10.18186 Journal of Thermal Engineering.323392 327445
10.18186 Journal of Thermal Engineering.323392 327445
10.18186 Journal of Thermal Engineering.323392 327445
ABSTRACT
In this paper, proper inverter topology for tropical region will be determined with temperature effect as a
considered factor. PV characteristics on certain temperature are obtained through mathematical equations in
MATLAB-Simulink and it will be represented by PV I-V and P-V characteristic curves. Based on the modeling,
some PV parameters such as temperature coefficients of short circuit current (I SC), open circuit voltage (VOC),
maximum output power (PMAX), and maximum power voltage (VMP) can be determined as high as 0.56%/oC,
-0.31%/oC -0.4%/oC, and 0.43 %/oC respectively. As the result, for tropical region, the central inverter is the
most proper inverter topology to supply 5 kW AC PV system, string inverter in the 2nd place, and AC module with
utility rates of 97.34%, 96.32 %, and 88.1% based on direct measurement on PV panel’s temperature respectively.
From NOCT approach, the utility rates are around 92.96 % for central inverter, 91.98% for string inverter and 91.8
% for AC module.
INTRODUCTION
Photovoltaic (PV) has been widely utilized to produce electricity from solar energy because of its
simplicity, low maintenance, and ecofriendly reason. Both of PV’s I-V (current-voltage) and P-V (power-voltage)
are shown by figure 1. PV V-I and P-V curve represent its power conversion characteristic at certain temperature
and irradiation. I-V curve shows the possible combinations of current and voltage output of a photovoltaic (PV)
device. On the other hand, P-V curve is obtained from voltage and current multiplication. VMP and IMP represent
PV’s maximum power voltage and maximum power current respectively [1].
PV electrical characteristic such as maximum output power (PMP), open circuit voltage (Voc), short
circuit current (ISC), maximum power voltage (VMP), and maximum power current (IMP) are strongly affected
by temperature fluctuation. Figure 2 shows temperature fluctuation effect on PV I-V curve. Blue curve and red
curve represents PV I-V curve before and after temperature rise respectively. Even though PV short circuit current
(ISC) slightly increase, open circuit voltage (VOC) are significantly decrease due to temperature rise. As the result,
temperature rise lead to reduction on PV maximum output power.
The standard test condition (STC) of PV specifies a PV module in temperature of 25°C, an irradiance of
1000 W/ m² with an AM 1.5 spectrum and zero wind speed [3]. On the other hand, ambient temperature in tropical
region can reach up to 40 °C [4]. Therefore, the standard test condition (STC) of PV cannot fully represent the
performance of PV in tropical region.
As a part of PV system, solar inverter is used to convert PV DC power to AC power. The basic inverter
topologies for PV are shown by figure 3. Figure 3.a shows central solar inverter topology where PV are arranged
series and parallel forming an array and connected to single inverter. Figure 3.b represent string inverter topology.
It consist of several PV module which are arranged in series forming a string. Every string is connected to one
string inverter. Figure 3.c shows AC-module concept. It consist a PV module which is connected to low-power
solar inverter [5]. These inverters play a key role in the energy efficiency and reliability since they operate the PV
array at the Maximum Power Point (MPP) [5]. Therefore, PV array maximum power point is one factor to be
considered to choose proper inverter topology.
However, due to its effect on PV electrical parameters, temperature rise on PV can also affect solar
inverter choosing. PV electrical parameters are important to determine PV solar inverter topology because different
inverter topology has different specification of power (P MP), open circuit voltage (VOC), short circuit current (ISC),
maximum power voltage (VMP), and maximum power current (IMP). Hence, temperature fluctuation on PV also
affect solar inverter topology choosing.
