1. bookVolume 39 (2021): Issue 2 (June 2021)
Journal Details
License
Format
Journal
eISSN
2083-134X
First Published
16 Apr 2011
Publication timeframe
4 times per year
Languages
English
Open Access

Influence of optimization techniques and nano surface coating materials on microstructure of iron-nickel-chromium alloy using wire EDM process

Published Online: 09 Aug 2021
Volume & Issue: Volume 39 (2021) - Issue 2 (June 2021)
Page range: 152 - 165
Received: 10 Feb 2021
Accepted: 14 Mar 2021
Journal Details
License
Format
Journal
eISSN
2083-134X
First Published
16 Apr 2011
Publication timeframe
4 times per year
Languages
English
Introduction

Wire EDM is an unconventional machining process. The machine has been used for machining a heavy hardened material with complicated shape in best surface quality. It is used for cutting the material by means of electro-thermal energy to produced spark in every few microseconds. The spark is produced between the tool (wire electrode) and work piece. Here, the tool acts as moving through the stationary work piece. By this process, the tool almost enclosed by spark with the temperature around 12,000 °C. In this environment, the great amount of heat is produced, and also the material has been removed by melting and evaporation [1]. Due to the high temperature environmental process, the tool and workpiece both are submerged in deionized water. The material surface and structure has been changed due to the rapidly changing thermal environment during machining operation. That was affecting the surface characterizes, which consist of all the aspects of surface finish, surface metallurgy and surface topography. The machined surface consists of two criteria. First is the top surface of workpiece, which is covered surface texture. It is one of the reasons affecting surface roughness. Another one is surface metallurgy, which produces recast layer when machining the work piece [2]. At that time, the recast layer forms due to solidification by rapid cooling. This surface is called as white layer because it appears white in colour when observed in electron microscope images.

This layer was formed by thermal, chemical and mechanical effects on the material surface to remove material. It is also controlled by some phenomenon such as machining parameters (pulse on/off time, wire feed/speed, coolant pressure, etc.) and work piece [3, 4]. Literally, when the machining the surface was subjected to local thermal stress, the period of pulse on time, and the spark intensity with the coolant increases the temperature. From that, the current intensity and pulse on time affect the machined material surface roughness and hardness [5, 6]. Most of researchers are focusing on the optimized parameters with the surface integrity of material [7, 8]. Dhobe et al. [7] found that the heat treatment is also one of the affecting parameter. And Kondapalli Siva Prasad et al. [17] has been found that the effect of grain size in WEDM process. Similarly, R. K. Garg et. al [9] studied that the relation between heat treatment and white layer on material during wire EDM process. The hardness and heat treatment of work piece has been affecting the machinability [12]. The EDM achieved highest MRR, SR, TWR and dimensional accuracy when workpiece and tool (electrode) done by cryogenic treatment [11]. Similarly, when wire EDM was machining the work piece after plasma nitriding, there were no significant changes of formation of white layer on surface [10]. The hardening with double tempering gives better surface roughness compared to single tempering [13]. Mehmet Altug et. al [14] observed that the material treatment also affected the wire EDM machinability.

In recent years, most of the researchers have been developing various techniques to investigate the wire EDM performance on various materials using various modelling and new optimization techniques. The performance and surface characteristic of wire EDM has been investigated [15]. Artificial intelligence techniques, such as fuzzy logic and ANN, were successfully applied to various machining processes [16]. It has been examined and provided optimized result for material removal rate and surface roughness. The SA and F-ANN approach has been used to wire EDM process to correlate the input parameters with output performance. It was observed that the wire EDM performance can be improved by using new technique approach [17]. The wire EDM neural network modelling has been analysing formation of residual stress during machining in metal matrix composite. It has been noted that pulse off time has significant characteristics on the formation of residual stress [18]. The PCA integrated with Taguchi approach was recommended to investigate the pulse on/off time, particle size, wire tension volume fraction, wire wear rate, surface roughness and recast layer thickness during machining in wire EDM [19]. In this way, advanced modelling techniques like support vector machine, PSO, RSM, regression and analysis have been employed in research area [20,22,23]. In the literature survey based has been to investigate wire EDM performance and analysis material surface of iron-nickel-chromium alloy through the hybrid optimization techniques (Taguchi and ANFIS).

Experimental details

In the present study, alloy 20 material and 2 mm thickness plate were discussed. The chemical composition of alloy 20 material is shown in Table 1. The alloy 20 material was applied HVOF molybdenum coating with the process temperature of 450 °C and then, plasma nitriding carried out maintaining the temperature of 500 °C in 12 hours [25]. Material specimen prepared as the dimensions of 150 mm × 150 mm × 2 mm was matched in material in the wire EDM.

Chemical composition of alloy 20 (in wt%)

Si S P Mn C Nb Mo Cu Cr Fe Ni
1 0.035 0.045 2 0.07 1 3 4 21 29.8 38
Experimental producer using hybrid optimization techniques

The experiment performance has been conducted on wire EDM as shown in Figure 1. It has brass wire with the diameter of 0.25 mm as cutting tool, deionized water consider to dielectric medium at constant atmospheric temperature. The flow chart diagram of the optimization techniques is shown in Figure 2. Three input variables as pulse on time (Ton), pulse off time (Toff) and wire feed (WT) has been used for machining on iron-nickel-chromium alloy and measured output performance as surface roughness (SR) and material removal rate (MRR). The experimental work has been optimized L27 orthogonal array by Mintab software to get error less input and output as shown in Table 2. The Table 3 shows three levels of input parameters. The L27 orthogonal array has been conducted in 3 × 3 level. The below equation shows S/N ratio (single to noise) input parameter.

Fig. 1

Cause and effect diagram for performance characteristics in WEDM

Fig. 2

Flow chart diagram of optimization techniques

Process parameters employed for EDM

Parameter Unit Factor Level

1 2 3
Pulse on time (μ sec) A 106 116 126
Pulse off time (μ sec) B 40 50 60
Wire feed (m/min) C 4 8 12

EDM process parameter of Taguchi L27 orthogonal array

Experimental trials Factor

A B C
1 1 1 1
2 1 1 2
3 1 1 3
4 1 2 1
5 1 2 2
6 1 2 3
7 1 3 1
8 1 3 2
9 1 3 3
10 2 1 1
11 2 1 2
12 2 1 3
13 2 2 1
14 2 2 2
15 2 2 3
16 2 3 1
17 2 3 2
18 2 3 3
19 3 1 1
20 3 1 2
21 3 1 3
22 3 2 1
23 3 2 2
24 3 2 3
25 3 3 1
26 3 3 2
27 3 3 3

The optimized output parameter founded in S/N ratio has been employed to conduct the input variables. The surface roughness and material removal rate has been predicted by Taguchi method to get minimum error and more accuracy.

