International Journal of Mechanical & Mechatronics Engineering IJMME-IJENS Vol:15 No:05 53

Prediction and Optimization of Tool Wear On A22E

(Bimetal Bearing Material) Using RSM and Genetic
Algorithm
1
R.Babu, 2 Dr. D. S. Robinson Smart, 3 Dr. G.Mahesh, 4M. Shanmugam
1
Research Scholar, Department of Mechanical Engineering, Karunya School of Mechanical sciences, Karunya University,
Coimbatore, Tamil Nadu, India.
2.
Professor, Department of Mechanical Engineering, Karunya School of Mechanical sciences, Karunya University, Coimbatore,
Tamil Nadu, India.
3
Professor, Department of Mechanical Engineering, Sree Sakthi Engineering
College, Coimbatore, Tamil Nadu, India.
4
Deputy General Manager (Operations), Bimetal Bearings Limited, Coimbatore, Tamil Nadu, India.
1
[email protected], [email protected], [email protected]
4
[email protected]

Abstract-- Tool wear is an important criterion in hard finish rates are usually chosen to give an economical tool life.
facing operation, which increases the temperature on the work Clearly, any tool or work material improvement that increases
piece, vibration, cutting force and decreases the surface tool life without causing unacceptable drops in production will
roughness. In this present study, spindle speed, feed rate, depth be beneficial. To form a basis for such improvements, efforts
of cut and end relief angle are taken as an input parameters. have been made to understand the behavior of the tool, how it
Experiment is conducted on A22E Bimetal bearing material
physically wears, the wear mechanisms, and forms of tool
using M42 HSS tool material in finish hard facing and tool wear
was measured using tool maker’s microscope. Design of failure. Cutting parameters, work piece material, cutting tool
Experiments (DoE) methodology is used to conduct the geometry have a crucial influence on the accomplishment of
experiment. Response Surface Methodology is used to predict the desired product quality, tool wear, surface roughness, cutting
tool wear. The second order mathematical model in terms of force and temperature rise in tool and work piece etc. During
machining parameters was developed. The direct and interaction turning and facing operation, friction between cutting tool and
effect of the machining parameters with tool wear were analyzed, work piece materials plays a key role in temperature rise, wear
which helped to select process parameter in order to reduce tool etc. Among the above, tool wear has been found to be a most
wear which ensures quality of the facing operation. influencing factor. CNC machining centers play a major role
in machining industry, even though spindle speed, feed rate,
Index Term-- Spindle speed, Feed rate, Depth of cut, End Relief depth of cut were already programmed before machining but
angle, Tool wear, Response Surface Methodology (RSM), the machining performance and product quality are not
Genetic Algorithm (GA). guaranteed up to standard level. Therefore, the optimum
1. INTRODUCTION turning and facing operation have to be accomplished.
In the recent scenario, due to the competitive marketing
strategy the major goal of the manufacturing industry is to The dynamic performance of rotating components is highly
increase productivity and product quality with lesser influential on efficiency of any rotating machine. Bearings are
production time. The quality of a product is strongly considered as the heart of rotating machinery [2]. Bearing play
associated with the condition of the cutting tool. Mainly a major role in industries and automobile sectors. A lot of
cutting tool plays a major role in manufacturing due to wear research is under processing by considering different
and breakage, so tool wear is found to have a direct impact on materials, tools etc and different combinations of machining
the quality of the product such as surface finish, dimensional parameters. The authors [3] developed a methodology to
accuracy and cost of the finished product is increased due to optimize hardened 52100 bearing steel using a low CBN
tool failure. The tool failure and the detection of tool failures content insert to optimize the crater wear rate. The author
are essential to improve manufacturing quality and to increase suggested that adhesion is the main wear mechanism. Flank
the productivity. The author [1] defined an empirical tool wear model [4] in finish turning of hardened 52100
relationship between tool life and cutting speed (VTn = C) and bearing steel, tool performance is evaluated by considering the
C are coefficients. parameters such as cutting speed, feed, and depth of cut. The
analysis of variance was carried out to investigate the cutting
Cutting tool life is one of the most important economic conditions, and suggested that cutting speed plays a dominant
considerations in metal cutting. In roughing operations, the role in determining the tool performance. The authors [5]
tool material, the various tool angles, cutting speed, and feed developed a mathematical model by considering cutting

