 907–916. https://doi.org/10.1016/j.asej.2016.01.006.
62. Prakash, D. B., & Lakshminarayana, C., (2017). Optimal siting of capacitors in radial
distribution network using whale optimization algorithm. Alexandria Engineering
Journal, 56(4), 499–509.
Non Commercial Use

CHAPTER 15
Solar Photovoltaic Powered Automatic

Irrigation System for the Agriculture Sector
HIMANSHU SHARMA and PANKAJ KUMAR
Department of EEE, SRMIST, Delhi NCR Campus, Ghaziabad,
Author Copy
Uttar Pradesh, India
ABSTRACT
The chapter was based on a self-contain home automatic system that was
solar powered, providing plant watering of the necessary repeatability and
moisture from a distinct water reservoir. It also able to alert the users in any
kind of unusual situation. Inside a home is an extreme environment for many
plants, as wetness was usually low, sunlight light levels vary significantly,
and temperature can vary widely. This chapter contains an automatic self-
watering system that sprinkles water to the plants when the soil moisture
sensor, attached to the device, measures a small amount of soil moisture and
overrides manual irrigation to keep the soil moisture level adequate. It was
like hiring somebody to provide water for your plants.
15.1 INTRODUCTION
It was required to build an automatic solar power plant watering system

which was irrigating the plant automatically after some time. This chapter
would explain the hardware and software implementation of a solar power
automatic power plant watering system. The plant watering system was a kind
of automatic system which irrigates the plant after every time when sensor
finds that soil was wet [1]. This system was a type of an embedded system,
Non Commercial Use

as it was a combination of software and hardware aimed for a specific task.

The circuit used here was electronic one that comes as a small electronic
board which itself was an embedded system, the IC used here was “16F876,”
which was interfaced with a PCB containing a PIC micro-controller [2].
Block diagram in Figure 15.1 gives an overview of the system, the Auto-
matic Solar Power Plant Watering System send the information to the micro-
controller through a PWM signal from the sensor which was dig in the soil to
check the moisture of the soil [3]. The signal to the micro-controller checks
the signal and send signal according to that to the motor to run or not. For
more details of micro-controller module refer to data-sheet of 16F876 [4].
The supply to the motor and the micro-controller was given by the battery.
Author Copy
The battery was re-chargeable. The battery was charged by the solar panel.
The solar panel gives the maximum voltage 17.5 V and the battery provides
the 12 V. The difference in the voltage was adjusted by the resistance put in
between.
FIGURE 15.1 Block diagram of system.
This project was mostly suitable for those kinds of people who like
to travel and want to go to for long trip. Also, such a system removes the
need to water very often when you are at home. This device may be helpful
Non Commercial Use

Solar Photovoltaic Powered Automatic Irrigation System 297
for florists’ people and other who like green environment at home but not
remember to water them frequently to water them.
It was simple and easy to use for plants of various sizes and varieties.
Similar devices are excellent for fields of vegetation in locations where the
soil lacks water or where there has been a severe drought, in addition to
potted plants. The device works with your plants as they are and so there
was no need to repot. It was also convenient because it provides an easy
way to water those hard to reach plants. This equipment may be used to
automatically fertilize while watering and supports fruitful development
since fertilizer can be added to the water supply [5].
Author Copy
15.1.1 AIMS
• To store the power supply in the battery from the solar panel from the
solar energy.
• To use the power supply to run the system to irrigate the soil when
required.
• Writing suitable assembly language program for the system to perform
correct water supply operation.
• Design of output display unit, i.e., PCB with LED indicators and
suitable driver circuit.
15.1.2 OBJECTIVES
• Theoretical study of an automated solar powered plant watering

system.
• Construction of a working model of an automated solar powered plant
watering system.
• Representation and explanation of system output.
• Exploration of advantages and disadvantages of the system.
15.1.3 DELIVERABLES
• Advance study and analysis of the automated solar powered plant

watering system.
• Working model.
• Results achieved from the system.
Non Commercial Use

15.2 BRIEF HISTORY OF SOLAR PANELS
15.2.1 1880–1960: EARLY STAGE OF SOLAR ENERGY

 (1883) by the Charles Fritts: An American inventor. According to

his description the first solar cells made from Selenium wafers [5].
 (1904) Wilhelm Hallwachs: He explains that the photosensitivity
can also be made by the combination of the copper and cuprous
oxide [6].
 (1932) Audobert and Stor: Discover that cadmium sulfide (CdS)
has the phovoltic effect after the nobel price given to Elbert einstine
Author Copy
for his theory on photoelectric effect.
 (1957) Hoffman Electronics: Achieved 8% efficient photovoltaic
cells and after one year later he got 9%.
 (1960) Hoffman Electronics: Achieves 14% efficient photovoltaic
cells.
 (1963) Japan: That was the biggest achievement that the japans
installs a 242-watt. That was the world’s largest array at that time to
collect the solar power from the sun.
15.2.2 1970–1990: INTERMEDIATE STAGE
 (1977) World: In that time total photovoltaic manufacturing produc-

tion was not more then 500 kilowatts and that time it was first time it
was increase more then 500 kilowatts.
 (1982) World: After five years the worldwide photovoltaic produc-
tion was increase till 9.3 megawatt.
15.2.3 1990–2007: ADVANCE STAGE
 (1991) America: In this year the president of America redesignates

Institute as the National Renewable Energy Laboratory.
 (1996) World: The most advance solar powered airplane was made
on which 3,000 super solar cells was embedded from which it took
the energy to fly.
 (1998) World: Remote control solar power aircraft.
Non Commercial Use

15.3 LITERATURE SURVEY
Solar Photovoltaic technology is widely used to produce electricity [7] from

the incident solar irradiation. This produced electricity can be widely used
for the irrigation system. The various works has been performed based on
the solar energized automatic [8–15] irrigation system. In this regards, Bolu
et al. [8] implemented a solar photo-voltaic energized micro-controller based
smart irrigation system which detects the moisture level and operates the
pump when the moisture level goes below the threshold value. Al-Ali et al.
[9] have produced a compendium of the usages of solar photo-voltaic [16].
According to the Abayomi-Alli et al. [10] smart solar photovoltaic system is
Author Copy
also an viable alternative for the irrigation which can be taken as a pathway
for the sustainable [17] agriculture. However, Mahalakshmi et al. [11] has
focused on the distant monitoring and control of solar powered irrigation
system for agricultural needs. The work presented by the author has used
internet of things (IoT) technology for this purpose.
In a similar work, Tamoor et al. [12] has focused their study for remote
rural area. The scholars investigated the socioeconomic as well as environ-
mental impact after the integration of solar photovoltaic system [18] for
remote agricultural needs. However, Rehman et al. [13] have implemented
three different sensors namely moisture, temperature and humidity sensors to
efficiently working of solar powered automatic irrigation system. Likewise,
Rout et al. [14] have also used IoT technique to design solar photovoltaic
energized smart watering system for agriculture purposes. The work can also
be assessed from the existing literature in which role of hybridization [15] of
solar power technology along with other technologies like diesel generator,
battery [19–21], etc., are implemented to power the automatic irrigation
system. This shows a wide scope of solar photovoltaic resource for powering
a irrigation system across the globe.
15.4 HARDWARE STRUCTURE
15.4.1 MICROCONTROLLER
The microcontroller module as shown in Figure 15.2 will decode the

incoming PWM signal and convert the information in electric form. To make
the microcontroller to work according to the requirement a suitable program
Non Commercial Use

was required. The microcontroller used was PIC 18F1320 for more details
refer to datasheet.
In this system embedded C was used for the development of the program,
but microcontroller understands only machine codes for that some kind of
tool was required that means a compiler, which will convert a high level
language like embedded C into the required machine codes. Embedded C
was used because as it was easy to understand and level of abstraction of
program was much higher as compared with assembly language. For more
information about embedded C look at reference.
MPLAB along with CCS C compiler was used to develop the program.
ICD2 kit was used to transfer the machine codes into the flash memory of
Author Copy
the microcontroller. To get familiar with CCS C compiler have a look at
reference.
15.4.2 SOLAR PANEL PUMPING SYSTEM
Two basic components were used in the solar powered pumping system.
The first component which was used was the combination of the solar panel
and the chargeable battery. The solar cell was the smallest element of the
solar panel. To the production of the direct current (DC) when the solar cell
exposed to the light its two specially prepared layers do the special roles. Its
gives the DC current to the battery which was then turns the DC pump on
which give the supply of water to the soil. The rating of the voltage and the
current at the output was calculated. The peak power was also calculated
by multiplying of the voltage and the current of the solar panel. It was also
depend on the irradiation and temperature from the sun on solar panel. The
irradiation and temperature receives on the solar panel surface was 1,000 W/
m2 at 25°C. The amount of the DC produced by the solar panel was directly
proportional to the light intensity striking on the solar panel [22, 23]. It could
be explained like this if the intensity of the light was halved the output of the
DC was half but the value of the direct voltage was reduce very little [24].
The ratings of the solar panel used are shown in Table 15.1.
TABLE 15.1 Characteristics Data of the Solar Panel

Maximum power 2.4 W
Maximum voltage 17.5 V
Maximum current 137 mA
Non Commercial Use

15.4.3 BATTERY SYSTEM
The battery which was selected was rechargeable lead acid battery in the
Hardware. There are many benefits of this battery such as easy to maintain,
cheap, available in the market in the different variety of the size. They can
also withstand for one day with their only 80% of the rated capacity. There
are other batteries also available such as gel cell that have gel instead of
the electrode. They are portable and easy to use. They can be use upside
down and in any way because they are seal packed. Another type of the
battery which could be use was nickel cadmium plates used with the solar
panel system. Although, the lead acid battery has high initial cost but the life
Author Copy
cycle cost was very low. There were some advantages of the NiCd batteries
like it is long lasting but it needs less maintenance. It can withstand in the
extreme conditions also. NiCd could tolerate more to complete discharge.
Car batteries was also an option but car batteries were not use to store the
power from the solar panel because the design of the car batteries were made
like that they produce cold cracking ampere for short period of time. Battery
banks were also the good option while charging the battery by the solar
panel. In the battery banks series of 12 V batteries could be connect in series
to get the desire voltage at the end. They are rechargeable. In the end after
going through all the available batteries the lithium ion battery was selected
due to its much advantage over other available batteries like it was easy to
use, portable and easy to handle. The battery 2 Ah storage capacities was
selected.
15.4.4 PUMP
The pump was the essential component of the project. The pump which was
used was especially selected to use the solar power efficiency. The convention
pumps need alternative current that generates the supply. The pump which
was use was the solar pump which was especially design for the DC supply
input from the battery. It was especially selected to work effectively during
low-light conditions, at reduced voltage, without stalling or overheating.
There were various types and size of pumps was available. The especially 12
V DC pump was selected to get the application for the low volume. The low
volume pumps keep the cost of the system down and use the less electricity
from the chargeable battery. Some pumps are fully submarine but the pump
which was used was not. The pump which was used was used two pipes
Non Commercial Use

connected and takes charge from the battery. If the pump which was used
under water the benefit would be it eliminates potential priming and freezing
problems. The range of the solar panel used would use between 12 and 36 V
DC. Some pumps could be use in the project could be positive displacement
pumps. It seals the water in cavities inside the pump and forces it up. They
design in such a manner that they maintain their lift capability all through the
solar day slowly if they were connected directly with the solar panel. If the
lighting conditions were fine it runs fast but if the lighting condition were not
good it run slow. The speed of the pump if connected directly depends upon
the lightning conditions available. The required system was not required on
the pumping vertically. So the pump which was used was the simple 12 V
Author Copy
DC pump.
Pumping was the mechanical work. The amount of water pump by the
pump and the length of the pipe used before and after that were the two main
factors which determine the rating of the pump. If the length of the pipe was
increase the required power of the pump was increase. By the appropriate
measurement it was found that the used pump of 12 V was alright. The fric-
tion of the pipe with the water was also reducing by the use of the smooth
pipe to increase the efficiency. The choice of pump depends on water volume
needed, efficiency, price and reliability.
15.5 PROBLEM ANALYSIS
15.5.1 THE SYSTEM DIAGRAM
As shown in Figure 15.2, the “Solar Powered Automatic Irrigation System”

can detect and sense the moisture level in the soil and water the plants
when needed. When the moisture level drops below the set level, the water
pump will automatically turn on and water the plant until the set moisture
level is reached. The sensor’s probes in the ground record the moisture
content through the conductivity established between the probes. When soil
moisture and mineral levels are high, conductivity is established and the
corresponding conductivity level is indicated by the LEDs. The energy to
pump the water is produced from the solar panel which stores in the battery
system. The irrigation system, which is commonly used in open fields, can
fully exploit the sun’s resource and use solar energy to power itself and even
the water pump.
Non Commercial Use

Author Copy
FIGURE 15.2 The system block diagram.
15.5.2 SPECIFICATION OF THE PROJECT (FIGURE 15.3)
15.5.2.1 INPUT
• Solar power energy.
15.5.2.2 OUTPUT
• Plant watering after every time when sensor finds that soil was wet
(for simplicity LED display board used).
15.5.2.3 PROCESS
• Solar Batter charger charges the battery, and the battery gives the
power supply to the system.
• The sensor senses the moisture in the soil and supply the water supply
when required.
• LED glow up when water supply required and stay off when not required.
Non Commercial Use

Author Copy
FIGURE 15.3 The input, output, and the process.
They were two types of connection which could be done. The first type
was when the solar panel and the pump directly connected with each other
and the second method was when the battery was intermediate between
them. They are called battery coupled and direct couple. A variety of factors
must be considered in determining the optimum system for a particular
application.
There were some drawbacks for the using of the battery system. First
was the efficiency of the overall system can be reduced because the voltage
on which the whole system works was dictated by the battery directly not
with the solar panel? It depends upon the amount of the battery charged.
The voltage supplied by the batteries can be one to four volts lower than the
voltage produced by the panels when the sun is at its maximum. This reduced
efficiency can be minimized by using an appropriate pump controller that
boosts the battery voltage supplied to the pump. But the connection was
straight and forward. There was no pump controller was used.
15.5.3 BATTERY-COUPLED SOLAR PUMPING SYSTEMS
Battery-couple water pumping system required the direct coupling of the

solar panel and the battery. The battery was connected with the project then.
The hour charges the batteries and the batteries in turn power the pump
whenever water is needed. Using batteries spreads pumping over a longer
Non Commercial Use

period of time by providing a constant operating voltage to the pump’s DC

motor. Thus, the system can provide a constant source of water for the project
even during the night and in low light.
15.5.4 DIRECT-COUPLE SOLAR PUMPING SYSTEMS
This system was not used in the project but could be used if the system not
required the battery and the amount of energy coming from the sun was good
it means there would be constant source of the sunlight. It could be used
straight in the sunlight.
Author Copy
15.6 PROBLEM IMPLEMENTATION
Because it combines hardware and software components, users can change

requirements depending on seasons, plant types or water requirements. In
addition, it does not require a highly skilled person to control the system.
All that is required is knowledge about the plant, which must be provided in
simple steps. This system finds many uses on farms owned by large corpora-
tions that produce crops for the world market. Installation requires very little
cost and minimizes labor costs for the company as no labor is required to
water the plants by testing soil moisture. Only one person is needed to operate
the machine in the control room. The use of solar energy saves a large amount
of electrical energy.
15.6.1 HARDWARE SPECIFICATION
• Port B0 was connected with the 12 V DC pump.

• Port B1 was connected with the electrodes.
• Port B2 was connected with the Led
In the hardware implimentation as shown in Figure 15.4,
• If the soil is wet then signal goes to microcontroller and PIN B
becomes 1. Then eventually pump stops running and the led glow.
• If the soil is dry then signal goes to microcontroller and PIN B
becomes 0. Then eventually pump start working and the led stop.
The flow chart of the above working has been shown in Figure 15.4.
Non Commercial Use

Author Copy
FIGURE 15.4 The flow chart of working.
15.6.2 PROGRAM USED
#include <16F877A.h>
#use delay(clock=4000000)
// macro-definition
#define motor PIN_B0
#define sensor input (PIN_B1)
#define led PIN_B2
#define on 0x00 // for on condition
#define off 0xff // for off condition
// main part of the program

void main()
{
int count;
// setup_oscillator(OSC_4MHZ);
Non Commercial Use

count=0;
output_high(motor); // to turn on the pump
// to flash LED
output_low(led);
delay_ms(500);
output_high(led);
while(1) // infinite loop
{
while(input(sensor)==0); // to checking the moisture of the soil
Author Copy
output_high(motor); // to start the motor
output_high(led); // turn on LED
delay_ms(250); // delay
// if soil was wet
while(input(sensor))
{
output_low(led); // led was stop glowing
output_low(motor); // turn off motor
delay_ms(250); // delay
}
}
}
15.6.3 CIRUIT DIAGRAM
In Figure 15.5, the ciruit diagram has been shown. The power has been
through by the transister “7805.” The 5 V supply from outside has been given
to the microcontroller. The PDS was used to use the microcontroller and the
power was given from outside. The optocoupler used to transfer the signal
from one electronic equipment to another with out making any connection.
The microcontroller and the optocoupler both connected with the 5 V supply.
The transister was used to drive the motor properly. The base of the transister
connected with the optocoupler and the collector was connected with the
pump. The emitter was connected with the voltage of 12 V. The transister
was giving the voltage and the current to the motor properly. The leads are
labeled base (B), collector (C), and emitter (E). These terms refer to the
Non Commercial Use

internal operation of a transistor, but they are not much help in understanding
how a transistor was used, so just treat them.
Author Copy
FIGURE 15.5 The flow chart of working.
15.7 CONCLUSION
The proposed work undertakes the feasible solution to a problem of the

unsustainable plants irrigation system. There is a rising need for technically
advanced irrigation practices to save thousands gallons of water that would or
else go to waste. The proposed hardware system uses solar panels to power
itself, making it a self-contained independent hardware model. The use of
clean renewable energy in engineering events cannot be underestimated if
hydrocarbons are not used in a sustainable manner because of the energy
disaster we are heading towards. The working model also eliminates the need
for manpower to irrigate slowly and intensively; the opportunity cost saved
allows the farmer to focus on other technical improvements and maximizing
crop yield.
Due to the cheap building materials and further cost optimization when
the model comes in the marketplace, it finds lots of applications in a broad
range. The scalability of the proposed work would be significantly easier
since the model can be used on both small and large farms. So, this model
would not only confine the market of larger and more affluent farms, but also
minor farms with limited resources. The system is also easy to use, and no
Non Commercial Use

multifaceted training is necessary on the part of farmers to use this device.

