Abstract
Enhancing the surface characteristics of biomedical implants is crucial for increasing their compatibility with biological implants and minimizing their corrosion rate. Despite the possibility of such surface alterations, micro electro discharge (μED) milling has not been widely used in this area. This study examines the processing of biodegradable magnesium alloy using μED milling, with an emphasis on analysing machining time (MT) and surface roughness (Ra). The impact of pulse on-time (T on ), voltage (V) and tool rotational speed (TRS) is assessed using Spearman correlation coefficient plots to determine their effects on the responses. Four machine learning (ML) methods viz. support vector regression (SVR), extreme gradient boosting (XGB), Random Forest (RF) and Gaussian Process Regression (GPR) are utilized to forecast these outcomes. Among these ML methods GPR exhibits the highest accuracy in predicting outcomes for the collected data. A parametric analysis is conducted using GPR based ML model. The process parameters are then optimized by a preference-based Marine Predators Algorithm (PBMPA) to attain Pareto optimum settings. The optimized surface is analysed using scanning electron microscopy revealing structural characteristics that endorse their use in feasibility for biodegradable medical implant applications.

This publication has 49 references indexed in Scilit: