학술논문

Open-Circuit Fault Diagnosis Method of ANPC Three-Level Inverter Based on Joint Optimized Echo State Network
Document Type
Conference
Source
2023 IEEE 2nd International Power Electronics and Application Symposium (PEAS) Power Electronics and Application Symposium (PEAS), 2023 IEEE 2nd International. :1912-1915 Nov, 2023
Subject
Power, Energy and Industry Applications
Legged locomotion
Fault diagnosis
Simulation
Fault detection
Switches
Feature extraction
Wavelet analysis
3-Level ANPC inverter
open-circuit fault diagnosis
wavelet packet
echo state network
Language
Abstract
A method based on Joint Optimized Echo State Network for inverter diagnosis is proposed. Firstly, the characteristics of the leg voltages under different fault conditions are extracted by wavelet packet analysis, and then low dimensional and discriminating feature vectors are extracted by Linear Discriminant Analysis. Next, Echo State Network optimized by an improved PSO is adopted to classify feature vectors of the wavelet packet energy spectrum. Simulation results prove the proposed method in this paper has more accuracy and faster detection speed than other models in the comparison experiment.