150405-2727-IJMME-IJENS © October 2015 IJENS

IJENS
 International Journal of Mechanical & Mechatronics Engineering IJMME-IJENS Vol:15 No:05 54

parameters using CBN cutting tools in finish hard turning radius and tool wear on residual stress distribution in hard
operation of hardened 52100 bearing steel. The authors turning of bearing steel JIS SUJ2. The results obtained in this
suggest that the adhesion is the dominant wear mechanism study shows that the tool nose radius affects the residual stress
within the range of cutting conditions. The performance of distribution significantly [17]. The authors [18] conducted an
PCBN tool in the finish turning of GCr15 bearing steel was experiment to find the CBN tool wear in hard turning, results
investigated to determine the influence of the work piece adhesion is the dominant wear mechanism in turning hardened
hardness on changes in cutting temperature and tool wear 52100 bearing steel with a hardness of 62 HRC using a low
characteristics [6]. An experimental investigation has been CBN content tool. An experimental study was carried out by
done to clarify the effects of tool nose radius and tool wear on using CBN tool during the turning of AISI 52100 bearing
residual stress distribution of hard turning of bearing steel (JIS steel. The flank wear rate is low when turning at low cutting
SUJ2) by using CBN tools with different nose radius, results speed; the flank wear rate becomes quite large and sensitive to
the tool nose radius affects the residual stress distribution the feed rate [19]. The model was proposed and conducted an
significantly[7]. The authors [8] conducted a simulation experiment in turning hardened 52100 bearing steel using the
analysis of high speed hard turning of AISI 52100 bearing KD050 low CBN content insert. The CBN tool crater wear
steel to study the cutting parameters of speed, feed, cutter rate is formulated and validated over a wide range of cutting
geometry and work piece hardness results cutting force and conditions [20]. The authors [21] presented a predictive model
feed force increase with increasing feed, tool edge radius, based on the artificial neural network (ANN). An experiment
negative rake angle and work piece hardness. The authors [9] was conducted on hard machining of 52100 bearing steel, and
conducted an experiment to study the hard turning by using the numerical results shows that more compressive residual
CBN tool of AISI 52100 bearing steel. The main objectives stress in both axial and circumferential direction of the
focus on tool wear and forces. The relationship between machined surface were obtained if higher values of the feed
cutting parameters are analyzed using RSM by considering rate were chosen. The authors [22] proposed a model and
cutting speed, feed rate and depth of cut and machining output validated through experimentally in finish turning of hardened
variables such as surface roughness, cutting forces were 52100 bearing steel using a low CBN content tool, results
analyzed and modeled. The depth of cut plays a major role suggest that adhesion is the dominant wear mechanism within
that influence cutting forces as compared to the feed rate and the range of conditions that were investigated. An
cutting speed. The authors [10] suggested that the cutting experimental investigation was carried out by using PCBN
speed is the most influencing factor which effects tool life of inserts.58 to machine finish hard turning of AISI 52100
turning of hardened 100Cr6 and PCBN cutting tools. The bearing steel and suggested that the chamfer angle has a great
experiment was conducted using CBN insert of finish turning influence on the cutting forces and tool stresses. All cutting
of hardened 52100 bearing steel. The objective of this study is force components increase with an increase of the chamfer
to present a analytical methodology to find the CBN tool flank angle [23]. An experimental investigation has been conducted
wear rate [11]. An orthogonal hard turning tests were to clarify the effects of tool nose radius and tool wear on
conducted to study the effects of flank tool wear and cutting residual stress distribution in hard turning of bearing steel JIS
parameters (cutting speed and feed rate), on white and dark SUJ2 and suggested that the tool wear increases, the residual
layer formation in hardened AISI 52100 bearing steel, using stress machined surface increases [24].
PCBN inserts [12]. An extensive study has been performed to
investigate the tool-wear mechanisms of CBN cutting tools in Various researchers have developed the surface roughness
finish turning of the X155CrMoV12 (AISI D2) cold work predictive models for the conventional hard turning of bearing
steel, X38CrMoV5 (AISI H11) hot work steel, 35NiCrMo16 materials. From the literature sources, it is found that the
hot work steel and 100Cr6 bearing steel (AISI 52100). A large machining of A22E (BIMETAL BEARING MATERIAL)
variation in tool-wear rate has been observed in machining of metal matrix composite is an important area of research, there
these steels [13]. The main objective of the present study is to is a limited research available in hard turning and facing
investigate the effects of process parameters (cutting speed, operation of bearing materials on surface roughness.
feed rate and depth of cut) on performance characteristics (tool
life, surface roughness and cutting forces) in finish hard BIMETAL Bearings are primarily used in engines of
turning of AISI 52100 bearing steel with CBN tool. The automobiles. The bearings are meant to reduce the friction
optimization was carried out using ANOVA and RSM. The between the moving parts of the engine crankshaft or crank
results show that feed rate and cutting speed strongly influence shaft. The construction of Bimetal metal bearing consists of
the surface roughness and tool life [14]. The authors [15] three layer structure consisting of strong steel backing. The
concluded that the cutting force was 50% larger and feed and back provides bearing rigidity and its press fit under severe
thrust forces were 100% larger when turning AISI 52100 ball conditions of increased temperature and cycling loads, a thin
bearing steel of hardness 63 HRC as compared to turning the inter layer and lining of an aluminium alloy. Bimetal bearings
same material having hardness 32 HRC. The authors [16] are most often are produced as a combination of the following
conducted an experiment to model the relationship between materials:
tool wear and cutting parameters in turning processes. An  Steel with white metals
experiment was conducted to clarify the effects of tool nose  Steel with sintered bronze

