Correlating the Experimental and ANN Modeled Tribological and Mechanical Behaviour of Nano SiC—Cenosphere Particulate Reinforced Al7075 Composite Fabricated by Ultrasonic Assisted Stir-Casting Route

Aluminum-based particulate composites, particularly those with nanoparticle reinforcements, are widely studied and utilized across various fields. The current study compares the mechanical and wear behavior of nanoparticle-reinforced Al7075 composites with predictions from an Artificial Neural Network (ANN) model. Two lots of composites were fabricated, reinforcing 0.6, 1.2, and 1.8 wt % of nano-sized Silicon Carbide (SiC) and Cenosphere (CS) by ultrasonic cavitation-assisted stir casting. The microstructure, dispersion of reinforcements, and wear surfaces were captured using a Scanning Electron Microscope (SEM). During the mechanical behaviour study, tensile, compression, and hardness tests were performed. The mechanical responses of the nano-reinforcements were theoretically calculated considering various strengthening mechanisms. The sliding wear study used three input factors–speed, distance, and load–each at three levels. SEM images confirmed uniform reinforcement dispersion. Cenosphere reinforcements had a greater impact on mechanical behavior than SiC. Cenosphere reinforcements significantly enhanced the mechanical behavior compared to SiC. Experimental yield strength exceeded theoretical values, with load transfer being the dominant strengthening mechanism, followed by dislocation strengthening. Al7075/Cenosphere composites exhibit a 71.9% lower wear rate, while Al7075/SiC shows a 45.93% lower wear rate compared to the pure Al7075 alloy. An ANN regression wear model was found to be 90.01% accurate compared to the experimental results.