Fault diagnosis method of Ningxia photovoltaic inverter based on wavelet neural network

Abstract

Accurate fault diagnosis is the premise to ensure the safe and reliable operation of photovoltaic three-level inverter. A fault diagnosis method based on wavelet neural network is researched in the paper. First of all, the topology and the fault characteristics of three-level inverter are analyzed, the fault features are analyzed for three-level inverter when single and double IGBTs fault, the eigenvectors of phase voltage, the upper bridge arm and the lower bridge arm voltage are extracted by three-layer Wavelet Package Transform, the BP neural network is designed for training data and testing. The simulation model is built by Matlab/Simulink, the simulation results show that the method can accurately diagnose for various fault circumstances.