150405-2727-IJMME-IJENS © October 2015 IJENS

IJENS
 International Journal of Mechanical & Mechatronics Engineering IJMME-IJENS Vol:15 No:05 55

 Steel with cast bronze

 Steel with aluminium alloys

Fig. 2. Single point cutting tool

Fig. 1. Aluminium alloy bearing (BIMETAL Bearing)

In the present study, an attempt has been made to investigate IV. OPTIMIZATION BY USING GA
the tool wear by considering the process parameters such as A Genetic Algorithm (GA) is a search technique used in
spindle speed, feed rate, depth of cut and end relief angle in computing to find true or approximate solutions to
hard facing operation using Response surface Methodology optimization and search problems. Genetic algorithms are
(RSM) approach. This methodology helps to obtain best categorized as global search heuristics. Genetic algorithms are
possible cutting conditions and tool geometry in dry facing of a particular class of evolutionary algorithms that use
A22E bearing material using M42 HSS tool material. The techniques inspired by evolutionary biology such as
mathematical model is developed using Design Expert 6.0 inheritance, mutation, selection, and crossover. The evolution
package and also been tested by the analysis of variance test usually starts from a population of randomly generated
(ANOVA). individuals and happens in generations. In each generation, the
fitness of every individual in the population is evaluated,
III. INFLUENCE OF END RELIEF ANGLE ON SINGLE multiple individuals are selected from the current population
POINT CUTTING TOOL (based on their fitness), and modified (recombined and
The angle between front surface of the tool & line normal to possibly mutated) to form a new population.
base of the tool is known as End relief angle. End relief angle The most important components in a GA consist of:
is used to minimize physical interference. If the relief angles
 Individual - Any possible solution
are too large, the cutting edge will be weakened and in danger
 Population - Group of all individuals
of breaking. If the cutting relief angle is too small, it causes
 Search Space-All possible solutions to the problem
the wear on the flank of the tool, thereby, significantly
 Chromosome-Blueprint for an individual
reducing the tool life. In general, for hard and tough material,  Trait - Possible aspect (features) of an individual
the relief angle should be 6 to 8 degrees for HSS tools and 5 to  Allele - Possible settings of trait (black, blond, etc.)
7 degrees for carbide tools. For medium steels, mild steels,  Locus - The position of a gene on the chromosome
cast iron, the relief angle should be 8 to 12 degrees for HSS  Genome - Collection of all chromosomes for an individual
tools and 5 to 10 degrees for carbide tools. For ductile
materials such as copper, brass, bronze and aluminium, ferritic V. RESPONSE SURFACE METHODLOGY (RSM)
malleable iron, the relief angle should be 12 to 16 degrees for Response Surface Methodology (RSM) is a collection of
HSS tools and 5 to 14 degrees for carbide tools. The authors mathematical and statistical techniques that are useful for the
modelling and analysis of problems in which a response of
finally concluded that larger relief angle generally tend to
interest is influenced by several variables and the aim is to
produce a better surface finish [25]. optimize this response [26]. The RSM method was well
adapted to develop an analytical model for the complex
problem.

