Performance Prediction in EDM Process for Al 6061 Alloy Using Response Surface Methodology and Genetic Algorithm

The Electric Discharge (EDM) method is a novel thermoelectric manufacturing technique in which materials are removed by a controlled spark erosion process between two electrodes immersed in a dielectric medium. Because of the difficulties of EDM, determining the optimum cutting parameters to improve cutting performance is extremely tough. As a result, optimizing operating parameters is a critical processing step, particularly for non-traditional machining process like EDM. Adequate selection of processing parameters for the EDM process does not provide ideal conditions, due to the unpredictable processing time required for a given function. Models of Multiple Regression and Genetic Algorithm are considered as effective methods for determining the optimal processing variables of Electrical Discharge Machining. The material removal rate (MRR) and tool wear (Tw) were investigated using the process variables of pulse on time (Ton ), pulse off time (Toff) , and current intensity (Ip) . The established empirical models were used to perform Genetic Algorithm (GA) to maximize (MRR) and minimize (Tw). The optimization results were utilized to establish machining conditions, validate empirical models, and obtain optimization outcomes. The optimal result that appears in this work was the pulse on (176.261 μ s), pulse off (39.42 μ s), and current intensity (23.62 Amp.) to maximize the MRR to (0.78391 g/min) and reduce tool wear to (0.0451 g/min).


Introduction
Machining Aluminum alloy using traditional machining technologies has problems such as high cutting temperatures and a high tool wear ratio. Aluminum is employed in a variety of industries, including automobiles, and aerospace. Aluminum alloy, on the other hand, has some advantages due to its low cost, low density, availability, and manufacturability [1].
AA 6061 Aluminum alloy is a precipitationhardened variant of the 6000 series Al alloys that are widely used. It's a heat-treatable extruded alloy with medium to high strength properties [2]. Electrical discharge machining (EDM) is a non-traditional machining technique that is commonly used to machine die surfaces [3]. EDM is a popular production technique for difficult-to-machine materials and complex geometries. Wire and electrode EDM are the two primary types of EDM processes. The manufacturing process setup includes the electrode material, the geometry of wire diameter or the electrode, and the energy transfer parameters of voltage V with its polarity, pulse current intensity I, and pulse on time (Ton) and pulse off time (Toff) [4] [5]. Several authors attempted to machine several materials by utilizing the electrical discharge machining process. S. Ranjith et al. [6] examined the influence of EDM machining variables (pulse-on duration, current, pulse-off duration, and spark gap voltage) on MRR and Tw of silicon nitride-titanium nitride ceramic composites with the copper electrode. From the results, it has been shown that the current is a highly important factor among other parameters on both MRR and TWR. Higher material removal rate is obtained when pulse-on time and current is higher, whereas lower EWR result from high gap voltage and low current. Huu-Phan et al. [7] applied Multi-response optimization based on Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) to examine the impact of lowfrequency vibration on material removal rate MRR and Surface Roughness Ra in EDM process. From the experimental results, in the low-frequency vibration aided EDM technique, the process performance accuracy has been enhanced to around 86.6 percent. By enhancing the quality of the machining surfaces, TOPSIS was able to improve the low-frequency vibration-assisted EDM process' performance. The material removal rate can be increased with low-frequency vibration due to the controlled spark energy. Mandeep and Sthitapragyan [8] examined the effects of machining parameters such as pulse-on time (T-on), pulse-off time (Toff), current (I), and voltage (V) on the MRR of Aluminum based composite material. The aluminum metal matrix composite was cast with 200 mesh size (Avg. size 75 mm) particles (20 percent SiCp and 8% Grp). Response surface methodology RSM created the design matrix and mathematical models. The experimental results indicate that the pulse-on time and current are both major factors that directly affected (increased) the material removal rate, according to the analysis. Finally, at a high level of "pulse-off time," the MRR is minimal, but "voltage" has no significant effect on MRR. Mandeep et al. [9] determined the optimum process parameters (Peak current (I), voltage (V), Ton, and tool material) that affect the MRR and Ra of EDM during the machining of a new hybrid aluminum metal matrix composite. From the analysis of variance (ANOVA), the MRR increased as the Ton and pulse current increased, however, the MRR declined as the voltage increased. Reduced Ra, on the other hand, could only be achieved with low I, V, and pulse length. It was also revealed that the electrodes used in EDM had a substantial impact on MRR and surface roughness. Chinmayee et al. [10] investigated the influence of input parameters (discharge current, pulse-on-time, and open circuit voltage) on MRR, Tw and radial overcut of AA 7075 aluminum-6% red mud metal matrix composites (AMMC) in the EDM process. From the results, it has been observed that pulse on-time and discharge current have a crucial influence on machinability characteristics of (AMMC) by providing useful information with less deviation to improve the accuracy of the EDM parts. Ramanuj et al. [11] carried out an experimental investigation to study the effect of (voltage, pulse on time, and current) on the (Ra) and (MRR) of Ti-6Al-4V ELI using EDM. Multi response Grey Relation Analysis (GRA) technique has been utilized for optimizing the machining parameters. From the results, it has been presented that the MRR and Ra were directly proportional to discharge current. Sagar and Pravin [12] provided an experimental investigation on the effect of (pulse on duration, discharge voltage, capacitance, and electrode rotation speed) on side gap width, MRR, and taper ratio when drilling Titanium alloy (Ti6Al4V) with copper tungsten (CuW) electrode. The experimental results demonstrate that capacitance, discharge voltage, and electrode rotation speed all have an impact on MRR parameters, whereas pulse on time, capacitance, and electrode rotation speed influence side gap width. On the other hand, capacitance and pulse on duration were the affecting parameters on the taper ratio. R.Rajesh and M. Dev Anand [13] calculated the optimum operating parameters namely; working voltage, oil pressure, spark gap, Pulse On Time, and Pulse Off Time, affecting the MRR by developing genetic algorithm and multiple regression models with an empirical model by conducting experiments based on the Grey Relational Analysis, which were used to obtain the greatest value for MRR in electric discharge machine. Tzeng et al. [14] analyzed MRR, electrode wear ratio, and workpiece surface finish on process parameters during the manufacture of SKD61 by electrical discharge machining (EDM). A hybrid method including a back-propagation neural network (BPNN), a genetic algorithm (GA), and response surface methodology (RSM) were proposed to determine optimal parameter settings of the EDM process.
In this paper, the influence of EDM parameters (pulse on time Ton (μs), pulse off time Toff (μs) and current intensity Ip (A)) was studied to determine their influence on the Material Removal Rate (MRR) and Tool Wear Rate (Tw) values based on Response Surface Methodology (RSM) and Genetic algorithm.

