- Email: [email protected]
BCCS Approach for the Parametric Optimization in Machining of Nimonic-263 alloy using RSM
Available online at www.sciencedirect.com
ScienceDirect Materials Today: Proceedings 5 (2018) 14416–14422
www.materialstoday.com/proceedings
ICAFM_2017
BCCS Approach for the Parametric Optimization in Machining of Nimonic-263 alloy using RSM T. Sathish* a
Research Scholar, Department of Mechanical Engineering, St. Perter’s University, Tamil Nadu, India
Abstract Machining of nickel-based alloys is one of the testing undertakings in the current research knowledge. Wire Electrical Discharge Machine (WEDM) is a propelled machine apparatus, broadly used to cut hard metals like nickel, titanium and other super composite alloys. Determination of WEDM process parameters to yield the coveted level of execution measures like metal removal rate and surface roughness is significant from quality and financial perspectives. In the present work an endeavor has been made to examine the impact of WEDM process parameters, for example, pulse on time, pulse off time, peak current and servo voltage in machining of Nimonic-263 alloys. A focal composite face focused outline of Response Surface Methodology (RSM) has been utilized for exploratory arrangement. The noteworthiness of process parameters are assessed by ANOVA investigation. Scientific forecast models are produced for metal removal rate and surface roughness by response surface methodology. In this paper a novel Hybrid Bee Colony Cuckoo Search (BCCS) optimization algorithm is incorporated to boost the process performance. The outcome of RSM and BCCS is compared and it shows the execution of the BCCS is superior to that of RSM. The machining information gathered for the Nimonic-263 amalgam will be valuable for the organization respectively. © 2017 Elsevier Ltd. All rights reserved. Selection and/or Peer-review under responsibility of ICAFM’17. Keywords:bee colony, cuckoo search, Nimonic-263 alloy, Response Surface Methodology (RSM), Analysis of Variance (ANOVA), Wire Electrical Discharge Machine (WEDM)
1. Introduction Utilizations of nickel based alloys are expanding step by step because of their unrivaled properties, for example, hardness at hoisted temperatures and high resistance to corrosion. Machining of nickel based compounds is one of the testing errands to the makers in the current past. Nimonic-263 is a nickel-chromium-cobalt-molybdenum * Corresponding author. Tel : +91 9952047764 E-mail address:[email protected] 2214-7853© 2017 Elsevier Ltd. All rights reserved. Selection and/or Peer-review under responsibility of ICAFM’17.
T. Sathish/Materials Today: Proceedings 5 (2018) 14416–14422
14417
composite uniquely implied for use in high temperature and high-quality applications [6, 7]. This material is for the most part utilized as a part of gas turbine hot segment materials. In the current past a few specialists were contributed their endeavors towards building up the numerical models for the execution measures of WEDM, and attempted to improve the same with various advancement systems [810]. Spedding and Wang [1], completed examinations in light of RSM plan. Scientific models are produced utilizing RSM and furthermore connected a multi-layered encourage forward neural system is utilized to show the WEDM execution. Cutting rate, surface finish and waviness were chosen as the execution measures, and the created model is utilized to process execution expectation and parameter advancement. The models created by RSM and additionally ANN are looked at and discovered both are giving precise outcomes. A full factorial trial configuration was utilized by Nihat [2], to concentrate the variety of cutting exhibitions with WEDM prepare parameters. Relapse Analysis was utilized to create scientific models for cutting velocity and surface finish. Nourishment and Bhattacharyya (2006) completed a test examination to decide the parameter setting, Taguchi strategy was utilized to streamline WEDM parameters. Mahapatra and Patnaik [3], endeavored to decide the critical machining parameters for execution measures like metal removal rate, surface finish and kerf. Utilizing Taguchi's parametric plan noteworthy machining parameters influencing the execution measures are recognized as release current, beat length, beat recurrence, wire speed, wire strain and dielectric liquid stream rate. Numerical models are created by methods for nonlinear relapse investigation for MRR, SF, and Kerf. At last Genetic calculation is utilized to enhance the WEDM procedure with Multipledestinations. Muthu et al., [5] exhibited the streamlining of WEDM process parameters of Incoloy800 super combination with different execution qualities, for example Grey-Taguchi technique. The variety of yield reactions with process parameters were scientifically demonstrated utilizing non-direct