This paper discuss about PV solar inverter choosing on tropical region that has high temperature
characteristic. The proper inverter topology can be determined by analyzing the effect of temperature and
irradiation on the PV panel. The effect of temperature and irradiation will be represented by PV I-V and P-V
Figure 2. The effect of temperature rise on the PV I-V characteristics curve [2]
1260
Journal of Thermal Engineering, Research Article, Vol. 3, No. 3, pp. 1259-1270, July, 2017
curves. PV characteristics on certain temperature would be analyzed using Solar panel modeling through
MATLAB Simulink based on mathematical equations that form PV characteristic curve.
PV MODELING
The modeled PV module is represented as diagrams blocks those are formed by mathematical
current’s equations of photovoltaic such as reverse saturation current (Irs), saturation current (Io), and photocurrent
(Iph). The current equation are shown by equation 1, 2, and 3 respectively [6].
Isc
(q)(Voc)
[exp( )]
( Ns)( K )(n)(T ) (1)
T 3 ( q)(Eg) 1 1
Io= Irs( ) exp [ ( − )]
Tref (n)(k ) Tref T (2)
G
Iph= [Isc+ K (T − 298)]
1000 W m− 2 (3)
PV’s output current is obtained by using both photocurrent (Iph) and saturation current (Io), which is
represented by equation 4 [5].
The modeled PV-A is represented by block diagram in figure 4. PV-A modeling is conducted to obtain
PV characteristic curve. As input variable, input voltage (Vin), normalized radiation, and temperature are used.
Input voltage is represented as ramp function which have gradient (m) of 1. In STC, output current (I PV) is equal
to PV’s STC short circuit current (ISC_STC) if Vin equal to 0 V. On the other hand, (IPV) is equal to 0 A if Vin is equal
to PV’s STC open circuit voltage (VOC_STC). The block diagram also shows the modeled PV-A power and current
with the value of 255.4 W and 8.89 A respectively. Technical specification of the modeled PV-A are shown by
table 1.
MODELING VERIFICATION
Verification is needed in order to ensure that modeled PV-A has same characteristic as the actual one.
Figure 4 and figure 5 shows modeled PV-A I-V and P-V characteristic curve respectively as the result of MATLAB-
Simulink simulation. From the simulation, value of short circuit current (ISC), open circuit voltage (VOC) and
maximum output power (PMP) can be obtained. The parameters which are obtained from the simulation are
compared to the actual parameters from PV datasheet. Comparison of the parameters are shown in table 2.
1261
Journal of Thermal Engineering, Research Article, Vol. 3, No. 3, pp. 1259-1270, July, 2017
Figure 5. Modeled PV-A Current-Voltage (I-V) characteristic curve under Standard Test Condition
(STC)
Figure 6. Modeled PV-A Power-Voltage (P-V) characteristic curve under Standard Test Condition (STC)
Table 2 shows that all parameters has percentage of error less than 0.2 %. It means that PV modeling has been
done properly.
1262
Journal of Thermal Engineering, Research Article, Vol. 3, No. 3, pp. 1259-1270, July, 2017
NOCT 20
T _ cell T _ ambient xS (5)
80
Jakarta is chosen as representation for tropical region. Monthly temperature data of Jakarta are shown by
the table 3.
Based on equation 5, maximum and minimum PV cell temperature in Jakarta can be obtained. The
maximum and minimum PV cell temperature are 58.14 oC and 58.25 oC respectively.
As a comparison, actual temperature measurement is also conducted on PV cell. The measurement result
is shown by table 4. Based on real measurement Maximum and Minimum cell temperature are 38.9 oC and 67.9
o
C respectively.
The different results between NOCT approach and actual measurement can occur because actually,
relation between ambient temperature and cell temperature is a complex thing. Both of results are used for further
calculation and analysis.
Table 3. Jakarta ambient temperature data
Month Temperature (oC)
January 26.89
February 27.03
March 27.35
April 27.51
May 27.25
June 27.15
July 27.52
August 27.92
September 28
October 27.33
November 26.97
December 27.37
Average 27.36
1263
Journal of Thermal Engineering, Research Article, Vol. 3, No. 3, pp. 1259-1270, July, 2017
INVERTER SPECIFICATION
Calculation and analysis are conducted to obtain proper inverter topology to supply 5 kW AC load in
tropical region. Central inverter, string inverter, and AC-Module are analyzed. Table 5 shows Inverter specification
which are used in calculation.
