Optimization of WEDM machining attributes of Al-Si1MgMn-GNP/TiB 2 hybrid composite using Taguchi,ANOVA,and machine learning approaches

Abstract

The current research explores the impact of Wire-cut Electrical Discharge Machine (WEDM) parameters on the machining attributes of ultrasonically dispersed Al-Si1MgMn-GNP/TiB₂ hybrid nano-composite, fabricated through an UT-assisted bottom pouring stir casting technique. The study examined five key process parameters of servo voltage (SV), pulse-on-time (T_ON), pulse-off-time (T_OFF), peak current (I_P), and variable frequency (VF), assessing their influence on material removal rate (MRR), surface roughness (SR), and kerf width (KfW). Experiments were designed using Taguchi’s L18 orthogonal array (OA), and multiple responses were optimized using both Taguchi and Analysis of Variance (ANOVA) approaches. The findings revealed that spark voltage (SV) and T_ON were the most influential factors affecting MRR and KfW, while SV and VF had a significant impact on SR. To predict the machining outcomes, artificial neural network (ANN) models were developed, achieving high predictive accuracy with a mean square error (MSE) of 0.3244 and a correlation coefficient (R) of 0.9868. Surface morphology was examined using scanning electron microscopy (SEM), which revealed features such as voids, pockmarks, and debris accumulation on the machined surfaces.