Taguchi method

Taguchi Method has been applied to experimental design through method employed orthogonal array to manage the process affecting parameters and their levels. It probably has to conduct all possible combination such as factorial design, Taguchi method combination test pair. These allow for collecting the data to find factors affecting the most output with least possible experimentations, which has been saving resource and time. ANOVA (analysis of variables) on the data collected to be used choosing new parameters for performance characterizes optimized. In order to find the study parameters, cause and effect diagram to determine potential parameters that affect machining characteristics (SR and MRR) was constructed. From the cause and effect diagram and literature review of wire EDM, we were choosing three input parameter for this investigation. In this investigation, L27 orthogonal array was with three control factors, that is, pulse on time, pulse off time and wire feed. Signal to noise ratio was obtained by using Minitab software. SR considered as lower is better (LB) and MRR considered as higher is better (HB) for determining optimized machining parameters. The loss function logarithmic transformation can be used to calculate S/N ratio.

Back propagation neural network

Back propagation is a special type of neural network and it is also named as back propagation error method. It is mostly used for multi layered neural networks. In artificial neural network, one of the methods is back propagation neural network used to calculate gradient descent that is needed to calculate the weight of the network. This method uses a derivative of the squared error function with respect to the weights of the network. The back propagation neural network, otherwise backward propagation error method, mainly look for error function, even then having the minimum value of weight space is known as rule or gradient descent. The weights that can minimize the error function are then considered to be a solution.

BPNN is a feed forward neural network. Back-propagation, an abbreviation for “backward propagation of errors” is a common method of training ANN. In 1986, Rumelhart et al. had described a new supervised learning procedure known as back propagation neural network (BPNN), which is used for linear as well as non-linear classification. From a desired output, the network learns from many inputs. In BPNN, the errors are back propagated to the input layer. BPNN is a supervised algorithm in which error difference between the desired output and calculated output is back propagated. The procedure is repeated during learning to minimize the error by adjusting the weights through the back propagation of error. As a result of weight adjustments, hidden units set their weights to represent important features of the task domain. It has the capability to establish the connection between HVS parameter values and FIS output values by adjusting the neural network weights and bias before and after embedding watermark. Owning to use of neural network, watermark can be extracted without the host signal, and therefore, reduce the boundary in practical applications.

Cumulative experimental data of EDM process.

SI.NO Pulse on time T ON (μ sec) Pulse off time T OFF (μ sec) Wire feed WF (m/min) Material removal rate MRR (mm/min) Surface roughness SR (μm)
1 106 40 4 0.68 1.385
2 106 40 8 0.71 1.255
3 106 40 12 0.505 1.175
4 106 50 8 0.48 1.35
5 106 50 12 0.405 1.265
6 106 50 4 0.2 1.105
7 106 60 12 0.315 1.35
8 106 60 4 0.165 1.14
9 106 60 8 0.165 1.08
10 116 40 12 2.405 2.57
11 116 40 4 2.205 2.39
12 116 40 8 0.75 1.435
13 116 50 4 1.54 2.525
14 116 50 8 0.58 1.38
15 116 50 12 0.66 1.72
16 116 60 8 0.47 1.63
17 116 60 12 0.625 1.85
18 116 60 4 0.5 1.625
19 126 40 8 3.4 2.9
20 126 40 12 1.26 1.85
21 126 40 4 2.2 2.66
22 126 50 12 0.91 2.08
23 126 50 4 1.66 2.635
24 126 50 8 1.54 2.445
25 126 60 4 1.34 2.855
26 126 60 8 1.345 2.56
27 126 60 12 0.29 1.295

BPNN consists of three layers: 1) input layer, 2) hidden layer, and 3) output layer. The number of hidden layers, and the number of hidden units in each hidden layer depends upon the complexity of the problem as shown in Figure 3.

Fig. 3

BPNN model used for process parameter analysis and optimization techniques

During this step, error is calculated by difference between the targeted output and actual output of each output unit. This error is back propagated to the previous layer, that is, hidden layer. For each unit in the hidden layer N, error at that node is calculated. In a similar way, error at each node of previous hidden layer, that is, N-1 is calculated. These calculated errors are used to correct the weights so that the error at each output unit is minimized. Forward and backward steps are repeated until the error is minimized up to the expected level. The training algorithm of back propagation includes four stages: 1. initialization of weights, 2. feed forward, 3. back propagation of errors, 4. updating the weights and biases. The bias acts like weights on the connection from units whose output is always 1. During initialization of weights, some arbitrary values are given initially to give some output by feed forwarding through the layers. So, the difference between the obtained and actual values is calculated as error and back propagated. High initial weight will result in 90 a faster learning rate. But weights may oscillate. If initial weights are too small, then learning rate will be slow (Table 5). For best results, initial weights may be considered between −0.5 to 0.5 or −1 to 1. In back propagation method first, the output of hidden layer is calculated by the formula: Hi=f(wijXiaj)j=1,2,,l {H_i} = f\left( {\sum {{w_{ij}}{X_i} - {a_j}} } \right)j = 1,2, \ldots ,l where and f represent the output of hidden layer and the incentive function of neurons, l is the neuron number of hidden layer, n is the neuron number of the input layer, is the weight factor between input-layer and hidden layer, is threshold value. The next step is to predict the output values by using the formula: Ok=Hiwikbkk=1,2,,m {O_k} = \sum {{H_i}{w_{ik}} - {b_k}k = 1,2, \ldots ,m} where is threshold value, m is the neuron number of the output layer. Then, according to the prediction error calculated by the difference between the predicted output and the expected output, weight factor and threshold value can be updated as follows: ωij=ωij+ηHi(1Hi)X(i)ωikeiki=1,2,,n:j=1,2,,l \matrix{{{\omega _{ij}} = {\omega _{ij}} + \eta {H_i}\left( {1 - {H_i}} \right)X\left( i \right)\sum {{\omega _{ik}}{e_{ik}}} } \hfill \cr {i = 1,2, \ldots ,n:j = 1,2, \ldots ,l} \hfill \cr } ωij=ωik+ηHiekj=1,2,,m {\omega _{ij}} = {\omega _{ik}} + \eta {H_i}{e_k}\,j = \,1,2, \ldots ,m a1=aj+ηHi(1Hi)x(i)ωikekj=1,2,,l {a_1} = {a_j} + \eta {H_i}\left( {1 - {H_i}} \right)x\left( i \right)\sum {{\omega _{ik}}{e_k}j = 1,2, \ldots ,l} bk=bk+ekk=1,2,,m {b_k} = {b_k} + {e_k}\,k = 1,2, \ldots ,m