150405-2727-IJMME-IJENS © October 2015 IJENS

IJENS
 International Journal of Mechanical & Mechatronics Engineering IJMME-IJENS Vol:15 No:05 56

Table I
PROCESS FACTORS AND THEIR LEVELS

Variables Unit Coded Variable Level

Lowest Low Centre High Highest
-2 -1 0 +1 +2
Spindle Speed rpm 400 500 600 700 800

Feed Rate mm/rev 0.04 0.06 0.08 0.1 0.12

Depth of Cut mm 1 1.2 1.4 1.6 1.8

End Relief Angle Degree 8 10 12 14 16

Table II
EXPERIMENTAL DESIGN MATRIX AND RESPONSE FACTORS

Run Spindle Speed Feed Rate (B) Depth of Cut End Relief Angle Tool Wear(T)
(A) mm/rev (C) (D) (Observed Value) mm
rpm mm degree
1 500 0.06 1.6 14 0.131
2 500 0.06 1.2 10 0.156
3 600 0.12 1.4 12 0.254
4 500 0.1 1.2 10 0.103
5 700 0.06 1.2 14 0.142
6 600 0.08 1.4 12 0.032
7 700 0.1 1.2 14 0.122
8 600 0.08 1.4 16 0.145
9 600 0.08 1.4 12 0.058
10 500 0.06 1.6 10 0.126
11 700 0.1 1.2 10 0.163
12 600 0.08 1.4 8 0.068
13 700 0.06 1.6 14 0.112
14 700 0.1 1.6 14 0.147
15 500 0.1 1.6 10 0.139
16 600 0.08 1.4 12 0.058
17 500 0.1 1.2 14 0.138
18 400 0.08 1.4 12 0.147
19 700 0.06 1.6 10 0.036
20 500 0.1 1.6 14 0.211
21 700 0.1 1.6 10 0.105
22 600 0.08 1.4 12 0.029
23 600 0.08 1.8 12 0.057
24 600 0.06 1.4 12 0.084
25 700 0.06 1.2 10 0.038
26 800 0.06 1.4 12 0.104
27 600 0.08 1.4 12 0.142
28 600 0.04 1.4 12 0.128
29 500 0.06 1.2 14 0.248
30 600 0.08 1 12 0.157

With this analytical model, an objective function with and response through statistical fitting method. The response,
constraint can be created, and computation time can be saved. tool wear (T) can be expressed as function of process
The RSM seeks for the relationship between design variable

150405-2727-IJMME-IJENS © October 2015 IJENS

IJENS
 International Journal of Mechanical & Mechatronics Engineering IJMME-IJENS Vol:15 No:05 57

parameters. The end relief angle (D), spindle speed (A), feed VI. EXPERIMENTAL SETUP
rate (B) and depth of cut (C) as shown in Eq. (1) The test specimen Bimetal Bearing of size 95 mm diameter
and thickness 2 mm are selected for experimental purpose.
Tool wear (T) = ψ(Diu, Aiu, Biu, Ciu) + eu-----(1) The outer side of bimetal bearing consists of steel and inner
side is made up of aluminum. Bimetal bearing is softest and it
where, consists of 6 - 20% tin, 1% copper, 2- 4% silicon and highly
ψ - response surface, strengthened by nickel and other elements. These type of
eu - residual, bearings used in the passenger cars with low and medium load
u - no of observations in the factorial gasoline engines. The experiments were conducted on special
experiment type of CNC lathe with M42 HSS single point facing tool
iu - represents level of the ith factor in the under dry condition. The machining operations were carried
uth observation. out as per the conditions provided by the design matrix. The
When the mathematical form of ψ is unknown, this function tool wear was measured using tool maker’s microscope on the
can be approximated satisfactorily within the experimental flank surface of the single point turning tool specimen. The
region by polynomials in terms of process parameter variable. machining of Bimetal Bearing experimental set-up is shown in
The selected design matrix, is a three-level, four factor central Fig.2.
composite rotatable factorial design (CCD) consisting of 30
sets of coded conditions. The ranges of all the parameters were
fixed by conducting trial runs. This was performed by varying
one of the parameters while retaining the rest of them as
constant values. The upper limit of a given parameter was
coded as (+2) and the lower limit was coded as (–2).The
intermediate levels of -1, 0, +1 of all the variables have been
calculated by interpolation. Thus, all the 30 experimental runs
to allow the estimation of the linear, quadratic and two way
interactive effects of the process parameters.

