Abstract: Objective: To enhance the fault warning function and accuracy of urban rail transit vehicles, and further realize intelligent operation and maintenance of the vehicles, an intelligent diagnosis and fault warning system for the vehicle braking system is constructed. Method: According to the extensively collected operational data and fault data of urban rail transit vehicle braking system on the ground end, the causes, symptoms, and characteristics of various brake system faults in the operational data are thoroughly studied, intelligent diagnosis algorithm models based on both mechanism and machine learning are constructed. By efficiently and quickly transmitting vehicle data to the ground intelligent operation and maintenance platform through vehicle\|ground linkage, the two types of intelligent diagnosis algorithm models are integrated into the ground platform to achieve intelligent fault diagnosis and fault warning of vehicle operational data. Maintenance suggestions are automatically generated based on the fault diagnosis and fault warning results, to reduce the on\|route failure rate of vehicles and improve the availability and maintenance efficiency of urban rail transit vehicles. Result &Conclusion: Through actual application on a certain urban rail transit line, it has been verified that the system can warn and locate most of the fault points in urban rail transit vehicle braking system, and improve the reliability and maintenance efficiency of the above vehicle braking system.