Application of Improved Neural Network in the Automotive Engine Fault Diagnosis

Abstract

Back Propagation neural network is a network, which is a multilayer feedforward network of according to the error back propagation algorithm training, i.e., BP neural network. The good nonlinear mapping ability of BP neural network can be a good application in engine fault diagnosis, but the traditional BP network has the trend of forgetting old samples during the training process when learning new samples ,and exists the drawback of low training accuracy. Therefore, a model of improved BP neural network is constructed. A neural network algorithm of increased state feedback in the output layer is designed in this paper. The simulation results show the proposed algorithm can effectively improve the BP neural network training accuracy, and achieve misfire diagnosis more accurately.