Experimental investigation and ML-aided analysis of tool wear and material removal in micro machining of Al-Si7Mg/SiC AMCs

Abstract

This work examines the micro milling capability of the aluminum–silicon–magnesium/silicon carbide (Al-Si7Mg/SiC) aluminum matrix composites (AMCs), which have been produced using the stir casting technique. The main interest can be seen in the measurement of the material removal rate (MRR), tool wear, and tool life with different micro machining conditions. Three important variables of input were used in the experiments of micro milling, namely depth of cut (DoC) (0.5, 1.0, and 1.5 mm), feed rate (FR) (1, 2, and 3 mm/min), and the tool diameter (TD) (0.3, 0.4, and 0.5 mm). The objective of the study was to have knowledge of the effects of these parameters on the efficiency of machining and tool performance, along with prediction using machine learning (ML). It was observed that higher MRR, longer tool life, and lower tool wear are desirable for micro milling of Al-Si7Mg/SiC AMCs. Balancing these parameter combinations brings about efficient, cost-effective, and high-quality micro milling of AMCs. In addition to the experimental analysis, ML models such as gradient boosting, random forest, and stacked ensemble were employed to predict machining responses, demonstrating strong predictive capability, and potential for enhancing process optimization.