A holistic research based on RSM and ANN for improving drilling outcomes in Al–Si–Cu–Mg (C355) alloy

Abstract

The unique properties of Al–Si-based alloys make them suitable for components that demand structural integrity and wear resistance. This study was conducted to investigate the microstructure, mechanical, and drilling properties of a commercial alloy belonging to the Al–Si casting alloy group and containing approximately 4.5–5.5% Si (Al–5Si–1Cu–Mg). Drilling experiments were conducted with an 8 mm uncoated HSS (High-Speed Steel) drill across a range of cutting speeds (V) and feed rates (f) while maintaining a consistent depth of cut (DoC) parameters. Microstructural analysis using optical microscopy and SEM identified key phases within the alloy, including α-Al, eutectic Si, β-Fe (β-Al5FeSi), and π-Fe (π-Al8Mg3FeSi6) inter-metallics. Statistical analyses of the effects of V and f on thrust force (Fz), surface roughness (Ra), and torque (Mz) were performed using Response Surface Methodology (RSM), Artificial Neural Networks (ANN), and Analysis of Variance (ANOVA). The ANOVA results highlighted the significance of both V and f on the measured outputs, with optimal performance observed at a V of 125 m/min and f of 0.05 mm/rev (confidence level: 95%, P < 0.05). Additionally, predictive models based on RSM and ANN were developed for Fz, Ra, and Mz.