relapse investigation technique. Ideal levels of process parameters were recognized utilizing GRA and the moderately huge parameters were resolved utilizing ANOVA. Shandilya et al. [4], endeavored to examine WEDM handle execution as far as cutting width (kerf) utilizing RSM. The analysis of variance was completed to concentrate the impact of process parameters on process execution. Numerical models have likewise been produced for reaction parameter and properties of the machined surface have been analyzed by utilizing SEM. Response Surface Methodology was utilized by Rao and Pawar (2009) to create numerical models for cutting pace and surface roughness. An artificial honey bee state (ABC) algorithm was then connected to locate the ideal blend of process parameters with a target of accomplishing most extreme machining speed for a coveted estimation of surface finish. 2. BCCS algorithm for the parametric optimization in machining of Nimonic-263 alloy The Nimonic - 263 compound coordinating is an unpredictable errand, thus delicate processing method is infiltrated to foresee the parameter for the coordinating. In the proposed procedure a hybrid system by consolidating bee colony and cuckoo search algorithms is utilized for the better optimisation parameter. The steps incorporated into the proposed method is as per the following; Step 1: Initialization The initial candidate solution or food source or the initial value to optimize are initialized based on the proposed objective. Step 2: Employee bee phase The employee bee phase evaluate and compare different solution by incorporating a fitness function for evaluating the solution of each candidate solution. This function decode the string and validate the objective function for each candidate solution. Step 3: Onlooker bee phase The onlooker bee phase choose elite food source to optimal and enhance the source of food. This phase reach to appropriate location at low loss of power and high voltage profile, which in turn optimize the velocity of population by utilizing the following equation (1) below.
Vi , j = xi , j + Φi , j ( xi , j − xk , j )
(1)
14418
T. Sathish/ Materials Today: Proceedings 5 (2018) 14416–14422
Where, k is the neighborhood solution of i and Φ is a random number in the range of [-1, 1], Vi , j is the neighborhood solution of M i , j . Step 4: Selection For the updation of process, the optimal fitness is choosen in the selection phase. In this phase the selection is undergone by the probability function for the selection of fitness and probability for selection is evaluated from the following equation (2) below.
probabilit y =
Φ n
(2)
Φ
i =1
Step 5: CS based Scout bee phase At the point when the onlooker bee step has not refined obviously better alternatives, leave from the specific choices and make the arbitrary number of scout bee order while utilizing the Cuckoo Search Optimization. Initialization: The initialization phase starts with the setting of cuckoo search parameter. These parameters consist of the number of nests ‘ n ’, the step size parameter ‘ α ’, iteration probability ‘ Pa ’ and the maximum number of iteration as termination criteria. Generate Initial Nests or Eggs of Host Birds: The initial locations of the nests are specified by the set of random values assigned to each variable as: Di(,0j) = Round x j ,min + rand x j ,max − x j ,min (3)
(
(
))
th
Where, ‘ Di(,0j) ’ is the initial value of the j variable for the i th nest; ‘ x j ,min ’ and ‘ x j ,max ’ are the minimum and th the maximum allowable values for the j variable; ‘ rand ’ is a random number in the range of [0, 1]. Due to the discrete nature of the problem round function is accomplished. Step size evaluation:In this step, the step sizes for individual parent weights obtained in step 3 are evaluated. The following expressions are used to find out the step size.
(
)
t Sz = α S Dt − Dbest ⋅r
(4) t
Where, ‘ S z ’ is the Step size, ‘ α ’is Step size parameter ( α = 0.01 ), ‘ D t ’ is the current parent weight, ‘ Dbest ’is the best solution so far, ‘r’ is a random number from a standard normal distribution [0,1] and ‘S’ is step. The step S is found out by using the Mantegna’s algorithm, shown in the below equation. u S= 1 (5) v β Where, ‘ β ’ is a parameter arising in the interval [1, 2] which we choose in our system as 1.5 (i.e. β = 3 2 ) and u and v are normal distributions, which are estimated as follows.
u ~ N(0,σ u2 ) , v ~ N(0,σ v2 ) πβ Γ(1 + β ) Sin 2 σu = Γ (1 + β ) β ⋅ 2( β −1) 2 2
(6) 1
β
, σv = 1
(7)
Generation of New Solution: In this generation step, the new upgraded solution or weights are produced for comparing initial weights in view of the cuckoo search algorithm (Levy Flight). The new weights are created by utilizing the step and the measured values acquired in the step 3. The new weights are produced by utilizing the expression given below.
