Cryogenic treatment analysis in WEDM of Al2024/Al 2 O 3 /W composite

Increasing demand for aluminum matrix composites (AMCs) in the manufacturing sector leads to the development of squeeze casted hybrid AMCs bearing superior mechanical attributes. Excessive hardness due to squeeze pressure and the presence of reinforcement particles makes the machining of AMCs challenging and results in excessive tool wear and inadequate surface finish. Wire electric discharge machining (WEDM) process has been preferred over conventional manufacturing processes for the machining of squeeze casted Al2024/Al 2 O 3 /W hybrid composite. This study investigates the selection of the most pertinent wire electrode for the precise machining of Al2024/Al 2 O 3 /W. Simple zinc-coated (NT) and cryogenic treated (CT) brass wire electrodes have been selected for machining. Scanning electron microscope (SEM) analysis of CT wire shows that cryogenic treatment enhances the soundness and stability by reducing microcavities in its microstructure. To analyze the effects of cryogenic treatment, the machining performance of CT wire has been compared with NT wire. Besides wire type, the influence of key input variables has also been analyzed on the imperative response measures including cutting speed (CS), surface roughness (SR) and kerf width (KW) to make the machining process more effective. Microstructural analysis of NT wire depicts a high Comparative analysis of both wire electrodes has declared that CT wire yields 26.96% and 15.10% superior results for CS and SR respectively, and 6.92% deprived results for KW than NT wire.


