Optimizing turning parameters of duralumin nano Cr₂C₃–MoS₂ using ANN and MOORA: A multi-objective approach

This investigation presents a comprehensive parametric optimization study for the precision machining of duralumin matrix composites reinforced with hybrid nano-scale particles comprising chromium carbide (Cr₂C₃) and molybdenum disulfide (MoS₂). The composite material was synthesized via liquid metallurgy stir casting methodology, incorporating 5 wt% nano-Cr₂C₃ as the primary reinforcement phase and 2 wt% MoS₂ as a solid lubricant additive to enhance tribological characteristics and machinability. The experimental framework employed a Taguchi L₂₇ orthogonal array design to systematically investigate the influence of critical machining parameters like cutting velocity, feed rate, and depth of cut on multiple response characteristics including surface roughness (Ra), tool vibration amplitude, and acoustic emission (AE) intensity. The multi-response optimization strategy integrated advanced computational intelligence techniques, specifically utilizing a Levenberg–Marquardt backpropagation artificial neural network (LM-BP-ANN) for predictive modeling and enhanced accuracy in handling non-linear optimization problems. The Multi-Objective Optimization on the basis of Ratio Analysis (MOORA) technique was implemented to establish the optimal parametric combination that simultaneously minimizes surface roughness, vibration, and acoustic emission. Statistical analysis of variance (ANOVA) was performed on the MOORA-derived performance index (Yᵢ) it revealed that feed rate exhibits the most pronounced influence on the combined response characteristics, contributing 71.05 % to the total variation. The optimization results indicate that the optimal machining conditions for achieving minimal surface roughness, vibration amplitude, and acoustic emission are: feed rate of 0.20 mm/rev, depth of cut of 0.75 mm, and cutting speed of 80 m/min.