A modified artificial neural network to predict the tribological properties of Al-SiC nanocomposites fabricated by accumulative roll bonding process

Abstract

This work highlights the major influence of SiC concentration and plastic deformation on boosting wear rates of aluminum composites supplemented with SiC particles comprising varied volume fractions (0–4%). It also shows how a basic neural network model augmented with a particle swarm optimizer can forecast wear rates and coefficients of friction for complicated composites. According to the experimental findings, increasing the quantity of accumulative roll bonding (ARB) enhances SiC particle dispersion homogeneity. Increasing the number of cycles and introducing additional SiC particles helped to reduce the wear rate and increase the friction coefficient. After nine ARB cycles, the Al-4 wt.% SiC nanocomposite had the best improvement in both wear rate and friction coefficient. The same sample was also used in efforts to enhance the characteristics of hardness, and it was selected as having the highest level of hardness, which has grown by 139%. All of the generated composites evaluated at four different wear loads were able to be predicted by the proposed model with great accuracy, with determination coefficient R2 values of 0.9768 and 0.9869 for the frictional coefficient and wear rates, respectively.