Machine learning-based model to predict von Mises stress and chip reduction coefficient developed during dry turning of EN36C steel.

This study comprehensively investigates the determination of chip reduction coefficient (CRC) and von Mises stress (VMS) during dry turning of Nickel-Chromium case-hardened steel (EN36C), renowned for its high surface hardness and core toughness. Machining parameters, including cutting speed (36-100 m/min), feed rate (0.49-0.86 mm/rev), and depth of cut (0.67-1.5 mm), were rigorously analyzed using Analysis of Variance (ANOVA) and Artificial Neural Networks. ANOVA identified cutting speed as the most influential factor, accounting for 52.04% of CRC and 35.04% of VMS variations, with feed rate and depth of cut also playing significant roles. ANN modeling achieved a correlation coefficient of 0.97, demonstrating excellent predictive accuracy for parameter optimization. Scanning Electron Microscopy revealed chip morphology, showing continuous chips under optimal conditions of high cutting speed (100 m/min), low feed rate (0.63 mm/rev), and moderate depth of cut (1.0 mm), minimizing stress and enhancing material removal efficiency. Brittle chips were observed at lower speeds (36 m/min) and higher feed rates, emphasizing the critical role of parameter selection. Optimal machining parameters significantly improved surface quality, reduced tool wear, and minimized operational stresses. This research offers a robust framework for machining process optimization, with implications for enhancing industrial efficiency and cost-effectiveness.