Experimental investigation and prediction of wear properties of Al/SiC metal matrix composites produced by thixomoulding method using Artificial Neural Networks

In this study, the wear properties of the SiC particle reinforced aluminium (A356) composite materials (MMCs), produced with thixomoulding method, were investigated both by experimental and Artificial Neural Network (ANN) model in order to determine the weight loss after the wear tests. Two different temperatures (590 °C and 600 °C) were used in production of the MMCs containing 5%, 10%, 15% and 20% SiC (vol%). The samples of MMC were tested at 2 ms⁻¹ constant sliding speed under 30 N and 60 N loads against four different sliding distances (500 m, 1000 m, 1500 m, and 2000 m). The results indicated that by increasing the production temperature increased the grain size of the MMCs was increased, but the hardness was decreased. The MMCs produced at 590 °C were found to have lower weight loss as compared with ones produced at 600 °C. In the theoretical prediction model of the MMCs, weight loss, SiC per cent, production temperature, applied weight and sliding distance were used as input values. After comparing the experimental results and the ANNs predicted data it was observed that R² was 0.9855. This shows that the developed prediction model has a high level of reliability.