Three-Phase Inverter Fault Diagnosis Strategy Based on Compressed Sensing and Wavelet Packet Decomposition
Chaobo Chen, Wenjie Li, Binbin Zhang, Song Gao, Xueqin Yang · 2021 China Automation Congress (CAC) · 2021
To solve the open circuit fault problem in the inverter system. a three-phase inverter fault feature extraction method based on compressed sensing and wavelet packet decomposition (CS-WPD) is proposed. Take the open circuit fault diagnosis of insulated gate bipolar type (IGBT) in three-phase inverter as an example. Firstly, a simulation model of the three-phase inverter topology is established, the phase voltage is adopted as the fault signal, and convert the three-phase voltage signal to two-phase through Clark transformation, and then use the CS algorithm to sample the data. Compression is used to improve the signal processing speed. Secondly, the WPD of energy entropy is used to extract the characteristics of the voltage signal. Finally, the BP neural network is optimized by the beetle antennae search algorithm (BAS) to implement fault diagnosis and compare it with the traditional BP neural network. Network identification methods are compared. The simulation results show that the accuracy and robustness of the system are effectively improved. The accuracy of IGBT tube fault diagnosis reaches 98.29%, the results show that the proposed method is feasible and effective.