Sustainable electric discharge machining using alumina-mixed deionized water as dielectric: Process modelling by artificial neural networks underpinning net-zero from industry

Abstract

The requirement for materials possessing both high strength and low density has garnered significant attention from industries and researchers in recent times. Among these materials, aluminum 6061 (Al6061) exhibits the desired properties. However, due to its diverse machining capabilities, powder-mixed electric discharge machining (PMEDM) has emerged as a viable option for cutting such materials. This method has been criticized for its high energy consumption and limited cutting efficiency. Furthermore, conventional dielectric (kerosene) employed in EDM has drastic environmental and operator's health concerns. To address the abovementioned issues, deionized water has been employed in this study which enhances the reusability of resources and minimizes the cost of the dielectric. Herein, to make the process sustainable, and to keep the environment free from hazardous fumes, generated during the machining process, deionized water has been used. In addition to that, to uplift the machining responses, alumina (Al2O3) nano-powder has been engaged. To conduct the study, response surface methodology (RSM) was employed. This investigation aimed to analyze the impact on the material removal rate (MRR), surface roughness (SR), and specific energy consumption (SEC) by using microscopy analysis, scanning electron microscopy (SEM), 3D surfaces profilometry, energy dispersive x-ray (EDX) analysis and after that, the machining responses are modelled using the artificial neural networks (ANN) technique. It was observed that by utilizing non-dominated sorting genetic algorithms (NSGA-II) an improvement of 87.42 % in MRR, 3.4 % better surface finish and 0.7 % better SEC have been obtained. Notably, CO2 emissions were found to be 94.27 % lower by using the deionized water as dielectric compared to those produced by kerosene oil.