Introduction
In engineering materials, aluminum is the most alluring material due to its high strength, low weight, formability, recyclability, and corrosion resistance capability. Inadequate wear resistance and deprived performance at elevated temperatures are major concerns of aluminum alloys [1]. These deficiencies are generally overcome by introducing reinforced particles in such alloys to produce aluminum matrix composites (AMCs) [2][3][4][5]. Reinforced particles in the metal matrix improve the mechanical characteristics of AMCs by modifying the microstructural attributes. These superior properties of AMCs have enhanced their demand in marine, automotive, defense, and aerospace applications [6]. AMCs are generally fabricated by stir casting process, in which preheated reinforced particles are introduced in the matrix material and mixed uniformly with a mechanical stirrer [7]. Nevertheless, researchers have successfully boosted the mechanical properties of AMCs via the squeeze casting process [8].
Although the addition of reinforced particles improves the wear resistance and other mechanical properties of AMCs [9], the machining of such materials is difficult by traditional means [10,11]. Reinforcement in AMCs has abrasive nature and retains a drastic threat to the tool life. Abrasion during machining erodes the tool material at a faster rate and causes early failure of the tool [12]. Moreover, reinforced particles are pulled out during machining and act as an abrasive cutting edge which deteriorates the surface finish of work material [13,14]. For this reason, wire electric discharge machining (WEDM) has been observed as an attractive substitute for the machining of AMCs because of no physical contact between tool and workpiece. In this machining process, workpiece material is eradicated by the series of electric sparks between tool and workpiece [15,16]. These electric sparks raise the temperature of the tool and workpiece interface up to 12000°C which melts and evaporates workpiece material [16]. A thin wire of molybdenum, tungsten, brass, or copper having a diameter of 0.05-0.3mm is typically used as a tool/electrode. Consequently, narrow kerfs are generated in the workpiece as compared to the conventional machining process which is also a perk of the WEDM process [17]. Dielectric fluid is generally used to maintain the insulating medium between electrode and workpiece. It also helps in flushing the eroded material from the machining zone due to high pressure [18]. Generally, the parts manufactured by composites or extremely hardened materials bearing intricate contours are machined precisely by the WEDM process with superior surface quality [19].
Keeping in view the demand in the manufacturing sector, researchers are focusing on the analysis of the working/machining behavior of hybrid AMCs due to their distinct properties. Therefore, several studies have been conducted to investigate the parametric effects on machining characteristics of various hybrid AMCs fabricated by different techniques. Muniappan et al., [20] described the effects of cutting parameters including V, I, TON, TOFF, TW, and FR on the CS of Al6061/SiC/graphite hybrid AMC prepared by the stir casting process. TOFF followed by TON, V and I was observed as a significant parameter. Satishkumar and Kanthababu [21] compared the machined surface quality of simple Al7075 alloy and Al7075/B4C/Al2O3 hybrid AMC fabricated by the stir casting process. It was noticed that simple Al7075 alloy at high TOFF and low TON values generate comparatively less SR. Ugrasen et al., [22] investigated the influence of process parameters (I, TON, TOFF, and bed speed) on the machining of Al2024/TiC/fly-ash hybrid AMC synthesized by stir casting technique. At lower values of TOFF and bed speed and higher values of I and TON, improved surface finish and dimensional accuracy were attained. Nag et al., [23] analyzed the impact of process parameters on the machining performance of A359/Al2O3/B4C hybrid AMC produced by electromagnetic stir casting process. Analysis of the material removal rate and SR values showed TON as the most significant input variable. Kumar et al., [24] employed Al6351/SiC/B4C hybrid AMC prepared by stir casting technique for the parametric optimization in the WEDM process. Their findings revealed that concentration of boron carbide particles has a significant impact on SR. Prakash et al., [25] examined the influence of V, TON, TOFF, FR, and concentration of two distinct reinforced particles on material removal rate of Al356/B4C/fly-ash hybrid AMC manufactured by the stir casting process. Their results declared that V and TON have considerable impact and concentration of reinforced particles has a negligible effect on material removal rate. Lal et al., [26] fabricated AA7075/SiC/Al2O3 hybrid AMC by inert gas-assisted electromagnetic stir casting technique to analyze the effects of input variables (TON, TOFF, I, and SW) on KW of AMC. From the literature, it has been observed that the machining performance of composites also depends upon the fabrication technique of the composite material [27]. Unfortunately, the machining of squeeze casted hybrid AMCs is scarcely reported.
To improve the machinability of hardened materials, cutting tools are treated cryogenically before machining. In cryogenic treatment, the material is kept at a very low temperature for a specific period resulting in improved electrical properties which are primarily imperative in electric discharge machining [28]. Superior electrical properties of electrode yield high material removal rate and surface finish [29]. Besides this, tool life and wear resistance of the alloys are also significantly enhanced with appropriate cryogenic treatment [30]. In this type of treatment, cooling time, soaking rate and temperature are the critical parameters that alter the properties of alloys [31][32][33]. Researchers have tried to optimize these parameters to achieve the desired material characteristics. Three particular sub-zero temperature ranges: (i) 223~193K (cold/cryogenic treatment), (ii) 193~113K (shallow cryogenic treatment) and (iii) 113~77K (deep cryogenic treatment) are mainly used for cryogenic treatment [34].
Tahir et al., [17] analyzed the effects of wire type (cold treated brass wire and non-cold treated brass wire) along with other input variables during machining HSLA steel. In wire type, cold treated brass wire yielded better surface finish and CS as compared to non-cold treated brass wire. Kapoor et al., [35] employed shallow cryogenic treated and non-treated brass wire for the machining of EN-31 steel. Their analysis indicates that shallow cryogenic treated wire has significantly reduced the SR. Kapoor et al., [36] compared the machining performance of non-cryogenic treated and deep cryogenic treated brass wire electrodes while cutting En-31 steel. Machining with deep cryogenic treated brass wire resulted in higher material removal rate than non-cryogenic treated brass wire. Nayak and Mahapatra [37] used deep cryogenic treated wire (Bronco cut-W wire) and workpiece material (Inconel 718) to improve the machining performance evaluated by the results of angular error, SR, and CS. Goyal [38] investigated the effects of NT and CT wire in the machining of Inconel 625. In their results, CT wire offered better results for material removal rate and SR. Previous studies have presented the cryogenically treated electrodes as the best electrode alternative for the machining of hardened alloys. While machining of squeeze casted AMCs with cryogenically treated wire electrode is not adequately addressed in the literature.
From the comprehensive literature review, it has been reported that researchers have used AMCs prepared by various techniques in the WEDM process. However, there is a need to explore the machining of squeeze casted hybrid AMCs by the WEDM process. Owing to excessive hardness and high amount of reinforcement particulates, machining of squeeze casted AMCs resulted in high electrode wear and SR [39]. Therefore, this research has been conducted to determine the appropriate wire electrode for the precise machining of the Al2024/Al2O3/W squeeze casted hybrid composite. As the material is squeeze casted, so, it is also desired to identify the optimum set of input variables along with the selection of the best suitable wire type. For this reason, four input variables including pulse duration, wire tension, wire feed rate, and wire runoff speed have been used to analyze the effects of input variables on cutting speed, surface roughness, and kerf width. To investigate the appropriate wire type for the machining of Al2024/Al2O3/W, cryogenically treated moly wire, simple moly wire, cryogenic treated zinc-coated brass wire, and simple zinc-coated brass wire electrodes have been selected. Box Behnken design of RSM has been adopted for experimental design and ANOVA has been performed for analysis of the effects of input variables on response measures. For multi-objective optimization, grey relational analysis has been applied in this study.