In addition to the environment, this benefit from lower water consumption in
the fields, the farmer also makes profit from a higher yield through optimized
irrigation. In addition to the positive environmental impact that this hardware
model would have, there will be huge economic benefits for farmers.
Given the needful demand for such a project, a win for all the key
stakeholders is obvious when using such a project. Aside from detecting
soil moisture, this device can also be used to test the difference between
mineral water and distilled water, the power of acids and any material that
involves the ionization thereof, making it a multi-purpose tool for measuring
the levels of various substances. As a responsible citizen of this planet, it is
Author Copy
our individual responsibility to take an active job in water conservation. And
so, this project is a result of that responsibility.
KEYWORDS
• cost optimization
• hardware specification
• optocoupler
• PIC micro-controller
• solar PV
• stakeholders
• transistor
REFERENCES
1. Balaji, V. R., & Sudha, M., (2016). Solar powered auto irrigation system. International
Journal of Emerging Technology in Computer Science & Electronics (IJETCSE), 20(2),
203–206.
2. Barman, A., Neogi, B., & Pal, S., (2020). Solar-powered automated IoT-based drip
irrigation system. In: IoT and Analytics for Agriculture (pp. 27–49). Springer, Singapore.
3. Priya, P., & Madhumitha, R., (2021). Modeling and simulation for automatic irrigation
system with PV solar tracking. Materials Today: Proceedings.
4. Khemissi, L., Khiari, B., Andoulsi, R., & Cherif, A., (2012). Low cost and high efficiency
of single phase photovoltaic system based on microcontroller. Solar Energy, 86(5),
1129–1141.
Non Commercial Use

5. Zhai, Z., Chen, X., Zhang, Y., & Zhou, R. (2021). Decision-making technology based
on knowledge engineering and experiment on the intelligent water-fertilizer irrigation
system. Journal of Computational Methods in Sciences and Engineering (pp. 1–20).
(Preprint).
6. Morison, W. L., (2004). Photosensitivity. New England Journal of Medicine, 350(11),

1111–1117.
7. Kumar, P., Pal, N., & Sharma, H., (2020). Performance analysis and evaluation of 10
kWp solar photovoltaic array for remote islands of Andaman and Nicobar. Sustainable
Energy Technologies and Assessments, 42, 100889.
8. Bolu, C. A., Azeta, J., Alele, F., Daranijo, E. O., Onyeubani, P., & Abioye, A. A., (2019).
Solar powered microcontroller-based automated irrigation system with moisture sensors.
In: Journal of Physics: Conference Series (Vol. 1378, No. 3, p. 032003). IOP Publishing.
9. Al-Ali, A. R., Rehman, S., Al-Agili, S., Al-Omari, M. H., & Al-Fayezi, M., (2001). Usage
Author Copy
of photovoltaics in an automated irrigation system. Renewable Energy, 23(1), 17–26.
10. Abayomi-Alli, O., Odusami, M., Ojinaka, D., Shobayo, O., Misra, S., Damasevicius,
R., & Maskeliunas, R., (2018). Smart-solar irrigation system (SMIS) for sustainable
agriculture. In: International Conference on Applied Informatics (pp. 198–212). Springer,
Cham.
11. Mahalakshmi, M., Priyanka, S., Rajaram, S. P., & Rajapriya, R., (2018). Distant
monitoring and controlling of solar driven irrigation system through IoT. In: 2018
National Power Engineering Conference (NPEC) (pp. 1–5). IEEE.
12. Tamoor, M., ZakaUllah, P., Mobeen, M., & Zaka, M. A., (2021). Solar powered automated
irrigation system in rural area and their socio economic and environmental impact.
International Journal of Sustainable Energy and Environmental Research, 10(1), 17–28.
13. Rehman, A. U., Asif, R. M., Tariq, R., & Javed, A., (2017). Gsm based solar automatic
irrigation system using moisture, temperature and humidity sensors. In: 2017
International Conference on Engineering Technology and Technopreneurship (ICE2T)
(pp. 1–4). IEEE.
14. Rout, K. K., Mallick, S., & Mishra, S., (2018). Solar powered smart irrigation system
using internet of things. In: 2018 2nd International Conference on Data Science and
Business Analytics (ICDSBA) (pp. 144–149). IEEE.
15. Tapiceria, P. A., & Magwili, E. G., (2021). Hybrid solar-hydrokinetic powered
automated irrigation system. In: 2021 IEEE 13th International Conference on Humanoid,
Nanotechnology, Information Technology, Communication and Control, Environment,
and Management (HNICEM) (pp. 1–5). IEEE.
16. Sharma, H., Kumar, P., Pal, N., & Sadhu, P. K., (2018). Problems in the accomplishment
of solar and wind energy in India. Problemy Ekorozwoju, 13(1).
17. Kumar, P., Sharma, H., Pal, N., & Sadhu, P. K., (2019). Comparative assessment
and obstacles in the advancement of renewable energy in India and China. Problemy
Ekorozwoju, 14(2).
18. Kumar, P., Pal, N., & Sharma, H., (2022). Optimization and techno-economic analysis
of a solar photo-voltaic/biomass/diesel/battery hybrid off-grid power generation system
for rural remote electrification in eastern India. Energy, 247, 123560.
19. Kumar, P., Pal, N., & Kumar, M., (2021). Hybrid operational deployment of renewable
energy—A distribution generation approach. In: Design, Analysis, and Applications of
Renewable Energy Systems (pp. 627–643). Academic Press.
Non Commercial Use

20. Kumar, P., Pal, N., & Sharma, H., (2021). Techno-economic analysis of solar photo-
voltaic/diesel generator hybrid system using different energy storage technologies for
isolated islands of India. Journal of Energy Storage, 41, 102965.
21. Kumar, P., Kumar, M., & Pal, N., (2021). AN efficient control approach OF voltage
and frequency regulation IN an autonomous microgrid. Revue Roumaine Des Sciences

Techniques-Serie Electrotechnique Et Energetique, 66(1), 33–39.
22. Hosenuzzaman, M., Rahim, N. A., Selvaraj, J., Hasanuzzaman, M., Malek, A. A., &
Nahar, A., (2015). Global prospects, progress, policies, and environmental impact of solar
photovoltaic power generation. Renewable and Sustainable Energy Reviews, 41, 284–297.
23. Kumar, P., Chandra, K. A., Patel, S., Pal, N., Kumar, M., & Sharma, H., (2020).
Operational challenges towards deployment of renewable energy. Resources, Challenges
and Applications, 129.
24. Kumar, P., Sikder, P. S., & Pal, N., (2018). Biomass fuel cell based distributed generation
Author Copy
system for Sagar Island. Bulletin of the Polish Academy of Sciences: Technical Sciences, 5.
Non Commercial Use

Non Commercial Use

Author Copy
CHAPTER 16
Optimal Parameter Estimation of

3-Phase Transmission Line Using a Grey
Wolf Optimization Algorithm
MUHAMMAD SUHAIL SHAIKH,1 ABDUL LATIF SHAH,2
Author Copy
SHAFIQ UR REHMAN MASSAN,2 RABIA ALI KHAN,2 MUNSIF ALI JATOI,4
SHUBASH KUMAR,3 and MANNAN HASSAN5
1
Guangdong, China
2
3
School of Electrical Engineering, Yanshan University, Qinhuangdao,
Republic of China
4
Salim Habib University, Karachi, Pakistan
5
School of Electrical Engineering, Southwest Jiaotong University, Chengdu,
Republic of China
ABSTRACT
A optimization, selects the best option from a list of possible options to meet a
predetermined set of criteria. In pursuit of the best and most optimal solution
for a particular issue, this theory is relevant to virtually all areas of natural as
well as social sciences, and it has applications across the board. It’s possible
to utilize the optimization domain to either maximize or decrease the value
of an actual function by selecting input values carefully and computing their
values. A significant field of applied mathematics is devoted to the application
of minimization and maximization with different restrictions on a wide variety
of objective functions. Many different optimization methods may be used
Non Commercial Use

constructed on the cost function and restrictions that are existing. As a result,
optimization methods are generally classified into the following categories:
discrete and continuous optimization, Unconstrained and Constrained
optimization, and none or multi-objective functions. Swarm intelligence is
a kind of artificial intelligence (AI), which mimics the intelligent behavior

of swarms, flocks, herds, and other groups of animals seen in nature. As a
result, these algorithms make use of the collective behavior of animals’
groupings. The whale optimization algorithm (WOA), dolphin echolocation
(DE), CS algorithm, bat algorithm (BA), Harmony Search (TS), artificial bee
colony (ABC) algorithm, grey wolf optimizer (GWO), fruit fly optimization
algorithm (FOA), the, hill-climbing, iterative local search, and other SI
Author Copy
algorithms are available. (This chapter will cover the idea of optimization,
as well as mathematical derivations and explanations of their applications in
real-world situations). Following that, it will examine the specific application
of optimization methods to one of the electrical engineering issues, namely,
the parameter estimation of overhead transmission lines, in more detail. It
is the transmission line, which is composed of inductance, capacitance, and
resistance, that is the most critical component of the power system. During the
transmission line design phase, these characteristics are very important. The
overhead transmission line parameter is computed using a new optimization
method known as gray wolf optimization, which is described in more detail
in the literature. GWO is a meta-heuristic algorithm that was recently created
based on natural inspiration. For testing, 3-phase transmission line test systems
are used. The command structure and hunting technique of gray wolves served
as inspiration for the suggested algorithm. The technique is used for several
optimization functions that differ in size and a lot of search agents participating.
The GWO algorithms provide optimal results that outperform those produced
by other algorithms previously in use. For the vast majority of these statistically
verified functions, the suggested method provided the most optimum solutions
that could be found. Following the results, it has been discovered that the
suggested method is more computationally efficient than the conventional
algorithm and exceeds it in terms of accuracy, resilience, and convergence
time. Moreover, the best optimal obtained using GWO for 3-phase capacitance
and inductance are 0.22436, 0.022935, 0.65915, and 0.34938.
16.1 INTRODUCTION
It is important to estimate line parameters since they are used in a variety of
power applications. The techniques of calculating line parameters may be
Non Commercial Use

Optimal Parameter Estimation of 3-Phase Transmission Line 315
divided into two categories: off-line methods and online approaches. Off-line
techniques measure the line characteristics based on the design of the tower
and the kind of wire used. However, because of the fluctuating nature of
the load and the changing weather conditions throughout the operation, the
measured line characteristics may prove to be incorrect. As a consequence of

the noise introduced during PMU measurements, estimate results that differ
from their real value have been seen [1]. The capacity of the transmission
line must be adequate to meet power flow changes. Transmission line
parameter estimation models are also commonly employed in a range of
network applications. Previously, power transmission parameters were
computed using physical transmission line features such as cable geometry,
Author Copy
tower structure, and length. These measures are subject to mistakes because
they overlook constantly changing operational elements including age, skin
effects, air temperature, and other environmental influences, which are all
taken into account. According to a recent trend, physicist-based strategies
are progressively being supplanted by measurement-based approaches.
Such measurements may be performed using a supervisory control and data
acquisition (SCADA) device, a phasor measuring unit (PMU) fault record.
Failure records are used in certain techniques for calculating parameter
estimates. To categorize parameters from reported fault transient recordings,
a phase domain strategy is provided, as well as relying on prony predicting
methods to detect fault transient features. Using state growth methods
necessitates a significant degree of redundancy in the computation as well
as subjective standards regarding parameters, which makes them impractical
for practical use. Furthermore, the computation of the parameter may be done
independently of the estimate of the current state of affairs. It is proposed to
use numerous measurement snapshots to do off-line parameter estimation,
including iterations for parameter estimate and states utilizing multiple
measurement snapshots. To estimate the parameters, a PMU technique is
created that uses the estimated condition of many measurement snapshots to
determine their values. As previously stated, several of the techniques make
use of numerous measurement snapshots to enhance repetition. To build a
well-conditioned coefficient matrix, the snapshot number must be maintained
as low as practicable. The limitations of parameter estimating strategies
based on distinct measurement snapshots stem from these competing factors.
Using dynamic parameter approximation, which is useful for future grids,
can alleviate these concerns [2].
Any power system’s principal goal is to provide high-quality, reliable,
and cost-effective energy to its customers. Furthermore, the energy system
Non Commercial Use

must achieve transmission, production, and supply of power systems,

since meeting changing consumer demand is difficult [3–5]. The use of
mathematical simulation is critical in the study of power systems. In the
domain of power systems, there are multiple complex problems with no
single solution. So, optimization approaches are employed to solve these

undetermined challenges. Multiple artificial intelligence (AI) optimization
methods have been adapted for different settings and issues. The main
advantage of AI is computation accuracy [6, 7]. There are several different
AI algorithms, including the ant lion optimizer (ALO), Hirschberg–Sinclair
algorithm (HS), dragonfly algorithm (DA), Tabu search (TS), compare and
swap algorithm (CAS). The overall costs and computing durations incurred
Author Copy
by the GWO method were lower than those incurred by other optimization
algorithms, and the algorithm demonstrated better convergence rates and
superior performance when compared to other optimization algorithms to
demonstrate the effectiveness.
To optimize engineering design issues, a gray wolf optimization method
(GWO) is being suggested. It all starts with creating an initial position for
the gray wolf population using a tent map, which distributes the population
fairly and prepares for a massive global search and rescue effort. Second,
to prevent the method from being trapped in a local optimal state, Gaussian
mutation perturbation is employed to perform various actions on the present
optimum solution.
The swarm intelligence method is capable of dealing with a wide range
of issues that traditional optimization approaches are unable to handle
effectively or efficiently. As a result, optimization methods have become
more popular in a broad range of fields of study. In other words, current
algorithms are capable of addressing certain problems well, but not all of
them do so consistently. Some novel heuristic algorithms are suggested
each year as a consequence of this, and ongoing research is conducted in
this area. Actual monetary worth is a novel population-based meta-heuristic
technique called grey wolf optimizer (GWO). Because the GWO method is
straightforward, versatile, and efficient, it may be employed efficiently in a
wide variety of real-world situations. There are many variants of the GWO
algorithm available for download. GWO algorithms were combined with
other investigative optimization approaches by several authors. As all of
these approaches depend on binary encoders to track down the gray wolf,
the information stored in genetic information is limited in its utility. In
every evolutionary method, the convergence rate is given precedence over
the quality of solutions while addressing an optimization issue, regardless
Non Commercial Use

of the technique. Generalized evolutionary computing technology (GWO)

produces superior outcomes in comparison to previous evolutionary
computer technologies. Real-time applications, on the other hand, are
primarily concerned with the convergence time [8–12].
A GWO is a recently developed competent meta-heuristic process that

mimics the public pyramid behavior of gray wolves [13]. Grey wolves live
in a group with an average number of participants ranging from 5 to 12
individuals. Grey wolves are regarded as excellent hunters, and their survival
and reproduction are dependent on the leadership chain and hunting etiquette.
Section 16.2 provides a thorough description of the GWO algorithm’s internal
structure. The GWO algorithm has been utilized to tackle issues in a variety
Author Copy
of engineering fields, including energy storage system sizing [17], optimum
dispatch of actual and reactive power [15, 16], fuzzy control system tuning
[18], and load frequency management [14]. Each new system is described in
its simplest form, after which it may be tweaked and hybridized with existing
algorithms to solve global or specialized functions [19–21]. Among other
things, the suggested MGWO algorithm is distinguished by its simplicity,
adaptability, and globalism. Scientific experts have suggested many changes
to the GWO algorithm to avoid slipping into the local optima trap. Ref. [22]
presents the GWO technique, which combines the GWO with Lévy Flight
(LGWO) to avoid local goals stagnation and increase performance. However,
this incorporation increased the complexity of the original GWO technique.
To be assessed utilizing different bundle conductor designs, transmission
line parameters must be analyzed using an optimization technique that
meets the following requirements. However, the complexity of the original
GWO method was increased as a result of this integration. Transmission
line parameters must be evaluated using an optimization method that has
the following criteria to be evaluated using various bundle conductor
configurations.
The following are the chapter’s major contributions:
• A detailed review of different optimization techniques is presented.
• The 3-phase transmission line’s parameters are estimated using GWO
on a separate bundle conductor.
• Four cases are studied while performing transmission line problems,
two bundle capacitance, 3 bundle capacitance, two bundle inductance,
and finally 3 bundle inductance.
• Best optimal solution considering the cases.
• Proposed GWO was compared with WOA available in the literature.
Non Commercial Use

16.2 LITERATURE REVIEW (DIFFERENT OPTIMIZATION

TECHNIQUES)
The field of computational intelligence (CI) includes meta-heuristic

algorithms based on nature. Mathematics CI (MathCI), Physicist CI

(PhyCI), bioinformatics CI (BioCI), and chemistry CI (ChmCI) are the four
main types of processes in CI [23]. Bioinformatics-based algorithms are
those that are based on biological data. The two main types of biology-based
meta-heuristic algorithms are evaluation-based algorithms and swarm-
based algorithms. Evaluation-based algorithms are the most common kind
of biology-based meta-heuristic algorithm. It has become widely accepted
Author Copy
in several engineering fields because of its benefits over other selections
of nature-based algorithms. Swarm meta-heuristic Algorithms have gained
widespread acceptance in a variety of engineering sectors due to their benefits
on classifications of nature-based algorithms. It has benefits over estimation-
based algorithms in that it keeps track of the search space throughout each
iteration and needs minimal operators for successful implementation, resulting
in rapid execution. For their part, the knowledge retained by evaluation-based
algorithms does not hold up whenever a new population is created, and this
results in a higher number of operators being required [24]. With exponential
growth in the issue size, swarm-based algorithms have been demonstrated
to be more efficient than other algorithms in addressing high-dimensional
across a wide area, combinations and nonlinear optimization problems
search region [25–27]. The academic community has been fascinated with
meta-heuristic algorithms for almost two decades, and it is no surprise that
they remain so. Particle swarm optimization (PSO) [28, 29], ant colony
optimization (ACO) [30, 31], artificial bee colony optimization (ABC) [32],
CS algorithm [33, 34], krill herd (KH) algorithm [35], bat-inspired (BA)
algorithm [36], firefly algorithm [37], genetic algorithm (GA) [38, 39], The,
evolution strategy (ES) [40], probabilistic incremental learning (PBIL) [41],
Genetic programming (GP) [42, 43], and biogeography-based optimizer
(BBO) [44, 45] are examples of another generation of evaluation-based
algorithms.
16.2.1 GWO
Recently [46], The GWO is a meta-heuristic algorithm that was created

to imitate the social configuration and Grey Wolves in the Wild Hunting
Non Commercial Use

Behavior. The GWO is composed of four types of Grey Wolves, each of

which has its unique hunting behavior (alpha, beta, delta, and omega) are the
four types of the wolf. Moreover, they utilized to approximate the leadership
hierarchy which is evolved as a result of the group’s struggle for survival
as a whole, which is modeled after the human brain. To carry out a collec-
tive hunt, it is frequently necessary to use 3 complex movements: searching
for prey, surrounding of prey, and attacking of prey. These maneuvers have
been described in detail in Ref. [14]. When it comes down to it, gray wolf
optimization has the following 3-distinct advantages: When associated with
other standard heuristic methods using a gravitational search method, Thus,
a greater rise in global optimization is possible since it allows for effective
Author Copy
exploitation and inquiry, as well as proactive local optimization prevention
and promising implementation in the face of uncertainty. A multi-objective
optimization technique based on differential evolution (DE) was used to
improve the size of a solar water pumping system which has since been
recommended [16]. Evolution programming (EP) is used to improve the
viable path supplied by a set of local activities, and the ABC approach is
also utilized as a local search method [17]. Many other areas have benefited
from GWO, for example, preventing a real power system from becoming
dark owing to failures in generating units or major transmission lines [18].
Furthermore, in Ref. [19], the optimal reactive power is determined to reduce
loss and voltage variation to the bare minimum.
This part models hunting methods as well as the social hierarchy to create
Grey Wolf Optimization and carry out work in the field.
16.2.1.1 INSPIRATION
Researchers believe gray wolves are top-of-the-food-chain predators, making

them top hunters. They would rather live in groups, and a distinct group of 5
to 12 candidates almost invariably follows a strict social leadership structure.
Because the alpha wolf should be followed by the rest of the pack, he is often
referred to as the dominating wolf. In the Mathematical model of hunting,
the GWO mechanism is outlined as below:
• Moving towards prey;
• Follow the prey till he moves, surrounding him, and harassing him
until he moves;
• Finally, attacking the prey.
Non Commercial Use

16.2.1.2 SOCIAL HIERARCHY
During mathematical modeling, make certain that the Grey Wolf Optimi-
zation is designed in such a way that the optimal solution in the alpha is
accepted. As a result, beta () is the second-best solution to the problem.

The delta () wolf is the third-best, while the remaining wolves are omega ()
wolves. The alpha, beta, and delta algorithms are responsible for the hunting
process.
16.2.1.3 TRACKING
Author Copy
As the first stage in Grey Wolf Optimization, they take down the gray wolf
and catch up with it during the hunting process. The mathematical Eqns. (1)
and (2) may be used to explain the wolf’s behavior.
   
X ( t +=
1) X p ( t ) − A.D (1)
   
=D C. X p ( t ) − X ( t ) (2)
 
where; X is the gray wolf position, t shows the number of iterations; X p
indicates the location of the prey; D denotes the distance between both prey
and wolf; A denotes the number of coefficients, which are computed as
follows:
   
A 2a. r1 − a
= (3)
where; r1 and r2 is the vector’s two random values within the range [0,1].
16.2.1.4 ENCIRCLING
The technique of GWO while attacking agents is shown in this attacking

prey illustration. As a consequence of the transmission line features a study,
the attacker’s behavior, which is dependent on leaders and their colleagues,
encircles the agent. Because capacitors and inductors are submissive to the
transmission line parameter, delta wolves compute it and assault the agent
farther down the transmission line parameter chain. The transmission line
parameter, also known as omega and scapegoat, is defined as the effects of
agents on end-users. We may update their positions by computing alpha,
  
beta, and delta, which are ( X α ), ( X β ), and ( X δ ), respectively.
Non Commercial Use

  

 X + X2 + X3
X ( t + 1) =1 (4)
3
   
= X 1 X α ( t ) − A1 . Dα (5)
   

X2
= X β ( t ) − A2 . DB (6)
   
X3
= X δ ( t ) − A2 . DB (7)
   
D= α C1 − X α ( t ) . X t (8)
   
D= C2 − X β ( t ) . X t (9)
Author Copy
β
16.2.1.5 ATTACKING
Grey wolves complete their hunt by attacking their prey after it stops moving.
The value of “a” in Eqn. (3) will drop linearly as a result of these kinds
of activities. This time also includes a change in the quantity of A, which
goes from [–a] to [a]. As an alternative, whenever |A| <1, the wolf’s location
becomes closer to the prey’s position. When the value of |A| >1, the pack of
wolves will run away from their target. As a result, in gray wolf optimiza-
tion, the update equation of “a” is the same as in Eqn. (10).
a= 2 − 2* t / Max iter (10)
where; ‘t’ is the current iteration.
16.2.2 WOA
WOA is a continuous optimization problem-solving method based on swarm

intelligence. When utilized in conjunction with current meta-heuristics
techniques, it has been demonstrated to improve performance [52, 53]. It is
simple to create and durable, for example, when compared to other swarm
intelligence systems, making it comparable to other nature-inspired algorithms
in terms of performance. The algorithm requires a lesser number of control
parameters; in reality, only one parameter (the time interval) needs to be fine-
tuned. The researchers have built a multi-dimensional search habitat for the
humpback whale population in WOA forages. It’s important to note that the
distance between humpback whales and their food is connected to the amount
Non Commercial Use

of objective cost in this model, hence individual whale locations are option
variables. Three operational strategies are employed to determine a whale’s
time-dependent position: shrinking around prey, bubble-net assault technique
(exploitation phase), and seeking for prey (exploration phase). The next
subsections explain various operating procedures, as well as a mathematical

expression for them.
16.2.2.1 ENCIRCLING PREY
When in their native habitat, humpback whales can recognize and circle their
Author Copy
prey. Because the ideal design is situated in the ideal location. Considering
that the search space is unknown at the onset, the WOA considers that the
current best answer is either the intended prey or something that is close by.
Other search agents are updating their locations around the top search agent
while the best search agent is being sought. The equations shown below
reflect the behavior.
16.2.2.2 SEARCHING AND ENCIRCLING PREY
Eqn. (11) may be used to demonstrate the process of looking for food (prey)
   
= D C. X rand − X (11)
   
X ( t + 1=
) X rand − A.D (12)
where; A and C are vectors of coefficients, as shown:
  
A 2a.r − a
= (13)
 
C = 2.r (14)
The value of a decreases linearly from 2 to 0, while the value of r is within
the range [0,1].
   
=D C. X * ( t ) − X (15)

where; D is the prey update location.
   
