Data-driven optimization of machining parameters for Hastelloy C276 using PSO and TLBO frameworks

Hastelloy C276 is renowned for its exceptional resistance to corrosion and elevated temperatures, rendering it a preferred material for aerospace and chemical processing applications. However, its high strength and work-hardening tendency pose significant challenges during machining. This study systematically investigates the machinability of Hastelloy C276 under four sustainable lubrication and cooling environments-dry machining, minimum quantity lubrication (MQL), nano-enhanced MQL (NMQL), and cryogenic CO₂ (CCO₂). Experiments were designed using a Taguchi L₁₆ orthogonal array, and the influence of cutting speed and feed rate on surface roughness, cutting force, tool wear, and cutting temperature was analysed using ANOVA. Compared to dry machining, cryogenic CO₂ cooling resulted in a reduction of surface roughness and cutting force by approximately 30-40%, along with a substantial decrease in tool wear and cutting temperature, whereas NMQL demonstrated moderate improvements due to enhanced lubrication at the tool-chip interface. ANOVA results revealed that feed rate and cutting speed were the most significant parameters, contributing up to 38.35% and 48.56% to variations in surface roughness and cutting temperature, respectively. To identify optimal machining conditions, Particle Swarm Optimization (PSO) and Teaching-Learning-Based Optimization (TLBO) algorithms were employed. Over 100 iterations, PSO achieved a higher optimization success rate of 83.6% compared to 79.1% for TLBO, while TLBO exhibited faster convergence with a computation time of 6.5 s against 9 s for PSO. The findings demonstrate that cryogenic CO₂-assisted machining combined with evolutionary optimization provides an effective and sustainable strategy for enhancing the machinability of Hastelloy C276.