D(t+1) = Dt + Sz
(8)
T. Sathish/Materials Today: Proceedings 5 (2018) 14416–14422
14419
Where, ‘ D ( t +1) ’ is the new weight, ‘Sz’ is step size and ‘ D t ’ is the current parent weight. From this step another set of optimized weights are obtained. Ending of CS process: In this phase, the termination criteria is tested and if it is converges, the iteration stops and if not the iteration continuous to get optimal result. Step 6: Termination Criteria The optimization process is iterated to maximum level and the iteration number converges to maximum, then it stops and keep the current best solution. 3. Results and Discussion WEDM of Electronica (India) made has been utilized to machine Nimonic-263 material with a pulse on time, pulse off time, peak current and servo voltage as info parameters. Gaps of 10 mm width are delivered on 120mm×110mm×18.5 mm workpiece. The endpoint of the present review is to upgrade the machining parameters for better execution. In this review, the execution measures are material expulsion rate and surface harshness. The levels of input parameters such as pulse on time, pulse off time, peak current and servo voltage are settled in view of the trial tries and are given in Table 1. Reaction surface strategy Central composite face provided food outline with 1 focus focuses has been utilized for trial arrange. The deliberate and computed reactions are given in Table 2 in the wake of leading the analyses according to the test arrange. ANOVA has been connected to know the critical parameters and their commitment. It has additionally been connected to demonstrate the reactions material expulsion rate and surface finish. These models are additionally used to anticipate and advancement. Proposed BCCS algorithm has been utilized to enhance reactions and these outcomes are contrasted with RSM outcomes. Table 1: Levels of input parameters Parameter
level 1
level 2
level 3
Pulse on time (Ton) μs
105
115
125
Pulse off time (Toff) μs
50
55
60
Peak current (Ip) A
10
11
12
Servo voltage (Sv) V
40
50
60
Material removal rate was calculated using the equation (9) and the surface roughness of the machined surface was measured by MarSurf M-400. The measurement of diameter is carried out by Coordinate Measuring Machine (CMM). πT d 2 − D 2 (9) MRR = mm 3 / min 4T Where, D is the diameter of the hole in mm, d is the diameter of the blank in mm, t is the thickness of work piece in mm and T be the Time taken for machining in min.
(
)
14420
T. Sathish/ Materials Today: Proceedings 5 (2018) 14416–14422
Table 2. Experimental plan and Responses Run order
Ton (μs)
Toff (μs)
Ip (A)
Sv (V)
MRR (mm3 / min)
SR (μm)
1
125
60
12
60
3.12412
2.01
2
125
60
10
60
0.359606
0.605
3
115
55
11
40
0.452898
0.525
4
115
55
11
50
0.380731
0.387
5
105
60
12
60
0.497021
0.537
6
125
55
11
50
0.327263
0.578
7
125
50
10
40
0.380919
0.484
8
115
55
12
50
2.447281
1.021
9
115
50
11
50
0.280316
0.501
10
105
50
10
40
0.446554
1.364
11
115
55
10
50
0.300545
0.514
12
105
60
10
60
0.426622
0.694
13
125
50
10
60
0.458812
0.46
14
115
55
11
50
0.310576
0.539
15
125
50
12
40
3.588263
2.027
16
105
60
10
40
0.529402
1.323
17
125
60
12
40
3.352623
2.089
18
105
55
11
50
0.396612
0.892
19
115
55
11
60
0.575236
0.76
20
115
60
11
50
0.514857
0.764
21
125
60
10
40
0.500181
1.081
22
105
50
12
40
1.768268
0.891
23
125
50
12
60
3.233984
1.85
24
105
60
12
40
0.948816
1.18
25
105
50
10
60
0.548839
0.863
26
105
50
12
60
0.771276
0.603
Analysis of variance has been applied on the experimental results for both metal removal rate and surface roughness are given Table 3 and Table 4 respectively.
