Research on Key Technology of Fault Diagnosis for Vehicle Network System

摘要

This paper analyzes the application classification and structure principle of vehicle network system. Aiming at the PID control algorithm of CAN node microcontroller, which is relatively poor in intelligence and accuracy, the design idea of CAN bus intelligent node based on BP neural network control algorithm is proposed. For the current network state prediction , the scheme is combined with the characteristics and requirements of vehicle fault diagnosis BP neural network fault diagnosis structure and algorithm design through the neural network fitting the corresponding relationship. At the same time, a PC diagnostic software is developed as an external diagnostic instrument. By the test of diagnosis software and body network system platform, the function of fault diagnosis is completed. The experimental results show that the method has a good denoising effect on complex vehicle fault signals. The BP network training of each system is effective, and the fault diagnosis method is achieved through the communication between the diagnosis software and the body network system platform