RESULT
PV modeling result
Temperature’s effect on PV characteristic curve
The first simulation is conducted in order to investigate temperature effect on PV performance, the
simulation is conducted under constant irradiance of 1000W/m2 and varying temperature. I-V and P-V
characteristic curves under temperature variation is shown by figure 7 and 8 respectively.
Figure 7. Modeled PV-A Current-Voltage (I-V) characteristic curve with temperature variation and constant
irradiance
Figure 8. Modeled PV-A Power-Voltage (P-V) characteristic curve with temperature variation and constant
irradiance
1264
Journal of Thermal Engineering, Research Article, Vol. 3, No. 3, pp. 1259-1270, July, 2017
Based on figure 8, temperature rise is inversely proportional to maximum power (P MP) and open circuit voltage
(VOC). On the other hand, figure 7 shows that temperature rise is slightly proportional to short circuit current (I SC).
As the temperature increases, the rate of the photon generation increases. Consequently the reverse saturation
current increases rapidly and this reduces the band gap. Hence this leads to marginal changes in the current but
major changes in voltage [8].
PV temperature coefficient
From the simulation which is conducted, temperature coefficients of short circuit current (I SC), open
circuit voltage (VOC), maximum output power (PMAX), and maximum power voltage (VMP) can be obtained. The
coefficients can be used to calculate ISC, VOC, PMAX, VMP in certain earth’s surface’s certain point with temperature
as parameter. The temperature coefficient can be obtained through the equation 6.
( X STC− X )
α= x 100 %perhundred
( X SCT ) x (Δ T ) (6)
Modeled PV-A is simulated to operate at temperature 25oC to 75oC. From the temperature variation, the
temperature coefficients of short circuit current (I SC), open circuit voltage (VOC), maximum output power
(PMAX), maximum point voltage (VMP) can be obtained as high as 0.056%/oC, -0.31%/oC and -0.39%/oC, -0.43
%/oC respectively.
Figure 9. Modeled PV-A Current-Voltage (I-V) characteristic curve with irradiance variation and constant
temperature
Figure 10. Modeled PV-A Power-Voltage (P-V) characteristic curve with irradiance variation and
constant
1265
Journal of Thermal Engineering, Research Article, Vol. 3, No. 3, pp. 1259-1270, July, 2017
PV Inverter Selection
Calculation on VOC and VMP values based on tropical region temperature characteristic are conducted to
determine numbers of PV which are used for each inverter topology. Value of V OC is used to determine maximum
PV voltage in low temperature and value of VMP is used to determine minimum PV’s voltage in high temperature.
Therefore, PV inverter can still operate effectively even though PV cell temperature are fluctuate. The calculation
is conducted based on real PV cell temperature measurement and NOCT approach.
VDC MAX
N MODULE MAX =
VOC MAX (8)
VDC MIN
N MODULE MIN =
VMP MIN (10)
Maximum number of string connected to Inverter is limited by output current of total PVs in a string. It is obtained
by equation 11.
I INV
N STRING =
IMP STC (11)
Maximum tolerable converted power of PVs which is represented by PV’s number is obtained by equation 12.
Po Po
% x100% (13)
P _ INV P _ INV
Based on the PV-A and inverters specification which are shown by table 1 and table 5, calculation on each inverter
topology are conducted by using equation 7 to equation 13. Result of calculation is shown by table 6, table 7, and
table 8.
The calculation result are used to choose some of PV strings and array combination on each inverter
topology. Output power and utility rates of arranged PV for each inverter topology are shown by table 9 to table
12. PV inverter utility rate is the key parameter to determine the proper inverter topology. It represents
compatibility between PV inverter and PV panel. Furthermore, due to temperature effect on PV maximum output
power, PV inverter utility rate is also affected by temperature fluctuation.