T. Sathish/Materials Today: Proceedings 5 (2018) 14416–14422
14421
Table 3. Analysis of Variance in Terms of Metal Removal Rate Source
SS
DOF
MS
F value
P-value
Percentage Contribution
Model
29.3208
9
3.257867
44.563324
< 0.0001
96.16369159
A- Ton
4.492364
1
4.492364
61.449614
< 0.0001
14.73364663
B- Toff
0.08323
1
0.08323
1.1384719
0.3018
0.26237672
C- Ip
13.8341
1
13.8341
189.23223
< 0.0001
45.37182225
D- Sv
0.216133
1
0.216133
2.956416
0.1048
0.708853345
AC
5.718684
1
5.718684
78.224046
< 0.0001
18.75561937
BC
0.126007
1
0.126007
1.7236097
0.2077
0.413266292
CD
0.24216
1
0.24216
3.31243
0.0875
0.794214331
A^2
0.012463
1
0.012463
0.1704738
0.6852
0.040875013
C^2
2.960149
1
2.960149
40.490926 < 0.0001
9.708427311
0.073106
Residual
1.169703
16
Cor. Total
30.49051
25
3.836285454 100
From the results, it is contemplated that Ton, Ip, association of Ton and Ip, Ip2 are critical model terms. Higher the beat on time, higher will be the vitality connected there by producing more measure of thermal energy among this period and it prompts higher metal removal rate. Top current is the measure of energy utilized as a part of release machining. Higher the pinnacle present, higher will be the vitality connected at machining and in this manner expanding the removal rate. From the ANOVA result, the R-Square, balanced R-square and forecasted Rsquare values were observed to be 96.2%, 94%and 90% respectively for the model. Confirmation tests have been conducted to check the effectiveness of BCCS for both MRR and SR, and the results were given in Table 4. Table 4. Confirmation Test results Response MRR (mm3/min) (Ton:125, Toff:50, Ip:12, Sv:40) SR (μm) (Ton:116, Toff:50, Ip:10, Sv:60)
Predicted value from BCCS
Experimental value
Deviation in percentage
3.6713
3.614
2
0.2618
0.282
7
4. Conclusion Utilization of response surface methodology is taken as a part of the present review to show WEDM execution measures of material removal rate and surface finish. The parameters influencing machining process are Pulse on time, pulse off time, peak current and servo voltage. The significance of process parameters have been recognized by applying analysis of variance examination for both metal removal rate and surface roughness. For metal removal rate it was found from the analysis of variance comes about that, Pulse on time, peak current and association impact of Pulse on time and peak current are more affecting than other model terms. In the present research an endeavor has been made to apply proposed optimization algorithm to improve the reactions. The ideal reaction values from response surface methodology and proposed BCCS algorithm are contrasted. It is found that, the aftereffects of BCCS are superior to that of response surface methodology. Because of extensive variety of advantages for
14422
T. Sathish/ Materials Today: Proceedings 5 (2018) 14416–14422
Nimonic-263, the machining information created without precedent for this work utilizing WEDM will be valuable to the manufacturing organizations. References [1] Spedding, T.A, Wang, Z.Q., 1997. Parametric optimization and surface characterization of wire electrical discharge machining process, Journal of Precision Engineering, 20, 5-15. [2] Nihat Tosun, 2003. The Effect of the Cutting Parameters on Performance of WEDM, KSME International Journal, 17, 816-824. [3] Mahapatra, S., Amar Patnaik, S., 2007. Optimization of wire electrical discharge machining (WEDM) process parameters using Taguchi method, International Journal of Advanced Manufacturing Technology, 34, 911–925. [4] Pragya Shandilya,.Jain, P.K., Jain, N.K., 2012. Parametric optimization during wire electrical discharge machining using response surface methodology, Procedia Engineering, 38, 2371 – 2377. [5] Muthu Kumar, V., Suresh Babu, A., Venkatasamy, R., Raajenthiren, M., 2010. Optimization of the WEDM Parameters on Machining Incoloy800 Super alloy with Multiple Quality Characteristics, International Journal of Engineering Science and Technology, 2, 162-183. [6] Abido, M.A., 2002. Optimal power flow using particle swarm optimization, Electrical Power and Energy Systems, 24, 563-571. [7] Bharathi Raja, S., Baskar, N., 2011. Particle swarm optimization technique for determining optimal machining parameters of different work piece materials in turning operation, International Journal of Advanced Manufacturing Technology, 54, 445–463. [8] Manna, A., Bhattacharyya, B., 2006, Taguchi and Gauss elimination method: A dual response approach for parametric optimization of CNC wire cut EDM of PRAlSiCMMC, International Journal of Advanced Manufacturing Technology, 28, 67–75. [9] Rao, R.V., Pawar, P.J., 2009. Modelling and optimization of process parameters of wire electrical discharging machining, Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture 223, 1431-1440. [10] Xin-She Yang, Nature-Inspired Optimization Algorithms, 1st Edition, 2014, page-99 Elsevier, London.
