| While the EMU is pursuing the gradual improvement in speed,safety is also a top priority.Traction motor is the main component of traction and braking,and its safety and reliability are of great importance.Three-phase squirrel cage induction motor is typically used in traction motor.According to its structure.In this paper,the stator inter-turn short circuit fault is selected as the principal research problem.The stator insulation system of traction motor of EMU is the weak link,among which the stator winding inter-turn short circuit fault is the most important.As the interturn short circuit fault diagnosis system of traction motor,this system aims to diagnose the inter-turn short circuit fault in the early stage to prevent the fault from deteriorating.In this paper,according to the existing motor fault diagnosis technology,this paper focuses on the EMU traction motor inter-turn fault diagnosis method,determined based on the stator current phase difference using artificial neural network technology for fault diagnosis method.The neural network is designed,and the training method of the neural network is studied,a complete neural network training method is proposed,and the training program of the neural network is developed.On the premise of complete fault diagnosis method is put forward,according to the requirements of this method is put forward train TCU system for inter-turn short circuit fault diagnosis of a complete technology solutions,including the setting method of fault diagnosis experiment condition,the data collection method,digital filtering method and failure characteristics of three phase current phase difference method is put forward.The method presented in this paper can be applied to the development of embedded TCU fault diagnosis program and the self-diagnosis of interturn short-circuit fault of the system. |
