Multiobjective optimization of EDM machining parameters of TIB₂ ceramic materials using regression and gray relational analysis

2.1 Selection of the workpiece

Titanium diboride material with dimensions 100 mm × 100 mm × 10 mm was selected as the workpiece, which was purchased from Domadia Private Limited (Mumbai, India). Various industries use titanium diboride due to its outstanding properties, including high hardness, melting point, and wear resistance [1,17].

2.2 Details of the equipment

Drilling with EDM, a non-traditional machining technique, uses electrical energy to remove material from a workpiece. The Baoma Suzhou DB703A machine performs EDM machining. Figure 1 depicts the EDM machine that drills holes in titanium diboride materials. Spark erosion drilling involves drilling a small hole into the workpiece using a rotating electrode that conforms to the desired hole geometry. It involves placing the workpiece and electrode in a dielectric fluid (usually DI water and kerosene) and using a high-frequency electrical discharge between them. The discharge creates a spark that vaporizes the material at the point of contact, creating a tiny void or hole. The electrode rotates, continuously removing the material, creating a straight or tapered hole. By adjusting the parameters of the electrical process, such as voltage and current, as well as the distance between the electrode and workpiece, the hole size can be precisely controlled, and the value can be displayed digitally. We considered the pulse switch-on time (T _on), discharge current (I _p), and flushing pressure as variable input parameters. Table 1 presents the levels and values of these variable parameters. We selected the input parameter values based on a literature review and the device capacity. We drilled holes in the titanium diboride material using hollow electrode rods made of brass, copper, tungsten, and tungsten with a diameter of 3 mm. We listed and compared the input parameters to identify the optimal processing parameters and influencing factors. The dielectric fluid consisted of distilled water with a specific flushing pressure to remove the resulting residue. After the machining conditions reached a steady state, we retrieved the voltage and operating current values from the machine monitor and recorded them while drilling the sample. The drilling length was 4 mm in all tests. As shown in Figure 2, we created 64 holes, 16 of which used brass, copper, tungsten, and tungsten copper electrodes. We recorded the working time for each test with various input parameters. We also measured the weight of the sample and tool electrode after each drilling operation to calculate the MRR and TWR.

Figure 1

A hole-drilling EDM machine.

Figure 2

TiB₂ workpiece with EDM hole.

2.3 Selection of electrodes

In this study, brass, copper, tungsten, copper, and tungsten were selected as electrode materials. The diameter of the electrode was 3 mm, and the length was 400 mm. They were selected to test the performance metrics such as MRR, overcut, TWR, and SR. The electrode materials for this study were selected based on their melting temperature, hardness, physical properties, and chemical properties [12].

2.4 Dielectric fluids

Due to its high dielectric strength, DI water can tolerate higher voltages without malfunctioning. This property is crucial for maintaining stable EDM processes, especially when working with complex geometries or high-performance machines. DI water can replace oil-based dielectric fluids, resulting in commonly machined parts with improved surface finishes. This is particularly helpful when surface quality is important. DI water, among many other dielectric fluids, is less harmful to the environment [18]. It does not contain any hazardous substances or additives that could make disposal difficult. With regard to specific machining requirements, such as the material being machined, the desired surface finish, the speed of the machining process, and safety considerations, DI water has its advantages.

4.1 MRR

4.2 TWR

4.3 Surface roughness (SR)

4.4 Overcut (OC)

6.1 MRR

Factors such as electrode materials and dielectric fluid influence the MRR in EDM. According to the experimental results, the MRR in the copper electrodes with DI water removes more material than the dielectric liquid kerosene due to the high thermal conductivity and cooling properties of the DI water [22]. The EDM process improves the MRR of titanium materials by increasing the capacity and voltage [23]. Brass electrodes with DI water achieve a moderate MRR compared to copper electrodes and tungsten–copper composite electrodes produce better MRR in DI water. Using kerosene as the dielectric for the tungsten electrode results in a lower MRR compared to DI water. Figures 3 and 4 show the MRR for the titanium diboride material using four different electrodes and DI water dielectric fluid media with variations in the input parameters. In general, copper electrodes outperform brass, tungsten, and tungsten–copper electrodes in terms of the MRR. The thermal conductivity of the copper tool is higher (402 W·(m·K)⁻¹ compared to those of brass (161 W·(m·K)⁻¹, tungsten copper electrodes (240 W·(m·K)⁻¹, and tungsten electrodes (174 W·(m·K)⁻¹. Higher thermal conductivity would result in a faster rate of heat conduction. The spark discharge then ignites faster than with electrodes made of brass or tungsten. Because of the higher heat dissipation rate, shallow craters form on the workpiece surface, resulting in less heat transfer to the workpiece [18]. The thermophysical properties of the materials used as electrodes have a significant impact on EDM performance [24,25]. This reduces the MRR. From Figure 3, we can see that increasing the T _on, spindle speed, and I _p increases the MRR of the TiB₂ material from 2.011 to 3.918 mm³·min⁻¹ for copper electrodes. Similarly, the MRR improves by 3.3944, 3.3696, and 2.64078 mm³·min⁻¹ for the brass, tungsten copper, and tungsten electrodes.

