Modelling of brake friction material performance at elevated temperatures

Abstract

Temperature sensitivity of friction materials has always been a critical aspect of their reliable functioning. The loss of a friction material's performance at elevated temperature cannot be easily predicted. The influence of composition and manufacturing parameters can be adjusted in synergy with the brake's operation conditions by embedding artificial intelligence. The influences of formulation, manufacturing, and operation conditions of brake friction materials have been investigated by means of artificial neural networks. A two hidden-layer neural network model trained by the Bayesian Regulation algorithm is shown to be able to accurately predict the complex influences on the friction material performance at elevated temperatures.