Advancements and challenges in modelling and simulations of micro-electrical discharge machining (micro-EDM): a review

Micro-Electrical Discharge Machining (micro-EDM) is a non-contact manufacturing technology enabling the precise fabrication of complex micro-features in hard-to-machine electro-conductive materials. The declination of the micro-EDM in various configurations, including sinking, micro-wire, drilling, milling, welding, and reverse, suits several applications in the industrial, aerospace, biomedical, and electronics fields. Despite its potential, the micro-EDM process is challenging, as intertwined electrothermal, mechanical, and fluidic phenomena occurring at the micro-scale, along with its stochastic nature, significantly affect its stability, accuracy, and performance. To deepen the understanding of these mechanisms and optimize the process, advanced modeling approaches are now being used in conjunction with experiments. In this regard, this review article aims to examine the literature on the exclusive micro-EDM modelling and simulations, highlighting how numerical analyses, based on various techniques (Finite Element Method-FEM, Finite Volume Method-FVM, and Computational Fluid Dynamics -CFD), are capable of addressing the critical aspects of the process. At the same time, the article will also identify the complexities, limitations, and unresolved challenges in micro-EDM modelling and simulations, which still require further attention. For instance, it has been noted that simplified assumptions regarding material description, plasma discharge radius, energy partition to the electrodes, heat transfer modes, debris motion, and molten material re-solidification frequently hinder accurate performance prediction. As a result, model corrections are applied using real-time data from micro-EDM experiments and monitoring. Moreover, future directions point toward fully predictive, multi-physics-informed models to enable an adaptive and high-precision process, which require integration among Deep learning, Machine learning, and Artificial Intelligence (AI) with monitoring and multi-physics modelling.