Xi = 2(2X − (Xmax + Xmin)) ------------- (2)

(Xmax − Xmin)

Where,
Xi - The required coded value of a variable X
X - Is any value of the variable from X min to
Xmax.
Xmin - Is the lower limit of the variable
Xmax - Is the upper limit of the variable
The intermediate values coded as −1, 0, and 1.
Fig. 2. MACHINING OF BIMETAL BEARING AND TOOL.

150405-2727-IJMME-IJENS © October 2015 IJENS

IJENS
 International Journal of Mechanical & Mechatronics Engineering IJMME-IJENS Vol:15 No:05 58

Table III
ANOVA TABLE FOR THE PREDICTION OF SPINDLE VIBRATION
Analysis of variance table [Partial sum of squares - Type III]
Sum of Mean F p-value
Source Squares df Square Value Prob > F
Model 0.072 14 5.173E-003 3.11 0.0184 significant
A-Spindle Speed 9.322E-003 1 9.322E-003 5.60 0.0318
B-Feed Rate 6.370E-003 1 6.370E-003 3.83 0.0693
C-Depth of cut 3.825E-003 1 3.825E-003 2.30 0.1502
D-End Relief angle 0.012 1 0.012 7.28 0.0165
AB 4.865E-003 1 4.865E-003 2.92 0.1079
AC 4.556E-005 1 4.556E-005 0.027 0.8708
AD 3.306E-005 1 3.306E-005 0.020 0.8898
BC 4.064E-003 1 4.064E-003 2.44 0.1389
BD 1.785E-003 1 1.785E-003 1.07 0.3167
CD 1.563E-006 1 1.563E-006 9.391E-004 0.9760
A^2 5.808E-003 1 5.808E-003 3.49 0.0814
B^2 0.026 1 0.026 15.77 0.0012
C^2 2.703E-003 1 2.703E-003 1.62 0.2218
D^2 2.635E-003 1 2.635E-003 1.58 0.2274
Residual 0.025 15 1.664E-003
Lack of Fit 0.016 10 1.621E-003 0.93 0.5724 not significant
Pure Error 8.745E-003 5 1.749E-003
Cor Total 0.097 29

VII. DEVELOPMENT OF M ATHEMATICAL MODEL Tool Wear (T) = +2.43364 -2.43646E-003* A -19.19687* B -
The analysis is carried out with the experimental data using 1.03552* C -0.023240* D +8.71875E-003 * A * B -8.43750E-
Design Expert V 9.0.4 software of state ease. The model 005 * A * C -7.18750E-006 * A * D +3.98438 * B * C -
is checked for its adequacy using ANOVA (analysis of
0.26406 * B * D +7.81250E-004 * C * D +1.45521E-006 * A2
variance). Table VIII shows ANOVA table for the prediction
of tool wear (T). It is observed from the Table VIII that the +77.31771 * B2 +0.24818 * C2+2.45052E-003 * D2.
model is significant and the lack of fit is not significant which
infers the significance of the model. Values of Prob> F less Where,
than 0.05 indicate the model terms as in significant and the A = Spindle speed in rpm
values greater than 0.10 indicate the model terms as not B = Feed rate in mm/rev
significant. The Model F-value of 3.11 implies that the model C = Depth of cut in mm
is significant. There is only a 0.25% chance that an F- D = End Relief angle in degree
value this large could occur due to noise. The "Lack of Fit F-
value" of 0.93 implies the Lack of Fit is not significant relative VIII. RESULTS AND D ISCUSSION
to the pure error. There is a 12.05% chance that a "Lack of Fit A mathematical model was developed by DoE based
F-value" this large could occur due to noise. Non-significant Response Surface Methodology (RSM) to predict the tool
lack of fit is good. The Fig.5 shows the Predicted Vs Actual wear by relating it with process parameters such as spindle
model. The regression equation obtains from the Design speed, feed rate, depth of cut and end relief angle. The residual
Expert software in terms of actual factors are given: plot is shown in Fig.3.

