Integrated cryogenic, statistical and ANN approach for performance optimization of Al6061/SiC/Gr hybrid composites

This work investigates the synergistic influence of silicon carbide (SiC)–graphite (Gr) hybrid reinforcement and deep cryogenic treatment (DCT) on the mechanical and tribological performance of Al6061 hybrid metal matrix composites (HMMCs). Composites containing a fixed 13 wt% SiC and varying Gr content (1–3 wt%) were fabricated by stir casting. Based on mechanical evaluation, the 13 wt% SiC–2 wt% Gr composition was identified as optimal and subjected to DCT at − 196 °C for 24 h. Cryo-treated composites exhibited improvements of ~ 8% in ultimate tensile strength, ~ 20% in yield strength, and ~ 8% in hardness compared to untreated counterparts, accompanied by a reduction in ductility. Dry sliding wear behavior was assessed using a pin-on-disc tribometer following a Taguchi L16 experimental design, considering load, sliding speed, sliding distance, and reinforcement content. Analysis of variance revealed applied load as the dominant factor governing wear scar depth (WSD) and coefficient of friction (COF). SEM and EDAX analyses indicated refined microstructures, improved reinforcement dispersion, and phase transformations induced by DCT, accounting for the enhanced properties. An artificial neural network (ANN) model was developed to predict WSD and COF, demonstrating high predictive accuracy and clear distinguishing between untreated and cryo-treated composites. The results confirm DCT as an effective post-processing strategy for improving the tribo-mechanical performance of Al6061/SiC/Gr HMMCs, supported by integrated experimental, statistical, and machine learning approaches.