Effects of machining parameters on spectral entropy of acoustic emission signals in the electro erosion

Understanding and optimizing mechanical manufacturing processes is essential for sustainable industrial development. Among unconventional machining methods, electrical discharge machining (EDM) distinguishes itself by its capability to remove material through successive electrical discharges submerged in a dielectric fluid. EDM encompasses intricate phenomena influenced by machine parameters, dielectric choice, and the materials involved. Unlike conventional machining, EDM operates with the tool electrode in close proximity to, but not in physical contact with, the workpiece, achieving material removal through localized overheating. This study focuses on monitoring EDM phenomena during the machining of AISI H13 steel, exploring variations in machining parameters and electrode materials (electrolytic copper and graphite). Acoustic emission (AE) signals and machine learning (ML) are employed for experimental characterization and data analysis. Spectral entropy is applied to AE signals, quantifying inherent signal uncertainty. The findings reveal remarkable accuracy (97.7%) and underscore the superior control achieved with graphite electrodes in managing machining phenomena compared to electrolytic copper electrodes. Graphical abstract