Fabrication of Al2024/Al2O3/W through squeeze casting technique
To analyze the effects of wire type and input variables on the machining performance of hybrid composite, Al2024/Al2O3/W has been fabricated by the squeeze casting process. The chemical composition of the matrix material used in the preparation of hybrid AMC has been given in Table 1. For the desired assortment of mechanical properties of composite, Al2O3 (3wt%) and W (1.5wt%) particles are used for the preparation of hybrid AMC having an average particle size of 10 nm and 200 nm respectively. Squeeze casting setup used for the preparation of hybrid AMC specimens has been displayed in Fig. 1. Figure 1. Hybrid AMC specimens' preparation through the squeeze casting process and its schematic illustration  Fig. 2(a). It can be seen that nanoparticles are distributed uniformly throughout the metal matrix, however, some clusters may exist due to high squeeze pressure. Microstructure of the composite material has also been analyzed on a higher resolution to visualize the dispersion of smaller sized Al2O3 particles (Fig. 2b).

Cryogenic treatment of wire electrodes
For the machining of the squeeze casted Al2024/Al2O3/W hybrid composite, two distinct wires including molybdenum and zinc-coated brass wires of diameter 0.3 mm have been selected. To enhance the electrical and mechanical properties, selected wires have been cryogenically treated using a temperature-controlled electrical chamber (CTT-SC-7520-02FI). Soaking of both wires has been performed at -70°C for 24h employing a ramp rate of 2°C/min. After cryogenic treatment, electrical conductivities of treated and non-treated wires have been tested on a portable Kelvin Bridge tester. UTS of selected wires has been measured on a universal testing machine Tungsten particles Aluminum oxide particles as per E8 ASTM standards. Results of electrical conductivity and UTS of tested wires are given in Table 2. It can be seen that UTS of molybdenum wire has increased to 1765 MPa while its electrical conductivity has been reduced from 2.08×10⁷ to 2.02×10⁷ S/m. In comparison with UTS, electrical conductivity is a fundamental property of the electrode for electric discharge machining. Reduced electrical conductivity of the electrode results in insufficient electrical spark density which effects the material removal process. This insufficient electrical conductivity declines CS and produces non-uniform and ineffective pulses which eventually generate poor surface finish [39]. Therefore, cryogenically treated molybdenum wire seems to be inappropriate for electric discharge machining hybrid AMC. On the other hand, the UTS of zinc-coated brass wire has been slightly decreased from 727 to 686 MPa. However, the electrical conductivity of cryogenic treated zinc-coated brass wire has been raised up to 15.6 × 10⁶ S/m with a percentage increment of 24.8%. Therefore, zinc-coated brass wires have been observed as an attractive option for machining. A framework has been developed for the selection of the best suitable wire electrode for the machining of the squeeze casted Al2024/Al2O3/W hybrid composite shown in Fig. 3. The machining performance of the wire electrode has been analyzed on the basis of CS, SR, and KW. Keeping wires' treatment results in view, cryogenic treated zinc-coated brass wire has been selected for the machining of composite material.  To have a deeper insight into the effects of cryogenic treatment, the microstructure of simple zinc-coated and cryogenic treated wires has been analyzed on a scanning electron microscope (SEM). The micrograph of NT wire reveals the presence of microcavities and pores in its structure (Fig. 4a). These microcavities offer hindrance in the flow of electrons which declines thermal and electrical conductivity. While the cryogenic treatment has increased the soundness and stability by reducing micro-cavities can be seen in Fig. 4b. As temperature decreases, atomic bonding in brass start to abate, and crystal structure regress towards its original phase. Residual stresses in the brass wire structure are also minimized in cryogenic treatment [40]. Machining with CT wire depicts the intense electric discharge in the machining zone due to improved thermal and electrical conductivity as compare to NT wire. The same phenomenon of the machining of AMC with NT and CT wires are shown in Fig. 5a

Experimentation details
To investigate the effects of wire type and input variables on the machining of hybrid AMC a series of experiments has been conducted on WEDM. In wire type, cryogenic treated and simple zinc-coated brass wires have been selected for experimentation. Besides wire type, effects of input variables including pulse duration, wire tension, wire feed rate, and wire runoff speed on cutting speed, surface roughness, and kerf width have also been analyzed.
Pilot experimentation has been conducted to explore the appropriate ranges of input variables for machining of Al2024/Al2O3/W hybrid composite. Selected ranges of input variables have been given in Table 3. Machining conditions other than input variables are provided in Table 4.

Experimental design 2.4.1 Response surface methodology
In previous studies, numerous statistical techniques including Factorial design [41], Taguchi method [42], Response surface methodology [43], Genetic algorithm [44], Fuzzy logic [45], and Artificial neural network [46] Wire electric discharge machining of AMCs hence, RSM is regarded as the most effective approach for analysis [47]. RSM is normally adopted to investigate the nature of the relationship between input variables and response measures. RSM describes this relationship as follows: where R is the response measure, f is the response surface, and TON, FR, Sw, and TW are the input variables.
Depending upon the nature of their relationship, RSM can use the second-order polynomial function for the empirical modeling due to its capability of analyzing the interaction effects of multiple variables. The quadratic regression model of R is exemplified below: where ao, ai, aij, and aii are the coefficients of constant, linear, interaction, and quadratic terms respectively.
Box Behnken design of RSM has been used for the design of experiments in this study. The design offers 30 number of experiments with 4 input variables. The complete design has been repeated for both cryogenic treated and simple zinc-coated brass wires separately, therefore, a total of 60 experiments have been conducted in this research. Results of CS, SR, and KW against each set of input variables for both wires are given in Table 5.

Results and discussions
This section covers the analysis of response measures by ANOVA, empirical modeling of individual response measures, analysis of the effects of input variables using 3D response surface graphs, microstructural analysis, multi-objective optimization using GRA, and performance comparison of NT and CT wire electrodes.

Main and interaction effects analysis
The effects of input variables on the response measures are presented in the form of 3D response surface graphs.
The 3D surface of the graph depicts the main as well as the interaction effects of two input variables simultaneously while keeping other variables constant at their mid-levels [48]. Contour lines on the base of the graph depict the reflection of 3D surface. Contour lines with distinct colors present the response values range from minimum to maximum.

Cutting speed
The effects of FR, TON, TW, and SW on the CS for both NT and CT wires are shown in Fig. 7a-d. It has been observed that CS increases with an increment in both TON and FR (Fig. 7a-b). The value of CS is smaller at the lower level of TON as discharge energy is provided for a shorter time span which enables the reinforced particles to protect the metal matrix from melting with their strong bonding. With an increase in TON, the discharge energy is supplied for more time to eradicate material from the work surface. This intense discharge energy melts and evaporates aluminum matrix easily and release the reinforced particles, this is the reason that a slight increase in TON raises the CS promptly. Similar trends of TON has been reported by Kavimani et al., [49]. High CS has been observed at a higher level of FR. As the increase in FR value raises the chance of new wire to expose and create clear and higher intensity sparks. This intense spark energy is responsible for material erosion by broadening and deepening the cutting zone [50].
Fig. 7c-d shows the effects of SW and TW on CS. It has been noticed that CS enhances by an increase in SW with a nonlinear trend. As increment in SW increases the rate of spark discharge which leads to the high amount of heat energy delivered to the cutting zone ultimately enhances the rate of melting and evaporation of work material [51]. From Fig. 7c-d, it has been observed that increase in TW improves the CS of composite material due to increased straightness of wire. While low CS has been reported at lower TW because wire vibrates and lags behind during cutting [52]. Within the suitable range of TW, CS increases with TW.
Comparatively analysis of the performance of both wire electrodes in Fig. 7 shows that CS of CT wire is high during the cutting of squeeze casted hybrid AMC. As in cryogenic treatment, the crystal structure of the electrode gets refines and makes the electron flow uniform and easier. The absence of any obstacles in the structure of electrode material enhances the electron flow per unit time which improves the electrical conductivity of the cryogenically treated material. Owing to the high electrical conductivity of cryogenically treated brass wire, a high amount of energy is supplied to the machining zone which melts and evaporates matrix material at a faster rate. Intense sparking is produced removes more material from the machining zone and the same phenomenon can be observed in Fig. 4b. It releases the reinforced particles from the matrix swiftly which increases the material removal rate and ultimately enhances the CS.

Surface roughness
Surface graphs for NT and CT wires presenting the effects of input variables on SR are shown in Fig. 8a-d. It is cleared from Fig. 8a-b that SR is maximum at a higher level of Ton and FR. Increase in feed rate, the wire gets renewed and recovered hastily during machining. This introduction of non-worn fresh wire apparently enhances the sparking efficiency which creates larger and deeper craters [50]. While the increase in TON increases the time duration of electric discharges which melts more workpiece and produces large-sized craters. These large-sized craters create irregularities on the machined surface eventually increase the SR [53]. As a result of micro-voids in the molten pool high SR is originated [54]. On the other hand, SR declines with an increase in TW which is actually due to the decrease in wire amplitude at higher tension. Due to high TW, the wire gets straightened and fluid begins to flow with higher speed around the wire which prevents the debris from adherence and keeps the workpiece surface clean. This ablution of the surface is responsible for low SR at higher TW [52].
From response surface graphs in Fig. 8, it has been observed that CT wire yields less SR as compared to NT wire.
As earlier discussion shows that CT wire has high electrical conductivity, which is also responsible for the reduction in SR. Because higher electrical conductivity of the electrode widens the discharge channel resulting in easier flushing of debris and released reinforced particles from the machining zone and keeps the machined surface cleaner. Moreover, this enlarged discharged channel distributes the spark energy uniformly in all orientations which produce small and uniform sized craters on the machined surface as compare to NT wire and the same behavior of CT wire can be realized in Fig. 5a-b. Therefore, CT wire results in superior surface quality for composite material.

Kerf width
KW is the gap created as a result of the cutting process that increases or decreases with variation in machining parameters. Change in KW with variation in input variables are presented in Fig. 9a-d. In this study, it has been observed that KW of hybrid AMC reduces with decrease in FR and increase in Ton (Fig. 9a-b). However, TON has a significant impact on KW as compared to FR. As the increase in TON offers more time to discharge energy for melting and vaporizing the work material. At a low level of TON, discharge energy is not sufficient to remove work material easily, while at higher values of TON, intense discharge energy melts the substrate metal and break the Al2O3 and W particles' bound from the matrix. Thus removal of reinforcement particles increases the KW significantly [24]. It has been observed that an increase in FR decreases the KW because a higher feed rate does Wire speed Wire tension not let the wire electrode to distribute heat energy uniformly in the cutting zone. Therefore, the ineffective distribution of heat energy in the machining zone results in reduced KW [55].
From Fig. 9c-d, it can be clearly seen that KW is minimum at higher values of TW and SW. When SW is low, the amplitude of wire vibration is comparatively higher which reduces the dielectric fluid flow around the wire and resulted in inefficient eradication of debris and reinforced particles from the discharge gap. Moreover, the debris particles which adhered around the wire also generate high KW. When the wire is running at a higher speed the debris and reinforced particles get discarded from the wire surface and resulted in reduced KW [56]. From the 3D surface graph, it can be seen that KW decreases as tension in the wire increases. At low TW, the amplitude of the wire is comparatively high resulted in the wider kerf. Average vibration amplitude decreases with an increase in TW which ultimately yields narrow KW [57].
From Fig. 9 it has been observed that CT wire generated wider kerf than NT wire in AMC. This increase in kerf is due to the higher electrical conductivity of CT wire. The high electrical conductivity of the electrode produces intense discharge energy to the machining zone and removes a high amount of composite material. Improved flushing using CT wire prevents debris and residue material from adherence and keeps the spark gap cleaner.
Moreover, the melting of matrix material releases the large-sized reinforced particles from the machining zone which also increases the spark gap abruptly. The actual phenomenon of the widening of spark gap is illustrated in Fig. 5 schematically. This increase in spark gap resulted in high kerf width of composite material with CT wire.