ScienceDirect Materials Today: Proceedings 5 (2018) 14416–14422
www.materialstoday.com/proceedings
ICAFM_2017
BCCS Approach for the Parametric Optimization in Machining of Nimonic-263 alloy using RSM T. Sathish* a
Research Scholar, Department of Mechanical Engineering, St. Perter’s University, Tamil Nadu, India
Abstract Machining of nickel-based alloys is one of the testing undertakings in the current research knowledge. Wire Electrical Discharge Machine (WEDM) is a propelled machine apparatus, broadly used to cut hard metals like nickel, titanium and other super composite alloys. Determination of WEDM process parameters to yield the coveted level of execution measures like metal removal rate and surface roughness is significant from quality and financial perspectives. In the present work an endeavor has been made to examine the impact of WEDM process parameters, for example, pulse on time, pulse off time, peak current and servo voltage in machining of Nimonic-263 alloys. A focal composite face focused outline of Response Surface Methodology (RSM) has been utilized for exploratory arrangement. The noteworthiness of process parameters are assessed by ANOVA investigation. Scientific forecast models are produced for metal removal rate and surface roughness by response surface methodology. In this paper a novel Hybrid Bee Colony Cuckoo Search (BCCS) optimization algorithm is incorporated to boost the process performance. The outcome of RSM and BCCS is compared and it shows the execution of the BCCS is superior to that of RSM. The machining information gathered for the Nimonic-263 amalgam will be valuable for the organization respectively. © 2017 Elsevier Ltd. All rights reserved. Selection and/or Peer-review under responsibility of ICAFM’17. Keywords:bee colony, cuckoo search, Nimonic-263 alloy, Response Surface Methodology (RSM), Analysis of Variance (ANOVA), Wire Electrical Discharge Machine (WEDM)
1. Introduction Utilizations of nickel based alloys are expanding step by step because of their unrivaled properties, for example, hardness at hoisted temperatures and high resistance to corrosion. Machining of nickel based compounds is one of the testing errands to the makers in the current past. Nimonic-263 is a nickel-chromium-cobalt-molybdenum * Corresponding author. Tel : +91 9952047764 E-mail address:[email protected] 2214-7853© 2017 Elsevier Ltd. All rights reserved. Selection and/or Peer-review under responsibility of ICAFM’17.
T. Sathish/Materials Today: Proceedings 5 (2018) 14416–14422
14417
composite uniquely implied for use in high temperature and high-quality applications [6, 7]. This material is for the most part utilized as a part of gas turbine hot segment materials. In the current past a few specialists were contributed their endeavors towards building up the numerical models for the execution measures of WEDM, and attempted to improve the same with various advancement systems [810]. Spedding and Wang [1], completed examinations in light of RSM plan. Scientific models are produced utilizing RSM and furthermore connected a multi-layered encourage forward neural system is utilized to show the WEDM execution. Cutting rate, surface finish and waviness were chosen as the execution measures, and the created model is utilized to process execution expectation and parameter advancement. The models created by RSM and additionally ANN are looked at and discovered both are giving precise outcomes. A full factorial trial configuration was utilized by Nihat [2], to concentrate the variety of cutting exhibitions with WEDM prepare parameters. Relapse Analysis was utilized to create scientific models for cutting velocity and surface finish. Nourishment and Bhattacharyya (2006) completed a test examination to decide the parameter setting, Taguchi strategy was utilized to streamline WEDM parameters. Mahapatra and Patnaik [3], endeavored to decide the critical machining parameters for execution measures like metal removal rate, surface finish and kerf. Utilizing Taguchi's parametric plan noteworthy machining parameters influencing the execution measures are recognized as release current, beat length, beat recurrence, wire speed, wire strain and dielectric liquid stream rate. Numerical models are created by methods for nonlinear relapse investigation for MRR, SF, and Kerf. At last Genetic calculation is utilized to enhance the WEDM procedure with Multipledestinations. Muthu et al., [5] exhibited the streamlining of WEDM process parameters of Incoloy800 super combination with different execution qualities, for example Grey-Taguchi technique. The variety of yield reactions with process parameters were scientifically demonstrated utilizing non-direct relapse investigation technique. Ideal levels of process parameters were recognized