Evaluating and Optimising Tribological Parameters of Enhanced Two-Step Stir Cast Al6061/Nano-SiO₂ Composite Using Machine Learning Techniques

This study aims to analyse the potential of a two-step stir casting technique in producing Al6061/nano-SiO₂ composites to enhance wear resistance in dry sliding conditions. This approach then aims to enhance the dispersion of the nano-SiO₂ reinforcements to achieve enhanced anti-wear performance. Using the L27 orthogonal array, wear behaviour influenced by the different loads (5 N, 10 N and 15 N) has been investigated. For microstructural characterisation of the composite, scanning electron microscopy (SEM), energy-dispersive X-ray analysis and X-ray diffraction were conducted to observe the dispersion of reinforcements and any signs of interfacial activities. A mathematical model using response surface methodology was also incorporated to determine the composites’ optimum wear rate and COF for the three distances: 1000 m, 2000 m and 3000 m. The findings indicated that adding the nano-SiO₂ affected wear rate and COF, further supported by SEM images of the worn surfaces revealing that the 3 wt%. Thus, nano-SiO₂ composite shows enhanced wear resistance and hardness of the surface. To further examine the impact of wear parameters under various loads, an artificial neural network (ANN) was used. Furthermore, to improve the composites’ surface properties, they were also treated using heat treatment techniques, such as precipitation hardening, quenching and ageing. This treatment increased microhardness by 58% for the 3-wt% composite.