Application of integrated entropy-COPRAS and ANN approaches for maximizing wire EDM machinability attributes of Al6082-T6/GNP/TiB2 composites

Abstract Wire Electrical Discharge Machining (WEDM) utilizes electrical sparks to cut conductive materials precisely. It employs a thin, electrically charged wire to erode the material, creating complex shapes with high accuracy, tight tolerance, and minimal material waste. The current study investigates the effect of WEDM process parameters on the various output properties of machining a hybrid Aluminium metal matrix nanocomposite (HAMMNC). A series of experiments were conducted on a composite constructed using Taguchi’s standard L 18 Orthogonal Array (OA) design. The acquired responses of Material-Removal-Rate (MRR), Surface-Roughness (SR), and Kerf-Width (K f W) were optimized using integrated Entropy-COPRAS and artificial neural network (ANN) algorithms respectively. The relative significance ( Q i )values obtained for the alternatives are analyzed with the Taguchi method. The outcomes showed that the optimal setting of WEDM process variables is achieved at Servo Voltage: high volts; Pulse-on-time: 60 μs; Pulse-off-time: 16 μs; Peak Current: 6 Amp; Variable frequency: 18 Hz; Speed: 50 RPM; Sim Speed: 30 RPM respectively. The models developed from ANN were adequate and accurate from the results of error histogram and regression plots and they are best suited for the prophecy of future responses. The surface morphology studies confirmed the appearance of pockmarks, voids, and globules of debris on the Wire EDMed surfaces.