Surface morphology of workpiece
The surface morphology of the machined specimens has been analyzed using SEM shown in Fig. 10a  m/min, SW=8 m/min, and TW=7 g. From the above comparison, it has been seen that NT wire results in inferior surface morphology than CT wire in both cases. Surface morphology of specimens machined with NT wire depicts micro-voids, globules, and large craters while CT wire resulted in small and shallow sized craters and few debris particles welded on the surface have also been noticed. The reason behind the remarkable performance of CT wire is the enlarged discharge channel which prevents the melted material from resolidification on the work surface and makes the flushing easy due to its superior electrical conductivity.  Table 5. From Fig. 11a-b, it can be observed that low energy parameters (TON=3 µs, FR=13 m/min, SW=11 m/min, and TW=10 g) at Exp. No. 08 has generated narrow kerf. When the comparison has been made between Fig. 11a and 11b, it can be seen that CT wire has produced a relatively higher KW value (0.340 mm) than NT wire (0.318 mm). However, high energy parameters

Wire surface analysis
In the machining of squeeze casted hybrid composite, the effects of wire's cryogenic treatment and machining variables have also been analyzed on wire surface. For this mean, micrographs of NT and CT wire have been compared at Exp. No. 04. In Fig. 12, the surface integrity of both wires shows the difference in erosion patterns during the machining of squeeze casted hybrid composite. Micrograph of eroded NT wire in Fig. 12(a) depicts some micro-cracks and clusters of voids on its surface. Besides this, larger and deeper craters have also been observed on its surface after machining. On the other hand, a shallow and small-sized crater has been seen on the CT wire's surface (Fig. 12b). At the same cutting conditions (Exp. No. 04) wear performance of wires shows that NT wire degenerates more swiftly as compared to CT. During machining, high temperature in the machining zone decreases the thermal conductivity of the wire electrode which melts and evaporates wire material and deteriorates the wire structure. Depreciation in wear behavior of CT wire is certainly due to the enhanced thermal and electrical conductivity attained during cryogenic treatment ultimately results in a comparatively smooth wire surface [58].  Figure 12. Microstructure of (a) NT and (b) CT wire after experimentation

Analysis of variance
To analyze the effects of input variables, the design of experiment has been formulated using a statistical technique. Each technique involves randomization as a basic principle in experimental design. These techniques randomly allocate the experimental units as well as conducts experiments randomly. This randomization avoids biases in the conclusions for input variables. After designing the experiment, results are analyzed with Analysis of variance (ANOVA). ANOVA distributes the total variation into accountable sources of variations (input variables) in the experiments [59]. ANOVA also estimates the percentage contribution of input variables and their interactions into the corresponding response measures. To predict the response measures, empirical models for each response measure have been developed using experimental results. Statistical significance of developed empirical models, main, interaction, and quadratic terms has also been assessed in this study [60]. The coefficient of determination (R²) ascertains the linear correlation of variables and response measures in the regression analysis and its higher values depict the adequacy of developed empirical models [61]. An adequate empirical model has the ability to predict the response measures with great accuracy.

Cutting speed
ANOVA of the CS results for both NT and CT wire electrodes have been provided in Table 6. Fit summary shows that all input variables have a significant impact on CS for both electrodes using 95% confidence level. The analysis shows that empirical models of CS for both wire electrodes are significant and adequacy measures (R², adjusted R², and predicted R²) are also in reasonable agreement. This reveals that developed models for CS can

Surface roughness
ANOVA results in Table 6 explain that empirical models of SR for NT and CT wire electrodes are significant as their probability values are less than 0.05. It is also observed that all input variables have also a significant effect on SR. Adequacy measures have also been focused during the analysis to check the adequacy of empirical models.
Numerical values all coefficients are closer to 1 which shows the competitive prediction capabilities of developed SR models for both wires are presented in Eqs. 5 and 6.

Kerf width
ANOVA of KW results for NT and CT wires has been presented in Table 6.

Multi-objective optimization
WEDM is a susceptible and delicate machining technique that requires intensive surveillance because of its sensitivity toward numerous electrical and non-electrical control parameters. Slight variation in a parameter may result in a convoluted impact on response measure. Therefore, it is inevitable to perform machining without an optimum set of input variables. Mostly, the optimization of a single response is not possible without deteriorating the other responses. However, application of the multi-objective optimization technique has been observed very helpful in the optimization of responses having conflicting nature [62]. In order to increase the CS of workpiece material, both SR and KW also need to be considered together.