BPNN parameters and interlayer details

Name Number No. MRR SR
parameter. Show 50 1st 9 9
Maximum parameter. Epochs 1000
Number of input parameter 3 2nd 8 10
Number of output parameter 1

One of the more popular activation functions for backpropagation networks is the sigmoid, a real function sc: IR → (0, 1) defined by the expression. The constant c can be selected arbitrarily and its reciprocal 1/c is called the temperature parameter in stochastic neural networks. An alternative to the sigmoid is the symmetrical sigmoid S(x).

Generally, there are two types of learning methods as follows: 1. sequential learning or pre-pattern method, 2. batch learning or pre-epoch method. In sequential learning, a given input pattern is broadcasted forward and then the error is calculated and back propagated, the weights are reorganized until we get the targeted output. In batch learning, the weights are reorganized only after the entire set of training network has been offered to the network. Thus, the weights are updated after every epoch.

Adaptive neuro-fuzzy interference system (ANFIS)

Due to changes in network environment, in a dynamic environment in need of an optimization algorithms, which learn from the experimental results and find the optimum solution to the end user. So many soft computing techniques are evolved such as artificial neural network, fuzzy logic system, and genetic algorithms, as shown in Figure 4.

Fig. 4

Adaptive neuro-fuzzy interference system (ANFIS)

To minimize output errors in fuzzy modelling system, it needs trial steps and auxiliary calculation for adjusting the fuzzy membership functions. Because there are no proper methods to convert the human idea into the knowledge part of the fuzzy system, which leads the system output inadaptability nature. In the case of artificial neural network, it having good adaptability of output based on the input parameters and it also support the non-linear correlation between the input and output parameters. To improve the performance combining the advantages of both FIS and ANN, it is called as ANFIS. This ANFIS supports the knowledge and adaption features.

Jang combines the advantages of fuzzy interference system (FIS) with artificial neural network (ANN) and named as adaptive neuro fuzzy interference system (ANFIS), which consists of multi-layer approaches. ANFIS consists of fuzzification, rule, normalization, defuzzification and output layers, as shown in the figure.

Jang developed network system (ANFIS) as a rule based system and it learns/decides the rules from the training data. Two different types of methods are used in FIS such as Mamdani and Takagi-Sugeon-Kang (TSK). In our system, we use the TSK approach to derive the optimized results. Here rules are mainly focusing on the input output relationship such as if this is the input parameter then what will be the output; simply if-then rules are used for correlating the input and output values.

IFaisX1andbisY1THENf1=p1a+q1b+r1IFaisX2andbisY2THENf2=p2a+q2b+r2..IFaisXnandbisYnTHENf1=pna+qnb+rn \matrix{{IF\,a\,is\,{X_1}\,and\,b\,is\,{Y_1}\,THEN\,{f_1} = {p_1}a + {q_1}b + {r_1}} \cr {IF\,a\,is\,{X_2}\,and\,b\,is\,{Y_2}\,THEN\,{f_2} = {p_2}a + {q_2}b + {r_2}} \cr { \ldots \ldots ..} \cr {IF\,a\,is\,{X_n}\,and\,b\,is\,{Y_n}\,THEN\,{f_1} = {p_n}a + {q_n}b + {r_n}} \cr }

The values of p, q and r constant that can be evaluated during the training process based on the training data. Next step is to form fuzzy rules by combining these functions by the ANFIS. In this layer, it receives input from the external source and it assigns the member function to all the input values based on the weights. This is an adaptive layer so based on the input fuzzy set assign a member function to each input values: O11=μXi(a)fori=1,2,n O_1^1 = {\mu _{Xi}}\left( a \right)\,f\,or\,i\, = \,1,2, \ldots n

Layer 2 – rule layer

In this model, there are two type of rules can be applied by the system such as AND and OR rule. Our proposed work adopts the AND rule to provide the relative strength information about input values, considered as firepower of all the fuzzy rules. Product of all the input membership value is considered as output of this layer.

Oi2=wi=μXi(a)*μyi(a) O_i^2 = {w_i} = {\mu _{Xi}}\left( a \right)*\mu {y_i}\left( a \right)

Layer 3 – normalization layer

Third layer is a Normalized layer. This is to normalize all the firing strength based on each rule's weight to all the rules’ weights.

It can be calculated as: Oi3=wi O_i^3 = wi

Layer 4 – defuzzification

Defuzzification is an adaptive layer, which uses the first order function to form fi = pix + qiy + ki. The derived output this layer is in the form of: Oi4=wifi=wi(pix+qiy+ki. O_i^4 = {w_i}{f_i} = {w_i}\left( {{p_i}x + {q_i}y + {k_i}} \right..

Layer 5 – output layer

This is an output layer. From this layer, we get the final output of this network by sum of all the input layers by using the formula: Oi5=fi=Σwifi=Σ[wi(pix+qiy+ki]. O_i^5 = fi = \Sigma {w_i}{f_i} = \Sigma \left[ {{w_i}\left( {{p_i}x + {q_i}y + {k_i}} \right.} \right].

Finally, by having ANFIS, the input values are mapped with output values to get the optimized results. This method follows the values that are grouped into two set such as training data and checking data. In ANFIS, a combination of the back propagation and least square methods are used to train the input values. The result of the training data is prediction of error in a minimized manner.

