Forecasting Tribological Properties of Wrought AZ91D Magnesium Alloy Using Soft Computing Model

The wear characteristics of wrought magnesium alloy AZ91D is assessed by varying the wear test parameters namely sliding velocity, sliding distance and normal load in the pin-on-disc tribometer. The experimental results are used to develop a statistical model, and soft computing models based on artificial neural network and Sugeno–Fuzzy logic to predict the wear rate of AZ91D alloy. Sugeno–Fuzzy model had the highest accuracy in prediction and hence used to study the effect of wear test parameters on the wear rate of AZ91D alloy. It is observed that the wear rate increases with decrease in load, increase in sliding velocity, and increase in sliding distance.