150405-2727-IJMME-IJENS © October 2015 IJENS

IJENS
 International Journal of Mechanical & Mechatronics Engineering IJMME-IJENS Vol:15 No:05 59

from the figure that at lower spindle speed and higher depth of
The interaction effect of these process parameters on tool wear cut has a significant influence. The tool wear increases when
were discussed below. Fig. 4 shows the interaction effect of the spindle speed is higher and low depth of cut. Fig. 6 shows
feed rate and spindle speed on tool wear. As the increase in the interaction effect of spindle speed and relief angle on tool
spindle speed from 600 rpm to 800 rpm the tool wear value is wear. From the Figure
low, whereas the feed rate has the
inverse relationship on the tool wear. Therefore, higher
spindle speed and feed rate between 0.06mm to 1mm has to be
chosen as low tool wear. Fig.5 shows the interaction effect of
spindle speed and depth of cut on tool wear. It is evidenced

Fig. 3. Residuals plot

150405-2727-IJMME-IJENS © October 2015 IJENS

IJENS
International Journal of Mechanical & Mechatronics Engineering IJMME-IJENS Vol:15 No:05 60

Fig. 4. Surface interaction plot of Spindle Speed and Feed rate

Fig. 5. Surface interaction plot of Spindle Speed and Depth of cut

150405-2727-IJMME-IJENS © October 2015 IJENS

IJENS
International Journal of Mechanical & Mechatronics Engineering IJMME-IJENS Vol:15 No:05 61

Fig. 6. Surface interaction plot of Spindle Speed and End Relief Angle

Fig. 7. Surface interaction plot of Feed rate and Depth of cut

150405-2727-IJMME-IJENS © October 2015 IJENS

IJENS
 International Journal of Mechanical & Mechatronics Engineering IJMME-IJENS Vol:15 No:05 62

Fig. 8. Surface interaction plot of Feed rate and Relief angle

Fig. 9. Surface interaction plot of depth of cut and relief angle

is noted that the both spindle speed and relief angle are tool wear. The tool wear is low when the feed rate is low and
influencing the tool. The tool wear is significantly higher depth of cut is higher. From the Fig.8 the higher feed rate and
between 12° to 16° and also the similar result is observed higher relief angle, increases the tool wear whereas lower feed
between 400 rpm to 500 rpm. From the result, the industry rate and lower relief angle decrease the tool wear. It is
preferring to fix the relief angle between 8° to 10°. Fig. 7 interesting to observe that if the relief angle between 8° to 10°
shows the interaction effect of feed rate and depth of cut on

150405-2727-IJMME-IJENS © October 2015 IJENS

IJENS
 International Journal of Mechanical & Mechatronics Engineering IJMME-IJENS Vol:15 No:05 63

and depth of cut between 1.4 mm to 1.6 mm will produce Table IV

GA PARAMETERS
lower tool wear as shown in Fig.9.
Parameters Parameters
VIII. EVALUATION OF GA RESULTS setting setting
In this present study, the optimization wear was carried out to Population size 100
minimize the cutting parameter based on Genetic Algorithm Scaling function Rank
methodology. The minimization of tool wear by using GA can Scaling function Rank
be expressed by the equation Function Stochastic uniform
Mutation function Gaussian
Minimize: T (A, B, C, D) Mutation rate 0.1
Crossover function Scattered
Within ranges of cutting parameters, Crossover rate 1.0
Generations 1000
400 rpm ≤ A ≤ 800 rpm

0.04 mm/rev ≤ B ≤ 0.12 mm/rev IX. VALIDATION OF THE MODEL

Table II shows that a regression model developed using CCD
1 mm ≤ C ≤ 1.8 mm by the RSM of DoE. Fig.10 shows the comparison of
predicted vs experimental value of Tool wear from RSM of
DoE. Table 5 shows the comparison of predicted vs
8 ° ≤ D ≤ 16°
experimental value of Tool wear. The GA-predicted optimum
conditions were further validated with physical measurements.
The number of the initial population size, the type of selection
The percentage of error is found to be within ±2 % which
function, the crossover rate, mutation rate and the generations
shows the validity of the model. The experimental results of
to be considered to get the best optimal results. The Table 4
tool wear with the optimum cutting parameters (as predicted
shows the GA operators. By solving the
by GA) shows the good agreement. Figure 10 shows the
optimization problem using MAT LAB 7.0, GA predicts the
performance of fitness value with generation and the best
optimum tool wear as 0.0180 mm for the machining of
individual performances of variables in coded form.
Bimetal bearing in the selected cutting condition range. The
GA-predicted tool wear value (best fitness function) is
expected to be lower than the minimum tool wear value of the
experimental and regression models.

