Multi Information Fusion and Fault Diagnosis System for Motor Drive System in High Speed Train

As the motor drive system is a multi-variable, nonlinear system, it's very difficult to diagnose the faults accurately by traditional diagnosis methods. So a multi-feature information fusion intelligent monitoring and fault diagnosis algorithm is proposed in this paper which combing the time-domain and frequency-domain information for feature fusion, multi neural network (MNN) for diagnosis fusion. The signal conditioning and fault diagnosis boards were developed and then integrated to the diagnostic system on the train. The algorithm was realized in the system and the system was installed and validated on the train. The diagnosis results showed that the intelligent fusion method was much more accurate and time-saving compared with traditional methods.