Gene expression programming for parametric optimization of an electrochemical machining process

Due to its capability to generate intricate shape profiles on various electrically conductive difficult-to-machine work materials, electrochemical machining (ECM) process has found immense applications in aerospace, automotive, die making, artillery and surgical instruments manufacturing industries. Like other machining processes, the performance of an ECM process with respect to various product quality characteristics is also influenced by its different input parameters. This paper proposes the novel application of an almost unexplored evolutionary algorithm in the form of gene expression programming for parametric optimization of an ECM process while treating electrolyte concentration, electrolyte flow rate, applied voltage and tool feed rate as the input parameters, and material removal rate (MRR) and average surface roughness (Ra) as the responses. It is noticed that an optimal parametric intermix as electrolyte concentration = 15.1417 g/l, electrolyte flow rate = 6.1846 l/min, applied voltage = 15.94709 V and tool feed rate = 0.99995 mm/min would lead to simultaneous optimization of both the responses. A comparative analysis of its optimization performance in relation to accuracy and variability of the derived solutions, and computational speed against genetic algorithm (GA), particle swarm optimization (PSO), ant colony optimization (ACO) and artificial bee colony (ABC) optimization techniques validates its superiority over the others. At the derived optimal combination, MRR is increased by 26.14, 3.64, 12.50 and 12.55%, and Ra is reduced by 1.28, 21.97, 21.03 and 7.86% against GA, PSO, ACO and ABC techniques. An optimal Pareto front is also developed to determine the optimal parametric intermixes for having maximum MRR and minimum Ra values. The scatter plots would assist the process engineers in investigating the influences of the input parameters of the considered ECM process on its responses.