Fig. 10. PREDICTED VS EXPERIMENTAL VALUE

150405-2727-IJMME-IJENS © October 2015 IJENS

IJENS
 International Journal of Mechanical & Mechatronics Engineering IJMME-IJENS Vol:15 No:05 64

Fig. 10. shows the performance of fitness value with generation and the best individual performances of variables in coded form.

Table V
OPTIMIZED PROCESS PARAMETER PREDICTED BY GA

Trial Spindle Feed Depth of End relief Confirmatory test for %

Speed rate cut angle Surface roughness error
(A) (B) (C) (D) Predicted Experimental Value
GA model
rpm mm/rev mm Degree µm µm
1 700 0.08 1.6 12 0.114 0.113 0.87
2 500 0.06 1.2 8 0.084 0.081 0.35
3 600 0.10 1.4 10 0.138 0.136 1.44

X. CONCLUSION
From the regression model developed using Central REFERENCE
Composite Design (CCD) by the Response Surface [1] Taylor, F. W, On the Art of Cutting Metals, The American Society
of Mechanical Engineers, (1906) New York, USA
Methodology (RSM) of Design of Experiments (DoE), the
[2] Devendiran. SK, Manivannan. S, Rajeswari, C, Joshua Michael
experimental investigation has been done for facing operation Amarnath and Apoorv Prasad, Bearing fault diagnosis using
considering, the factors such as Spindle Speed, Feed Rate, CWT, BGA and Artificial Bee Colony Algorithm, International
Depth of cut, and End relief angle to predict tool wear of Journal of Mechanical & Mechatronics Engineering, 15 (03),
2015,1-16.
Bimetal Bearing, the following conclusions are drawn:
[3] Yong Huang and Steven Y.Liang, Modelling of CBN tool crater
wear in finish hard turning, The International Journal of
The Spindle Speed and Feed rate are the most important Advanced Manufacturing Technology, 24, ( 9-10), 2004, 632-639.
parameters to be considered for tool wear compared to the [4] Yong Huang and Steven Y.Liang, Effect of Cutting Conditions on
Tool Performance in CBN Hard Turning, Journal of
other factors such as depth of cut and end relief angle. The
Manufacturing Process, 7(1), 2005, 10-16.
better tool life was obtained at the end relief angle (8°-10°) [5] Yong Huanga and Ty G. Dawson, Tool crater wear depth modeling
and spindle speed (600 rpm to 800 rpm). The tool wear is in CBN hard turning, Wear, 258(9), 2005, 1455–1461.
minimum value at the region of feed rate (0.06mm - 0.1 [6] Liu. X.L, Wen. D.H, Li. Z.J, Xiao. L. and Yan. F.G. Cutting
temperature and tool wear of hard turning hardened bearing steel,
mm).The tool wear is minimum at the region of depth of cut
Journal of Materials Processing Technology, 129(1–3), 2002, 200–
(1.4 mm – 1.6 mm). Further GA predicts the optimum tool 206.
wear as 0.0180 mm for the machining of Bimetal bearing in [7] Meng Liu, Jun-ichiro Takagi and Akira Tsukuda, Effect of tool
the selected cutting condition range. The confirmatory result nose radius and tool wear on residual stress distribution in hard
turning of bearing steel, Journal of Materials Processing
shows good arguments for experimental vs predicted value.
Technology 150, 2004, 234–241.
[8] Qian. L , and M.R. Hossan, Effect on cutting force in turning
hardened tool steels with cubic boron nitride inserts, Journal of
Material Processing Technology, 191(1-3),2007,274-278.

150405-2727-IJMME-IJENS © October 2015 IJENS

IJENS
 International Journal of Mechanical & Mechatronics Engineering IJMME-IJENS Vol:15 No:05 65

[9] Boucha. K, Athmane. M. Y, Mabrouki. T and Rigal. F. J.

