Fault diagnosis of CBTC On-board equipment based on improved GA-BP

CBTC On-board equipment is a core component to ensure the safety of subway trains and improve transportation efficiency, and it is of great significance to quickly and effectively diagnose its fault types. Aiming at the complexity of the fault data of CBTC vehicle equipment, an intelligent fault diagnosis method of CBTC vehicle equipment based on the improved GA-BP neural network model is proposed, and the most critical ATP type faults in the CBTC system are tested for fault diagnosis. Firstly, the genetic algorithm is used to optimize the BP neural network; secondly, the rough set theory is used to reduce the ATP-type fault features to reduce the complexity of the fault features. Compared with the unoptimized BP neural network, the fault diagnosis model based on the improved GA-BP neural network has a shorter training period, and the mean square error of training is only 0.00033064, which is significantly lower than the first two neural networks.