Results and discussion
Hybrid optimization of process parameter studies

The Taguchi model was used for the optimization of signal response. The performance output, surface roughness (SR) was the type “lower the better” and material removal rate was the type “higher the better”. The S/N ratio was examined by loss function logarithmic transformation, which was given by Xiong [29] as shown in the following equation: SNratio=|10log10[1ni=1nyi2] {S \over N}ratio = | - 10{{log}_{10}}[ {{1 \over n}\sum\limits_{i = 1}^n {y_i^2} } ]

The ANOVA model was conducted using the MINITAB-19 software in order to examine the significant input parameters. The three types of parameters were the deciding factor of WDM: pulse on time, pulse off time, wire feed as shown in Figure 5(a) & 5(b). Higher F value and lower P value mention the importance degree of every input parameter on material removal rate and surface roughness of hybrid treated iron-nickel-chromium alloy at 95% confidence level as shown in Table 6 and 7.

Fig. 5

(a) Plot diagram of material removal rate; (b) Plot diagram of surface roughness SR (μm)

Regression ratio of material removal rate

Source DF Adj SS Adj MS F-Value P-Value
Regression 3 13.1260 4.3753 21.32 0.000
Pulse on time t on (μ sec) 1 7.8095 7.8095 38.06 0.450
Pulse off time t on (μ sec) 1 5.0545 5.0545 24.63 0.300
Wire feed 1 0.2620 0.2620 1.28 0.250
Error 23 4.7195 0.2052
Total 26 17.8454

Regression ratio of surface roughness SR (μm)

Source DF Adj SS Adj MS F-Value P-Value
Regression 3 1.96406 0.65469 16.14 0.000
Pulse on time t on (μ sec) 1 1.76964 1.76964 43.62 0.420
Pulse off time t on (μ sec) 1 0.08050 0.08050 1.98 0.470
Wire feed 1 0.11392 0.11392 2.81 0.210
Error 23 0.93314 0.04057
Total 26 2.89719

The BPNN gives better performance. Hence, we were required to develop confidence level and overall best fit model performance, as shown in Figure 6. For that, we had to input parameter in ANFIS. The figure shows the ANFIS model simulation results in the form of graphical rules. It is noted that the performance output probability was maximum by varying input parameters and also by applying the if-then rules:

If (pulse on time is low) and (wire feed is low) and (signal to noise ratio is high), then (surface roughness is high).

If (pulse on time is moderate) and (wire feed is moderate) and (signal to noise ratio is high), then (surface roughness is high).

If (pulse on time is moderate) and (wire feed is low) and (signal to noise ratio is high), then (surface roughness is high).

If (pulse on time is moderate) and (wire feed is low) and (signal to noise ratio is moderate) then (surface roughness is high).

If (pulse on time is high) and (wire feed is moderate) and (signal to noise ratio is high), then (surface roughness is high).

If (pulse on time is high) and (wire feed is low) and (signal to noise ratio is high), then (surface roughness is high).

Fig. 6

The best fitness value estimated by BPNN model of (a) Material removal rate (b) Surface roughness SR (μm)

The Figure 7 shows that the correlation of the input parameters to output performance of wire EDM. From this, we can clearly concluded that the output performance as material removal rate was high when the pulse on time and wire feed became maximum and also surface roughness was high while the pulse on time and wire feed became minimum. Similarly, the pulse off time was minimum the performance output as material removal rate became high and surface roughness was less. The BPNN error calculation methods were implemented at the output parameters such as experimentation output parameters value (EOP) and prediction of back propagation neural network output parameters (BPOP), as shown in Table 8 (a) and 8 (b).

Fig. 7

Plot diagram of material removal rate of ANFIS model

(a) 1st EDM Process parameter of alloy 20 materials; (b) 2nd EDM Process parameter of alloy 20 materials

No MRR (mm/min) SR (μm)

EOP BPOP ERROR EOP BPOP ERROR
1 0.68 0.5643 11.57 1.385 1.3627 2.23
2 0.71 0.7451 −3.51 1.255 1.2712 −1.62
4 0.48 0.4921 −1.21 1.35 1.3215 2.85
10 2.405 2.4184 −1.34 2.57 2.3954 17.46
14 0.58 0.5683 1.17 1.38 1.4725 −9.25
17 0.625 1.0514 −42.64 1.85 2.3891 −53.91
22 0.91 0.9244 −1.44 2.08 2.0176 6.24
23 1.66 1.7587 −9.87 2.635 2.7871 −15.21
25 1.34 1.3572 −1.72 2.855 2.8142 4.08
No MRR (mm/min) SR (μm)

EOP BPOP ERROR EOP BPOP ERROR
3 0.505 0.6383 −13.33 1.175 1.1254 4.96
6 0.2 0.1662 3.38 1.105 1.1031 0.19
9 0.165 0.1675 −0.25 1.08 1.1182 −3.82
12 0.75 0.7593 −0.93 1.435 1.4258 0.92
15 0.66 0.6247 3.53 1.72 1.7467 −2.67
18 0.5 0.7124 −21.24 1.625 1.6024 2.26
20 1.26 1.4972 −23.72 1.85 1.4358 41.42
24 1.54 1.5524 −1.24 2.445 2.5123 −6.73
26 1.345 1.3571 −1.21 2.56 2.5917 −3.17

The graph shows the details of the predicted value obtained by Taguchi, back propagation neural network, fuzzy logic system modelling and hybrid ANFIS model. To investigate the scattering about the agreement line (45 degrees line), one line was plotted in the range within the 2% error and then two more lines, which was Taguchi, BPNN and fuzzy logic, were over, then the range of 4,6,12% error was conceived. It was calculated that the predicted value accommodated by Taguchi, BPNN model and fuzzy model generated over the range 99% of accuracy, which obtained the reliability of models as shown in Table 9. The hybrid ANFIS model dominated the Taguchi, BPNN model and fuzzy logic model, and the reason of the maximum of hybrid ANFIS model predicted values were plotted on the agreement line.