utilizing GRA and the moderately huge parameters were resolved utilizing ANOVA. Shandilya et al. [4], endeavored to examine WEDM handle execution as far as cutting width (kerf) utilizing RSM. The analysis of variance was completed to concentrate the impact of process parameters on process execution. Numerical models have likewise been produced for reaction parameter and properties of the machined surface have been analyzed by utilizing SEM. Response Surface Methodology was utilized by Rao and Pawar (2009) to create numerical models for cutting pace and surface roughness. An artificial honey bee state (ABC) algorithm was then connected to locate the ideal blend of process parameters with a target of accomplishing most extreme machining speed for a coveted estimation of surface finish. 2. BCCS algorithm for the parametric optimization in machining of Nimonic-263 alloy The Nimonic - 263 compound coordinating is an unpredictable errand, thus delicate processing method is infiltrated to foresee the parameter for the coordinating. In the proposed procedure a hybrid system by consolidating bee colony and cuckoo search algorithms is utilized for the better optimisation parameter. The steps incorporated into the proposed method is as per the following; Step 1: Initialization The initial candidate solution or food source or the initial value to optimize are initialized based on the proposed objective. Step 2: Employee bee phase The employee bee phase evaluate and compare different solution by incorporating a fitness function for evaluating the solution of each candidate solution. This function decode the string and validate the objective function for each candidate solution. Step 3: Onlooker bee phase The onlooker bee phase choose elite food source to optimal and enhance the source of food. This phase reach to appropriate location at low loss of power and high voltage profile, which in turn optimize the velocity of population by utilizing the following equation (1) below.
Vi , j = xi , j + Φi , j ( xi , j − xk , j )
(1)
14418
T. Sathish/ Materials Today: Proceedings 5 (2018) 14416–14422
Where, k is the neighborhood solution of i and Φ is a random number in the range of [-1, 1], Vi , j is the neighborhood solution of M i , j . Step 4: Selection For the updation of process, the optimal fitness is choosen in the selection phase. In this phase the selection is undergone by the probability function for the selection of fitness and probability for selection is evaluated from the following equation (2) below.
probabilit y =
Φ n
(2)
Φ
i =1
Step 5: CS based Scout bee phase At the point when the onlooker bee step has not refined obviously better alternatives, leave from the specific choices and make the arbitrary number of scout bee order while utilizing the Cuckoo Search Optimization. Initialization: The initialization phase starts with the setting of cuckoo search parameter. These parameters consist of the number of nests ‘ n ’, the step size parameter ‘ α ’, iteration probability ‘ Pa ’ and the maximum number of iteration as termination criteria. Generate Initial Nests or Eggs of Host Birds: The initial locations of the nests are specified by the set of random values assigned to each variable as: Di(,0j) = Round x j ,min + rand x j ,max − x j ,min (3)
(
(
))
th
Where, ‘ Di(,0j) ’ is the initial value of the j variable for the i th nest; ‘ x j ,min ’ and ‘ x j ,max ’ are the minimum and th the maximum allowable values for the j variable; ‘ rand ’ is a random number in the range of [0, 1]. Due to the discrete nature of the problem round function is accomplished. Step size evaluation:In this step, the step sizes for individual parent weights obtained in step 3 are evaluated. The following expressions are used to find out the step size.
(
)
t Sz = α S Dt − Dbest ⋅r
(4) t
Where, ‘ S z ’ is the Step size, ‘ α ’is Step size parameter ( α = 0.01 ), ‘ D t ’ is the current parent weight, ‘ Dbest ’is the best solution so far, ‘r’ is a random number from a standard normal distribution [0,1] and ‘S’ is step. The step S is found out by using the Mantegna’s algorithm, shown in the below equation. u S= 1 (5) v β Where, ‘ β ’ is a parameter arising in the interval [1, 2] which we choose in our system as 1.5 (i.e. β = 3 2 ) and u and v are normal distributions, which are estimated as follows.
u ~ N(0,σ u2 ) , v ~ N(0,σ v2 ) πβ Γ(1 + β ) Sin 2 σu = Γ (1 + β ) β ⋅ 2( β −1) 2 2
(6) 1
β
, σv = 1
(7)
Generation of New Solution: In this generation step, the new upgraded solution or weights are produced for comparing initial weights in view of the cuckoo search algorithm (Levy Flight). The new weights are created by utilizing the step and the measured values acquired in the step 3. The new weights are produced by utilizing the expression given below.
