Prediction of Wear Characteristics of AA2219-Gr Matrix Composites Using GRNN and Taguchi-Based Approach

Aluminium matrix composites are widely used in many applications due to its numerous advantages. Experimental investigation of wear characteristics and prediction of wear is the order of the day. The present study examines the aluminium alloy AA2219-graphite (Gr) composites in terms of its wear characteristics in dry sliding condition. The conventional stir casting method was adopted to fabricate the composites with reinforcements in proportion of 1.5, 3 and 4.5 wt% to determine its capability as self-lubricating material in dry sliding conditions. Taguchi Method was utilized to study each parametric influence on the responses of wear test. A generalized regression neural network (GRNN) is used to predict the wear characteristics of AA2219-Gr composites at different factors and levels based on Taguchi L27 orthogonal array optimized outcomes. The GRNN extrapolations were confirmed by the responses of the study and the predictions were on par with the obtained results.