Statistical analysis of surface roughness and cutting forces using
response surface methodology in hard turning of AISI 52100
bearing steel with CBN tool, Int. Journal of Refractory Metals &
Hard Materials, 28, 2010, 349–361.
[10] Banga, G.C and Abrao, A.M. Turning of hardened 100Cr6 bearing
steel with ceramic and PCBN cutting tools. J. Mater. Process.
Technol. 144, 2003.143–237–241.
[11] Yong Huang and Steven Y. Liang, Modeling of CBN Tool Flank
Wear Progression in Finish Hard Turning, Transactions of the
ASME, 126, 2004, 98- 106
[12] Attanasio, C. Cappellini, G. Rotella, R. M’Saoubi and Umbrello.
D. Tool wear effects on white and dark layer formation in hard
turning of AISI 52100 steel, Wear, 286–287, 2012, 98–107.
[13] Gérard Poulachon, Bandyopadhyay. B.P, Jawahir. I.S. Sébastien
Pheulpin and Emmanuel Seguin , Wear behavior of CBN tools
while turning various hardened steels, Wear 256, 2004, 302–310.
[14] Samir Khamel1, Nouredine Ouelaa and Khaider Bouacha,
Analysis and prediction of tool wear, surface roughness and cutting
forces in hard turning with CBN tool, Journal of Mechanical
Science and Technology 26 (11), 2012, 3605-3616.
[15] Konig. W. Komanduri. R. Tonshoff. H. K. and Ackershott. G.
Machining of hard materials, Annals of the CIRP, 33 (2), 1984,
417-428.
[16] Liu. XL, Wen. DH, Li. ZJ, Xiao. L and Yan. FG. Cutting
temperature and tool wear of hard turning hardened bearing steel.
Journal of Material Processing Technology, 129(1–3), 2002, 200–
206.
[17] Meng Liu, Jun-ichiro Takagi and Akira Tsukuda, Effect of tool
nose radius and tool wear on residual stress distribution in hard
turning of bearing steel, Journal of Materials Processing
Technology 150, 2004,234–241.
[18] El-Helieby. S.O.A and Rowe. G.W. A quantitative comparison
between residual stresses and fatigue properties of surface-ground
bearing steel (EN31), Wear, 58, 1980, 155–172.
[19] Lin Z-C and Chen D-Y. A study of cutting with a CBN tool.
Journal of Materials Processing Technology 49, 1995, 149–164.
[20] Yong Huang and· Steven Y. Liang, Modelling of CBN tool crater
wear in finish hard turning, International Journal of Advanced
Manufacturing Technology, 24, 2004,632–639.
[21] Umbrello. D, Ambrogioa. G, Filice, L.and Shivpuri, R. An ANN
approach for predicting subsurface residual stresses and the desired
cutting conditions during hard turning. Journal of Materials
Processing Technology, 189, 2007, 143–152.
[22] Yong Huang and Dawson. Ty G. Tool crater wear depth modeling
in CBN hard turning, Wear. 258, 2005, 1455–1461.
[23] Kurt, A. and Ulvi, S., “The effect of chamfer angle of
polycrystalline cubic boron nitride cutting tool on the cutting
forces and the tool stresses in finishing hard turning of AISI 52100
steel,” Materials and Design, 26(4), 2005,351-356.
[24] Meng Liu, Jun-ichiro Takagi and Akira Tsukuda, Effect of tool
nose radius and tool wear on residual stress distribution in hard
turning of bearing steel, Journal of Materials Processing
Technology, 150 (3), 2004, 234–241.
[25] Edward G. Hoffman,, Christopher J., McCauley.,and Muhammed
Iqbal Hussain., Shop Reference for Students and Apprentices, New
York : Industrial Press, 2nd ed. 2000.
[26] Logesh. K, Muralinath. P, Poyyamozhi. N, Dilip Raja.N and
Yassar Dawood .A. Optimization of Cutting Factors Which
influence Temperature on Tip of Tool in response to Surface
Finish for an Aluminium Alloy AL6063 during Turning Process on
CNC Lathe, International Journal of Mechanical & Mechatronics
Engineering IJMME-IJENS, 14 (06),2014,72-75.

150405-2727-IJMME-IJENS © October 2015 IJENS

IJENS