D(t+1) = Dt + Sz
(8)
T. Sathish/Materials Today: Proceedings 5 (2018) 14416–14422
14419
Where, ‘ D ( t +1) ’ is the new weight, ‘Sz’ is step size and ‘ D t ’ is the current parent weight. From this step another set of optimized weights are obtained. Ending of CS process: In this phase, the termination criteria is tested and if it is converges, the iteration stops and if not the iteration continuous to get optimal result. Step 6: Termination Criteria The optimization process is iterated to maximum level and the iteration number converges to maximum, then it stops and keep the current best solution. 3. Results and Discussion WEDM of Electronica (India) made has been utilized to machine Nimonic-263 material with a pulse on time, pulse off time, peak current and servo voltage as info parameters. Gaps of 10 mm width are delivered on 120mm×110mm×18.5 mm workpiece. The endpoint of the present review is to upgrade the machining parameters for better execution. In this review, the execution measures are material expulsion rate and surface harshness. The levels of input parameters such as pulse on time, pulse off time, peak current and servo voltage are settled in view of the trial tries and are given in Table 1. Reaction surface strategy Central composite face provided food outline with 1 focus focuses has been utilized for trial arrange. The deliberate and computed reactions are given in Table 2 in the wake of leading the analyses according to the test arrange. ANOVA has been connected to know the critical parameters and their commitment. It has additionally been connected to demonstrate the reactions material expulsion rate and surface finish. These models are additionally used to anticipate and advancement. Proposed BCCS algorithm has been utilized to enhance reactions and these outcomes are contrasted with RSM outcomes. Table 1: Levels of input parameters Parameter
level 1
level 2
level 3
Pulse on time (Ton) μs
105
115
125
Pulse off time (Toff) μs
50
55
60
Peak current (Ip) A
10
11
12
Servo voltage (Sv) V
40
50
60
Material removal rate was calculated using the equation (9) and the surface roughness of the machined surface was measured by MarSurf M-400. The measurement of diameter is carried out by Coordinate Measuring Machine (CMM). πT d 2 − D 2 (9) MRR = mm 3 / min 4T Where, D is the diameter of the hole in mm, d is the diameter of the blank in mm, t is the thickness of work piece in mm and T be the Time taken for machining in min.
(
)
14420
T. Sathish/ Materials Today: Proceedings 5 (2018) 14416–14422
Table 2. Experimental plan and Responses Run order
Ton (μs)
Toff (μs)
Ip (A)
Sv (V)
MRR (mm3 / min)
SR (μm)
1
125
60
12
60
3.12412
2.01
2
125
60
10
60
0.359606
0.605
3
115
55
11
40
0.452898
0.525
4
115
55
11
50
0.380731
0.387
5
105
60
12
60
0.497021
0.537
6
125
55
11
50
0.327263
0.578
7
125
50
10
40
0.380919
0.484
8
115
55
12
50
2.447281
1.021
9
115
50
11
50
0.280316
0.501
10
105
50
10
40
0.446554
1.364
11
115
55
10
50
0.300545
0.514
12
105
60
10
60
0.426622
0.694
13
125
50
10
60
0.458812
0.46
14
115
55
11
50
0.310576
0.539
15
125
50
12
40
3.588263
2.027
16
105
60
10
40
0.529402
1.323
17
125
60
12
40
3.352623
2.089
18
105
55
11
50
0.396612
0.892
19
115
55
11
60
0.575236
0.76
20
115
60
11
50
0.514857
0.764
21
125
60
10
40
0.500181
1.081
22
105
50
12
40
1.768268
0.891
23
125
50
12
60
3.233984
1.85
24
105
60
12
40
0.948816
1.18
25
105
50
10
60
0.548839
0.863
26
105
50
12
60
0.771276
0.603
Analysis of variance has been applied on the experimental results for both metal removal rate and surface roughness are given Table 3 and Table 4 respectively.