X ( t +=
1) X * ( t ) − A.D (16)
If A≥1 then Eqn. (11) and (12) are used. It means that the whale is
searching for prey. If A<1, then Eqns. (15) and (16) are used, which means
that the whale already searched for food and now attacking the prey.
Non Commercial Use

Where; ‘t’ is the number of iterations currently being performed; ‘X’ denotes
the vector position; and X* denotes the best possible value.
16.2.2.3 POSITION OF SPIRAL UPDATE
Eqn. (17) shows the new position:

  
  X * ( t ) − A.D ifp < 0.5
X ( t + 1) = bl  (17)
 D.e .cos ( 2π l ) + X ( t ) ifp ≥ o.5 ##
*
Author Copy
where; p is a number between [0,1], l is between [–1,1], and b is a constant.
16.2.3 MFO
It was Mirjalili [54] who first suggested the moth–flame optimization (MFO)
method. Specifically, it falls under the category of population-based meta-
heuristic methods. A random night butterfly is created in the solution space,
the fitness values of each moth are calculated, and then the moth with the
greatest fitness value is marked with a flame using MFO. A spiral movement
function is used to update the moths’ locations to attain flame-decorated ideal
positions, As a result, each moth’s location is updated, and new positions are
generated, all of which are repeated until the termination requirements have
been reached. Table 16.1 lists the MFO’s features.
TABLE 16.1 Features of MFO

Algorithm Description MFO Elements
Decision variable Moths position in every dimension
Solutions Moth position
Initial solutions Moth random position
Current solutions Moth current position
New best solution New position
Best optimal solution Flames position
Fitness function The distance between moth and flame
New generating process for solution A spiral path towards the flame
Creating the first population of Moths: As previously stated in Ref.

[55], Mirjalili believed that each moth can fly in one of four dimensions:
Non Commercial Use

one-dimensional, two-dimensional, 3-dimensional, and hyperdimensional

space. The following is an expression for the collection of moths:
 MO1,1 MO1,2   MO1, d 
 MO MO2,2   MO2, d 
 2,1
M =        (18)
 
      
 MOn ,1 MOn ,2   MOn , d 

where; ‘n’ is the number of moths; and ‘d’ is the number of columns of
matrix ‘M.’
Author Copy
The MFO algorithm’s remaining components are mostly flames. Each
flame is represented by a 3-dimensional matrix, with fitness function vectors
connected with each flame.
 F1,1 F1,2   F1, d 
F F2,2   F2, d 
 2,1
F =       (19)
 
      
 Fn ,1 Fn ,2   F2, d 

where; ‘n’ is the number of moths and d is the dimension of the moth matrix.
It should be mentioned that both moths and flames are effective remedies
in this situation. It is the manner we handle and modifies them for each itera-
tion that distinguishes them from one another. A search agent is a creature
that moves throughout the search area, While flames are the greatest habitat
for moths thus far. To put it another way, think of flames as flags or pins
that moths deposit in their search area as they fly around looking for food.
It’s because of this that each moth searches for an alternative to the current
solution near a flame (and changes the flag accordingly). Because of this
procedure, a moth’s best option is always within reach.
16.2.3.1 UPDATING THE MOTHS’ POSITIONS
MFO uses three-types of functions for convergence of the global optimiza-

tion of the optimization problems which are defined below:
MFO = (I, P, T ) (20)
where; I is a function; ‘P’ is the flight of the moth in search of space; and ‘T’
is the stopping criteria.
Non Commercial Use

Mi = S(Mi, Fj) (21)

where;
• Mi is the number of ith moths;
• Fj indicates the number of jth flames;

• S is the twisting function and can be expressed as follows:
S ( M i , Fj ) Di .ebt .cos ( 2π t ) + Fj
= (22)
where;
• Di is the distance between moth and flame;
Author Copy
• b is a constant value [1];
• t is a random number within the range of [–1,1];
• Di is the distance can be calculated as:
Di = |Fj – Mi| (23)

The moth’s position is obtained by Eqn. (23) and the distance between
moth and flame is controlled by t. The twisting motion of the moth
describes how to update its position around the flame. Updating the posi-
tion about diverse locations to contain this situation, Eqn. (26) is utilized
in this regard.
 N −1 
Flame number = round  N − I *  (24)
 T 
where;
• I is the number of iterations;
• N is the total amount of flames;
• T represents the number of iterations.
Another kind of meta-heuristic is a first-generation method that may
operate in conjunction with the top secondary heuristics to efficiently find
and provide the best or just slightly better than optimal solutions for opti-
mization issues. All meta-heuristics must search in both depth and breadth
throughout their search process. Developing an appropriate diversification
strategy is critical to dealing with this situation effectively. Meta-heuristic
algorithms are used to find excellent results to optimize issues in situations
when the complexity of the problem being addressed or the amount of search
time accessible prevents the usage of precise optimization algorithms from
being used.
Non Commercial Use

16.2.4 TABU SEARCH (TS)
Tabu exploration is a meta-heuristic method that involves randomly

exploring the entire solution space. Tabu search was developed by Fred
Glover and colleagues as a way of locating or controlling local optima.

Using the Tabu search has the advantage of effectively locating the history
of the stop cycle and avoiding getting caught in local minima using custom-
izable remembers [56]. The higher-level heuristic algorithm is solved, and
efforts are made to determine whether the solution is good. Nara et al. [57]
proposed a Tabu search to find the ideal position in terms of DG with the
fewest losses in distribution networks or systems. According to the conclu-
Author Copy
sions of this study, the author investigated how much the power system’s
loss percentage may be reduced following optimal DG allocation on the
demand side of the system. This technique is based on the premise that the
total number of DGs and total DG producing capacity are both equal to or
greater than 1. When the NSGA II is employed as a pointing out technique,
a multi-objective Tabu search is compared to the NSGA II, and MOTS
delivers a much superior result in terms of processing time, an enticing
aspect, especially as the investigation progresses and the passage of time
becomes more critical [4].
16.2.5 PARTICLE SWARM OPTIMIZER
It is one of the most frequently employed stochastic optimization methods,

and it is known as particle swarm optimization (PSO). Developed by Dr.
Eberhart and Dr. Kamedy in 1995, this method is especially well suited for
convivial breeding or nesting in bird colonies or gregarious breeding in bird
colonies, or gregarious breeding in bird colonies. To increase the voltage
profile and decrease power losses as well as Total Harmonic Distortion
(THD), a PSO-based technique is employed as a distribution network solu-
tion tool. It also reduces THD. The primary objective of this technique is
to decrease total harmonic distortion (THD) while concurrently improving
the voltage profile. Smart PSO is, by its nature, very comparable to GA in
terms of functionality. A random population will be used to start the system
and subsequently, an upgraded population will be used to achieve optimal
outcomes, based on the outcome of PSO and GA. The Proposed approach
has provided a solution of better quality than the GA technique, which was
previously used. Contrasting PSO to GA, the time required to complete the
Non Commercial Use

task is much less, and the method can be implemented to a real network with
a high degree of ease. By focusing on bus voltage limitations as the major
constraints, the recommended technique is one way to reduce system power
losses. To reduce total power losses, the optimal placement and capacity
of distributed generation (DG) must be determined by comparing the new

network to the old one. As well as this, the upgraded power system’s line
voltage consistency is later connected, and the DG is recognized by the line
voltage consistency signal. Applications for the real-time power system
utilize the system.
Author Copy
16.2.6 PARTICLE SWARM OPTIMIZER
The ant colony transmits the algorithm for ant interbreeding with a single
inhalation. The ants disperse pheromones on the earth’s surface to choose
a suitable path and then operate following the colony’s social organization.
The ant colony method, often known as a beginning algorithm, is an optimi-
zation approach. At the start of each reiteration of a procedure, new artificial
ants are utilized to recognize or construct the process. Every single person
participating in inventing a solution for performing the traveling action from
the lowest point to the highest point on a plot chart while avoiding a barrier
that does not tour the highest point that has already been visited. At the end of
each iteration, the pheromone price is altered based on the quality of the ants’
solutions, to bias the ants’ future time iterations to construct the best feasible
solution in comparison to the previous. An ACO approach for locating DG in
the distribution network. The suggested ACO algorithm has been tested and
shown to minimize operational expenses while improving electricity quality
and customer service dependability. DG increases electricity quality and
customer service dependability while cutting operational costs. To reach the
distributed system’s goal, an ant colony employs optimization dependability
to execute a distributed system’s optimal distribution generation location
plan. To solve the problem, the ACO method was adopted. They discovered
that the ACO technique outperforms both the external genetic algorithm (GA)
and the need for normal computing time addition. This has the potential to
cause ACS to persist for a long time, and it should be taken carefully since it
may lead to much more solved possibilities in the future. However, because
of the frequent response complication size and the computational efficiency
of the system, extra time is not needed for the procedure under investigation
to be completed.
Non Commercial Use

16.2.7 SIMULATED ANNEALING (SA)
Meta-heuristics are a global optimization problem of finding a decent

approximation to the global ideal of a job and useful when the location
and range are distinct. Annealing guarantees that the natural optimization
process happens with the least energy. And in 1985, Vlado Cerny described it
to Scott Kirkpatrick, C. Daniel Gelatt, and Mario P. Vecchi. The well must be
solved using simulated annealing cold. What are the beginning heat intensity
and cooling procedures for the great significance of simulated annealing? It
is possible to achieve more than standard solutions effectively and similarly
as well as simply one goal using simulated annealing.
Author Copy
16.2.8 GENETIC ALGORITHM (GA)
A kind of optimization technique is the GA, which uses natural progression

mechanisms including mutation, option crossover, and inheritance to solve an
optimization problem. When an encoding candidate is solved to the ideal diffi-
culty supplied for a great solution, the genetic approach first builds randomly
residents (or chromosomes) of the string. The general health of the residents is
assessed at the end of each cycle to build a new dweller. The algorithm’s next
or upcoming iteration is used in the new modified population. A good health
plane or flat is obtained by repeating this approach until an extreme measure
or number of generations is reached. DG placement and sizing GA The DG is
normally nearer the substation. Moving DG closer to the load center minimizes
power loss and has additional advantages, according to the suggested algo-
rithm. By reducing power losses and improving power quality, this research
endeavor provides the greatest answer for a group’s size or upgrade. The
suggested multi-objective technique minimizes all functions simultaneously.
Have recommended a site and size for the DG. The method is evaluated on an
IEEE 69 bus system and compared to nonlinear optimization. According to the
DG study, improper DG location and size would reduce system stability and
dependability, increasing system losses. This strategy minimizes power losses
and DG investment costs while optimizing DG placement and size.
16.2.9 HYBRID OPTIMIZATION TECHNIQUES
Generating GAs is very useful for figuring out the optimal size and loca-
tion for distributed generating systems. The GA, on the other hand, can
Non Commercial Use

sometimes produce low-quality results, making it inefficient when exact

results are required. The evolutionary algorithm is employed in concert with
other heuristic optimization approaches to overcome this constraint. The
equivalence along traditional GA was solved by integrating GA with SA to
uncover the method’s ability to construct a fantastic class. A solution based

on GA and fuzzy logic that takes into consideration the different forces of the
objective function is one of the applications of multi-objective issues. To get
a better result, researchers used additional optimization methods with GA.
The transmission and distribution of energy from generators to consumers is
the primary function of the electrical transmission and distribution system
Many power applications need an accurate estimation of line parameters
Author Copy
[49–58]. In a deregulated environment, the overhead transmission line
system (OHTL) is critical [47, 48].
This section will present, analyzes, and discusses the results. A Computer
with a 64-bit Intel(R)-Core (TM) i7-4610M CPU @ 3.00 GHz and 12 GB
RAM is set up with the Windows 10 operating system. Finally, the following
section is organized:
16.3.1 CASE 01: THE CAPACITANCE WITH A TWO-BUNDLE

CONDUCTOR FOR A 3-PHASE TRANSMISSION LINE
Initially, the suggested GWO was used to calculate 3-phase transmission line
characteristics, and a value of 0.22436 was determined as the optimal capaci-
tance per unit length for use with the two bundle conductors. Figure 16.1(a)
shows the convergence curve between the GWO and WOA for transmission
line parameter estimation when the iteration number is 100. As you can see
in Table 16.2, the optimum solution to the transmission line problem may be
found.
16.3.2 CASE 02: THE CAPACITANCE WITH A 3-BUNDLE CONDUCTOR

FOR A 3-PHASE TRANSMISSION LINE
In this case, we have analyzed capacitance per unit length considering 3 bundle
conductors. Figure 16.1(b) displays the suggested algorithm’s convergence
Non Commercial Use

characteristic, with the ideal value for this example being 0.022935 when
the number of search agents is 20 and the number of iterations is 100. As
shown in Table 16.2, using GWO provides the best possible outcomes when
compared to those produced by using WOA. An algorithm’s reliability and
superiority over an existing one must be demonstrated.
16.3.3 CASE 03: THE INDUCTANCE WITH A TWO-BUNDLE

CONDUCTOR FOR A 3-PHASE TRANSMISSION LINE
For the two bundle conductors, the recommended GWO was utilized to
Author Copy
compute 3-phase transmission line inductance. Figure 16.1(c) shows the
convergence curve for the estimate of transmission line parameters using
GWO and WOA at 100 iterations. As shown in Table 16.2, the transmission
line issue may have an optimal solution and the best optimal value is 0.65915.
16.3.4 CASE 04: THE INDUCTANCE WITH A 3-BUNDLE CONDUCTOR

FOR A 3-PHASE TRANSMISSION LINE
Here, we looked into inductance per unit length for 3 bundle conductors. Figure
16.2(d) illustrates the suggested algorithm’s convergence characteristic, with
the most optimum value being 0.34938 for 20 search agents and 100 iterations.
Figure 16.1(d) shows the suggested method’s convergence. Table 16.2
compares the results acquired using GWO to those obtained using WOA. To
show the GWO algorithm’s supremacy over the WOA algorithm’s reliability.
16.4 CONCLUSION
This chapter covered a variety of optimization techniques for increasing the size
and position of DG in the distribution system, including analytical, heuristic,
and hybrid methodologies. In the appendix, there is a brief comparison of
different optimization strategies. The research found that analytic methods
perform poorly when dealing with complicated network issues. While heuristic
methods provide a simple answer. However, it may occasionally result in low-
quality outputs and more processing time. A combined strategy that incorpo-
rates two or more heuristic techniques may provide a high-quality result. This
chapter discusses the parameter computation of the overhead transmission
line (OHTL) power system using the suggested GWO algorithm. Three-phase
Non Commercial Use

FIGURE 16.1 Three-phase transmission line convergence characteristics – (a) the Author Copy
capacitance with the two-conductor bundle; (b) the capacitance with the 3-conductor bundle;
(c) the inductance with the two-conductor bundle; and (d) the inductance with a 3-conductor
bundle.
transmission lines models, the method was applied to 06 functions, respectively.

The total costs and calculation durations encountered while using the GWO
method were smaller than those attained when utilizing different optimization
procedures with a greater convergence rate, such as WOA. The collected results
confirmed that the suggested GWO algorithm outperforms the other methods
in terms of exploitation. For power systems, the suggested method is both
quick and robust. Grey Wolf Optimization’s convergence feature shows its
ability to tackle a tough issue. The suggested method’s findings are compared
to the traditional approach for 3-phase transmission line systems. The findings
Non Commercial Use

Apple Academic Press 332
TABLE 16.2 Comparative Analysis of 3-Phase Transmission Line

Parameters Capacitance per Unit Length (μF) Inductance per Unit Length (mH)
GWO WOA Parameters GWO WOA
Control variable Control variable obtained Control variable Control variable obtained
obtained by GWO by WOA obtained by GWO by WOA
Case 01 –0.013912 –5.4542e-07 Case 03 0.0051506 0.00031172
0.00098618 –1.4406e-07 1.05 1.01
0.881 0.82 0.34048 0.008144
–0.074601 0.82 –0.46989 1.01
0.2162 0.82 –0.011444 1.01
Best Optimal Value Best Optimal Value Best Optimal Value Best Optimal Value
0.22436 0.22475 0.65915 0.66063
Case 02 0.94 0.89 Case 04 –0.00052943 0.019337
0.94 0.89 2.5 0.078428
–0.42275 0.31239 –0.047056 0.077129
0.94 0.89 0.084181 0.02629
Non Commercial Use

–0.39658 0.27428 0.012898 0.089032
Best Optimal Value Best Optimal Value Best Optimal Value Best Optimal Value
0.022935 0.023897 0.34938 0.42958
Author Copy
demonstrate Grey Wolf Optimization’s efficacy in resolving the transmission

line parameters issue. Finally, the GWO method outperforms other algorithms
in terms of resilience, computing effort, and avoidance of premature conver-
gence. The GWO algorithm’s convergence accuracy has been significantly
improved by evaluating 06 functions and comparing the results with ALO

and DA. With 100 maximum iterations and 20 populations, it achieves a 90%
success rate for all optimization functions, demonstrating GWO’s efficacy in
addressing difficult problems. The best optimal values for 3-phase capacitance
and inductance are 0.22436, 0.022935, 0.65915, and 0.34938 are the best ideal
values for 3-phase capacitance. For the time being, we shall concentrate on two
problems. On the one hand, we will broaden GWO’s applicability to additional
Author Copy
real-world engineering issues. On the other hand, to tackle optimization issues
more efficiently, we shall create new meta-heuristic optimization techniques.
KEYWORDS
• algorithms
• hybrid optimization
• objective function
• optimization
• parameter estimation
• simulated annealing
• transmission line
REFERENCES
1. Zhang, Y., & Yuan, L., (2020). Kalman filter approach for line parameter estimation
for long transmission lines. In: 2020 IEEE Power and Energy Conference at Illinois
(PECI). IEEE.
2. Ren, P., Ali, A., & Lev-Ari, H., (2019). Tracking transmission line parameters in power
grids observed by PMUs. In: 2019 IEEE Milan PowerTech. IEEE.
3. Ansari, M. M., et al., (2020). Planning for distribution system with grey wolf optimization
method. Journal of Electrical Engineering & Technology, 15(4), 1485–1499.
4. Ansari, M. M., et al., (2019). A review of technical methods for distributed systems
with distributed generation (DG). In: 2019 2nd International Conference on Computing,
Mathematics and Engineering Technologies (iCoMET). IEEE.
5. Ansari, M. M., et al., (2020). Considering the uncertainty of hydrothermal wind and
solar-based DG. Alexandria Engineering Journal, 59(6), 4211–4236.
Non Commercial Use

6. Wolpert, D. H., & William, G. M., (1997). No free lunch theorems for optimization.
IEEE Transactions on Evolutionary Computation, 1(1), 67–82.
7. Zheng-Ming, G., & Juan, Z., (2019). An improved grey wolf optimization algorithm
with variable weights. Computational Intelligence and Neuroscience, 2019.
8. Luo, Q., et al., (2016). A novel complex-valued encoding grey wolf optimization algorithm.
Algorithms, 9(1), 4.
9. Zhang, S., & Yongquan, Z. H. O. U., (2018). Grey wolf optimizer with ranking-based
mutation operator for IIR model identification. Chinese Journal of Electronics, 27(5),
1071–1079.
10. Zhang, S., Qifang, L., & Yongquan, Z., (2017). Hybrid grey wolf optimizer using
elite opposition-based learning strategy and simplex method. International Journal of
Computational Intelligence and Applications, 16(2), 1750012.
11. Zhang, S., et al., (2016). Grey wolf optimizer for unmanned combat aerial vehicle path
Author Copy
planning. Advances in Engineering Software, 99, 121–136.
12. Raj, S., & Biplab, B., (2018). Reactive power planning by opposition-based grey wolf
optimization method. International Transactions on Electrical Energy Systems, 28(6),
e2551.
13. Mirjalili, S., Seyed, M. M., & Andrew, L., (2014). Grey wolf optimizer. Advances in
Engineering Software, 69, 46–61.
14. Muro, C., et al., (2011). Wolf-pack (Canis lupus) hunting strategies emerge from simple
rules in computational simulations. Behavioral Processes, 88(3), 192–197.
15. Xu, Y., et al., (2016). An adaptively fast fuzzy fractional order PID control for pumped
storage hydro unit using improved gravitational search algorithm. Energy Conversion
and Management, 111, 67–78.
16. Muhsen, D. H., Abu, B. G., & Tamer, K., (2016). Multiobjective differential evolution
algorithm-based sizing of a standalone photovoltaic water pumping system. Energy
Conversion and Management, 118, 32–43.
17. Contreras-Cruz, M. A., Ayala-Ramirez, V., & Hernandez-Belmonte, U. H., (2015).
Mobile robot path planning using artificial bee colony and evolutionary programming.
Applied Soft Computing, 30, 319–328.
18. Mahdad, B., & Srairi, K., (2015). Blackout risk prevention in a smart grid based flexible
optimal strategy using grey wolf-pattern search algorithms. Energy Conversion and
Management, 98, 411–429.
19. Sulaiman, M. H., et al., (2015). Using the gray wolf optimizer for solving optimal reactive
power dispatch problem. Applied Soft Computing, 32, 286–292.
20. Jayabarathi, T., et al., (2016). Economic dispatch using hybrid grey wolf optimizer.
Energy, 111, 630–641.
21. Yang, B., et al., (2017). Grouped grey wolf optimizer for maximum power point
tracking of doubly-fed induction generator-based wind turbine. Energy Conversion and
Management, 133, 427–443.
22. Nishmitha, B., et al., (2018). Computation of transmission line parameters using
MATLAB. Int. J. Res. Sci. Innovation, 5(4), 2321–2705.
23. Xing, B., & Wen-Jing, G., (2014). Innovative Computational Intelligence: A Rough
Guide to 134 Clever Algorithms, 22–28.
24. Mirjalili, S., & Andrew, L., (2016). The whale optimization algorithm. Advances in
25. Beheshti, Z., & Siti, M. H. S., (2013). A review of population-based meta-heuristic
algorithms. Int. J. Adv. Soft Comput. Appl., 5(1), 1–35.
Non Commercial Use