Results of BPNN-ANFIS optimization

Sl. No. Input parameters Output parameters

Pulse on time (μ sec) Pulse off time (μ sec) Wire feed (m/min) Optimal value (mm) Machining performance
1. 110 80 10 0.184 Material removal rate
2. 106 60 10 1.95 Surface roughness

While in case of Taguchi, BPNN and fuzzy modelling, some more predicted values were crossing the 2% of error, as shown in Figure 8. Consequently, the Taguchi, BPNN and fuzzy models were dominated by the hybrid ANFIS model. The Taguchi gives the largest mean error predicted value, which was 4.56% followed by fuzzy logic within the mean error 2.78%; BPNN model gave 1.52% mean error and hybrid ANFIS employed mean error of 0.5%. ANFIS Taguchi hybrid algorithm is made to be optimized and develop advantages, while it can solve nonlinear functions and approximation problems. In addition, the hybrid algorithm is a flexible and robust tool that can engage predictive and confidence interval prediction 99.79% in hybrid ANFIS model. It is very close to the actual value (Table 10).

Fig. 8

Plot diagram of surface roughness of ANFIS model

EDM process parameters and error of confirmation experiment

Input parameters Output parameters

Pulse on time (μ sec) Pulse off time (μ sec) Wire feed (m/min) Material removal rate (mm/min) Surface roughness (μm)
Experiment predicted by 106 60 8 0.155 1.08
BPNN-ANFIS model 105.7 59 8 0.1675 1.1182
Error % 1.5 −3.5
Metallurgical and mechanical properties

The analysis of metallurgical and mechanical properties of iron-nickel-chromium alloy and hybrid treated iron-nickel-chromium alloy was carried out after machining in wire EDM. Ganapathy Srinivasan et.al [24] investigated the hybrid treated iron-nickel-chromium alloy before machining; it had deposition compound layer form as presence of MoxN, FexN, CrN, NiXN, and thin FCC M phase. That layer produced compressive residual stress on the material surface. The Figure 9 (a) and (b) shows the SEM analysis of iron-nickel-chromium alloy before and after machining in wire EDM. Figure 9(a) clearly shows the formation of recast layer on the surface of the material. The recast layer was formed as shrinkage of melted material and rapid solidification that induced tensile residual and high thermal stress, which produced micro cracks on surface [25]. Y.S Tarng et. al [26] observed that the recast layer lead to changes of microstructure as an austenite (√/− Fe). This phase was not observed in the material before machining. The tensile residual stress of recast layer increased the effect of maximum shear stress, which lead to indentation depth and plastic deformation [27]. These affect the mechanical properties of material after machining. Figure 9 (c) shows that the SEM analysis of hybrid treated iron-nickel-chromium alloy after machining in wire EDM same machining parameters. It clearly shows that no evidence was found regarding the formation of recast layer on the surface after machining. This effect is believed to be due to the fact that the tensile and thermal stress effect was balanced by the presence of compressive residual stress in hybrid treated material [28]. And also the crack propagation and micro holes were not noted in the machined surface. Due to this reason, there were no significant changes in the mechanical properties.

Fig. 9

SEM analysis a) Alloy 20 materials b) Coating materials c) Hybrid treated alloy 20 microstructure after machining

Conclusion

By the experimental investigation and optimization of wire EDM process parameters for hybrid coating on iron-nickel-chromium, we conclude:

The hybrid optimization techniques gives better optimized result when compared to Taguchi, BPNN and fuzzy logic.

The error was calculated. The BPNN method was used in material removal rate and surface roughness. BPNN appears to acquiesce an error percentage of 4, 8 and 12, respectively, indicating its high degree of consistency.

In the confirmation test, the errors of −16.29 and 12.58 % were noticed between the experiment result and the predicted result by BPNN-GA model for material removal rate and surface roughness, respectively.

The hybrid surface treated specimen after wire EDM machining observed that the formation of recast layer on machining surface was minimum as compared to the without treated material.

Formation of pore and cracks were less in the hybrid treated machining specimen due to the presence of twin boundaries and grain boundary thickness.

Fig. 1

Cause and effect diagram for performance characteristics in WEDM
Cause and effect diagram for performance characteristics in WEDM

Fig. 2

Flow chart diagram of optimization techniques
Flow chart diagram of optimization techniques

Fig. 3

BPNN model used for process parameter analysis and optimization techniques
BPNN model used for process parameter analysis and optimization techniques

Fig. 4

Adaptive neuro-fuzzy interference system (ANFIS)
Adaptive neuro-fuzzy interference system (ANFIS)

Fig. 5

(a) Plot diagram of material removal rate; (b) Plot diagram of surface roughness SR (μm)
(a) Plot diagram of material removal rate; (b) Plot diagram of surface roughness SR (μm)

Fig. 6

The best fitness value estimated by BPNN model of (a) Material removal rate (b) Surface roughness SR (μm)
The best fitness value estimated by BPNN model of (a) Material removal rate (b) Surface roughness SR (μm)

Fig. 7

Plot diagram of material removal rate of ANFIS model
Plot diagram of material removal rate of ANFIS model

Fig. 8

Plot diagram of surface roughness of ANFIS model
Plot diagram of surface roughness of ANFIS model

Fig. 9

SEM analysis a) Alloy 20 materials b) Coating materials c) Hybrid treated alloy 20 microstructure after machining
SEM analysis a) Alloy 20 materials b) Coating materials c) Hybrid treated alloy 20 microstructure after machining

EDM process parameter of Taguchi L27 orthogonal array

Experimental trials Factor

A B C
1 1 1 1
2 1 1 2
3 1 1 3
4 1 2 1
5 1 2 2
6 1 2 3
7 1 3 1
8 1 3 2
9 1 3 3
10 2 1 1
11 2 1 2
12 2 1 3
13 2 2 1
14 2 2 2
15 2 2 3
16 2 3 1
17 2 3 2
18 2 3 3
19 3 1 1
20 3 1 2
21 3 1 3
22 3 2 1
23 3 2 2
24 3 2 3
25 3 3 1
26 3 3 2
27 3 3 3

Cumulative experimental data of EDM process.