Figure 3

MRRs for different electrode materials and dielectric fluids.

Figure 4

Influence of input parameters on the MRR: (a) copper electrodes, (b) brass electrodes, (c) tungsten–copper electrodes, and (d) tungsten electrodes.

6.2 TWR

The TWR of the TiB₂ material using different electrode materials is shown in Figure 5. Kerosene-based dielectric fluids exhibit a higher TWR than DI water and dielectric fluids. The higher viscosity of kerosene-based dielectric fluids causes deposit accumulation at the tool–electrode interface, which in turn leads to increased tool wear due to abrasion and erosion [22]. Improved EDM heat dissipation in DI water results in less wear and lower thermal stress on the tool. Kerosene, on the other hand, has lower thermal conductivity and can, therefore, increase the temperature around the tool, leading to increased wear and thermal degradation. Chemical reactions with the tool material can lead to chemical wear. Additives and contaminants in kerosene-based dielectric fluids can increase these reactions. The possibility of chemical reactions with the tool material is usually lower when using DI water because it is chemically pure. DI water’s dielectric fluid strength surpasses that of kerosene, enabling it to sustain the EDM process at lower voltages, thereby reducing the energy transfer to the tool and consequently reducing wear [26]. The TWR increases with increasing I _p, T _on, and spindle speed, as shown in Figure 6(a)–(d) for four different electrode materials. In the case of tungsten, brass, and tungsten–copper electrodes, the copper electrode material is the softest. It also conducts heat well, which means that it wears down faster than the other three electrode materials when titanium diboride materials are machined. During the experimental study, copper electrodes achieved the highest TWR at 0.863 mm³·min⁻¹, followed by brass electrodes at 0.594 mm³·min⁻¹, tungsten copper at 0.391 mm³·min⁻¹, and tungsten at 0.298 mm³·min⁻¹.

Figure 5

TWRs for different electrode materials and dielectric fluids.

Figure 6

Influence of input parameters on the TWR: (a) copper electrodes, (b) brass electrodes, (c) tungsten–copper electrodes, and (d) tungsten electrodes.

6.3 SR

The Mitutoyo SJ 210 surface Roughness tester is used to measure the SR of EDM holes. The SR of the TiB₂ material using different electrode materials is shown in Figure 7. When DI water is used to machine the titanium diboride material, it can result in minimal SR, particularly when combined with harder electrode materials like tungsten, tungsten–copper, and brass. Copper electrode materials are soft and, therefore, can produce higher SR compared to harder electrode materials. Due to the high material removal, the SR of tungsten and tungsten–copper is lower compared to the copper electrode material [27]. Figure 8(a)–(d) shows the interaction diagram for different electrode materials versus the input parameters. The SR value of the TiB₂ material increased with increasing I _p, T _on, and spindle speed. The highest SR value of 9.635 μm is achieved for the copper electrode when T _on = 6, I _p = 7, S = 100, and FR = 10, followed by an SR value of the brass electrode (7.46475 μm), tungsten–copper electrode (6.919 μm), and tungsten electrode (5.9230 μm).

Figure 7

SR for different electrode materials and dielectric fluids.

Figure 8

Influence of input parameters on the SR: (a) copper electrodes, (b) brass electrodes, (c) tungsten–copper electrodes, and (d) tungsten electrodes.

6.4 Overcut

Based on the experimental results, it was found that a DI water-based dielectric medium produces less overcut compared to a kerosene-based dielectric medium due to its better dielectric strength, as shown in Figure 9. The kerosene-based dielectric medium produces a high overcut due to the high viscosity of kerosene and lower thermal properties [28]. Hard materials such as tungsten and tungsten–copper electrodes combined with DI water-based dielectric medium produce a lower amount of overcut compared to softer electrode materials such as copper and brass [29]. Figure 10(a)–(d) displays the overcut interaction diagram for various input parameters, dielectric fluids, and electrode materials. By selecting the appropriate electrode material and dielectric fluid medium, one can reduce the overcut of the EDM hole, considering the machining materials used. We observed a higher overcut rate for copper electrodes with a dielectric fluid value of 0.616 mm, followed by brass, tungsten–copper, and tungsten electrodes.

Figure 9

Overcut for different electrode materials and dielectric fluids.

Figure 10

Influence of input parameters on the overcut: (a) copper electrodes, (b) brass electrodes, (c) tungsten–copper electrodes, and (d) tungsten electrodes.