26. Hazir, E., Emine, S. E., & Kücük, H. K., (2018). Optimization of CNC cutting parameters
using design of experiment (DOE) and desirability function. Journal of Forestry Research
29(5), 1423–1434.
27. Song, M., & Dongmei, C., (2018). An improved knowledge-informed NSGA-II for
multi-objective land allocation (MOLA). Geo-spatial Information Science, 21(4),

273–287.
28. Zeugmann, T., et al., (2011). Particle swarm optimization. Encyclopedia of Machine
Learning, 1(1), 760–766.
29. Shi, Y., & Russell, C. E., (1999). Empirical study of particle swarm optimization.
Proceedings of the 1999 Congress on Evolutionary Computation-CEC99 (Cat. No.
99TH8406). (Vol. 3). IEEE.
30. Stützle, T., et al., (2011). A concise overview of applications of ant colony optimization.
Wiley Encyclopedia of Operations Research and Management Science, 2, 896–911.
Author Copy
31. Dorigo, M., & Gianni Di, C., (1999). Ant colony optimization: A new meta-heuristic.
Proceedings of the 1999 Congress on Evolutionary Computation-CEC99 (Cat. No.
99TH8406) (Vol. 2). IEEE.
32. Karaboga, D., & Bahriye, B., (2007). A powerful and efficient algorithm for numerical
function optimization: Artificial bee colony (ABC) algorithm. Journal of Global
Optimization 39(3), 459–471.
33. Chiroma, H., et al., (2017). Bio-inspired computation: Recent development on the
modifications of the cuckoo search algorithm. Applied Soft Computing, 61, 149–173.
34. Xin-She, Y., & Suash, D., (2014). Cuckoo search: Recent advances and applications.
Neural Computing and Applications, 24(1), 169–174.
35. Singh, G. P., & Abhay, S., (2014). Comparative study of krill herd, firefly and cuckoo
search algorithms for unimodal and multimodal optimization. International Journal of
Intelligent Systems and Applications in Engineering, 2(3), 26–37.
36. Xin-She, Y., (2010). A new metaheuristic bat-inspired algorithm. Nature Inspired
Cooperative Strategies for Optimization (NICSO 2010) (pp. 65–74). Springer, Berlin,
Heidelberg.
37. Xin-She, Y., & Xingshi, H., (2013). Firefly algorithm: Recent advances and applications.
International Journal of Swarm Intelligence, 1(1), 36–50.
38. Deb, K., et al., (2002). A fast and elitist multiobjective genetic algorithm: NSGA-II.
IEEE Transactions on Evolutionary Computation, 6(2), 182–197.
39. Dasgupta, D., & Zbigniew, M., (2013). Evolutionary Algorithms in Engineering
Applications. Springer Science & Business Media.
40. Zhao, C., et al., (2009). Independent tasks scheduling based on genetic algorithm in
cloud computing. In: 2009 5th International Conference on Wireless Communications,
Networking and Mobile Computing. IEEE.
41. Knowles, J., & David, C., (1999). The pareto archived evolution strategy: A new baseline
algorithm for pareto multiobjective optimization. Proceedings of the 1999 Congress on
Evolutionary Computation-CEC99 (Cat. No. 99TH8406) (Vol. 1). IEEE.
42. Khandelwal, M., et al., (2017). Function development for appraising brittleness of
intact rocks using genetic programming and non-linear multiple regression models.
Engineering with Computers, 33(1), 13–21.
43. Koza, J. R., Forrest, H. B., & Oscar, S., (1999). Genetic programming as a Darwinian
invention machine. European Conference on Genetic Programming. Springer, Berlin,
Heidelberg.
Non Commercial Use

44. Bhattacharya, A., & Chattopadhyay, P. K., (2010). Application of biogeography-

based optimization to solve different optimal power flow problems. IET Generation,
Transmission & Distribution, 5(1), 70–80.
45. Sun, S., et al., (2018). Fast object detection based on binary deep convolution neural
networks. CAAI Transactions on Intelligence Technology, 3(4), 191–197.

46. Mirjalili, S., Seyed, M. M., & Andrew, L., (2014). Grey wolf optimizer. Advances in
47. Shaikh, M. S., et al., (2022). Optimal parameter estimation of 1-phase and 3-phase
transmission line for various bundle conductor’s using modified whale optimization
algorithm. International Journal of Electrical Power & Energy Systems, 138, 107893.
48. Mastoi, M. S., et al., (2021). Research on power system transient stability with wind
generation integration under fault condition to achieve economic benefits. IET Power
Electronics.
Author Copy
49. Shaikh, M. S., et al., (2021). Optimal parameter estimation of overhead transmission line
considering different bundle conductors with the uncertainty of load modeling. Optimal
Control Applications and Methods.
50. Shaikh, M. S., et al., (2021). Parameter estimation of AC transmission line considering
different bundle conductors using flux linkage technique estimation des paramètres d’une
ligne de transmission à courant alternatif en tenant compte de différents conducteurs de
faisceau à l’aide de la technique de liaison de flux. IEEE Canadian Journal of Electrical
and Computer Engineering.
51. Shaikh, M. S., et al., (2021). Application of grey wolf optimization algorithm in
parameter calculation of overhead transmission line system. IET Science, Measurement
& Technology, 15(2), 218–231.
52. Mirjalili, S., & Andrew, L., (2016). The whale optimization algorithm. Advances in
53. Rana, N., Muhammad, S. A. L., & Haruna, C., (2020). Whale optimization algorithm:
A systematic review of contemporary applications, modifications and developments.
Neural Computing and Applications, 1–33.
54. Mirjalili, S., (2015). Moth-flame optimization algorithm: A novel nature-inspired
heuristic paradigm. Knowledge-Based Systems, 89, 228–249.
55. El-Ghazali, T., (2009). Metaheuristics: From Design to Implementation (Vol. 74). John
Wiley & Sons.
56. Nara, K., et al., (2001). Application of tabu search to optimal placement of distributed
generators. In: 2001 IEEE Power Engineering Society Winter Meeting; Conference
Proceedings (Cat. No. 01CH37194) (Vol. 2). IEEE.
57. Maciel, R. S., & Padilha-Feltrin, A., (2009). Distributed generation impact evaluation
using a multi-objective tabu search. In: 2009 15th International Conference on Intelligent
System Applications to Power Systems. IEEE.
58. Shaikh, M. S., et al., (2020). Analysis of underground cable fault techniques using MATLAB
simulation. Sukkur IBA Journal of Computing and Mathematical Sciences, 4(1), 1–10.
Non Commercial Use

CHAPTER 17
Network Reconfiguration-Based
Outage Management for Reliability
Enhancement of Microgrid: A Hardware
in Loop Approach
Author Copy
SHRUTI PRAJAPATI, SONAL, SOURAV KUMAR SAHU, and
DEBOMITA GHOSH
Department of Electrical and Electronics Engineering, BIT Mesra, Ranchi,
Jharkhand, India
ABSTRACT
The demand of energy with the sprawl of power system network has posed a
challenge for reliable power. Also, the rising fuel cost adds pressure to power
demand. With increasing attention to reliable power supply and distributed
generation (DG), the concept of microgrids (MGs) has become increasingly
important. MG is an aggregation of distributed energy resources (DERs)
and connected loads, facilitating remote applications during abnormal
scenarios. Hence, MGs are providing reliable, affordable, and secure power
supply. However, the external uncertainties may lead to sudden outages of
components leading to mal-operation. Therefore, the outage management
and reliable operation of MG should be a priority for enhancing the quality
power. Reliability is coined as the probability of a system or component to
perform its intended role adequately for a specified time period. Thus, to
know the practicality of these MGs with optimal injections, firstly steady
state modeling is presented using distribution load flow (DLF) analysis.
Load outage management using network reconfiguration to enhance the
network reliability is further proposed. The proposed method estimates and
Non Commercial Use

compares the reliability of both IEEE 33-bus base and reconfigured network
with underground cables and overhead lines for variable weather conditions.
Finally, the optimal topology that minimizes the customer based reliability
indices is selected as the best fit topology. Careful contemplation of both
the networks tested in MATLAB and Typhoon hardware in loop (HIL) for
healthy and possible outage conditions gives an incredibly comprehensible
depiction of reconfigured network efficacy in terms of enhanced outage
management and reliability indices.
17.1 INTRODUCTION
Author Copy
Microgrids are evolving with the advent of power system and integration
of DERs. The major objective of these MGs is to meet the ever increasing
energy demand. Consequently, the orientation of network plays a vital role
to meet the demand in an uninterrupted and reliable manner. Reconfiguration
of network is very essential in wide aspects of power system application.
The reconfiguration of any power system network depends on the specific
objective such as voltage profile improvement, loss reduction in network
leading to improvement of local power injection, minimizing the line current,
fault current minimization, etc. Junlakarn et al. [1] showcased the importance
of reconfiguration in improving voltage profile and reducing the losses. Zhao
et al. [2] and Ali et al. [3] presented issues related to fault management of
DC microgrids, including fault detection, location, identification, isolation,
reconfiguration, and comprehensive fault management of MGs. Song et al.
[4] established the importance of establishing small power grids through
distributed generation (DG) in the accident on the power grid. Liao et al.
[5] analyzed the influence of line impedance, load demand and network
topology on voltage unbalance caused by DERs. Alam et al. [6] explained the
operation of MGs in co-ordination with DERs for enhancing the reliability
and resiliency of the power supply. Ahmed et al. [7] classified MGs in terms
of their intended application and the control in terms of their operating
principle and performance.
Munzo-Delgado et al. [8] established the fact that reconfiguration
done only for network loss reduction may cause detrimental effect on
voltage profile and voltage stability. Penuela Meneses et al. [9] prioritized
reconfiguration to reduce the power quality issues through amalgamation of
multi-objective function into a single objective function. Although authors
in the above literatures improved the network parameters in distinct ways,
which indirectly improved the reliability of network, but Li et al. [10] used
Non Commercial Use

Network Reconfiguration-Based Outage Management 339
network reconfiguration to improve the availability of power in case of

any abnormalities in the network. Farzin et al. [12]; Xin et al. [13]; and
Nazmul Huda et al. [14] detailed the reliability evaluation methods based
on the customer precedence and also analyzed the reliability for line
switching operation in the power system. Arefi et al. [15] explained an

analytical reliability evaluation technique for reliability assessment of the
network, and network reconfiguration is considered as a post-fault scenario.
Esmaeilian et al. [16] proposed a hybrid method for reconfiguration of
the network to boost the network performance and to reduce the run time
of the distribution network which resulted in decrease in power loss due
to DG integration. Abdul Rahim et al. [17] presented an optimal network
Author Copy
reconfiguration technique considering limitations of the protection system to
reduce the impact of the integration of distributed generators. Wu et al. [11];
and Paterakis et al. [18] enhanced the service reliability, by analyzing the
distribution system reconfiguration problem by utilizing the mixed-integer
linear programming method to reduce the power losses and to improve
the reliability indices of the system. de Quevedo et al. [19] introduced a
framework that incorporated the possibility of creating intentional islanding
in case of faults and highlights the benefits of network reconfiguration in
reliability improvement of the network.
To analyze the practical implementation of these MGs with optimal
injections, steady state modeling and analysis is significant. Thus, DLF is an
effective method. Sahu et al. [20] coined the DLF method and presented a
very effective way to handle the high R/X ratio by topological matrices. This
method showcased very effective results for unbalanced 3-phase systems as
well. Unlike the traditional load flow, the ‘Y’ matrix is not required in this
method. Due to all these topological matrix advantages and non-repeated use
of the matrices, the convergence of this method is faster as compared to the
previously used load flow methods.
Real-time modeling environment is very effective and essential for the
various power system application such as fault detection, power manage-
ment and power injection studies. Ghosh et al. [21] verified the power system
parameters values for a microgrid with highly penetrated DERs under of
real-time environment. Sahu et al. [22] also showcased the importance of
real-time environment for verifying the proposed fault detection technique
within microgrid. Kumar et al. performed the the unbalance studies in
microgrid for highly penetrated solar photovoltaic (SPV) in the Typhoon
HIL based real-time environment. Osama et al. [23]; Selim et al. [24] cross
verified the application of Typhoon HIL based real-time environment with
the effectiveness of inhouse developed prototype of PMU. Due to the high
Non Commercial Use

proximity to the real-time scenarios, the application of Typhoon HIL real-

time environment is very crucial for any power system result validation.
This chapter addresses the network outage management for the reliability
enhancement of microgrid. In this chapter, the DLF technique is used for
load flow analysis of distribution system. It uses two matrices bus injection
to branch current (BIBC) and branch current to bus voltage (BCBV). BIBC
matrix consists of bus injection and branch current relationship matrix and
BCBV matrix comprising of branch current and bus voltage matrix. Both the
matrices are used for solving the load flow of a system, and hence there is no
requirement to build the admittance matrix, i.e., the Jacobian matrix, and LU
which is usually required in the conventional load flow technique. Hence, the
Author Copy
proposed method is better than the conventional load flow solution method.
The DLF method is thus highly tailor-fitted for the steady-state analysis of
microgrid. Typhoon HIL real-time simulation platform is used for the valida-
tion of DLF. For this work, a real-time hardware configuration of 602+ is
used for the modeling of both normal and reconfigured IEEE-33 bus network.
Load outage management using network reconfiguration is analyzed for
MG reliability enhancement. To deal with the uncertainties involved in MG
operability condition, the Monte Carlo simulation is performed. Several case
studies are carried out and the analysis is done for both underground cables
and overhead lines under variable weather conditions for the considered
IEEE 33 bus base and reconfigured network. Careful contemplation of both
the networks tested in Typhoon HIL for healthy as well as all possible outage
conditions gives an incredibly comprehensible depiction of reconfigured
network efficacy in terms of enhanced outage management and reliability
indices.
The chapter is organized as follows. Section 17.2 details steady-state
analysis of microgrid, Section 17.3 provides modeling methodology of
microgrid using Typhoon HIL for real-time analysis, Section 17.4 presents an
overview for reliability assessment of the modeled microgrid, Section 17.5
reflects case studies, simulation, and result analysis. Section 17.6 concludes
the work.
17.2 STEADY-STATE ANALYSIS OF MICROGRID
Steady-state modeling is the primary stage to know the practicality of the

power system network. Selim et al. [25]; Baran et al. [26]; Sahu et al. [20];
and Schweitzer et al. [28] performed the load flow analysis for possible
power injection.
Non Commercial Use

To proceed further with the above-mentioned injections, it is very

essential to know if the network parameters, such as voltage and currents
are within the specified limit or not. Below are the steps that are involved
in calculating the currents and the voltages of the network required to be
analyzed for injection.

 Step 1: This step involves writing the KCL equations for each
branch considering every node current down the radial network,
starting from branch 1 to the last branch. These KCL equations for
the branches can be expressed as in Eqn. (1) [18]:
[ B ](( n −1)×1)
= [ S ](( n −1)×( n −1)) × [ I i ](( n −1)×1) (1)
Author Copy
where; [S]((n–1)×(n–1)) represents square matrix containing ones and
zeros. For a continuously connected radial distribution system, the
elements in the main diagonal and above it is ones and below the main
diagonal, all elements are zeros. This matrix is called BIBC matrix.
[Ii]((n–1)×1) represents the line current variable of (n–1) branches.
 Step 2: In this step voltage deviation of each node with respect to
the reference node is calculated by using the branch currents and
the corresponding impedances of the lines. Eqn. (2) represents the
voltage deviation matrix.
[ ∆V ](( n −1)×=
1)
[ Z ](( n −1)×( n −1)) × [ B ](( n −1)×1) (2)
Now, the value of [B] can be put in Eqns. (2) and (3) can be obtained
as follows:
[ ∆V ](( n −1)×=
1)
[ Z ](( n −1)×( n −1)) × [ S ](( n −1)×( n −1)) × [ I i ](( n −1)×1) (3)
Again, [Z]((n–1)×(n–1)) × [S]((n–1)×(n–1)) = [DLF], thus Eqn. (3) can be

represented as in Eqn. (4)
[ ∆V ]((=
n −1)×1)
[ DLF ](( n −1)×( n −1)) × [ I i ](( n −1)×1) (4)
The currents in the network can be found out by Eqn. (5), as follows:
 P + jQ * 
 I  (( n −1)×1) =  v t  
t i i
i (5)
 i   (( n −1)×1)
where; I is node current at ith node and tth itration and V is the voltage
at ith node and tth iteration.
From Eqns. (6) and (7), the voltage can be found as follows:
Non Commercial Use

 ∆V t +1  =
( ( n −1)×1)
[ DLF ](( n −1)×( n −1)) ×  I it  (( n −1)×1) (6)
The effective change in voltage, i.e., ΔV as that of the previous
iteration can be calculated from Eqn. (6) and the final voltage at ith
node can be calculated from Eqn. (7).

V t +=
1
 V t  +  ∆V t +1  (7)
( ( n −1)×1)   ( ( n −1)×1)  ( ( n −1)×1)
The difference in voltage ‘D’ between two consecutive iterations can

be calculated from Eqn. (18), as follows:
D V t +1 − V t
= (8)
Author Copy
The LF convergence can be achieved by Eqn. (9), where the conver-
gence criteria ‘E,’ should be set according to the user.
D>E (9)
This process will continue until the preset convergence ‘E,’ is achieved.
Upon calculating the voltages and currents in the network and comparing
them to the predefined standards, the realization of the network can be done.
The load flow analysis values are considered to be optimistic boundaries
for the real-time analysis. Further, Typhoon HIL-based real-time simulation
platform is used for validation.
17.3 MICROGRID MODELING METHODOLOGY USING TYPHOON

HIL FOR REAL TIME ANALYSIS
Real-time analysis of network aids in immediate precautionary measures

during the time of outages and guide in proper decision making. Since the
recent past, power networks have grown exponentially in terms of size and
technology due to the incorporation of active microgrids. As a result, utility
companies must strive to ensure that the customer’s reliability considerations
are satisfied.
To ensure the reliability to the customers, an accurate result analysis
needs to be performed. To collect the required result, in this work HIL 602+
real-time simulator is used. Typhoon HIL 602+ real-time simulator is a six
core (FPGA) device with two ARM Cortex A53 64 bit processors. The ARM
processors are used to handle the I/O signal of the simulator and FPGA
processes the plant part for higher speed real-time analysis. The 602+ is also
having 16 analog I/O capable of transferring signals at a rate of 2 MS/s. The
Non Commercial Use

real-time simulator is also having 32 digital I/O with the capability of trans-
ferring 280 MS/s. Most importantly, the simulator runs on its own software
platform to avoid compatibility issues.
Due to the mentioned advantages, Typhoon HIL real-time simulation
platform is used for the modeling of both the normal configuration and the
reconfigured IEEE-33 bus network. Both the networks are modeled in the
real-time environment for the analysis. For further calculation of the reli-
ability parameters, the voltage, currents, and the power delivered must be
calculated with the atmost accuracy. The followed procedure as provided in
Figure 17.1, ensure industry-standard reliability in network dynamic param-
eter measurement. It depicts the overall work flow for accurate calculation
Author Copy
of the network parameters.
FIGURE 17.1 Flowchart for HIL modeling and validation of microgrid network.
Non Commercial Use

17.4 OVERVIEW FOR RELIABILITY ASSESSMENT OF MICROGRID
The main challenge faced by electric power utilities in developing countries

today is that the power demand is increasing at a rapid rate. Also, electric
supply is dependent on the component health, weather condition, unpredict-

able DERs nature, and the related concerns. This has resulted in the need for
more extensive justifications for the MG modeling, planning, and improve-
ments in its operation with respect to customer demands. The customer
failure statistics reveal the system’s behavior, by depicting the gap in power
supply to the customer. Reliability aims to maximize the performance by
minimizing interruptions of the customers and providing a continuous power
Author Copy
supply. This section presents the overview of reliability evaluation of MG
considering the effect of outages and managing its reliable operation under
various outage scenarios and selecting the topology which is robust to these
disturbances.
17.4.1 COMPUTATION OF RELIABILITY INDICES
The reliability indices SAIFI, SAIDI, CAIDI, and AENS reflect MG

performance under the impact of outages. These reliability indices include
frequency of outages, measures of outage duration, response time, and
system availability. These are vital for any corrective action planning to
be taken in response to outages experienced. Incorporation of appropriate
strategies protect the MG from further degradation and keeps it intact for a
longer duration.
1. System Average Interruption Frequency Index (SAIFI): It is one of
the most basic reliability indices; SAIFI indicates the average number
of interruptions encountered by the customers over the span of one
year. SAIFI can be enhanced by reducing the number of interruptions.
It also depicts the recurrence of fault. Its unit is (interruption/system
customer/year) [3, 5, 29].
∑j∈LT
λ j * Lj
SAIFI = failure/customer/year (10)
∑ j∈LT
Lj
where; λj denotes the failure at node j; Lj is the number of customers

being served at load point j; and LT is the total load points of the
considered network.
Non Commercial Use

2. System Average Interruption Duration Index (SAIDI): It is

the ratio of the annual duration of interruptions (sustained) to the
number of consumers. It doesn’t show the state of the system;
while it shows how quickly utility can rejuvenate the supply after
an outage takes place in the network. Its unit is (hours/system

customer/year) [3, 5].
SAIDI =
∑ j∈LT
 j * Lj
hour/customer/year (11)
∑ j∈LT
Lj
where; Uj denotes the duration of outage per year at load point j; Lj
Author Copy
is the number of customers being served at load point j; and LT is the
total load points of the considered network.
3. Consumer Average Interruption Duration Index (CAIDI): It is the
ratio of the total duration of interruptions to the number of customer
interruptions encountered over the span of a year. CAIDI indicates how
much longer do an average interruption may last. Its unit is (hours/
customer interruption) [3, 5, 27].
CAIDI =
∑ j∈LT
 j * Lj
hour/failure (12)
∑ j∈LT
λ j * Lj
where; Uj denotes the duration of outage per year at load point j; λj

is the failure at node j; Lj is the number of customers being served at
load point j; and LT is the total load points of the considered network.
4. Average Energy Not Supplied (AENS): It is an index that provides
information regarding energy not supplied to the customers.
AENS =
∑ P *U j∈LT aj j
kW/hour (13)
∑ L j∈LT j
where; Uj denotes the duration of outage per year at load point j; Paj is
the average load in kW at node j Lj is the number of customers being
served at load point j; and LT is the total load points of the considered
network.
Based on the above discussed reliability metrics, Monte-Carlo simula-
tion is used to find the distribution of reliability indices for the considered
MG and assess the effectiveness of the network to supply power to the
customers.
Non Commercial Use

