Multi-Objective Optimization and Prediction of Responses Using GRA and ANN in Micro-EDM of Ti6Al4V Using Different Tool Materials

Abstract

An experimental investigation was conducted to evaluate the machinability of a titanium alloy (Ti6Al4V) using copper (Cu), tungsten carbide (WC), and graphite (C) tools. Voltage (V), capacitance (pF), pulse-on time (Ton), and pulse-off time (Toff) were considered as the input machining parameters, whereas the material removal rate (MRR) and tool wear rate (TWR) were considered as the output machining parameters. A Taguchi L[Formula: see text] orthogonal array and gray relational analysis (GRA) were utilized to design and optimize the machining parameters for both responses. Artificial neural network (ANN) analysis was performed to predict the experimental outcomes. Scanning electron microscopy (SEM) and energy-dispersive spectroscopy (EDS) were used to assess the surface morphology and determine the elemental composition of the machined surface. The results indicated that the optimum machining conditions for the copper tool were 150 V, 1000 pF, 15[Formula: see text]s (Ton), and 15[Formula: see text]s (Toff). However, the optimal machining conditions for the WC were 200 V, 100 pF, 25 [Formula: see text]s (Ton), and 10[Formula: see text]s (Toff), and the optimal conditions for the C were 200 V, 1000 pF, 20[Formula: see text]s (Ton), and 25[Formula: see text]s (Toff), respectively. The highest MRR achieved using the WC tool was 9.4510 mg/s, whereas the TWR of the Cu, WC, and C tools were 1.1039 mg/s, 1.0307 mg/s, and 1.2796 mg/s, respectively. The results showed that machining with the graphite tool had a higher TWR than machining with the Cu and WC tools.