Modeling the sliding wear and friction properties of polyphenylene sulfide composites using artificial neural networks

Abstract

In the present study artificial neural network (ANN) approach was used for the prediction of wear and friction properties of polyphenylene sulfide (PPS) composites. Within an importance analysis the relevance of characteristic mechanical and thermo-mechanical input variables was assessed in predicting the response variable (specific wear rate and coefficient of friction). The latter is believed to be of help for a better understanding of the wear process with these materials. An optimal brain surgeon (OBS) method was applied to prune the ANN architecture by identifying and removing irrelevant nodes in its structure. The goal was minimizing the training computational cost and improving prediction. Finally, the optimized ANN was utilized to gain knowledge for the tribological properties of new material combinations, which were not tested. The quality of prediction was good when comparing the predicted and real test values.