Multi-Objective Optimization of Material Removal Rate and Surface Roughness in Ultrasonic Vibration-Assisted EDM Using NSGA-II, GPR, and AHP

Ultrasonic Vibration-Assisted Electrical Discharge Machining (UV-EDM) constitutes a promising technique for improving machining efficiency and surface quality, particularly when working with difficult-to-machine materials. This study presents a comprehensive multi-objective optimization approach for UV-EDM applied to 90CrSi steel, aiming to maximize the Material Removal Rate (MRR) while minimizing Surface Roughness (Ra). The experimental data were collected under varying process parameters, including the peak current, pulse-on time, and ultrasonic vibration amplitude. A Gaussian Process Regression (GPR) model was developed to accurately predict MRR and Ra. These predictive models were then integrated into the Non-Dominated Sorting Genetic Algorithm II (NSGA-II) to perform Pareto-based optimization. Additionally, the Analytic Hierarchy Process (AHP) was employed to rank the Pareto-optimal solutions based on decision-makers’ preferences. The results demonstrate the effectiveness of combining GPR and NSGA-II for modeling and optimizing UV-EDM, while the use of AHP enables a rational selection of optimal machining conditions. This hybrid methodology offers valuable insights into enhancing productivity and surface integrity in precision machining applications.