Grey relational analysis
where is the ith normalized value of jth experiment, is the measured ith response value of jth experiment and is the desired/targeted response value. In the case of CS, larger the better function is desirable, and smaller the better objective is preferred for both SR and KW.
To determine the relationship between optimum values and normalized values, grey relational co-efficient has been practiced. Grey relational co-efficient has been calculated using relation (12) given below: where is the grey relational coefficient of the ith normalized value of the jth experiment. is the distinguishing co-efficient and its value is generally taken between 0 and 1. In this study, distinguishing co-efficient is taken at 0.5. Δ is the deviation from the desired/targeted value which is expressed as in Eqs. (13)(14)(15) [64]: The next step of GRA is the computation of GRG which is the weighting sum of all grey relational coefficients.
GRG represents the relationship between input variables and all response measures simultaneously. It is presented in Eq. (16) below: where is the GRG of the jth experiment and n is the number of response measures (in current study n=3). High GRG value represents the optimal set of input variables for all response measures simultaneously and it is ranked as 1 st . For the current study, Table 7 presents the grey relational coefficients, GRG, and their ranking for NT and CT wire electrodes.

Comparative performance analysis of wire electrodes
Although the response surface graphs have clearly revealed the remarkable performance of the CT wire for CS and SR. To further envision the efficiency of the CT wire electrode, optimum values of CS, SR, and KW have been compared. The percentage increase and decrease of response values of CT wire with respect to NT wire have been displayed in Fig. 13. As maximum CS values are desired, therefore, the comparison has been made at Exp. No. 04 (provided in Table 5). CT wire yields 26.96 % high CS than NT wire. For Ra, the minimum values of both wires are compared at Exp. No. 13. CT wire produces relatively 15.10 % less SR, so the percentage improvement is denoted by a negative sign in Fig. 13. CT wire on the other hand has generated 6.92% higher KW when its minimum values are analyzed at Exp. No. 08. Based on the above discussion, it can be accomplished that the CT wire electrode is the best candidate for the machining of squeeze casted hybrid AMC. Figure 13. Performance comparison of NT and CT wire electrodes

Conclusion
This research aims to investigate the selection of potential wire electrode (wire type) and its effects on the machining performance of the squeeze casted Al2024/Al2O3/W hybrid composite in the WEDM process. To examine the influence of wire type, the machining performance of the cryogenic treated zinc-coated brass wire electrode was compared with simple zinc-coated brass wire. Besides this, the effects of input variables including pulse duration, wire tension, wire feed rate, and wire runoff speed were assessed in terms of cutting speed, surface roughness, and kerf width. Experimental results were analyzed using ANOVA and mathematical models were developed for the response measures (CS, SR, and KW) of both electrode wires. Further, the obtained results of SR and KW were supported by the microstructural analysis. Surface characteristics of both wire electrodes were also evaluated using SEM analysis. GRA was used for the multi-objective optimization of response measures.
Based on experimental consequences and findings from their respective discussion, some salient conclusions are  Cryogenic treatment declined the electrical conductivity of molybdenum wire and improved UTS to a limited extent. On the other hand, the electrical conductivity of zinc-coated brass wire was significantly escalated by 24.8% with cryogenic treatment which depicted the CT brass wire as the potential candidate for machining of squeeze casted Al2024/Al2O3/W.
 ANOVA showed that TON as the most significant input variable for the CS, SR, and KW while FR, TW, and SW were the subsequent substantial input variables for both wires.
 Surface morphology analysis of machined surfaces of AMC characterized that high energy parameters with NT wire generate a surface with a high concentration of micro-voids, globules, deep and larger craters which were reduced by practicing low energy parameters with CT wire.
 SEM analysis revealed that there exists a high concentration of micro-voids and deep craters on the surface of NT wire while shallow craters and a quite smooth surface have been observed in CT wire after machining.
 Multi-objective optimization based on GRA presented TON=3 µs, FR=13 m/min, SW=11 m/min, and TW=10 g as an optimal set of input variables to achieve 75.1% and 72.5% overall results with NT and CT wire respectively by trading off up to a limited extent.
 Comparative analysis of the optimum values of performance measures of wire electrode divulged that CT wire electrode offers 26.96% and 15.10% superior results for CS and SR respectively while 6.92% deprived results for KW than NT wire.
This research proposed suitable wire electrode for the squeeze casted hybrid AMC in the WEDM process, which can be helpful for the practitioners in the machining of such type of material on the industrial scale. Future research can be conducted to analyze the effects of suggested wire type on recast layer thickness, microhardness, and infusion of wire material on the workpiece surface. That study will provide an in-depth analysis of the surface morphology of the workpiece machined with the recommended wire electrode.