SI.NO Pulse on time T ON (μ sec) Pulse off time T OFF (μ sec) Wire feed WF (m/min) Material removal rate MRR (mm/min) Surface roughness SR (μm)
1 106 40 4 0.68 1.385
2 106 40 8 0.71 1.255
3 106 40 12 0.505 1.175
4 106 50 8 0.48 1.35
5 106 50 12 0.405 1.265
6 106 50 4 0.2 1.105
7 106 60 12 0.315 1.35
8 106 60 4 0.165 1.14
9 106 60 8 0.165 1.08
10 116 40 12 2.405 2.57
11 116 40 4 2.205 2.39
12 116 40 8 0.75 1.435
13 116 50 4 1.54 2.525
14 116 50 8 0.58 1.38
15 116 50 12 0.66 1.72
16 116 60 8 0.47 1.63
17 116 60 12 0.625 1.85
18 116 60 4 0.5 1.625
19 126 40 8 3.4 2.9
20 126 40 12 1.26 1.85
21 126 40 4 2.2 2.66
22 126 50 12 0.91 2.08
23 126 50 4 1.66 2.635
24 126 50 8 1.54 2.445
25 126 60 4 1.34 2.855
26 126 60 8 1.345 2.56
27 126 60 12 0.29 1.295

(a) 1st EDM Process parameter of alloy 20 materials; (b) 2nd EDM Process parameter of alloy 20 materials

No MRR (mm/min) SR (μm)

EOP BPOP ERROR EOP BPOP ERROR
1 0.68 0.5643 11.57 1.385 1.3627 2.23
2 0.71 0.7451 −3.51 1.255 1.2712 −1.62
4 0.48 0.4921 −1.21 1.35 1.3215 2.85
10 2.405 2.4184 −1.34 2.57 2.3954 17.46
14 0.58 0.5683 1.17 1.38 1.4725 −9.25
17 0.625 1.0514 −42.64 1.85 2.3891 −53.91
22 0.91 0.9244 −1.44 2.08 2.0176 6.24
23 1.66 1.7587 −9.87 2.635 2.7871 −15.21
25 1.34 1.3572 −1.72 2.855 2.8142 4.08

Regression ratio of material removal rate

Source DF Adj SS Adj MS F-Value P-Value
Regression 3 13.1260 4.3753 21.32 0.000
Pulse on time t on (μ sec) 1 7.8095 7.8095 38.06 0.450
Pulse off time t on (μ sec) 1 5.0545 5.0545 24.63 0.300
Wire feed 1 0.2620 0.2620 1.28 0.250
Error 23 4.7195 0.2052
Total 26 17.8454

Results of BPNN-ANFIS optimization

Sl. No. Input parameters Output parameters

Pulse on time (μ sec) Pulse off time (μ sec) Wire feed (m/min) Optimal value (mm) Machining performance
1. 110 80 10 0.184 Material removal rate
2. 106 60 10 1.95 Surface roughness

EDM process parameters and error of confirmation experiment

Input parameters Output parameters

Pulse on time (μ sec) Pulse off time (μ sec) Wire feed (m/min) Material removal rate (mm/min) Surface roughness (μm)
Experiment predicted by 106 60 8 0.155 1.08
BPNN-ANFIS model 105.7 59 8 0.1675 1.1182
Error % 1.5 −3.5

BPNN parameters and interlayer details

Name Number No. MRR SR
parameter. Show 50 1st 9 9
Maximum parameter. Epochs 1000
Number of input parameter 3 2nd 8 10
Number of output parameter 1

Process parameters employed for EDM

Parameter Unit Factor Level

1 2 3
Pulse on time (μ sec) A 106 116 126
Pulse off time (μ sec) B 40 50 60
Wire feed (m/min) C 4 8 12

Regression ratio of surface roughness SR (μm)

Source DF Adj SS Adj MS F-Value P-Value
Regression 3 1.96406 0.65469 16.14 0.000
Pulse on time t on (μ sec) 1 1.76964 1.76964 43.62 0.420
Pulse off time t on (μ sec) 1 0.08050 0.08050 1.98 0.470
Wire feed 1 0.11392 0.11392 2.81 0.210
Error 23 0.93314 0.04057
Total 26 2.89719

Chemical composition of alloy 20 (in wt%)

Si S P Mn C Nb Mo Cu Cr Fe Ni
1 0.035 0.045 2 0.07 1 3 4 21 29.8 38

Ablyaz, T.R. and Belinin, D.S., 2014. Wire electrical discharge machining of items after plasmatic surface hardening. Middle-East Journal of Scientific Research, 19(8), pp.1096–1098. AblyazT.R. BelininD.S. 2014 Wire electrical discharge machining of items after plasmatic surface hardening Middle-East Journal of Scientific Research 19 8 1096 1098 Search in Google Scholar

Altug, M., Erdem, M. and Ozay, C., 2015. Experimental investigation of kerf of Ti6Al4V exposed to different heat treatment processes in WEDM and optimization of parameters using genetic algorithm. The International Journal of Advanced Manufacturing Technology, 78(9–12), pp.1573–1583. AltugM. ErdemM. OzayC. 2015 Experimental investigation of kerf of Ti6Al4V exposed to different heat treatment processes in WEDM and optimization of parameters using genetic algorithm The International Journal of Advanced Manufacturing Technology 78 9–12 1573 1583 10.1007/s00170-014-6702-x Search in Google Scholar

Azam, M., Jahanzaib, M., Abbasi, J.A., Abbas, M., Wasim, A. and Hussain, S., 2016. Parametric analysis of recast layer formation in wire-cut EDM of HSLA steel. The International Journal of Advanced Manufacturing Technology, 87(1–4), pp.713–722. AzamM. JahanzaibM. AbbasiJ.A. AbbasM. WasimA. HussainS. 2016 Parametric analysis of recast layer formation in wire-cut EDM of HSLA steel The International Journal of Advanced Manufacturing Technology 87 1–4 713 722 10.1007/s00170-016-8518-3 Search in Google Scholar

Bosheh, S.S. and Mativenga, P.T., 2006. White layer formation in hard turning of H13 tool steel at high cutting speeds using CBN tooling. International Journal of Machine Tools and Manufacture, 46(2), pp.225–233. BoshehS.S. MativengaP.T. 2006 White layer formation in hard turning of H13 tool steel at high cutting speeds using CBN tooling International Journal of Machine Tools and Manufacture 46 2 225 233 10.1016/j.ijmachtools.2005.04.009 Search in Google Scholar

Çaydaş, U., Hasçalık, A. and Ekici, S., 2009. An adaptive neuro-fuzzy inference system (ANFIS) model for wire-EDM. Expert Systems with Applications, 36(3), pp.6135–6139. ÇaydaşU. HasçalıkA. EkiciS. 2009 An adaptive neuro-fuzzy inference system (ANFIS) model for wire-EDM Expert Systems with Applications 36 3 6135 6139 10.1016/j.eswa.2008.07.019 Search in Google Scholar

