Traction power supply system fault diagnosis model based on Bayesian network considering the uncertainty of information

Abstract

Fault diagnosis is of great significance to maintain safe and stable operation of the traction power supply system. Fast diagnosing fault to restore power supply is important. A kind of TPSS fault diagnosis model based on Bayesian network considering the uncertainty of information is proposed in this paper. The fault diagnosis model inherits the capability of dealing with uncertain information of Bayesian network and reduces the uncertainty from original source. A fault simulation model is proposed to produce sample data for learning parameters. Accuracy of fault diagnosis model and fault simulation model are verified by an example of failure event.