Affiliation:
1. School of Computer Science and Engineering, Xi’an Technological University, Xi’an 710021, China
2. Beijing Aerospace Automatic Control Institution, Beijing 100854, China
Abstract
This study addresses the challenges of limited fault samples, noise interference, and low accuracy in existing fault diagnosis methods for three-phase inverters under real acquisition conditions. To increase the number of samples, Wavelet Packet Decomposition (WPD) denoising and a Conditional Variational Auto-Encoder (CVAE) are used for sample enhancement based on the existing faulty samples. The resulting dataset is then normalized, pre-processed, and used to train an improved deep residual network (SE-ResNet18) fault diagnosis model with a channel attention mechanism. Results show that the augmented fault samples improve the diagnosis accuracy compared with the original samples. Furthermore, the SE-ResNet18 model achieves higher fault diagnosis accuracy with fewer iterations and faster convergence, indicating its effectiveness in accurately diagnosing inverter open-circuit faults across various sample situations.
Subject
Electrical and Electronic Engineering,Computer Networks and Communications,Hardware and Architecture,Signal Processing,Control and Systems Engineering
Reference24 articles.
1. Fault Diagnosis Technique for Three-level Inverters Based on ICEEMDAN-FE and Support Vector Machine;Cao;Locomot. Electr. Drive,2023
2. Chen, T. (2021). Research on Open-Circuit Fault Diagnosis Method for Three-Phase Voltage Source Inverters. [Ph.D. Thesis, University of Science and Technology Beijing].
3. A Review of Fault Diagnosis Methods for Multilevel Inverters;Song;Micromotors,2019
4. Diagnosing Open-Circuit Faults in Three-Level Grid-Connected Inverters Based on an Adaptive Sliding-Mode Observer;Xu;Trans. China Electrotech. Soc.,2023
5. Early Fault Parameter Identification for Grid-Connected Inverters Based on a Model;Fan;J. China Three Gorges Univ. (Nat. Sci.),2022
Cited by
6 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献