Experimental Work
A CHEMER EDM machine model (CM-323 C) as shown in Figure 1, is used to implement the experimental work.

Work piece material
For evaluating the optimum values of process variables, a Copper electrode was used to Machine twenty samples of AA 6061 Aluminum alloy with dimensions (15×15×5mm) as shown in Figure 2. Table 1 illustrates the Chemical Composition of AA 6061 Alloy.

Selection parameters and their levels
Process variables are the parameters that influence Material Removal Rate (MRR) and Tool Wear Rate (Tw) of machined surface and include the current intensity (Ip), pulse on time (Ton), and pulse off time (Toff). Table 2 shows the parameters values and their levels that were used in the experiments.

Design and Analysis of Experimental Work
To find which input parameters generate the optimum output and to identify the influence of input parameters that enhance the developed qualities of the part above the obtained qualities, RSM was adopted to in this study to develop statistical and mathematical models due to its reliable performance [15,16]. The central composite design is the most frequent way of building a quadratic model of a response surface (CCD). Experiments are conducted with RSM and a Central Composite Design (CCD) matrix with Two-level factorial (full factorial) where it consists of 8 cube points, 6 center points, and 6 axial points with α=1.68179.
Experimental design is an important stage in creating a response surface model using MINITAB software [15]. Material Removal Rate (MRR) and Tool Wear Rate (Tw) are calculated from equations (1,2) [17,18] after experimentation, as shown in Table 3.

Analysis of Variance
On the basis of Table 3

Data Means
It is clear from the main effects plot of Figure 3 that the material removal rate highly increases with the increase of current intensity, and decreases with the increase in pulse on time and pulse off time, the reason behind this is that the discharge energy increases with the increase of pulse on time and peak current leading to a faster cutting rate. With the decrease in the pulse off time, the number of discharges within a given period becomes more which leads to a higher material removal rate.
To depict the input variables relationship between (Ton, Toff, and Ip) and the output ( Table 4 shows the overall significance of the mathematical model, with (R-Sq) determining the fit value between predicted and experimental findings. The (R-Sq(adj)) value indicates that the independent variables (Ton, Toff, and Ip) recorded (98.23) percent of the variance in the dependent variable (Y), with the remainder due to random error. It is also clear from the main effects plot of Figure  4 that the most influencing factor on tool wear rate is the peak current; tool wear rate is minimum at lower currents.
To depict the input parameters relationship between (Ton, Toff, and Ip) and the output  Table 5 shows the overall significance of the mathematical model, with (R-Sq) determining the fit value between predicted and experimental findings. The (R-Sq(adj)) value indicates that the independent variables (Ton, Toff, and Ip) recorded (75.22) percent of the variance in the dependent variable (Y), with the remainder due to random error. Figure 4 shows the Tw Main Effects Plot.

GA Results
Genetic algorithm is a probabilistic search technique that generates a new population from an iterative collection (called a population) of mathematical objects (typically fixed-length binary character Strings), each with a fitness value [19]. The genetic algorithm analyzes the experimental and expected values by using the intersection technique and the best value is supplied in the schedule below. The best results were obtained after choosing the fitness function for a total length of the string of 18 and then transferring the results to MATLAB after setting up GA parameters and presenting them in Table 6.  Providing better reproductive opportunities through offspring gives more possible solutions so Table 7 shows that the increase in the crossover value caused improvement in the results until it reaches the optimal values, where reading No (15)

Conclusions
This research presents a realistic method for optimizing EDM cutting parameters based on GA. The machining parameters included pulse on, pulse off, and current intensity Ip. The EDM method yields results such as metal removal rate and tool wear.
The response surface method (RSM) uses statistical and mathematical methods for issue modeling and analysis to locate the input variables that generate the optimum response. Finally, GA was able to determine the best circumstances. That is, between experimental data, pulse on (176.261 μs), pulse off (39.42 μ s), current intensity (23.62 Amp.) to maximize the MRR to (0.78391 g/min) and reduce tool wear to (0.0451 g/min). The machining current of the EDM process is the most influential factor revealed by the response