17.4.2 MONTE-CARLO SIMULATION FOR RELIABILITY EVALUATION
Monte-Carlo simulation (MCS) is a computational technique that estimates

quantitative analysis using randomness of parameters in a range of values
that might be deterministic in principle. This allows proper decision-making

and reduces the risk of failure owing to external or environmental factors.
When any line interrupts, a part of the network will be out of service, as
a result the customers will experience disruption. Hence, reliability indices
AENS, SAIFI, SAIDI, and CIADI can be appropriately estimated to reflect
the loss of energy and customer-related impact using Eqns. (10)–(13). Hence,
it gives the performance measure of MG in terms of reliability indices. MCS
Author Copy
creates random operational history based on the average component failure
rate per unit time, and average repair time and with the different outage cases
with different probability of occurrences under normal and abnormal weather
conditions. Thus, the probability of failure can be estimated by using the
expression of conditional probability P, as given by Eqns. (14)–(16) [25, 26].
P ( l , w ) = {OH ,UG, aw, nw} (14)
The conditional probability of line failure with respect to weather varia-
tion is given by Eqn. (15).
P ( l|nw ) *P ( aw )
P ( l |w ) = (15)
P(nw)
The conditional probability under normal weather with respect to line
type is given by Eqn. (16).
P ( nw|OH ) *P (UG )
P ( nw|l ) = (16)
P(OH )
Here; P(l|w) denotes the conditional probability of line type with respect
to weather, P(l) and P(w) are the probabilities of failure of line and occurrence
of abnormal weather scenario, respectively. The updated joint probability of
line with respect to weather scenarios, P(l|ws) is given by Eqns. (17)–(18).
P(l , ws )
P ( l|ws ) = (17)
P( ws )
P ( l|ws )
Pnew = (18)
∑( P ( l|ws )
l |ws )
The value of joint probability of line with variation of weather condi-

tion can be constructed from the combined probabilities, which satisfies the
Non Commercial Use

condition that the probability values should be greater than 0, as given by

Eqn. (19).
For any event , P ( l|ws ) ≥ 0 (19)
This conditional probability is capable of estimating the system parameters

of failure and repair rate according to the weather as well as line type for a
specific network under consideration. Figure 17.2 represents the flowchart
for reliability analysis of MG taking into consideration MCS method.
Author Copy
FIGURE 17.2 Flowchart for Monte-Carlo simulation.
Non Commercial Use

For different outage scenarios, the average duration, its frequency, average
interruption time as well as system availability are calculated. After initializing
the beginning of MCS the consequetive simulation steps are referred to as
S=s+1. For this, the reliability indices SAIFI, SAIDI, CAIDI, and AENS
are evaluated. Scenario-wise different values of indices are calculated for

variable weather conditions and different types of lines tested on both IEEE
33 bus base and reconfigured network. This assessment can be used for MG
planning and operation utilizing the obtained reliability indices indicating a
comparison between normal and reconfigured network topologies.
Author Copy
17.5 ANALYSIS AND RESULTS
This section summarizes and analyzes the results for reliability-oriented

outage management concerning the description of the previous sections and
its related mathematical expressions.
To claim the aforesaid concept, the IEEE 33-bus network with normal
configuration and reconfiguration is tested with DLF initially. This is
followed by modeling and testing the two networks in Typhoon HIL soft-
ware as in Figure 17.3 with optimal DG location for real power integration
as reflected in Table 17.1.
FIGURE 17.3 Base and reconfigured IEEE 33-bus model with DGs in typhoon HIL.
Initially, both the networks are tested under healthy conditions, followed
by interruptions at all possible locations in both the network to analyze in
which network the customers are least affected. This is also extended by
varying the type of lines and the lines exposed to variable weather conditions.
Non Commercial Use

TABLE 17.1 IEEE 33-Bus Optimal DG Location and Capacity [13]

SL. No. Bus No. Real Power (kW)
1. 12 913.05
2. 24 882.86
3. 30 1079.05
When the rated power is injected in both the base and reconfigured IEEE
33-bus networks, the reconfigured network results in low power consumption
without derating the load, which is a considerable amount, i.e., 192.69 kW.
The reduction in network loss, inherently improved network efficiency. This
Author Copy
improvement is about 5.18% of the total network load and this difference in
power can be observed from Figure 17.4. With the injections as per Table 17.1,
there is a possibility of an increase in the voltage profile in the microgrid. To
ensure that the voltage profile remains within the acceptable operational range,
simulation is performed to find the bus voltages.
FIGURE 17.4 Power consumption for IEEE 33 bus base and reconfigured network without
derating the loads.
Figure 17.5 shows the real-time voltage profile results for the base as well
as reconfigured IEEE 33-bus distribution system. The radius in Figure 17.5
shows the voltage in volts and the labels on circumference denote the nodes
Non Commercial Use

of the network. It can be observed from Figure 17.5, that the voltage profile
for reconfigured as well as for base network is within acceptable operational
range for the smooth operation, with some additional improvement for the
reconfigured network.
Author Copy
FIGURE 17.5 Real-time voltage profile results for base and reconfigured IEEE-33 bus
system.
This is followed by creating different possible outages at all possible

locations in both IEEE 33-bus base and reconfigured networks. For creating
outages it is necessary to partition the core of the simulator. The six cores
of the 602+ is partitioned according to the placement of the SPV location
outage. Due to the effective core partition, the sampling rate can be made
high. In this work, a high sampling rate, i.e., 1.6 µsec is chosen for capturing
all the dynamic variations involved. Table 17.2 shows the result for different
possible outages due to different interruptions created in both the base and
reconfigured network for comparative analysis of kW outage.
From Table 17.2 and also from Figure 17.6 it can be observed that in
most of the interruption conditions, reconfigured network kW outages are
less than the normal network kW outages with OH lines under both normal
and abnormal weather conditions. In the case of underground cables also the
reconfigured configuration proved to be a better approach.
Non Commercial Use

TABLE 17.2 HIL based Comparative Assessment for Different Possible Outages in Base
and Reconfigured IEEE 33-Bus Network
Line Outage of Base Outage of Reconfigured Remark
Outages Network (kW) Network (kW)
Line2 3710.56049 3737.1665 Base network is better

Line3 3254.76077 2123.35 Reconfigured network is better
Author Copy
Non Commercial Use

Author Copy
FIGURE 17.6 Comparative analysis of kW outage for IEEE 33-bus base and reconfigured
networks.
Further, a comparative analysis is done for the IEEE 33-bus base and
reconfigured system in terms of reliability indices (SAIFI, SAIDI, CAIDI,
and AENS). The absolute failure and repair rates considered are as shown in
Table 17.3 and are used for conditional probability computation pertaining
to normal and wind-related weather conditions for different types of lines.
These conditional probabilities obtained for different scenarios are further
used as the input parameters for MCS for the computation of reliability
indices considering outage cases in the network. Analyzing the reliability
indices of both the networks with OH lines and UG cables for variable
weather conditions, the observations are as presented in Figures 17.7–17.9,
respectively.
TABLE 17.3 Failure and Repair Rate Considering Normal and Wind Related Weather
Conditions for Different Types of Lines [20, 21]
Weather Condition Type of Lines Failure Rate (f/yr.) Repair Rate (hrs.)
Normal Overhead 0.79 4
Underground 0.15 6
Abnormal Overhead 2.23 15
Underground 0.15 6
Figure 17.7 represents a graphical comparison obtained for different OH

line disruptions created in IEEE 33-bus base and reconfigured system for
Non Commercial Use

analyzing the reliability indices under normal weather conditions. It can be

observed that reliability indices value evaluated for OH lines under normal
weather conditions shows improvement in kW outage as well as improve-
ment in AENS, SAIFI, SAIDI, and CAIDI values with the reconfigured
network. Also, these values are superior as compared to the OH line exposed
to abnormal weather conditions as in Figure 17.8. UG cables are on the other
hand are not exposed to the external environment so they are less susceptible
to disruption. However, it can be observed that reliability indices value
evaluated even for UG cables show an improvement in the reconfigured
network as in Figure 17.9.
Author Copy
FIGURE 17.7 Comparative analysis of reliability indices for IEEE 33-bus and reconfigured
33-bus for OH lines in normal weather.
17.6 CONCLUSION
The ever-increasing demand for electrical energy led to the development

of strategies for reliable outage management and enhanced quality power.
In this proposed work, the practicality of MG with optimal injections is
presented using its steady-state modeling. This is estimated by the DLF
method and results are supported by Typhoon HIL. The quick convergence
of DLF in case of a high R/X ratio makes it a very effective tool for this
analysis. The real-time data with very high resolution from Typhoon HIL
simulator includes all possible variations for effective deployment in the
MG and ensures high reliability of the scheme. Load outage management
Non Commercial Use

using network reconfiguration to enhance the network reliability is

further analyzed. The reliability assessment of both the network validates
the efficacy of the network. The results for SAIFI, SAIDI, CAIDI, and
AENS substantiates that in the IEEE 33-bus reconfigured network, for
different possible outage scenarios, as compared to the base network for

its efficacy.
Author Copy
33-bus for OH lines in abnormal weather.
33-bus for UG cables.
Non Commercial Use

KEYWORDS
• average energy not supplied (AENS)

• customer average interruption duration index (CAIDI)

• distributed load flow (DLF)
• distribution energy resources (DERs)
• microgrid (MG)
• Monte Carlo Simulation (MCS)
• network reconfiguration
Author Copy
• system average interruption duration index (SAIDI)
• system average interruption frequency index (SAIFI)
• typhoon hardware in loop (HIL)
REFERENCES
1. Junlakarn, S., & Marija, I., (2014). Distribution system reliability options and utility
liability. IEEE Transactions on Smart Grid, 5(5), 2227–2234.
2. Zhao, S., & Chanan, S., (2017). Studying the reliability implications of line switching
operations. IEEE Transactions on Power Systems, 32(6), 4614–4625.
3. Ali, Z., Yacine, T., Syed, Z. A., Mustafa, A. H., Muhammad, S., Chun-Lien, S., &
Josep, M. G., (2021). Fault management in DC microgrids: A review of challenges,
countermeasures, and future research trends. IEEE Access.
4. Jing-Hui, S., Xue-Rong, J., Hong-Kai, L., Hong-Bo, G., & Qing-Xin, L., (2008). Distributed
power generation technology and the effect of it on the accident of large-scale power grids.
In: 2008 China International Conference on Electricity Distribution (pp. 1–5). IEEE.
5. Liao, H., & Jovica, V. M., (2017). Methodology for the analysis of voltage unbalance
in networks with single-phase distributed generation. IET Generation, Transmission &
Distribution, 11(2), 550–559.
6. Alam, M. N., Saikat, C., & Arindam, G., (2018). Networked microgrids: State-of-the-art
and future perspectives. IEEE Transactions on Industrial Informatics, 15(3), 1238–1250.
7. Ahmed, M., Lasantha, M., Arash, V., & Manoj, D., (2020). Stability and control aspects
of microgrid architectures–A comprehensive review. IEEE Access, 8, 144730–144766.
8. Muñoz-Delgado, G., Javier, C., & José, M. A., (2016). Reliability assessment for
distribution optimization models: A non-simulation-based linear programming
approach. IEEE Transactions on Smart Grid, 9(4), 3048–3059.
9. Meneses, C.A. P., & Jose, R. S. M., (2013). Improving the grid operation and reliability
cost of distribution systems with dispersed generation. IEEE Transactions on Power
Systems 28(3), 2485–2496.
Non Commercial Use

10. Li, Z., Wenchuan, W., Xue, T., & Boming, Z., (2020). Optimization model-based
reliability assessment for distribution networks considering detailed placement of
circuit breakers and switches. IEEE Transactions on Power Systems, 35(5), 3991–4004.
11. Li, Z., Wenchuan, W., Boming, Z., & Xue, T., (2019). Analytical reliability
assessment method for complex distribution networks considering post-fault network

reconfiguration. IEEE Transactions on Power Systems, 35(2), 1457–1467.
12. Farzin, H., Fotuhi-Firuzabad, M., & Moeini-Aghtaie, M., (2017). Role of outage
management strategy in reliability performance of multi-microgrid distribution systems.
IEEE Transactions on Power Systems, 33(3), 2359–2369.
13. Xin, S., Cheng, Y., Zhang, X., & Wei, C., (2017). A novel multi-microgrids system
reliability assessment algorithm using parallel computing. In: 2017 IEEE Conference on
Energy Internet and Energy System Integration (EI2) (pp. 1–5). IEEE.
14. Huda, A. S. N., & Rastko, Ž., (2018). Study effect of components availability on
Author Copy
distribution system reliability through multilevel Monte Carlo method. IEEE Transactions
on Industrial Informatics, 15(6), 3133–3142.
15. Arefi, A., Gerard, L., Ghavameddin, N., & Behnaz, B., (2020). A fast adequacy
analysis for radial distribution networks considering reconfiguration and DGs. IEEE
Transactions on Smart Grid, 11(5), 3896–3909.
16. Esmaeilian, H. R., & Roohollah, F., (2014). Energy loss minimization in distribution
systems utilizing an enhanced reconfiguration method integrating distributed generation.
IEEE Systems Journal, 9(4), 1430–1439.
17. Rahim, M. N. A., Hazlie, M., Abdul, H. A. B., Mir, T. R., Ola, B., & Nurulafiqah, N. M.,
(2019). Protection coordination toward optimal network reconfiguration and DG sizing.
IEEE Access, 7, 163700–163718.
18. Paterakis, N. G., Andrea, M., Sergio, F. S., Ozan, E., Gianfranco, C., Anastasios, G. B.,
& João, P. S. C., (2015). Multi-objective reconfiguration of radial distribution systems
using reliability indices. IEEE Transactions on Power Systems, 31(2), 1048–1062.
19. De Quevedo, P. M., Javier, C., Andrea, M., Gianfranco, C., & Radu, P., (2017). Reliability
assessment of microgrids with local and mobile generation, time-dependent profiles, and
intraday reconfiguration. IEEE Transactions on Industry Applications, 54(1), 61–72.
20. Sahu, S. K., & Debomita, G., (2019). Hosting capacity enhancement in distribution
system in highly trenchant photo-voltaic environment: A hardware in loop approach.
IEEE Access, 8, 14440–14451.
21. Sahu, S. K., & Debomita, G., (2021). Spectral kurtosis-based fault detection for a highly
penetrated distributed generation: A real-time analysis. In: Advances in Smart Grid
Automation and Industry 4.0 (pp. 649–656). Springer, Singapore.
22. Sahu, S. K., & Debomita, G., (2021). Photovoltaic hosting capacity increment in an
unbalanced active distribution network. In: 2021 1st International Conference on Power
Electronics and Energy (ICPEE) (pp. 1–5). IEEE.
23. Kumar, R., Sourav, K. S., Debomita, G., & Sarbani, C., (2021). A multi HIL-based
approach for real-time phasor data monitoring using phasor measurement unit. In:
Advances in Smart Grid Automation and Industry 4.0 (pp. 623–630). Springer, Singapore.
24. Osama, R. A., Ahmed, F. Z., & Almoataz, Y. A., (2019). A planning framework for
optimal partitioning of distribution networks into microgrids. IEEE Systems Journal,
14(1), 916–926.
25. Selim, A., Salah, K., Ali, S. A., & Francisco, J., (2020). Optimal placement of DGs in
distribution system using an improved Harris hawk’s optimizer based on single-and
multi-objective approaches. IEEE Access, 8, 52815–52829.
Non Commercial Use

26. Baran, M. E., & Felix, F. W., (1989). Network reconfiguration in distribution systems
for loss reduction and load balancing. IEEE Power Engineering Review, 9(4), 101, 102.
27. Dalsgaard, M. T., & Guangya, Y., (2016). Reconfiguration of Power Distribution
Networks by Evolutionary Algorithm for Reliability Improvement. Technical report,
special course, DTU.

28. Schweitzer, E. O., David, W., Héctor, J. A. F., Demetrios, A. T., David, A. C., & David,
S. E., (2011). Line protection: Redundancy, reliability, and affordability. In: 2011 64th
Annual Conference for Protective Relay Engineers (pp. 1–24). IEEE.
29. Jae-Han, K., Ju-Yong, K., Jin-Tae, C., Il-Keun, S., Bo-Min, K., Il-Yop, C., & Joon-Ho,
C., (2014). Comparison between underground cable and overhead line for a low-voltage
direct current distribution network serving communication repeater. Energies, 7(3),
1656–1672.
Author Copy
Non Commercial Use

Non Commercial Use

Author Copy
CHAPTER 18
A Net Energy Meter-Based Approach

for Islanding Detection in Modern
Distribution Systems
SOHAM DUTTA,1 AKASH KUMAR PANDEY,2 SOURAV KUMAR SAHU,3
Author Copy
and PRADIP KUMAR SADHU4
1
Department of Electrical and Electronics Engineering, Manipal Institute
of Technology, Manipal Academy of Higher Education, Manipal, Karnataka,
India
2
Design Engineer, Larsen and Toubro, Chennai, India
3
Department of Electrical and Electronics Engineering, BIT Mesra,
Ranchi, Jharkhand, India
4
Department of Electrical Engineering, IIT(ISM) Dhanbad, Jharkhand, India
ABSTRACT
The era of smart grid has provided unprecedented opportunities, which

has taken the power industry to a new level assuring better reliability,
continuity and productivity. However, the grid system has to face challenging
uncertainties due to faults, voltage dip, islanding, etc. Islanding can be
defined as a situation during which distributed generation (DG) continue
to feed some portion of loads even after being disconnected from the main
grid. Island situations can be dangerous for the utility workers and users end
equipments. Net energy meter (NEM) has the ability to record flow of energy
in two directions. To feed the household loads, consumer utilizes power from
two sources i.e. from the grid and different DGs. If there is a surplus power
Non Commercial Use

generation by DGs, the excess power will be fed back to the grid through
a NEM. To deal islanding events with NEM, this chapter recommends an
advance technique of detecting islanding for doubly-fed induction generator
(DFIG) wind turbines based microgrid. Walsh Hadamard transform (WHT)
is used as the signal processing tool to generate the WHT coefficients of the
signals derived from net energy meter (NEM) and artificial neural network
(ANN) has been employed for the examination of these coefficients and
classification/detection of islanding events. The results obtained for various
islanding and non-islanding scenarios show that the presented method is
faster and effective. The simulation is performed in MATLAB/SIMULINK
environment.
Author Copy
18.1 INTRODUCTION
The era of smart grid has provided unprecedented opportunities, which

has booked the power business to a fresh level assuring better reliability,
continuity and productivity [1]. However, the grid system has to face
challenging uncertainties due to faults, voltage dip, islanding, etc. Islanding
can be defined as a situation during which distributed generation (DG) carry
on to nourish some percentage of loads even after being disconnected from
the main grid. Island situations can be risky for the service workers and users
end equipments [2].
Net energy meter (NEM) has the ability to record flow of energy in two
directions [3]. To feed the household loads, consumer utilizes power from
two sources, i.e., from the grid and different DGs. If there is a surplus power
generation by DGs, the excess power will be fed back to the grid through a
NEM.
To deal islanding events with NEM, this chapter recommends an
advance technique of detecting islanding for doubly-fed induction
generator (DFIG) wind turbines based microgrid. Walsh Hadamard trans-
form (WHT) is used as the signal processing tool to generate the WHT
coefficients of the signals derived from NEM and artificial neural network
(ANN) has been employed for the examination of these coefficients and
classification/detection of islanding events. The results obtained for
various islanding and non-islanding scenarios show that the presented
method is faster and effective. The simulation is performed in MATLAB/
SIMULINK environment.
Non Commercial Use

A Net Energy Meter-Based Approach for Islanding Detection 361
18.2 MODERN DISTRIBUTION SYSTEM
18.2.1 ADVANCEMENT OF ELECTRICAL TECHNOLOGY

Unlike any other developing country, India cannot ignore the importance
of power industry in pillaring its overall development. Moreover, to be
a developed nation with increased industrial productivity and large job
creation, India has to rely on power that can fuel the factories and provide
the opportunity to be a developed nation. As the productivity of all other
sectors mainly depends upon the energy fueling, every power industry have
to increase its output for supplying the required demand. Thus, there is
Author Copy
a need of shift in paradigm from the point of view of energy production
and energy transmission by using the latest power electronics and smart
grid technology [4]. It can result in effective and efficient generation and
utilization of energy that can result in development of power sector in any
country. The excessive usage of fossil fuels can result in high amount of
carbon dioxide (CO2) production that consistently results in global warming
increasing the average global temperature. This proves to be very dangerous
for the living beings and our animal planet. Hence, the Government has
taken various steps for the improvement in the field of renewable energy
sources (ES) [5].
18.2.2 DG
Power networks are becoming smarter progressively and concurrently,

the prominence of DGs is growing. The notion of DG has contributed
the power business to a fresh elevation that has guaranteed improved
dependability and stability [6]. The word DG implies power produced
by numerous decentralized sources for definite engagements. DGs have
decreased the dependence on main grid source because it performs as a
source in a micro-grid possessing the capability to cater small amounts
of loads like household loads. Nevertheless, as the infiltration of these
DGs has augmented in the grid system, several stimulating worries have
surfaced like faults, voltage dip and islanding that have a difference in
pattern from previously recognized conventional faults [7]. Thus, they
need to be identified and resolved.
Non Commercial Use

