A Real-Time Diagnosis Method for Contactor Faults in Electric Traction Drive System

The contactor is a key component in the precharging circuit. Once it malfunctions, it will interrupt the power flow in electric traction drive system (ETDS), resulting in parking. There are various reasons that can cause abnormalities in contactors, and not all fault causes require parking for maintenance. However, there is currently no effective method for tracing the cause of the malfunction, resulting in only indiscriminate emergency parking handling. Therefore, this article proposes a fault diagnosis strategy for contactor faults. First, the ETDS operating condition estimation model is proposed based on a combination of system operating mechanism and data-driven approach, and then the switch state of the contactor is derived using the correspondence between operating conditions and switch states. Second, the fault modes of the contactor are analyzed based on the interaction principle between control instructions and feedback instructions, and fault diagnosis rules are proposed. Third, in order to reduce the computational complexity of the diagnostic model, diagnosis enablement and optimal feature extraction mechanisms have been established, and the framework is being built including diagnosis enablement, operating condition estimation, and cause diagnosis. Finally, the effectiveness of the method has been verified based on the experimental platform.