1266
Journal of Thermal Engineering, Research Article, Vol. 3, No. 3, pp. 1259-1270, July, 2017
Table 9. PV maximum output power and inverter utility rates for central inverter topology based on direct
measurement on pv panel’s temperature
One String
Total
Configuration Po (W) Po/PINV (%)
Modules
A 10 2027.92 40.56
B 11 2230.71 44.61
C 12 2433.50 48.67
D 13 2636.30 52.73
E 14 2839.09 56.78
F 15 3041.88 60.84
Two Strings
Total
Configuration Po (W) Po/PINV (%)
Modules
G 20 4055.84 81.12
H 22 4461.43 89.23
I 24 4867.01 97.34
J 26 5272.6 105.45
K 28 5678.18 113.56
L 30 6083.76 121.67
Three Strings
Total
Configuration Po (W) Po/PINV (%)
Modules
M 30 6083.7696 121.675392
N 33 6692.14656 133.8429312
O 36 7300.52352 146.0104704
P 39 7908.90048 158.1780096
Q 42 8517.27744 170.3455488
R 45 9125.6544 182.513088
Based on table 9 and 10, there are some configuration with different PV strings and array combination
for central inverter topology. Table 9 shows that based on direct measurement on PV panel’s temperature,
configuration I is the best configuration for central inverter with total PV-A modules of 24, it have the best utility
rates i.e. 97.34% and based on NOCT approach, the best configuration for central inverter topology is configuration
H with utility rates of 92.96 % .
String inverter topology uses one PV’ string for each PV solar inverter PV. Table 11 and table 12 show
that the optimum number of PV module for string inverter is 12 modules with utility rates of 96.32 % based on
direct measurement on PV panel’s temperature and 11 modules with utility rates of 91.98 % based on NOCT
approach . In addition, two string inverter are needed to supply 5 kW AC system. Therefore, 24 PV modules and 22
1267
Journal of Thermal Engineering, Research Article, Vol. 3, No. 3, pp. 1259-1270, July, 2017
Table 10. PV maximum output power and inverter utility rates for central inverter topology based on NOCT
approach
One String
Total
Configuration Po (W) Po/PINV (%)
Modules
A 10 2112.63 42.25
B 11 2323.89 46.48
C 12 2535.15 50.70
D 13 2746.41 54.93
E 14 2957.67 59.15
F 15 3168.94 63.37
Two Strings
Total
Configuration Po (W) Po/PINV (%)
Modules
G 20 4225.25 81.12
H 22 4647.77 92.96
I 24 5070.3 101.41
J 26 5492.82 109.85
K 28 5915.35 118.31
L 30 6337.87 126.76
Three Strings
Total
Configuration Po (W) Po/PINV (%)
Modules
M 30 6337.88 126.75
N 33 6971.67 139.43
O 36 7605.45 152.10
P 39 8239.24 164.78
Q 42 8873.02 177.460
R 45 9506.82 190.14
Table 11. PV maximum output power and inverter utility rates for string inverter topology based on direct
measurement on PV panel’s temperature
Single String
Number of Modules Po (W) Po/PINV (%)
8 1605.44 64.22
9 1806.12 72.24
10 2006.8 80.27
11 2207.48 88.3
12 2408.16 96.32
13 2608.84 104.35
14 2809.52 112.38
15 3010.2 120.41
PV modules are needed for string inverter topology based on direct measurement on PV panel’s temperature and
NOCT approach respectively.
AC module topology use small capacity inverter which can only be utilize by a PV. Therefore, to supply
5 kWAC system, 24 AC module are needed. Table 5 shows that the specification of small inverter is enough to
compensate PV-A 255 Wp. By using equation 14 and equation 15, utility rates percentage of AC module topology
can be obtained, i.e. 88.1 % for real cell temperature measurement and 91.8% for NOCT calculation.
