TEACHING-LEARNING-BASED PARAMETRIC OPTIMIZATION OF AN ELECTRICAL DISCHARGE MACHINING PROCESS

Abstract

Due to several unique features, electrical discharge machining (EDM) has proved itself as one of the efficient non-traditional machining processes for generating intricate shape geometries on various advanced engineering materials in order to fulfill the requirement of the present day manufacturing industries. In this paper, the machining capability of an EDM process is studied during standard hole making operation on pearlitic SG iron 450/12 grade material, while considering gap voltage, peak current, cycle time and tool rotation as input parameters. On the other hand, material removal rate, surface roughness, tool wear rate, overcut and circularity error are treated as responses. Based on single- and multi-objective optimization models, this process is optimized using the teaching-learning-based optimization (TLBO) algorithm, and its performance is contrasted against firefly algorithm, differential evolution algorithm and cuckoo search algorithm. It is revealed that the TLBO algorithm supersedes the others with respect to accuracy and consistency of the derived optimal solutions, and computational efforts.