18.2.3 DG AS A FEASIBLE ENERGY SOURCE
Electricity production and its lossless grid circulation is mostly adminis-

trated by the Government which is the regulating authority of this matter.
Likewise, the electrical power providers have to adhere to guideline and

stability of demand-supply ratio. Nevertheless, there is always a chance of
power catastrophe and abnormalities of supply owing to several causes like
faults, natural disaster, etc. To deal with them, the power system needs an
alternative but unfailing sources like DG. Thus, the Government and public
sectors recently has concentrated on deregulating the power supply process
and DGs are the paramount response for the necessities. However, each coin
Author Copy
has two sides. The use of DG comes with its own hassle. For instance, its
ability to cope in the high competition power market [8]. Contemporary
growth in the area of technology, e.g., ultra-fast switching and competence
to cope up with the huge voltage level, has pressed the power industry to use
DG as a feasible energy source [9].
18.2.4 DG CLASSIFICATION
The DGs are broadly classified into two categories traditional and nontra-
ditional as shown in Figure 18.1. Traditional DGs are classified as sources
which work on combustion engines and whose whole output depends upon
the internal combustion. Some examples are gas microturbine and low speed
turbine. They are quite small in size and area but have a unique advantage of
being not dependent on geographical resources for their operation. Moreover,
they can be installed at any location [10]. Nontraditional are the one that
have the ability to produce electricity with assurance of zero emission and
eco-friendly. However, they have a disadvantage of producing DC power
which cannot be connected to the main grid directly. Thus, this dc power
is converted into ac by utilizing power electronics devices (for example,
rectifier or inverter) before connecting it to the main distribution system.
Due to this drawback, power sources such as photovoltaic and fuel cell uses
converters before getting connected to the grid.
18.2.5 DISTRIBUTION UNIT WITH MULTIPLE DGS
Since, DG or decentralized sources are known for producing electrical energy

apart from the main grid supply, they are placed nearer to the load center in a
Non Commercial Use

single unit or multiple unit as per suitability for maintaining the continuity of
supply. One such arrangements is shown in Figure 18.2. It basically provides
an alternate power generation unit for the upgradation and modernization of
previously running grid supply system [8–10]. Also, for the utility purposes,
multiple DGs can be connected simultaneously for supplying different kinds

of complex load demand. In the figure, loads are supported by main grid as
well as by the sources of DG through transformers and circuit breakers. Here
the purpose of DGs is to provide backup power during any uncertainties like
faults and islanding, etc. Hence, from above figure the advantages of DGs
can be better understood in case of faulty operation [7–10].
Author Copy
FIGURE 18.1 Classification of DG.
18.2.6 USEFULNESS OF DGS
The usefulness of the DGs can be can be summarized as follows:
Non Commercial Use

1. Better Reliability: DG are load located power resources that are

placed nearer to the demand site as a alternate power resources. Due
to this, DG provides increased power availability which is unaffected
from the impacts of transmission and distribution system faults which
further helps in decreasing the peak demand overload. Furthermore,

utility of multiple DG distribute the demand requirement equally on
all the ES and hence reduce the overload on main grid [8].
Author Copy
FIGURE 18.2 System consisting of multiple DGs.
Non Commercial Use

2. Continuity: As DG are located nearer to the load center and are flex-
ible in its area of application, an uninterrupted production in output
of DGs can be obtained. This is useful when there is any disturbance
like faults or by an act of terror. This gives DGs an additional advan-
tage over the main grid, i.e., despite being any fault, DGs continue to
feed power to the load center [9].
3. Flexibility: DG are much versatile when it comes to their location of
application as they can be placed anywhere close or a bit far from the
load center as per the requirement without affecting the quality and
continuity of output that are being obtained from the micro-grid [10].
4. Reduction in Peak Demand: DGs, by generating power, helps
Author Copy
the main grid in reducing its overall load demand during the peak
hours [11].
5. Improved Power Quality: DG reduces the power degradation by
supplying the additional support to the main grid and hence improves
the power quality and profile of voltage [11].
6. Reduced Transmission Losses: DGs being located near the load
center, the requirement of transmission and distribution lines reduce
to a much smaller extent as there is small distance between the
source and demand center. As a result of this, the transmission losses
associated with it is lesser than the main grid [12].
Apart from the various advantages associated with DG as listed above,
DGs also suffer some shortcomings. There are few challenges that must be
noted for DGs which will have impact on the whole working unit and the
output that is obtained from the DGs. Such issues are discussed in the next
section.
18.2.7 VARIOUS ISSUES ASSOCIATED WITH DGS
DGs have some particular negative aspects as followed [9, 11, 12]:
1. Setup Cost: Capital investment, i.e., the cost for setting up a DG
system is quite higher when compared to large central plants as DGs
requires advanced technologies. Moreover, in a hybrid system, the
flow of power is bi-directional which makes the whole connected
unit costlier due to varios controllers.
2. Reduction of Power Quality: Though, with large number of benefits
related to power and voltage stability, the idea of utilizing multiple
DGs as a substitute resource can work in a negative manner because
Non Commercial Use

of many factors. DGs affects the system working frequency. Thus,

for maintaining a constant frequency, suitable load frequency control
systems are needed with the potential of maintaining frequency
constant for multiple DG system.
3. Connection Issues: Connecting multiple DGs with the main transmission

line in hybrid grid mode requires a great effort and technological
advancement to maintain the stability of voltage and frequency.
18.2.8 MICRO-GRIDS
Author Copy
As the power scientists have progressed in the field of power engineering,
the production of power mainly relies on non-renewable energy resources
like diesel, coal, etc., as they constitute for more than 70% of total energy
production. However, on the other side, these non-renewable sources causes
huge expanse of pollution that contributes to numerous environmental issues
like global warming and healthiness related issues. Therefore, to provide
mankind the superior substitutes, the power researchers and engineers have
engrossed on renewable and non-conventional energy segment. As the usage
of nontraditional ES are on hike in today’s world, power engineers have
the responsibility and requirement to cope up with these changes in a very
positive manner and embrace the ways of renewable energy generation
techniques. In hunt for improved and quality power generation, one have to
adopt a clever and faster tactic that would be fulfilling the existing energy
obligation in a eco friendly means. It has to be done in such a way that after
shifting from one resources to the other, we can uphold the continuousness
of power supply in an efficient manner. For this requirements, micro-grid
were developed to attend the drive of generating and supplying the power
mainly by utilizing the natural resources. Micro-grids [13, 14] possess the
ability to work both as minor power grids those can operate in association
with several other minor grids as well as with the main-grid. Decentralized
sources mainly constitute of micro-grids that can operate on small scale for
energy production that can serve the purpose of operating various small
loads besides sending back excess power to the main grid. Decentralized
source planted for any specific purpose and having the capacity of small
scale energy production and which can also be stored for future usage are
known as micro-grids [13].
Micro-grids predominantly exploits the renewable form of energy like
wind and solar power for their input energy demand that can be converted
into electrical energy. The electrical energy that are produced by micro-grid
Non Commercial Use

can be deployed for small load demand and also can act as reserve storage
during the peak demand [15]. Thus, micro-grids are the conclusive answers
for our environment friendly energy requirement that can provide additional
support to the main grid in an efficient way for maintaining the continuity
of supply specially during the localized fault. In this chapter, for making a
hybrid micro-grid, wind turbine based decentralized source has been taken in
the form of a micro-grid through a 30 km transmission line, fastened to the
grid of 120 kV. Thus, from above-mentioned, the advantages of the micro-
grid can be stated as following [15]:
• Maintaining the steadiness in power supply during any localized fault
Author Copy
which disturbs the main supply of the power grid;
• Supplying the required reactive and active power demand during
island condition;
• Collaborating with the main grid to supply the peak load demand;
• Supplying the generated surplus power to the main grid.
18.3 CONCEPT OF ISLANDING IN POWER SYSTEM
18.3.1 ISLANDING PHENOMENON
Islanding is considered as a situation in which a DG or group of DGs remain

to feed some part of load in spite of being disconnected from the main source
as shown in Figure 18.3. For avoiding islanding, the concept of islanding
condition must be understand thoroughly and its impact on the system [16, 17].
System parameters, like voltage and frequency, vastly fluctuate after islanding.
Thus, islanding condition must be properly taken care of for the protection of
system. In some cases, islanding also affects the power quality and operation
of system. Islanding can be of two types intentional and unintentional [16]. In
intentional islanding, a section of unit is willingly isolated from the main grid
for some maintenance or removal of faults. This type of islanding generally
does not pose any problem as everything is preplanned. However, the main
concern is unintentional islanding, where a section is isolated unwillingly
which can be dangerous for the whole electrically islanded system.
18.3.2 DISADVANTAGES OF ISLANDING
There are negligible disadvantages of intentional islanding. However, inadver-

tent islanding welcomes various drawbacks. Some of them are as follows [18]:
Non Commercial Use

1. Dangerous for Utility Workers: Islanding is a sudden phenomenon

that adversely affects not only the isolated unit but it is life threatening
for the utility staff. This is because, during islanding, loads in the
islanded grid are energized from the DGs even when the main grid
supply is cut off which can be unsafe for the service workers. Thus, the
workers may get electrocuted.
Author Copy
FIGURE 18.3 Island scenarios.
Non Commercial Use

2. Instability of Voltage: In islanding instants, main grids are cut out

suddenly due to any of the reasons like circuit breaker opening or
faults, etc. The islanded part is unable to cope with the required
reactive power and that fluctuates the voltage of islanded section.
This instability in voltage of the working unit can have very negative
effect on the connected loads or user based appliances.
3. Fluctuation in Frequency: Grid connected systems operate at a
constant frequency but frequency of islanded section fluctuates due to
the inability of the islanded part to cope up with the required active
power. This fluctuation in frequency can be dangerous for the connected
loads as the loads are designed to operate on a constant frequency.
Author Copy
18.3.3 ISLANDING DETECTION TECHNIQUES
Islanding detection is vital for the security of service workers and utility
system connected to the grid. There are several approaches already been
given by scientists in the past that are largely categorized into classical
methods and modern methods [18]. Classical methods are divided into
active, passive hybrid and local methods as shown in Figure 18.4.
FIGURE 18.4 Classification of islanding detection methods.
18.3.3.1 CLASSICAL METHOD OF ISLANDING DETECTION
Classical methods of islanding detection are directly or indirectly related

with the measurement of magnitude of signal like voltage and frequency,
etc. Classical method is further divided into passive, active hybrid and local
methods [19].
Non Commercial Use

Passive approaches deal with assessment of parameter changes like

harmonic distortion, frequency, voltage, etc., with respect to the threshold rate.
These parameters are largely affected by islanding situation. Passive methods
have found advantage when there is a high power difference between genera-
tion and load demand [20]. Passive techniques are reliable, swift and do not
create any system variation, but they have high non-detection zone (NDZ).
Classical method of islanding detection has active technique as their
second classification [21]. Active method completely depends upon injecting
a minor disturbance or perturbation signal at point of common coupling
(PCC) and this perturbation signal will give noteworthy alteration in system
parameters in islanded system. The range of fluctuations is checked due to
Author Copy
disturbance signal and if this range lies outside the predefined value, then
islanding is detected otherwise, the whole process for detecting islanding is
repeated. Under the scenario of matched load and generation, this technique
becomes superior to the passive technique. Active technique, by using
disturbance signal, can detect islanding with high efficacy and lesser NDZ
[18, 22]. Nevertheless, they also reduce the power quality and upsurge the
total harmonic distortion.
Hybrid technique is a synthesis of passive and active methods of islanding
detection [23]. Hybrid technique firstly apply the technique discussed in
passive technique and then apply the active method for detecting islanding
condition [24]. Local approaches like transfer trip scheme and power line
carrier communication, etc., [25] utilizes a communication path between
the DG and utility section of islanded portion. Local methods are a bit
costly owing to their communication channel prerequisite rendering local
approaches inapt for a single DG system [18, 26]. Local approaches possess
extreme advantage when it is used in multiple system consisting of several
DGs as these approaches have zero NDZ.
18.3.3.2 MODERN METHODS OF ISLANDING DETECTION
Modern methods are relatively newer than the traditional methods of this
category. It detects islanding with signal processing tools and intelligent clas-
sifiers [27]. Modern methods are very advantageous as they have zero NDZ
and detect islanding very effectively. Island detection by signal processing
uses numerous signal processing methods for obtaining the features of signals
that are attained from the islanded section for detection of islanding instants.
Some of these techniques are Stockwell transform, wavelet transform,
time-time transform, mathematical morphology, etc. [18, 28–30]. Intelligent
Non Commercial Use

techniques are the latest techniques in the field of islanding detection method.
Classifiers like fuzzy logic, ANN and decision tree, etc., are used for detecting
islanding in an effective and efficient manner [31–35]. Classifier methods
have additional advantage of zero NDZ.
In all the above methods, a set of hardware and software needs to be

installed in the grid for detecting islanding conditions. It takes a lot of
time for installation of these devices. Besides time, these installments also
requires a lot of capital. Since, NEM are already present in modern day grid,
its functions can be leveraged for island detection. The hardware of NEM can
be employed for the acquisition of signals and its software can be used for
processing of these signals. However, islanding signals involves transients
Author Copy
that are not detectable by time analysis. Thus, an advanced signal processing
tool is required for analyzing these signals. In the present method, WHT is
used which is explained in the next section.
18.4 ISLANDING DETECTION EMPLOYING WALSH HADAMARD

TRANSFORM
18.4.1 NEED OF FAST ISLANDING DETECTION METHOD
Connecting DGs in the main stream transmission networks is the key for
providing continuous and efficient power for energy demand in the present
scenario [3–5]. DGs have been the most significant component in the field
of energy market and modern infrastructure. DG, on one side are acting as
a pillar to the energy demand and supply, but on the other side faults related
with the DGs like islanding could be proved fatal for the utility workers
and safety equipments connected to the system [7]. So, for the protection of
whole working unit, it is compulsory that islanding phenomenon is detected
and corrected in smallest possible time. Unable to detect islanding situation
in time could be dangerous for the working personnel, reduce power quality
for consumer loads or can cause false triggering of protective devices [18].
Hence, for providing protection against islanding now a day’s islanding
detection relays employing advance techniques are being used.
18.4.2 BLOCK DIAGRAM OF SIMULATION MODEL
In this chapter, the block diagram of the simulation model developed is

presented in Figure 18.5. It is important that the model reflects a real system
Non Commercial Use

in all vital parts. In the model main grid is connected to DFIG wind turbine
based micro-grid which is rotating at 15 m/sec through a 30 km transmis-
sion line and grounding transformer. The model also contains of a NEM
so that bidirectional energy flow can be measured which also serves as an
additional advantage for collecting samples of signals that can processed

through signal processor and trained by intelligent classifiers for the
purpose of islanding detection. A circuit breaker separates the micro-grid
from the main grid to demonstrate the islanding phenomenon. At PCC, the
main grid at a voltage of 120 kV is connected to a micro-grid to form a
hybrid grid system. Hence, the behavior of the simulated system represents
a real situation.
Author Copy
FIGURE 18.5 Block diagram of simulation model.
18.4.3 TEST SYSTEM CONSIDERED
The test model consists of six 1.5 MW wind turbines that cumulatively gives
9 MW power output connected with 25 kV system. Wind turbines have
DFIG as core working component composed of a wound rotor induction
generator and an AC/DC/AC insulated gate bipolar transistor (IGBT) based
pulse width modulation (PWM) converter. The grid operating at 60 Hz is
linked directly to the stator winding while the rotor is supplied at fluctuating
frequency through the AC/DC/AC IGBT-based PWM converter. The DFIG
technology enables us to pluck out maximal energy from the wind despite
of low wind speeds by regulating the turbine speed, while diminishing
mechanical burdens on the turbine during puff of wind.
The test system considered is shown in Figure 18.6. In this model, the
parameters considered are given in Table 18.1. The duration for which
samples are taken is in between t=10 ms to t=30 ms (i.e., 4,000 samples).
Various operating condition like islanding and faults are introduced after
20 ms.
Non Commercial Use

TABLE 18.1 Parameters of the Test System Considered

Parameters Value
Wind speed 15 m/s
Torque 1.2 p.u.

Reactive power output 0 MVAR
Sampling time 5 µs
Simulation time 40 ms
18.4.4 WALSH HADAMARD TRANSFORM
Author Copy
The WHT is a form of Fourier transform where a linear, symmetric and
orthogonal operation on 2i real numbers, as in Eqn. (1), is performed to get
Hi, which is a 2i× 2i matrix [36–38]. Here 1⁄√2 is normalization. WHT can be
used both recursively or by binary representation. In recursive method, WHT
matrix H1 of order 1×1 is generated by taking H0 =1 and then Hi is calcu-
lated using Eqn. (2), where ∅ denotes the Kronecker product. Similarly, the
Hadamard matrix for its (k, n)th entry is written by using Eqn. (3). In Eqn. (4)
and (5), we are denoting kj and nj as the binary digits, i.e., 0 or 1, of k and n,
respectively [39].
1  H i −1 H i −1 
Hi =   (1)
2  H i −1 − H i −1 
H=
i H1∅H i −1 , i > 1 (2)
1 ∑ j k jnj
( H i )=
k ,n i
(−1) (3)
2 2
∑
i −1
=k m=0
km 2m ki −1 2i −1 + ki − 2 2i − 2 +…+ k1 2 + k0
= (4)
∑
i −1
=n m=0
nm 2m ni −1 2i −1 +…+ n1 2 + n0
= (5)
18.4.5 NET ENERGY METER
NEM has the ability to record flow of energy in two directions [3, 40]. As
can be seen from the simulated model, NEM is placed at PCC for measuring
bi-directional power flow [21]. To feed the household loads or various types
of other connected loads, consumer utilizes power from two sources, i.e.,
Non Commercial Use

Apple Academic Press 374
FIGURE 18.6 Test system considered.
Non Commercial Use

Author Copy
from the grid and different DGs. If there is a surplus power generation by
DGs than the requirement, the excess power which is not required for the
connected loads on DG side will be fed back to the grid through a NEM
[41]. It can obtain power measurement in both directions and that provides
additional profit to the consumer end. The working formula of NEM is given
by Eqn. (6). Here, power fed to or taken from the grid is denoted by Pgrid,
PgrenDG denotes power produced by DG and Pconsumed is the power absorbed
by the load. To measure the equivalent power flow, voltage and current are
calculated by the NEM [42, 43]. The proposed algorithm uses this measured
voltage signals for processing through signal processors and detecting island
scenarios. This eliminates the additional requirement for a measuring device
Author Copy
separately for island detection.
Pgrid = Pconsumed – PgrenDG (6)
18.4.6 PROPOSED METHODOLOGY
The methodology required for islanding detection needs to be very reliable

and fast as islanding can cause serious damages to the connected loads and
could be proved life threatening for the utility workers. Thus, looking on
negative side of islanding and the damages it can cause to the working unit,
this chapter proposed a method of islanding detection employing WHT.
To demonstrate the applicability of WHT for island detection, normal,
islanding and faults scenarios are simulated. Samples of voltage signals are
collected from NEM at PCC for processing through signal processor, i.e.,
WHT and WHT coefficients obtained for the scenarios. The WHT coef-
ficients graphs obtained for the scenarios are plotted in Figures 18.7–18.11.
As seen in Figure 18.8, during islanding, the graph shows high density of
samples having lower magnitude with respect to normal scenario in which
density of lower magnitude samples are quite low as shown in Figure 18.7.
The graphs as obtained in Figures 18.9–18.11 for various faults show that
this density is different from the normal and islanding scenarios. Therefore,
it can be observed that WHT provides different results for above listed
scenarios. Thus, the flow chart of the proposed methodology is shown in
Figure 18.12. The ANN is properly trained by simulating various islanding
and non-islanding scenarios given in Table 18.2. The power mismatch is
ensured by load switching or load disconnection with respect to the DG
generation just before opening of the circuit breaker for demonstrating
islanding scenarios (Figure 18.12).
Non Commercial Use

Author Copy
FIGURE 18.7 WHT coefficients graph with no islanding.
FIGURE 18.8 WHT coefficients graph with islanding.
Non Commercial Use

Author Copy
FIGURE 18.9 WHT coefficients graph with LG fault.
FIGURE 18.10 WHT coefficients graph with LL fault.
Non Commercial Use

Author Copy
FIGURE 18.11 WHT coefficients graph with LLLG fault.
TABLE 18.2 Scenarios for ANN Training

Scenarios Number of Scenarios
Island with active power mismatch ranging from +10% to –10% 1,000
(keeping zero reactive power mismatch).
Island with reactive power mismatch ranging from +10% to 1,000
–10% (keeping zero active power mismatch).
Island with both active as well as reactive power mismatch 1,000
ranging from +10% to –10%.
Various type of faults with fault resistance ranging from 1,000
50 Ω to 200 Ω at various grid positions.
Switching on of capacitor ranging from 50 kVAR to 100 kVAR 500
at different locations.
Switching on of loads ranging from 50 kW to 100 kW at 500
different locations.
18.4.7 RESULTS AND DISCUSSION
To test the performance of the proposed method, the algorithm is tested

against various power system scenarios which are not considered during the
ANN training process. Table 18.3 shows the effectiveness of the proposed
algorithm in segregating island scenarios from various other conditions.
Non Commercial Use

Thus, it is seen that the method is able to detect island cases in a small time
span of 20 ms.
Author Copy
FIGURE 18.12 Flowchart of the proposed WHT-based methodology.
Non Commercial Use