T. Sathish/Materials Today: Proceedings 5 (2018) 14416–14422
14421
Table 3. Analysis of Variance in Terms of Metal Removal Rate Source
SS
DOF
MS
F value
P-value
Percentage Contribution
Model
29.3208
9
3.257867
44.563324
< 0.0001
96.16369159
A- Ton
4.492364
1
4.492364
61.449614
< 0.0001
14.73364663
B- Toff
0.08323
1
0.08323
1.1384719
0.3018
0.26237672
C- Ip
13.8341
1
13.8341
189.23223
< 0.0001
45.37182225
D- Sv
0.216133
1
0.216133
2.956416
0.1048
0.708853345
AC
5.718684
1
5.718684
78.224046
< 0.0001
18.75561937
BC
0.126007
1
0.126007
1.7236097
0.2077
0.413266292
CD
0.24216
1
0.24216
3.31243
0.0875
0.794214331
A^2
0.012463
1
0.012463
0.1704738
0.6852
0.040875013
C^2
2.960149
1
2.960149
40.490926 < 0.0001
9.708427311
0.073106
Residual
1.169703
16
Cor. Total
30.49051
25
3.836285454 100
From the results, it is contemplated that Ton, Ip, association of Ton and Ip, Ip2 are critical model terms. Higher the beat on time, higher will be the vitality connected there by producing more measure of thermal energy among this period and it prompts higher metal removal rate. Top current is the measure of energy utilized as a part of release machining. Higher the pinnacle present, higher will be the vitality connected at machining and in this manner expanding the removal rate. From the ANOVA result, the R-Square, balanced R-square and forecasted Rsquare values were observed to be 96.2%, 94%and 90% respectively for the model. Confirmation tests have been conducted to check the effectiveness of BCCS for both MRR and SR, and the results were given in Table 4. Table 4. Confirmation Test results Response MRR (mm3/min) (Ton:125, Toff:50, Ip:12, Sv:40) SR (μm) (Ton:116, Toff:50, Ip:10, Sv:60)
Predicted value from BCCS
Experimental value
Deviation in percentage
3.6713
3.614
2
0.2618
0.282
7
4. Conclusion Utilization of response surface methodology is taken as a part of the present review to show WEDM execution measures of material removal rate and surface finish. The parameters influencing machining process are Pulse on time, pulse off time, peak current and servo voltage. The significance of process parameters have been recognized by applying analysis of variance examination for both metal removal rate and surface roughness. For metal removal rate it was found from the analysis of variance comes about that, Pulse on time, peak current and association impact of Pulse on time and peak current are more affecting than other model terms. In the present research an endeavor has been made to apply proposed optimization algorithm to improve the reactions. The ideal reaction values from response surface methodology and proposed BCCS algorithm are contrasted. It is found that, the aftereffects of BCCS are superior to that of response surface methodology. Because of extensive variety of advantages for
14422
T. Sathish/ Materials Today: Proceedings 5 (2018) 14416–14422
Nimonic-263, the machining information created without precedent for this work utilizing WEDM will be valuable to the manufacturing organizations. References [1] Spedding, T.A, Wang, Z.Q., 1997. Parametric optimization and surface characterization of wire electrical discharge machining process, Journal of Precision Engineering, 20, 5-15. [2] Nihat Tosun, 2003. The Effect of the Cutting Parameters on Performance of WEDM, KSME International Journal, 17, 816-824. [3] Mahapatra, S., Amar Patnaik, S., 2007. Optimization of wire electrical discharge machining (WEDM) process parameters using Taguchi method, International Journal of Advanced Manufacturing Technology, 34, 911–925. [4] Pragya Shandilya,.Jain, P.K., Jain, N.K., 2012. Parametric optimization during wire electrical discharge machining using response surface methodology, Procedia Engineering, 38, 2371 – 2377. [5] Muthu Kumar, V., Suresh Babu, A., Venkatasamy, R., Raajenthiren, M., 2010. Optimization of the WEDM Parameters on Machining Incoloy800 Super alloy with Multiple Quality Characteristics, International Journal of Engineering Science and Technology, 2, 162-183. [6] Abido, M.A., 2002. Optimal power flow using particle swarm optimization, Electrical Power and Energy Systems, 24, 563-571. [7] Bharathi Raja, S., Baskar, N., 2011. Particle swarm optimization technique for determining optimal machining parameters of different work piece materials in turning operation, International Journal of Advanced Manufacturing Technology, 54, 445–463. [8] Manna, A., Bhattacharyya, B., 2006, Taguchi and Gauss elimination method: A dual response approach for parametric optimization of CNC wire cut EDM of PRAlSiCMMC, International Journal of Advanced Manufacturing Technology, 28, 67–75. [9] Rao, R.V., Pawar, P.J., 2009. Modelling and optimization of process parameters of wire electrical discharging machining, Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture 223, 1431-1440. [10] Xin-She Yang, Nature-Inspired Optimization Algorithms, 1st Edition, 2014, page-99 Elsevier, London.