Choi, K.K., Nam, W.J. and Lee, Y.S., 2008. Effects of heat treatment on the surface of a die steel STD11 machined by W-EDM. Journal of materials processing technology, 201(1–3), pp.580–584. ChoiK.K. NamW.J. LeeY.S. 2008 Effects of heat treatment on the surface of a die steel STD11 machined by W-EDM Journal of materials processing technology 201 1–3 580 584 10.1016/j.jmatprotec.2007.11.156 Search in Google Scholar

Dhobe, M.M., Chopde, I.K. and Gogte, C.L., 2013. Effect of heat treatment and process parameters on surface roughness in wire electro discharge machining. Int. J. Mech. Eng. & Rob. Res, 2(2), pp.275–281. DhobeM.M. ChopdeI.K. GogteC.L. 2013 Effect of heat treatment and process parameters on surface roughness in wire electro discharge machining Int. J. Mech. Eng. & Rob. Res 2 2 275 281 Search in Google Scholar

Dhobe, M.M., Chopde, I.K. and Gogte, C.L., 2013. Investigations on surface characteristics of heat treated tool steel after wire electro-discharge machining. Materials and manufacturing processes, 28(10), pp.1143–1146. DhobeM.M. ChopdeI.K. GogteC.L. 2013 Investigations on surface characteristics of heat treated tool steel after wire electro-discharge machining Materials and manufacturing processes 28 10 1143 1146 10.1080/10426914.2013.822976 Search in Google Scholar

Garg, R.K., Singh, K.K., Sachdeva, A., Sharma, V.S., Ojha, K. and Singh, S., 2010. Review of research work in sinking EDM and WEDM on metal matrix composite materials. The International Journal of Advanced Manufacturing Technology, 50(5–8), pp.611–624. GargR.K. SinghK.K. SachdevaA. SharmaV.S. OjhaK. SinghS. 2010 Review of research work in sinking EDM and WEDM on metal matrix composite materials The International Journal of Advanced Manufacturing Technology 50 5–8 611 624 10.1007/s00170-010-2534-5 Search in Google Scholar

Kapoor, J., Khamba, J.S. and Singh, S., 2012. The effect of machining parameters on surface roughness and material removal rate with cryogenic treated wire in WEDM. International Journal of Machining and Machinability of Materials 2, 12(1–2), pp.126–141. KapoorJ. KhambaJ.S. SinghS. 2012 The effect of machining parameters on surface roughness and material removal rate with cryogenic treated wire in WEDM International Journal of Machining and Machinability of Materials 2 12 1–2 126 141 10.1504/IJMMM.2012.048562 Search in Google Scholar

Khan, A.A., Ali, M.B.M. and Shaffiar, N.B.M., 2006. Relationship of surface roughness with current and voltage during wire EDM. Journal of applied sciences, 6(10), pp.2317–2320. KhanA.A. AliM.B.M. ShaffiarN.B.M. 2006 Relationship of surface roughness with current and voltage during wire EDM Journal of applied sciences 6 10 2317 2320 10.3923/jas.2006.2317.2320 Search in Google Scholar

Manevski, K., Børgesen, C.D., Li, X., Andersen, M.N., Abrahamsen, P., Hu, C. and Hansen, S., 2016. Integrated modelling of crop production and nitrate leaching with the Daisy model. MethodsX, 3, pp.350–363. ManevskiK. BørgesenC.D. LiX. AndersenM.N. AbrahamsenP. HuC. HansenS. 2016 Integrated modelling of crop production and nitrate leaching with the Daisy model MethodsX 3 350 363 10.1016/j.mex.2016.04.008486563527222825 Search in Google Scholar

Marafona, J.D. and Araujo, A., 2009. Influence of work-piece hardness on EDM performance. International Journal of Machine Tools and Manufacture, 49(9), pp.744–748. MarafonaJ.D. AraujoA. 2009 Influence of work-piece hardness on EDM performance International Journal of Machine Tools and Manufacture 49 9 744 748 10.1016/j.ijmachtools.2009.03.002 Search in Google Scholar

Morotti, S. and Grandi, E., 2017. Logistic regression analysis of populations of electrophysiological models to assess proarrythmic risk. MethodsX, 4, pp.25–34. MorottiS. GrandiE. 2017 Logistic regression analysis of populations of electrophysiological models to assess proarrythmic risk MethodsX 4 25 34 10.1016/j.mex.2016.12.002 Search in Google Scholar

Nain, S.S., Garg, D. and Kumar, S., 2018. Investigation for obtaining the optimal solution for improving the performance of WEDM of super alloy Udimet-L605 using particle swarm optimization. Engineering science and technology, an international journal, 21(2), pp.261–273. NainS.S. GargD. KumarS. 2018 Investigation for obtaining the optimal solution for improving the performance of WEDM of super alloy Udimet-L605 using particle swarm optimization Engineering science and technology, an international journal 21 2 261 273 10.1016/j.jestch.2018.03.005 Search in Google Scholar

Newton, T.R., Melkote, S.N., Watkins, T.R., Trejo, R.M. and Reister, L., 2009. Investigation of the effect of process parameters on the formation and characteristics of recast layer in wire-EDM of Inconel 718. Materials Science and Engineering: A, 513, pp.208–215. NewtonT.R. MelkoteS.N. WatkinsT.R. TrejoR.M. ReisterL. 2009 Investigation of the effect of process parameters on the formation and characteristics of recast layer in wire-EDM of Inconel 718 Materials Science and Engineering: A 513 208 215 10.1016/j.msea.2009.01.061 Search in Google Scholar

Prasad, K.S., Rao, C.S. and Rao, D.N., 2011. Establishing empirical relationships to predict grain size and hardness of pulsed current microplasma arc welded Inconel 625 sheets. Journal of Materials & Metallurgical Engineering, 1(3), pp.1–10. PrasadK.S. RaoC.S. RaoD.N. 2011 Establishing empirical relationships to predict grain size and hardness of pulsed current microplasma arc welded Inconel 625 sheets Journal of Materials & Metallurgical Engineering 1 3 1 10 Search in Google Scholar