The IEEE Standard 1547-2018 for the connection of DG with the main
grid states that each DG must detach itself from the islanded grid within 2s
of unintentional islanding. This time includes islanding detection algorithm
time and execution time of trip signal. Therefore, a technique having minimal
detection time is desirable. Thus, the present method meets this criteria.
TABLE 18.3 Results for Various Scenarios under WHT-based Method

Scenarios Results
Island with 50% power mismatch. Island
Island with 150% power mismatch. Island
Author Copy
Capacitor switching near PCC. Not island
LG fault near PCC with fault impedance angle = 45°. Not island
LL fault near PCC with fault resistance of 50 Ω. Not island
18.5 CONCLUSION
Modern grids are the irrefutable response for managing the complex power
grid, catering to the required dynamics to the power system to enhance its
efficacy. The usage of DGs have fortified the grid utility’s sustainability.
Hence, limitations like islanding instants must be dealt with rationally. The
penetration of hybrid grid system is on rise, owing to its advantages and
flexibilities that provides assurance for continuity of supply. Moreover,
hybrid grid composes of inverters and batteries that are crucial for
backup power management. On the other hand, besides the advantages of
interconnecting DG with the main supply grid, certain drawbacks of these
interconnections are there and one of them is islanding. Islanding, in most
interconnection is a decisive phenomenon as this can cause severe damages
to the grid management system by fluctuating the voltage and frequency
or would affect the working of connected loads and safety equipments or
can put life of workers to a great risk. Therefore, experts and scholars have
constantly dedicated there time in devising islanding detection approaches
which are exceedingly effective and quick such that the damaging effects
of islanding can be dodged to a huge degree. In continuation with the same
trajectory, the chapter is composed of a method which can detect islanding
in very short interval of time as per the IEEE standards. The method uses
signal processing tools such as WHT for islanding detection, which is
comparatively accurate and efficient with the previous available methods as
it can detect islanding in 20 ms. Furthermore, the use of NEM reduces the
Non Commercial Use

components requirement. This is because the hardware components of the

NEM is used for measuring different signals and the software part of NEM
is used for the processing of these signals with WHT. Thus, unlike other
island detection methods, no extra components are needed to be installed
for detection of islanding instants. This in turn reduces the overall cost for
island detection.
KEYWORDS
Author Copy
• distributed generation
• island detection
• islanding
• micro-grid
• net energy meter
• smart grid
• Walsh Hadamard transform
REFERENCES
1. Khalid, H., & Abdulfetah, S., (2021). Existing developments in adaptive smart grid
protection: A review. Electric Power Systems Research, 191, 106901.
2. Butt, O. M., Muhammad, Z., & Tallal, M. B., (2021). Recent advancement in smart grid
technology: Future prospects in the electrical power network. Ain Shams Engineering
Journal 12(1), 687–695.
3. Dutta, S., Debomita, G., & Dusmanta, K. M., (2016). Location biased nature of net
energy metering. In: 2016 International Conference on Computation of Power, Energy
Information and Communication (ICCPEIC) (pp. 350–355). IEEE.
4. Kappagantu, R., & Arul, D. S., (2018). Challenges and issues of smart grid implementation:
A case of Indian scenario. Journal of Electrical Systems and Information Technology,
5(3), 453–467.
5. Dutta, S., Pradip, K. S., Maddikara, J. B. R., & Dusmanta, K. M., (2020). Role of
microphasor measurement unit for decision making based on enhanced situational
awareness of a modern distribution system. In: Decision Making Applications in Modern
Power Systems (pp. 181–199). Academic Press.
6. Kakran, S., & Saurabh, C., (2018). Smart operations of smart grids integrated with
distributed generation: A review. Renewable and Sustainable Energy Reviews, 81,
524–535.
7. Singh, B., & Janmejay, S., (2017). A review on distributed generation planning.
Renewable and Sustainable Energy Reviews, 76, 529–544.
Non Commercial Use

8. Zsiborács, H., Nóra, H. B., András, V., László, Z., Zoltán, B., Kinga, M., & Gábor,
P., (2019). Intermittent renewable energy sources: The role of energy storage in the
european power system of 2040. Electronics, 8(7), 729.
9. HA, M. P., Phung, D. H., & Vigna, K. R., (2017). A review of the optimal allocation
of distributed generation: Objectives, constraints, methods, and algorithms. Renewable

and Sustainable Energy Reviews, 75, 293–312.
10. Bansal, R., (2017). Handbook of Distributed Generation. Switzerland Valsan SP, Springer
International Publishing, Switzerland.
11. Marwali, M. N., Jin-Woo, J., & Ali, K., (2004). Control of distributed generation systems-
Part II: Load sharing control. IEEE Transactions on Power Electronics, 19(6), 1551–1561.
12. Marwali, M. N., & Ali, K., (2004). Control of distributed generation systems-Part I:
Voltages and currents control. IEEE Transactions on Power Electronics, 19(6), 1541–1550.
13. Hussain, A., Van-Hai, B., & Hak-Man, K., (2019). Microgrids as a resilience resource
Author Copy
and strategies used by microgrids for enhancing resilience. Applied Energy 240, 56–72.
14. Hirsch, A., Yael, P., & Josep, G., (2018). Microgrids: A review of technologies, key drivers,
and outstanding issues. Renewable and Sustainable Energy Reviews, 90, 402–411.
15. Hossain Md, A., Hemanshu, R. P., Md Jahangir, H., & Frede, B., (2019). Evolution
of microgrids with converter-interfaced generations: Challenges and opportunities.
16. Wu, H., Yifei, Z., Rui, C., Kai, D., Deqiang, G., Shanglin, D., Xi, W., Yang, Y., & Hao, Y.,
(2019). A microgrid system with multiple island detection strategies. In: 2019 IEEE PES
Asia-Pacific Power and Energy Engineering Conference (APPEEC) (pp. 1–4). IEEE.
17. Yin, G., Sun, M., & Xiao, L. (2007). Review of island detection methods of distributed
generation [J]. Electronic Measurement Technology, 1.
18. Dutta, S., Pradip, K. S., Jaya, B. R. M., & Dusmanta, K. M., (2018). Shifting of research
trends in islanding detection method-a comprehensive survey. Protection and Control of
Modern Power Systems, 3(1), 1–20.
19. Dutta, S., Shailesh, V., Pradip, K. S., Maddikara, J. B. R., & Dusmanta, K. M., (2019).
Islanding detection in a distribution system: A pattern assessment based approach
using Concordia analysis. In: 2019 20th International Conference on Intelligent System
Application to Power Systems (ISAP) (pp. 1–5). IEEE.
20. Kulkarni, N. K., & Khedkar, M. K., (2021). Methods to detect the occurrence of an
unintentional island with passive approach: A review. Journal of The Institution of
Engineers (India): Series B, 1–21.
21. Verma, S., Soham, D., Pradip, K. S., Jaya, B. R. M., & Dusmanta, K. M., (2019).
Islanding detection using bi-directional energy meter in a DFIG based active distribution
network. In: 2019 International Conference on Computer, Electrical & Communication
Engineering (ICCECE) (pp. 1–4). IEEE.
22. Atram, S., Hitesh, M., Yogesh, L., & Bhombe, R. M. (2013). Review on Active Islanding
Detection Technique in Distributed Generation for Current Controlled Inverter. 28,
483–493.
23. Singh, S. K., Mayank, R., Mahiraj, S. R., & Tripurari, N. G., (2021). Hybrid islanding
detection technique for inverter based microgrid. In: 2021 2nd International Conference
for Emerging Technology (INCET) (pp. 1–5). IEEE.
24. Murugesan, S., & Venkatakirthiga, M., (2018). Hybrid analyzing technique based active
islanding detection for multiple DGs. IEEE Transactions on Industrial Informatics,
15(3), 1311–1320.
Non Commercial Use

25. Min-Sung, K., Raza, H., Gyu-Jung, C., Chul-Hwan, K., Chung-Yuen, W., & Jong-Seo, C.,
(2019). Comprehensive review of islanding detection methods for distributed generation
systems. Energies, 12(5), 837.
26. Chawda, G. S., Abdul, G. S., Mahmood, S., Sanjeevikumar, P., Bo Holm-Nielsen,
J., Om Prakash, M., & Palanisamy, K., (2020). Comprehensive review on detection
and classification of power quality disturbances in utility grid with renewable energy
penetration. IEEE Access, 8, 146807–146830.
27. Mahat, P., Zhe, C., & Bak-Jensen, B., (2008). Review of islanding detection methods
for distributed generation. In: 2008 Third International Conference on Electric Utility
Deregulation and Restructuring and Power Technologies (pp. 2743–2748). IEEE.
28. Dehghani, M., Mohammad, G., Taher, N., Kavousi-Fard, A., & Sanjeevikumar, P.,
(2020). False data injection attack detection based on Hilbert-Huang transform in AC
smart islands. IEEE Access, 8, 179002–179017.
Author Copy
29. Kolli, A. T., & Navid, G., (2020). A novel phaselet-based approach for islanding detection
in inverter-based distributed generation systems. Electric Power Systems Research, 182,
106226.
30. Ahmadipour, M., Hashim, H., Mohammad, L. O., & Mohd, A. R., (2019). Islanding
detection method using ridgelet probabilistic neural network in distributed generation.
Neurocomputing 329, 188–209.
31. Khamis, A., Yan, X., & Azah, M., (2017). Comparative study in determining features
extraction for islanding detection using data mining technique: Correlation and coefficient
analysis. International Journal of Electrical and Computer Engineering, 7(3), 1112.
32. Kumar, D., & Partha, S. B., (2018). Artificial neural network and phasor data-based
islanding detection in smart grid. IET Generation, Transmission & Distribution, 12(21),
5843–5850.
33. Kong, X., Xiaoyuan, X., Zheng, Y., Sijie, C., Huoming, Y., & Dong, H., (2018). Deep
learning hybrid method for islanding detection in distributed generation. Applied Energy,
210, 776–785.
34. Yu, S., & Lulin, Y., (2018). Islanding detection method based on S transform and ANFIS.
Journal of Renewable and Sustainable Energy, 10(5), 055503.
35. Admasie, S., Syed, B. A. B., Teke, G., Raza, H., & Chul, H. K., (2020). Intelligent
islanding detection of multi-distributed generation using artificial neural network based
on intrinsic mode function feature. Journal of Modern Power Systems and Clean Energy,
8(3), 511–520.
36. Sneha, P. S., Syam, S., & Ashok, S. K., (2020). A chaotic color image encryption scheme
combining Walsh–Hadamard transform and Arnold–Tent maps. Journal of Ambient
Intelligence and Humanized Computing, 11(3), 1289–1308.
37. Subathra, M. S. P., Mazin, A. M., Mashael, S. M., Garcia-Zapirain, B., Sairamya, N. J.,
& Thomas, G. S., (2020). Detection of focal and non-focal electroencephalogram signals
using fast Walsh-Hadamard transform and artificial neural network. Sensors, 20(17), 4952.
38. Prabha, K., & Shatheesh, S. I., (2020). A novel blind color image watermarking based
on Walsh Hadamard transform. Multimedia Tools and Applications, 79(9), 6845–6869.
39. Andrushia, A. D., & Thangarjan, R., (2018). Saliency-based image compression
using Walsh–Hadamard transform (WHT). In: Biologically Rationalized Computing
Techniques for Image Processing Applications (pp. 21–42). Springer, Cham.
40. Syafii, S., Muhammad, I. R., Lovely, S., & Irvan, Z., (2020). Web-based net energy
meter for grid connected PV system. TEM Journal, 9(1), 37–41.
Non Commercial Use

41. Nguyen, T. A., & Raymond, H. B., (2017). Maximizing the cost-savings for time-of-use
and net-metering customers using behind-the-meter energy storage systems. In: 2017
North American Power Symposium (NAPS) (pp. 1–6). IEEE.
42. Dutta, S., Debomita, G., & Dusmanta, K. M., (2016). Optimum solar panel rating for
net energy metering environment. In: 2016 International Conference on Electrical,

Electronics, and Optimization Techniques (ICEEOT) (pp. 2900–2904. IEEE.
43. Anderson, T., Daniel, C., Chloe, F., Harrison, H., Nina, M., Bailey, T., Andres, C., &
Arthur, S., (2021). Behind the meter: Implementing distributed energy technologies to
balance energy load in Virginia. In: 2021 Systems and Information Engineering Design
Symposium (SIEDS) (pp. 1–6). IEEE.
Author Copy
Non Commercial Use

Index
A electric vehicles (BEVs), 254, 255, 257

management system (BMS), 91, 92, 100,
Algorithms, 4, 16, 18, 39, 45, 67–69, 74, 75,
255, 256, 258–260
78, 79, 84–87, 100, 234, 269, 284, 285,
optimization, 259
314, 316–318, 320, 321, 325, 333
pack model, 99
Alternating current (AC), 5, 8, 9, 104, 105,
system, 301
Author Copy
155, 182, 229, 234, 272, 372
Bayesian regularization, 73
Aluminum tube, 24
Binary particle swarm optimization (BPSO),
Ambient temperature, 30, 31, 40, 196, 200,
269
260
Biogas, 39, 55, 56, 59–62
Ameliorated Harris Hawks optimization
Biogeography-based optimization (BBO),
(AHHO), 284
288, 318
Anaerobic bio-gasification, 59
Bioinformatics
Ant
algorithms, 318
colony optimization (ACO), 286, 318, 327
CI (BioCI), 318
lion optimizer (ALO), 316, 333
Biology meta-heuristic algorithm, 318
Artificial
Biomass energy, 37, 39, 55, 196
bee colony (ABC), 209, 287, 314, 318,
Block
319
chain technology, 61
intelligence (AI), 42, 56, 60, 70, 71, 74,
diagram (simulation model), 371
86, 269, 270, 289, 314, 316
Bluetooth, 39, 48, 53
neural network (ANN), 61, 71, 74, 360,
low energy (BLE), 48
371, 375, 378 Boost converter, 131, 139, 149
Authentic load forecasting, 84 Bosch Global Software Technologies
Automated (BGSW), 123, 125, 152
capacitor bank, 279 Branch current to bus voltage (BCBV), 214,
opportunities, 51 340
selfwatering system, 295 Buck
solar power plant watering system, 295–297 boost converters, 125, 134, 151, 152
Automotive industry, 251, 256, 263 converter, 105, 127, 136, 146
Average energy not supplied (AENS), Building
344–346, 348, 352–355 energy management system (BEMS), 37,
39, 41, 42, 45, 51, 53, 55, 61, 62
B integrated photovoltaic (BIPV), 58
Bat management information system (BMIS),
algorithm (BA), 209, 314, 318 43
inspired, 318 Bus
Battery, 39, 91–94, 96, 97, 99, 100, 126, injection to branch current (BIBC), 213,
155–157, 172, 174, 196, 197, 253, 214, 340, 341
255–263, 296, 297, 299–305 system, 208, 216, 218, 220–223, 284,
couple water pumping system, 304 286, 328, 350
Non Commercial Use

386 Index
C Continuous conduction mode (CCM),

103–105, 113, 117, 123–127, 131, 142,
Cable geometry, 315
151, 152
Cadmium sulfide (CdS), 298
Control algorithm, 6, 18, 186, 228, 248
Capacitance, 142, 143, 152, 155, 158, 160,
Conventional
161, 171, 172, 174
automotive industry, 256
deliverance, 171, 172, 174
electricity, 256, 257
Capacitor, 1, 3, 6, 9, 18, 91, 92, 94, 103, energy, 260
104, 106, 113, 122, 123, 126, 136, 157, resources, 2
163, 166, 173, 174, 242, 268–270, machine learning, 76
274–281, 283–290, 320, 378 two-wheelers, 262
allocation, 288 vehicles, 251, 252, 263, 264
Carbon Converter active space vectors, 239
Author Copy
dioxide (CO2), 45, 51, 59, 60, 361 Cost
monoxide, 45 analysis, 209, 223
Cascaded H-bridge benefit analysis, 41
inverter, 3 effective energy, 315
multilevel inverter (CHB-MLI), 1, 3, 4, function
9, 18 minimization, 180
Central Power Research Institute (CPRI), reduction, 244
155, 174 optimization, 308, 309
Charge Cuckoo search (CS), 286, 314, 318
discharge protocol, 164 algorithm (CSA), 286
efficiency, 155, 171, 172, 174 Cuk converter, 127, 142
Chemical oxygen demand (COD), 60 Cyclic voltammetry studies, 157
Chemistry CI (ChmCI), 318
Circuit averaging (CA), 126, 127, 142, 151, D
152 Data
Cleaner power production, 208 communication, 48, 53, 61
Climate, 69, 70, 251 load requirements, 48
Cloud software applications, 50 Dataset, 71
Compare swap algorithm (CAS), 316 Deep learning, 44, 68, 69, 72, 74, 75, 78,
Computational 86, 87
efficiency, 327 Delamination, 56
intelligence (CI), 318 Depth of discharge (DoD), 158
Connected vehicle technology, 263, 264 Design
Constant dc link
current (CC), 155, 156, 158–160, capacitors, 6
162–167, 169, 171, 174 voltage reference, 5
constant voltage (CCCV), 159, 163, interfacing inductor, 5
164, 171 maintenance, 270
load electrical facilities, 279 Dielectric medium, 157
voltage (CV), 105, 141, 155, 156, Diesel
158–160, 164, 166, 169, 173, 174 fuel-operated diesel generators, 196
Consumer generators, 196
average interruption duration index Differential evolution, 319
(CAIDI), 344, 345, 348, 352–355 Digital signal processors (DSPs), 180
electronics, 50, 94 Dimethylformamide (DMF), 163
Non Commercial Use

Index 387
Diode clamped inverter (DCMLI), 3 response, 180, 245, 246, 249

Direct waveforms, 15
axis current reference, 187
couple solar pumping systems, 305 E
current (DC), 1, 4–13, 15, 16, 18, Electric

103–105, 114, 120, 125–127, 141, 151, double-layer-capacitor (EDLC), 157
182, 197, 201, 229, 234, 258, 259, engineering issues, 314
300–302, 305, 338, 362, 372 generation, 39, 54, 55, 57, 58, 196, 197, 260
link voltage reference, 18 load forecasting, 79–83, 86, 87
loss calculation generation, 8 market pricing techniques, 283
load flow (DLF), 207, 209, 213, 214, 222, mobility, 252, 253, 256, 257, 260–264
337, 339–341, 348, 353, 355 motor, 228, 253, 255, 258, 259
normal irradiance (DNI), 22
Author Copy
power
torque control (DTC), 180, 238–240, buses, 263
245–248 cars, 261, 263
Discharge protocol, 155, 156, 164, 169, systems, 272
171–175 vehicles, 254, 257, 261, 263
Discontinuous conduction mode (DCM), production, 362
104, 126, 142 traction motor, 258, 259
Distributed utilization, 67
energy resources (DERs), 283, 337–339, vehicle (EV), 54, 91, 92, 94, 100, 103,
344, 355 125, 179–181, 184, 185, 189, 190,
generation (DG), 207–209, 211, 212, 251–264
215–220, 222, 223, 260, 261, 269, 270, technologies, 262
283, 285–287, 326–328, 330, 337–339, Electrochemical characteristics, 158
348, 349, 359–367, 370, 371, 375, 380, Electromagnetic
381 load torque, 190
load flow, 337 torque, 181, 184, 189, 231, 238, 239
network, 2, 268, 271, 287, 288, 326, 327, Electronic equipment, 307
339 Encircling prey, 322
static compensator (DSTATCOM), 1–4, End-user electric load demand applications,
6, 9, 10, 15, 18 205
system, 207, 208, 213, 267–271, 280, Energy
282, 283, 285, 287–290, 329, 330, 339, conservation, 69, 267
340, 349, 362, 364 consumption
deliver energy, 271 efficiency, 38
Dolphin echolocation (DE), 314, 319 optimization, 39
Domestic applications, 21, 34 efficiency, 39, 41, 42, 52, 54, 61, 155,
Doubly-fed induction generator (DFIG), 158, 160, 169, 171, 174
360, 372 insufficiency, 2
Dragonfly algorithm (DA), 316, 333 management, 38–41, 62
Duty cycle, 104, 105, 107, 116, 122, network enhancement, 70
124–127, 143, 151 resource consumption, 41
Dynamic sources (ES), 2, 21, 39, 55, 56, 58, 156,
load variations, 18 196, 197, 282, 318, 361, 364, 366
models (synchronous reference frame), 231 storage system, 155, 156, 158, 161, 172,
parameter approximation, 315 196, 255, 285, 317
performance, 9, 246, 247, 249 transmission, 361
Non Commercial Use