6.5 GRA grade analysis

Figure 11 illustrates the GRA of the titanium diboride material using various electrode materials. The first step in GRA is to normalize the experimental results to a common scale. The second step in GRA is to assign weights to each criterion based on their significance to the machining process. The third step is to calculate the GRC; this coefficient indicates how closely each alternative meets the desired performance levels. Due to the equal weighting of all parameters, the ξΔmax value is assumed to be 0.5 [16]. The fourth step is to grade the GRA; this grade reflects the overall performance of each alternative based on the calculated GRCs and their respective weights. The fifth step is to rank the different trials based on their gray relational grades, guiding the selection of optimal EDM parameters. We plot Figure 11 based on the grade value, an important criterion in GRA optimization.

Figure 11

GRA grade analysis of four different electrode materials.

Copper and tungsten electrodes have reliable grayscale values owing to their stable response to numerous parameters. For applications that involve stability and precision, the copper electrode is a perfect choice. Brass and tungsten–copper electrodes display greater variations in the EDM tests. Brass achieves top outcomes in certain tests but lacks uniformity and robustness. These indicate the sensitivity of each electrode’s performance to specific EDM parameter settings. The tungsten–copper electrodes offer balanced and competitive performance with relatively high gray values in several tests. Applications that require high material removal and durability can benefit from this technology. The tungsten electrode exhibits a constant performance with relatively small fluctuations, indicating reliability. The tungsten electrode is ideal for processing hard materials and attaining smooth surfaces. For example, brass performs extremely well in Experiment 3 but lags significantly behind in others. The tungsten–copper electrodes in Figure 11 illustrate a peak in Experiment 7 but fluctuate elsewhere. The highest average gray value across all trials identifies the optimal electrode for multi-target performance. Based on their stability and average GRG, we can categorize the performance of electrodes into tungsten–copper, tungsten–copper, copper, and brass. Depending on whether one prioritizes stability or peak performance, this ranking can cater as a guide to the material choice for specific EDM applications. For copper electrode trial 1, the highest-grade value of 0.779 indicates the optimal conditions with a favorable balance among MRR, TWR, SR, and overcut. Trial numbers 15 and 12 have the lowest rank, which leads to an unfavorable condition for machining. For brass electrode trial 4, the highest-grade value of 0.759 indicates that this parameter setting offers the best balance between high MRR, low TWR, smooth SR, and accurate overcut. Trial 16 displays a low gray value, suggesting that this parameter may not be as beneficial for optimal EDM performance. In tungsten copper electrode trial 1, we achieved the best balance of high TWR, low TWR, and accurate OC for this set of experiments. Trials 1–4 rank in the top 4, suggesting that the parameter settings in these trials provide more optimal EDM performance using a tungsten–copper electrode. Trial 1 yields the highest-grade value of 0.702 for the tungsten electrode, indicating the best overall EDM performance with this parameter set. This GRA helps us identify optimal parameter combinations for EDM machining with different electrode materials. The GRA determines T _on = 3, I _p = 4, S = 100, and DR = 10 as the optimal conditions for machining the TiB₂ material, which gives better MRR and lower or moderate TWR, SR, and OC.

6.6 Material characterization

We performed material characterization tests before and after drilling holes using EDM to determine cracks, porosity, and surface integrity of the workpiece. The DK7735 BORUI CNC wire EDM machine can cut samples with dimensions of 10 mm × 10 mm × 5 mm for SEM analysis. After cutting the sample, we can etch it to eliminate impurities and oxides from the workpiece. This smoothens the surface and improves the contrast between the different layers of a workpiece [15].

Table 11 displays the elemental composition and atomic weight percentage of the titanium diboride material. Figure 12 shows the SEM and EDS analysis of the workpiece before performing EDM machining. Figure 13(a) shows a sample at 250× magnification with a working distance (WD) of 6.9 mm and an accelerating voltage of 20.00 kV. According to the surface morphology, it appears that the sample has an uneven, porous structure with varying depths and irregular features. This type of texture could suggest a high-energy machining process, such as EDM, or an etching process, which leaves the surface with microcraters or pits. Figure 13(b) shows the acquisition of this image at 250× magnification, with an accelerating voltage of 20.00 kV and a WD of 6.9 mm. The surface has a rough and broken texture, similar to the first image, indicating the possible material removal through EDM. Some regions seem denser and darker, suggesting localized differences in the material composition or the presence of residues from the machining process. The EDS spot in Figure 13(c) suggests a potential elemental analysis point. The surface morphology is coarser and has more pronounced, jagged edges, which may indicate the effects of an electrical discharge. The EDS analysis can help identify the elemental composition at that specific point, which is useful for understanding the material changes or contaminant deposition. Figure 13(c) and (d) shows the formation of a small recast layer over the EDM hole area.

Figure 12

SEM analysis before machining and EDS analysis of the titanium diboride material.

Figure 13

SEM analysis after EDM machining: (a) copper electrodes, (b) brass electrodes, (c) tungsten–copper electrodes, and (d) tungsten electrodes.