Fault Diagnosis of Micro Grid Inverter Based on Wavelet Transform and Probabilistic Neural Network

This paper focuses on the fault diagnosis of IGBT switch for microgrid three-phase inverters. Based on the analysis of all kinds of fault mechanisms, firstly, the fault simulation model of microgrid inverter is built on MATLAB / Simulink to simulate the fault state of inverter switch under various faults. The load-side three-phase current fault data is collected. Then, wavelet transform is used to extract feature parameters, and normalize the process in the form of energy to construct the fault feature vector. Finally, the genetic algorithm-based probability neural network (PNN) is used for classification. The experimental results show that the method that combining wavelet transform and PNN can diagnose the faults of microgrid inverter accurately and effectively.