388 Index
Engine automobiles, 254 horizontal irradiance (GHI), 22, 23

Enhanced multilayer second-order general- optimization, 319, 324, 328
ized integrator (EMSOGI), 4 warming, 38, 156, 258, 361, 366
Equivalent wireless communication protocol, 48
circuit model (ECM), 91, 92, 96, 97, 100 Gravitational search
series resistance (ESR), 103, 104, 107, algorithm (GSA), 288
123, 126, 139, 143, 149, 152, 162 method, 319
E-rickshaw, 253 Gray wolf optimization, 283, 314, 316, 319,
E-scooters, 263, 264 321
Evolution programming (EP), 319 Green
energy, 37, 39, 41, 55–57, 61, 62
F house
effects, 256
Author Copy
Faster adoption manufacturing of electric
(FAME), 261 gas emissions, 61, 208, 258
Feeder reactance, 275 Internet of Things (G-IoT), 62
Feed-forward neural network, 75 Grey wolf optimizer (GWO), 283, 284, 287,
Field-oriented 313, 314, 316–320, 329–331, 333
control (FOC), 180, 189, 232, 233, 245–249 sine cosine and crow search algorithm
controllers, 227 (GWOSCACSA), 283
Firefly optimization, 269 Grid
Fixed capacitor group, 279 connected modes, 283
Flexibility, 365 management, 62, 380
Flower pollination algorithm (FPA), 288
Flux error, 238 H
Fly Hair tension moisture sensors, 44
capacitor inverter (FCMLI), 3 Hardware
optimization algorithm (FOA), 314 in loop (HIL), 338–340, 342, 343, 348,
Fossil fuel consumption, 260 351, 353, 355
Framework generation costs, 288 specification, 305, 309
Fruit fly optimization algorithm, 314 Harmonic distortion, 279, 370
Fuel cell (FC), 195–204 Harris Hawks optimization (HHO), 284
Fuzzy logic techniques, 42 Heat
energy, 22
G transfer, 23–25, 27, 31, 34
Gain margin (GM), 120, 143, 149 Heating ventilation air-conditioning
Gas microturbine, 362 (HVAC), 41, 43
Gasoline-powered vehicle, 254 Higher
Gauge workspace utilization, 51 intensity discharge lights, 273
Gaussian mutation perturbation, 316 level heuristic algorithm, 287, 326
Generation performance power electronic devices, 180
reference current, 187 power distribution systems, 3
unit vector template, 7 Hirschberg-Sinclair algorithm (HS), 316
Genetic algorithm (GA), 209, 285, 286, 318, Hoffman Electronics, 298
326–329 Hybrid
Geothermal energy, 55 electric vehicles (HEVs), 258
Global, 317 energy system, 195, 196, 198, 205
climate changes, 256 optimization, 269, 284, 333
electricity generation, 55 design, 269
Non Commercial Use

Index 389
of multiple energy resources K

(HOMER), 195, 197, 199, 200, 204
Kaggle database, 71, 78, 86
techniques, 328
Kirchhoff’s current law, 213
ultracapacitor, 175
Krill herd (KH), 318
systems, 156
Hyderabad Electric Supply Company
L
(HESCO), 280
Hydrogen Lead
fuel cell, 205, 256 acid batteries, 156, 157, 196, 255, 263
tank, 197, 198, 200, 203, 204 carbon hybrid ultracapacitor (Pb-C HUC),
Hysteresis predictive control approach, 240 155–158, 173, 175
oxide, 155, 156
I Least square linear curve fitting technique, 160
Author Copy
Li-ion cell model, 92, 99
Imperialistic competitive algorithm (ICA), Line modeling, 210, 223
286 Linear-quadratic regulator (LQR), 242
Improved Lithium-ion batteries, 91, 255, 260
performance management, 51 Load
second-order generalized integrator, 1 demand growth, 87
Indoor environment quality (IEQ), 43, 44 flow, 208, 269, 270, 339, 340, 342, 355
Induction motor (IM), 179–182, 184, 185, analysis, 223, 269, 340, 342
188–190, 229, 232, 259, 272, 273 modeling, 211
Infrared (IR), 21, 43, 45, 60, 275 Long
Instability, 103, 105, 123–125, 149, 151, range (LoRa), 39, 53, 56, 61, 255
152, 181, 369 Wide Area Network (LoRaWAN), 48
Institute of Electrical Electronics Engineers short-term memory (LSTM), 68, 71, 72,
(IEEE), 2, 10, 18, 48, 50, 207–209, 75–87
216, 222, 284, 286, 328, 338, 340, 343, term evolution (LTE), 39, 48
348–354, 380 for machines (LTE-M), 39
Insulated gate bipolar
junction transistors (IGBT), 6 M
transistor (IGBT), 6, 372 Machine
Intelligent power consumption, 54 discrete-time model, 243
Interior permanent-magnet synchronous learning (ML), 38, 45, 74, 76, 86
machines (IPMSM), 227, 229, 232, 233, Material tracking, 50
245, 246 Mathematical
Internal resistance, 156, 161–163, 169, 173, modeling, 104, 105, 124, 209, 227, 229,
174 320
International Energy Agency, 38 morphology, 370
Internet optimization, 289
connection, 48 MATLAB-Simulink, 2, 4, 18, 103–106, 108,
of energy (IoE), 41, 42, 61 123–127, 152, 179, 181, 182, 188
of things (IoT), 37–40, 42, 43, 48–56, Maximum torque per
58–62, 299 ampere (MTPA), 232, 246
remote monitoring system, 56 voltage (MTPV), 232
Irrigation system, 299, 302, 308 Mean
Island, 283, 339, 359–361, 363, 367–372, absolute percentage (MAPE), 68, 72,
375, 376, 380, 381 77–82, 84–87
detection, 369, 371, 375, 381 electric power production, 200
Non Commercial Use

390 Index
Meta-heuristic Network
algorithm, 287, 314, 318 architecture, 38
approach, 288 connectivity, 256
method, 208, 326 protocols used (smart buildings), 45
optimization algorithms, 284 reconfiguration, 337, 339, 340, 354, 355

techniques, 288, 321 Neural network development, 71
Methane, 59 Non
Microbial dynamics, 60 conventional resources, 208
Micro detection zone (NDZ), 370, 371
controller, 232, 296, 299 faradaic reactions, 158
grid (MG), 70, 156, 283, 337–340, 342, ideal, 124
344–348, 353, 355, 361, 365–367, 372, higher-order converters, 127
381 linear
Author Copy
Model load, 4, 7, 8, 18
predictive optimization, 328
control (MPC), 180, 189, 240–245, 248 rechargeable primary batteries, 253
current control (MPCC), 179–181, 184, technical losses (NTLS), 268, 271
185, 187–190, 242, 243, 245–248 customer management systems, 271
measurement, 244 trackbound transport systems, 228
torque control, 244 Novel heuristic algorithms, 316
reference adaptive system (MRAS), 179
current estimator, 190 O
Modified cultural algorithm (MCA), 288 Objective function, 180, 209, 212, 215, 285,
Modulation 289, 329, 333, 338
space vector theory, 249 Occupancy detection, 43, 44
vector, 237 Off-grid Biogas Power Generation Program,
Monetary smart grid activities, 67 59
Monitoring installation, 68 Off-line parameter estimation, 315
Monte-Carlo simulation (MCS), 340, Open
345–348, 352, 355 circuit
Moth-flame optimization (MFO), 323, 324 potential (OCP), 164
Motivation, 208 voltage, 96, 161
Multi loop configuration, 125, 151
dimensional search habitat, 321 Oppositional
layer perceptron (MLP), 68, 72–74, gray wolf optimization (OGWO), 283
78–87 Harris Hawk Optimization (OHHO), 284
level inverter, 1, 18 Optical sensors, 45
Optimal
N performance, 163, 165, 175
Narrow Band-Internet of Things (NB-IoT), positioning, 223
39 Optimization, 61, 68, 96, 97, 155, 158, 173,
National Aeronautics Space Administration 185, 207, 209, 214, 222, 240–242, 244,
(NASA), 198 259, 268, 269, 283–286, 288–290, 313,
Negative temperature coefficient (NTC), 43 314, 316–319, 321, 324–331, 333
Net algorithm, 96, 97, 209, 284
energy meter (NEM), 359, 360, 371–373, dependability, 327
375, 380, 381 methodologies, 284
zero energy buildings (nZEBs), 40 techniques, 268, 289, 290, 317, 330, 333
Non Commercial Use

Index 391
Optocoupler, 307, 309 generation, 60, 67, 68, 156, 208, 212,
Organic fluids, 24 360, 363, 366, 375
Orthogonal frequency-division multiplexing load fluctuations, 70
(OFDM), 48 loss reduction, 207, 269, 284, 290
Overhead transmission line system (OHTL), modulation, 259

329, 330 production consumption, 87
quality (PQ), 1–3, 18, 208, 211, 268, 328,
P 338, 365, 367, 370, 371
Paper target, 209 system
Parabolic trough collector (PTC), 23–29, economics, 268
32–34 Simulator Siemens Calculation
Parameter estimation, 92, 96, 98, 100, 314, (PSSSINCAL), 280
Predictive
Author Copy
315, 329, 333
Particle swarm flux-weakening algorithm, 246
optimization (PSO), 207–209, 214, 217, maintenance, 51
222, 223, 269, 284, 285, 318, 326 Pressure regulating valve (PRV), 141
optimizer, 285, 326, 327 Probabilistic incremental learning, 318
Peak inverse voltage (PIV), 3 Proportional-integral (PI), 8, 9, 12, 61, 151,
Performance characterization, 175 180, 187–189, 234, 243–245, 248
Permanent magnet, 229–231 Pulse width
synchronous automobiles, 228 modulation, 2, 234, 372
Phase space vector theory, 234
lock loops (PLL), 3, 4 modulator (PWM), 6, 9, 143, 179, 180,
margin (PM), 114, 120, 127, 143, 149, 232, 234, 235, 237, 296, 299, 372
229, 238
Phasor measuring unit (PMU), 315, 339 Q
Photovoltaic (PV), 1–4, 6, 8–12, 15, 18, 45, Quadrature-axis current reference, 187
55–58, 60, 105, 156, 195–205, 208, 256, Quantifiable building insights, 51
269, 289, 298, 299, 362
array, 1–4, 6, 10–12, 15, 18, 195, 200, R
201, 204
tracking system, 205 Radial distribution
Physical security, 52 network, 223, 285
PIC micro-controller, 296, 309 system, 207, 269, 341
Plugin hybrid electric vehicles (PHEVs), Radiofrequency (RF), 50
258 Railway traction drive system, 245
Point of, Ramp speed, 249
common coupling (PCC), 5, 370, 372, Rated voltage, 161, 162
373, 375, 380 Reactive power compensation, 268, 283
load (POL), 105 Real-time
Polyvinylidene fluoride (PVDF), 163 applications, 317
Power data, 10, 39, 40, 42, 87, 353
applications, 180, 229, 314, 329 electricity, 52
electronics devices, 259, 362 irradiation, 10
factor, 1, 4, 8, 15, 18, 126, 267–270, monitoring, 42, 56
272–275, 277–279, 283, 290 power system, 327
correction capacitor (PFCC), 278 Recurrent neural network (RNN), 68, 71,
flow equations, 284 72, 74–76, 78–87
Non Commercial Use

392 Index
Reduced Sine-cosine algorithm (SCA), 288

energy consumption, 51 Small signal modeling, 124, 152
operational costs, 52 Smart
Reflector, 21–23, 25, 28, 30, 33, 34 building, 37–40, 42–45, 53–57, 59, 61, 62
Renewable grid management, 37

energy grid, 38, 54–57, 67, 68, 70, 71, 87, 260,
charging infrastructure, 257 261, 359–361, 381
charging systems, 262 technology, 260, 361
resources (RESs), 22, 53, 196, 366 Social
fraction, 195, 197, 201, 205 configuration, 318
sources, 59, 61, 70, 366 hierarchy, 320
Resistance temperature detectors (RTD), 43 Sodium nickel chloride battery, 255
Right half plane (RHP), 122, 123, 127, 149, Software implementation, 295
Author Copy
152 Soil moisture sensor, 295
Root mean squared (RMSE), 68, 72, 77–82, Solar
84–87 cell, 205, 300
Rotor flux energy, 22, 34, 55, 56, 58, 59, 62, 297,
estimation, 184 298, 302, 305
orientation, 184, 189 irradiation, 9, 10, 18, 197, 299
microgrid, 155, 156, 158, 172, 175
S panel pumping system, 300
Safety assurance, 263, 264 photovoltaic (SPV), 156, 196, 299, 339,
Sample hold circuit (S-H), 6, 7 350
Second order generalized integrator, 3 resource, 299
Self-loop memory blocks, 76 power
Sensible climatic consignments, 69 applications, 155, 171, 173, 174
Sensitivity analysis approach, 209 automatic irrigation system, 302
Sensorless, 179–181, 184, 185, 189, 190 automatic power plant watering
Sensors network protocols, 62 system, 295
Shark smell optimization (SSO), 288 irrigation system, 299
Short meter, 56
range communication, 48 technology, 299
term load forecasting (STLF), 69–72, 86, PV, 3, 309
87 radiation, 13, 21, 23, 195, 196, 199, 200,
wave radiation, 22 205
Shunt thermal systems (STSs), 58
capacitor, 269, 275, 288 water pumping system, 319
location, 278 Space
compensation, 278 allocation management, 53
Sigfox, 39, 48, 53 vector modulation (SVM), 71, 234
Signal injection speed estimation tech- Stable closed loop operations, 6
niques, 181 Stainless steel reflectors, 23, 33, 34
Simulated Stakeholders, 309
annealing (SA), 287, 328, 329, 333 Standard
study, 13, 188, 190 optimal performance of HUCS, 172
Simulink test protocol, 155, 156, 171–175
design optimization, 197 State
parameter estimation, 96 of charge (SoC), 91–93, 97–100, 158, 197
Non Commercial Use

Index 393
space Tech-savvy smart society, 51

averaging (SSA), 126, 127 Temperature
modeling, 127, 152 coefficient, 198
Stator phase current, 190 correction resistance, 284
Steady data, 199, 205
state sensor, 24
modeling, 340 Thermal
performance, 18 efficiency, 22, 23, 25, 26, 28, 29, 32–34
response, 246, 249 performance, 24, 27, 34
waveforms, 15 runaway, 260
voltage discharge curve, 165 Time-delay compensation, 243
Steam engines, 254 Top-of-line voltage switching, 239
Stockwell transform, 370 Torque
Author Copy
Structural error, 189, 190, 238
engineering design optimization, 286 producing current, 243
optimization problems, 286 Total harmonic distortion (THD), 3, 10, 18,
Supercapacitors, 255 234, 326, 370
Supervisory control data acquisition Tracking, 320
(SCADA), 315 Transistor, 308, 309
Surface Transmission line, 53, 54, 273, 283, 284,
mount motor rotor designs, 229 314, 315, 317, 319, 320, 329–331, 333,
permanent magnet synchronous machine 366, 367, 372
(SPMSM), 229, 230 parameter, 314, 317, 320, 329
Swarm chain, 320
algorithms, 318 properties, 284
intelligence, 314
method, 316 U
systems, 321
Ultra-wideband (UWB), 39, 48, 53
meta-heuristic algorithms, 318
Switch mode power supplies (SMPS), 125, Uninterruptible power supplies (UPS), 2
126, 134 Unit vector template, 7, 18
Synchronous Universal serial bus (USB), 39, 50
AC machines, 229
converters, 104, 105, 124 V
machines, 229 Vector control (VC), 95, 180, 184
reference frame theory (SRFT), 3, 4, 16–18 Vehicular pollution, 263, 264
reluctance, 229 Ventilation system, 52
System Voltage
average interruption discharge characteristics, 164–167
duration index (SAIDI), 345, 355 fluctuations, 259, 271
frequency index (SAIFI), 344–346, profile, 207–209, 220, 221, 269, 280,
348, 352–355 286, 289, 290, 326, 338, 349, 350
integration, 279 improvement, 289, 290, 338
stability, 287
T regulation, 1, 271, 272, 274
Tabu search (TS), 287, 314, 316, 326 regulators, 125, 274
Teaching-learning-based optimization source
(TLBO), 284, 285 converter (VSC), 1
Technological breakthroughs, 263 inverter, 180, 190, 238
Non Commercial Use

394 Index
stability, 209, 268, 287, 338, 365 Whale

index (VSI), 180–183, 185, 186, 238, optimization algorithm (WOA), 288, 314,
288 317, 321, 322, 329–331
variation, 268, 272, 280, 287, 289, 319 time-dependent position, 322
Wireless fidelity (Wi-Fi), 39, 50, 53
W Workplace opportunities, 52
Walsh Hadamard transform (WHT), 360,
371, 373, 375–381 Z
Waste disposal system, 39 Zero current detector circuit (ZCD), 6, 7
Water conservation, 309 Zigbee, 50
Wavelet transform, 370 protocol, 50
Weighting factor tuning, 245 technology, 50
Author Copy
Non Commercial Use

Non Commercial Use

Author Copy
Raj
Sekhar
Krishna
Mahapatra
A Pragmatic Approach Towards Sustainable Development
Providing innovative, efficient, clean, and safe solutions and research for interfacing internet technology with
energy power grids for smart cities and smart transportation, this new volume discusses the use and
automation of electricity infrastructures for energy producers and manufacturers, integrating the
implementation of the Internet of Things (IoT) technology for distributed energy systems in order to optimize
energy efficiency and wastage. The volume offers a wide range of research on using IoT for energy solutions,
such as algorithms for the design and control of energy grids, investigations of thermal efficiency from solar
grids, energy for smart buildings using IoT, deep learning for electrical load forecasting, hybrid
ultracapacitors in solar microgrids, induction motor-driven electric vehicles, power loss reduction and voltage
improvement, and much more.
The
Author Copy
ABOUT THE EDITORS
Sheila Mahapatra, PhD, is Professor and Head, Electrical and Electronics Engineering
Department, and Assistant Dean (Research), Alliance College of Engineering and Design,
Internet of Energy
Alliance University, Bangalore, India. She has a total experience of 18 years in the field of engineering,
academics, and research. Her areas of interest include power systems optimization, FACTS devices,
renewable energy, energy economics, and sustainability. She has many research publications in SCI, SCIE,
and Scopus-indexed journals as well as book chapters and conference proceedings to her credit. She serves
Internet of Energy
as a reviewer for reputed journals.
The
Mohan Krishna S., PhD, is Associate Professor in the Department of Electrical and Electronics Engineering
at Alliance College of Engineering and Design, Alliance University, Bangalore, India. His research interests
include electric vehicles, smart homes and IoT-based building energy management systems, and state
observers for induction motors energy economics and sustainability. He has published his research in

journals as well as book chapters and conference proceedings. He is an advisory board member to the energy
section of Heliyon (Elsevier). He serves as Associate Editor of the International Journal of Smart Vehicles and
Smart Transportation and is also editing several books on electric vehicles, smart grids, and energy
sustainability.
B. Chandra Sekhar, PhD, is Specialist–EV in Transportation Business Unit at L & T Technology Services
Limited (LTTS), Bangalore, India. Prior to joining LTTS, he worked as a Technical Lead in the Automotive
Business Division at TATA Consultancy Services Limited, and as an R & D Engineer in the Solar Division at
Power One Micro Systems Pvt. Ltd,. Dr. Sekhar has published over 21 papers in peer-reviewed journals and
conferences. His research areas are hybrid electric vehicles, EV chargers, battery management systems,
power electronics converters, DC micro grids, LED drivers, and hybrid renewable energy systems.
Saurav Raj, PhD, is an Assistant Professor with the Electrical Engineering Department at the Institute of
Chemical Technology Mumbai, Marathwada Campus, Jalna, India. He has also worked as an Assistant
Professor at Alliance University, Bangalore, India, and as an Associate Editor of the European Journal of
Electrical Engineering. He has done extensive research work in power system optimization, reactive power
planning, swarm and evolutionary optimization techniques, FACTS devices, computational intelligence, and
renewable energy.
ISBN: 978-1-77491-418-2
90000
Sheila Mahapatra | Mohan Krishna S.

www.appleacademicpress.com 9 781774 914182 B. Chandra Sekhar | Saurav Raj
Editors APPLE ACADEMIC PRESS
Non Commercial Use

The Internet of Energy - Author Copy - 071023

Uploaded by

Copyright:

Available Formats

The Internet of Energy - Author Copy - 071023

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Internet of Energy - Author Copy - 071023

Uploaded by

Copyright:

Available Formats

Apple Academic Press

A Pragmatic Approach Towards

Sheila Mahapatra | Mohan Krishna S.

Non Commercial Use

Non Commercial Use

A Pragmatic Approach Towards

Non Commercial Use

Non Commercial Use

A Pragmatic Approach Towards

Non Commercial Use

Burlington, ON L7N 0A4, CANADA Abingdon, Oxon, OX14 4RN UK

© 2024 by Apple Academic Press, Inc.

Library and Archives Canada Cataloguing in Publication

CIP data on file with US Library of C

ISBN: 978-1-77491-418-2 (hbk)

Non Commercial Use

Sheila Mahapatra, PhD

Non Commercial Use

Mohan Krishna S., PhD

B. Chandra Sekhar, PhD

Non Commercial Use

in Electrical Engineering from the Central Power Research Institute (CPRI)

Technology, Bengaluru; NBKR Institute of Technology, Nellore, Andhra

Saurav Raj, PhD

Non Commercial Use

Non Commercial Use

1. Design and Control Using an ISOGI-Q Algorithm for Grid

2. Thermal Efficiency Investigation of a Solar Parabolic Trough

3. Smart Buildings Using IoT and Green Engineering...................................37

4. A Novel Electrical Load Forecasting Model Using a Deep

5. Battery Plant Model Development for BMS Application...........................91

6. Modeling of Constant Voltage Control in Synchronous

7. Determination of Open Loop Responses of Switched DC–DC

8. Evolution of Hybrid Ultracapacitors in Solar Microgrids.......................155

9. Speed Sensorless Model Predictive Current Control of Induction

Non Commercial Use

10. Performance Analysis of Hybrid Off-Grid Two-Axis Photovoltaic

11. Efficient Integration of Distributed Generation in Radial

12. Classical and Predictive Control of Interior Permanent Magnet

13. Electric Mobility...........................................................................................251

14. Power Loss Reduction and Voltage Improvement Through

15. Solar Photovoltaic Powered Automatic Irrigation System for

16. Optimal Parameter Estimation of 3-Phase Transmission Line

17. Network Reconfiguration-Based Outage Management for Reliability

18. A Net Energy Meter-Based Approach for Islanding Detection in

Non Commercial Use

Non Commercial Use

Munsif Ali Jatoi

Technology, Coimbatore, Tamil Nadu, India

Shafiq Ur Rehman Massan

Muhammad Shahid Mastoi

Non Commercial Use

Akash Kumar Pandey

Prudhvi Kiran Pasam

J. Pradeep Kumar Rao

Pradip Kumar Sadhu

Anwar Ali Sahito

Abdul Latif Shah