The maximum utility rates comparison on three different inverter topology are shown by figure 11 and
figure 12. It can be concluded that with high temperature characteristic of tropical region, central inverter is the
best topology to supply 5 kWAC system compared to other inverter topology because it have best utility rates based
on both of direct measurement on PV panel’s temperature and NOCT approach.
1268
Journal of Thermal Engineering, Research Article, Vol. 3, No. 3, pp. 1259-1270, July, 2017
Table 12. PV maximum output power and inverter utility rates for central inverter topology based on NOCT
calculation
Single String
Number of Modules Po (W) Po/PINV (%)
8 1672.5 66.89
9 1881.56 75.26
10 2090.62 83.62
11 2299.68 91.98
12 2508.74 100.35
13 2717.8 108.7
14 2926.86 117.075
15 3135.93 125.46
100
97,34 %
96,32 %
Maximum Utility Rates
98
96
94
92
90 88,1 %
88
86
84
82
Central Inverter String Inverter AC Moodule
Inverter Topology
Figure 11. Comparison of inverter topology maximum utility rates based on real temperature measurement
93,5
Maximum Utility rates
92,96 %
93
92,5
91,98 %
92 91,8 %
91,5
91
Central Inverter String Inverter AC Moodule
Inverter Topology
Figure 12. Comparison of inverter topology maximum utility rates based on NOCT approach
CONCLUSION
This research successfully accomplish PV-A modeling and the results are temperature coefficients of
short circuit current (ISC), open circuit voltage (VOC), maximum output power (P MAX), and maximum power
voltage (VMP) consecutively as high as 0.56%/oC, -0.31%/oC -0.4%/oC, and 0.43 %/oC. Central inverter is the
most proper inverter topology to supply 5 kW AC system in tropical area with utility rates of 97.34% and 92.96 %
for real cell temperature measurement and NOCT calculation. On the other hand, string inverter topology has
utility rates of 96.32 % and 91.98 % for real cell temperature measurement and NOCT calculation, and AC-
module topology has utility rates of 88.1 % and 91.8 % for real cell temperature measurement and NOCT
calculation.
NOMENCLATURE
Tref Referenced temperature (298 K)
Eg Semiconductor band-gap energy (eV)
Np Number of cells in parallel
1269
Journal of Thermal Engineering, Research Article, Vol. 3, No. 3, pp. 1259-1270, July, 2017
REFERENCES
[1]Solar Metric (2010) Guide to Interpreting I-V Curve Measurements of PV Arrays. Available on
http://www.solmetric.com/
[2]Bowden, Stuart, Honsberg, Christiana. (2013). Instructions on Photovoltaic. http://www.pveducation.org/
[3]A. Green, M. (1998). Solar Cells Operating Principles, Technology and System Applications. Kensington:
The University of New South Wales.
[4]Meteoblue (2015). General Climate Zone. Available on https://content.meteoblue.com/en/meteoscool/general-
climate-zones/tropical-zone
[5]Cramer, G.; Ibrahim, M.; Kleinkauf, W.: PV System Technologies – State of the Art and Trends in Decen-
tralised Electrification. reFOCUS, January/February 2004.
[6]Shivananda Pukhrem (2013). A PHOTOVOLTAIC PANEL MODEL IN MATLAB/SIMULINK. Available on
http://www.researchgate.net/publication/255721161
[7]Bowden, Stuart, Honsberg, Christiana. (2013). Nominal Operating Cell Temperature.
http://www.pveducation.org/.
[8]N.H. Zaini, M. Z. Ab Kadir, M. Izadi, N.I. Ahmad, M.A.M Radzi and N. Azis. (2015). The Effect of Temperature
on a Mono-crystalline Solar PV Panel. IEEE Conference on Energy Conversion (CENCON). Pages: 249 - 253,
DOI: 10.1109/CENCON.2015.7409548
1270