Uncertainty-based Prediction of Wear Resistance for Brake Pad under Braking Conditions

Brake pads are an important part of the brake pair, excessive wear will affect the braking performance of the car, which is very unfavorable to driving safety. The wear of brake pads is complex, non-linear and multi-dimensional. Therefore, this paper uses genetic algorithm to improve the initial weight and threshold of BP neural network, and builds a brake pad wear prediction model based on braking conditions with test data as samples. Moreover, considering the uncertainty and inaccuracy of data collection, the wear process is predicted with uncertainty. The simulation results show that the GA-BP neural network optimized by genetic algorithm has faster convergence speed and higher prediction accuracy than BP neural network, and the probability that the interval prediction value obtained by uncertain prediction contains the predicted value and the actual value is 100%, which improves the reliability of prediction results.