Fault Diagnosis for PWM Inverter Based on LMD

Focusing on the character of nonlinear and the non-stationary characteristics of the fault signal of inverter, this paper proposes a method of open-circuit fault diagnosis for PWM inverter which based on local mean decomposition(LMD). the simulation model of inverter driving control system is built, getting three-phase output current of different fault conditions. LMD method is used to decompose three-phase output current respectively. the feature vector of fault is constructed with energy value of the principal component and proportional coefficient of the forward current in each phase. Finally, the model of fault diagnosis is constructed with neural network to locate the leg of bridge with open-circuit fault of the inverter. According to the simulation and experiment, the results show that the method by using local mean decomposition for fault signal are good, it provides a new approach of fault diagnosis for PWM inverter with open-circuit .