Modified teaching learning based optimization for maximization of MRR in wire-cut EDM of Ti6Al4V alloy for sustainable production

Abstract

Wire-cut electrical discharge machining (WEDM) has emerged as prominent advanced machining process to machine electrically conductive difficult-to-machine materials to any intricate shape and size. Amongst Ti-alloys, Ti6Al4V is extensively used in diverse engineering applications and is popularly researched. In this work, maximization of material removal rate (MRR) is addressed as economic aspect of sustainable production during WEDM of Ti6Al4V alloy employing modified teaching-learning based optimization (M-TLBO) algorithm. A novel method for fitness curve fitting is illustrated to obtain global optima for maximization of MRR. Taguchi L16 OA is employed to perform WEDM experiments. It is observed that MRR at optimal cutting conditions improved by 27.51% as compared to its initial maximum value. The fitness curve constructed in the optimal search domain resulted in smooth U-shape curve. ANOVA result showed that current (56.58%) and pulse-off-time (23.57%) are highly dominant process parameters influencing MRR followed by pulse on time (11.66%) and wire speed (7.20%). Machined surface morphology is studied using SEM images. The proposed M-TLBO algorithm is found highly accurate and consistent during several runs conducted and converged faster taking less than ten iterations. Also, proposed novel approach for fitness curve fitting can be effectively applied in any optimization problem.Wire-cut electrical discharge machining (WEDM) has emerged as prominent advanced machining process to machine electrically conductive difficult-to-machine materials to any intricate shape and size. Amongst Ti-alloys, Ti6Al4V is extensively used in diverse engineering applications and is popularly researched. In this work, maximization of material removal rate (MRR) is addressed as economic aspect of sustainable production during WEDM of Ti6Al4V alloy employing modified teaching-learning based optimization (M-TLBO) algorithm. A novel method for fitness curve fitting is illustrated to obtain global optima for maximization of MRR. Taguchi L16 OA is employed to perform WEDM experiments. It is observed that MRR at optimal cutting conditions improved by 27.51% as compared to its initial maximum value. The fitness curve constructed in the optimal search domain resulted in smooth U-shape curve. ANOVA result showed that current (56.58%) and pulse-off-time (23.57%) are highly dominant process parameters influencing...