Rao, P.S., Ramji, K. and Satyanarayana, B., 2011. Effect of WEDM conditions on surface roughness: A parametric optimization using Taguchi method. International journal of advanced engineering sciences and technologies, 6(1), pp.41–48. RaoP.S. RamjiK. SatyanarayanaB. 2011 Effect of WEDM conditions on surface roughness: A parametric optimization using Taguchi method International journal of advanced engineering sciences and technologies 6 1 41 48 Search in Google Scholar

Rao, T.B. and Krishna, A.G., 2013. Simultaneous optimization of multiple performance characteristics in WEDM for machining ZC63/SiC p MMC. Advances in Manufacturing, 1(3), pp.265–275. RaoT.B. KrishnaA.G. 2013 Simultaneous optimization of multiple performance characteristics in WEDM for machining ZC63/SiC p MMC Advances in Manufacturing 1 3 265 275 10.1007/s40436-013-0029-y Search in Google Scholar

Bharathi, R.S., Shanmugam, N.S., Kannan, R.M. and Vendan, S.A., 2018. Studies on the Parametric Effects of Plasma Arc Welding of 2205 Duplex Stainless Steel. High Temperature Materials and Processes, 37(3), pp.219–232. BharathiR.S. ShanmugamN.S. KannanR.M. VendanS.A. 2018 Studies on the Parametric Effects of Plasma Arc Welding of 2205 Duplex Stainless Steel High Temperature Materials and Processes 37 3 219 232 10.1515/htmp-2016-0087 Search in Google Scholar

Sajjadi, S.A., Zolfaghari, G., Adab, H., Allahabadi, A. and Delsouz, M., 2017. Measurement and modeling of particulate matter concentrations: applying spatial analysis and regression techniques to assess air quality. MethodsX, 4, pp.372–390. SajjadiS.A. ZolfaghariG. AdabH. AllahabadiA. DelsouzM. 2017 Measurement and modeling of particulate matter concentrations: applying spatial analysis and regression techniques to assess air quality MethodsX 4 372 390 10.1016/j.mex.2017.09.006 Search in Google Scholar

Sidhu, S.S., Batish, A. and Kumar, S., 2013. Neural network–based modeling to predict residual stresses during electric discharge machining of Al/SiC metal matrix composites. Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture, 227(11), pp.1679–1692. SidhuS.S. BatishA. KumarS. 2013 Neural network–based modeling to predict residual stresses during electric discharge machining of Al/SiC metal matrix composites Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture 227 11 1679 1692 10.1177/0954405413492505 Search in Google Scholar

Singh, R. and Singh, B., 2011. Comparison of cryo-treatment effect on machining characteristics of titanium in electric discharge machining. International Journal of Automotive and Mechanical Engineering, 3, pp.239–48. SinghR. SinghB. 2011 Comparison of cryo-treatment effect on machining characteristics of titanium in electric discharge machining International Journal of Automotive and Mechanical Engineering 3 239 48 10.15282/ijame.3.2011.1.0020 Search in Google Scholar

Sone, T. and Masui, K., 1991. Application of ion nitriding to wire-electrical-discharge-machined blanking dies. Materials Science and Engineering: A, 140, pp.486–493. SoneT. MasuiK. 1991 Application of ion nitriding to wire-electrical-discharge-machined blanking dies Materials Science and Engineering: A 140 486 493 10.1016/0921-5093(91)90467-2 Search in Google Scholar

Srinivasan, R.G., Selvabharathi, R., Palani, S. and Karuppasamy, R., 2019. Influence of high-velocity oxygen fuel spraying and plasma nitriding on microstructure properties of iron-nickel-chromium alloy using hybrid surface heat treatment. Materials Research Express, 6(8), p.086584. SrinivasanR.G. SelvabharathiR. PalaniS. KaruppasamyR. 2019 Influence of high-velocity oxygen fuel spraying and plasma nitriding on microstructure properties of iron-nickel-chromium alloy using hybrid surface heat treatment Materials Research Express 6 8 086584 10.1088/2053-1591/ab1eee Search in Google Scholar

Tarng, Y.S., Ma, S.C. and Chung, L.K., 1995. Determination of optimal cutting parameters in wire electrical discharge machining. International Journal of Machine Tools and Manufacture, 35(12), pp.1693–1701. TarngY.S. MaS.C. ChungL.K. 1995 Determination of optimal cutting parameters in wire electrical discharge machining International Journal of Machine Tools and Manufacture 35 12 1693 1701 10.1016/0890-6955(95)00019-T Search in Google Scholar

Warren, A.W., Guo, Y.B. and Weaver, M.L., 2006. The influence of machining induced residual stress and phase transformation on the measurement of subsurface mechanical behavior using nanoindentation. Surface and Coatings Technology, 200(11), pp.3459–3467. WarrenA.W. GuoY.B. WeaverM.L. 2006 The influence of machining induced residual stress and phase transformation on the measurement of subsurface mechanical behavior using nanoindentation Surface and Coatings Technology 200 11 3459 3467 10.1016/j.surfcoat.2004.12.028 Search in Google Scholar

Williams, R.E. and Rajurkar, K.P., 1991. Study of wire electrical discharge machined surface characteristics. Journal of materials processing technology, 28(1–2), pp.127–138. WilliamsR.E. RajurkarK.P. 1991 Study of wire electrical discharge machined surface characteristics Journal of materials processing technology 28 1–2 127 138 10.1016/0924-0136(91)90212-W Search in Google Scholar

Xiong, G., Zheng, M., Li, D., Zhao, L., Wang, Y. and Li, M., 2010, October. Study on the influence of non-electrical parameters on processing quality of WEDM-HS and improvement measures. In International Conference on Computer and Computing Technologies in Agriculture (pp. 514–520). Springer, Berlin, Heidelberg. XiongG. ZhengM. LiD. ZhaoL. WangY. LiM. 2010 October. Study on the influence of non-electrical parameters on processing quality of WEDM-HS and improvement measures In International Conference on Computer and Computing Technologies in Agriculture 514 520 Springer Berlin, Heidelberg 10.1007/978-3-642-18369-0_61 Search in Google Scholar

Recommended articles from Trend MD

Plan your remote conference with Sciendo