Back to Search Start Over

A General Diode Open-Circuit Fault Diagnosis Method for Autotransformer Rectifier Unit Considering Grid Voltage Disturbance

Authors :
Yang, Chao
Zhang, Fanghua
Cheng, Long
Zhang, Zhaorong
Source :
IEEE Journal of Emerging and Selected Topics in Power Electronics; 2023, Vol. 11 Issue: 2 p2214-2226, 13p
Publication Year :
2023

Abstract

It is crucial for the reliable operation of the autotransformer rectifier unit (ATRU) to accurately diagnose diode open-circuit (OC) fault. However, grid voltage disturbance may lead to misdiagnosis by existing methods. This article takes the 18-pulse ATRU as an example and first analyzes its sag-based fault feature in the dc-link output voltage (DLOV) caused by diode OC fault under no grid voltage disturbance. Then, the reason for misdiagnosis under grid voltage disturbance is discussed. Subsequently, a general diode OC fault diagnosis method with strong immunity to grid voltage disturbance is proposed. In the proposed method, the DLOV waveform is divided equally according to the number of rectified pulses. Hausdorff distance (HD) algorithm is employed to extract the sag-based fault feature by calculating the similarity between the DLOV sampling sequence and the constructed template sequence. To be immune to grid voltage disturbance, the template sequence is adaptively generated as per the sampling sequence in each interval. Also, the generalized minimum failure mode is defined to normalize two sequences for good universality. With the help of the phase of grid single-phase voltage, OC faults can be detected and located according to the interval positions with the sag. Finally, simulation and experimental results confirm the strong immunity of the proposed method to grid voltage disturbance and its good universality to the different ATRUs.

Details

Language :
English
ISSN :
21686777
Volume :
11
Issue :
2
Database :
Supplemental Index
Journal :
IEEE Journal of Emerging and Selected Topics in Power Electronics
Publication Type :
Periodical
Accession number :
ejs62728780
Full Text :
https://doi.org/10.1109/JESTPE.2022.3220370