AUTHOR=Yang Hongxin , Peng Zishun , Xu Qijin , Huang Tingxuan , Zhu Xiangou 

TITLE=Inverter fault diagnosis based on Fourier transform and evolutionary neural network

JOURNAL=Frontiers in Energy Research

VOLUME=Volume 10 - 2022

YEAR=2023

URL=https://www.frontiersin.org/journals/energy-research/articles/10.3389/fenrg.2022.1090209

DOI=10.3389/fenrg.2022.1090209

ISSN=2296-598X

ABSTRACT=The traditional fault diagnosis methods for converter have the problems such as high eigenvector dimension, too many independent variables, complex neural network structure, low precision, slow response speed and poor anti-interference. Therefore, this paper proposes an inverter fault diagnosis method based on Fourier Transform and evolutionary neural network. This method combines the amplitude of low-frequency harmonic component of three-phase inverter output current which is obtained by Fast Fourier Transform (FFT) and the average value in a period of three-phase inverter output current into the fault eigenvector. This method uses evolutionary neural network trained by combining genetic algorithm (GA), ant colony optimization (ACO) algorithm and Back-propagation (BP) algorithm to realize fault diagnosis. This method can effectively resist noise interference and reduce the number of independent variables in the part of feature extraction, so that it can simplify the network model. In addition, this method can avoid the network training from trapping in local optima in the part of fault classification, with high accuracy and fast response speed. MATLAB/Simulink is used to build a three-phase inverter simulation model, LabVIEW is used to build a data collection platform for feature extraction. Eigenvectors under different fault conditions are collected to obtain 500 groups of data for network model training. The simulation results show that the proposed algorithm and method of fault feature extraction can effectively detect and locate the insulated-gate bipolar transistor (IGBT) open circuit (OC) faults in three-phase inverter and can be applied to online monitoring.