A robust data-driven method for open-circuit fault diagnosis of power switches in three-phase inverters with low-quality data
Machine learning (ML) techniques have shown great potential for power converter fault diagnosis. However, the data measured by the diagnostic processor may be corrupted in real-world applications, which would degrade the performance of ML-based diagnostic models. This article proposed a robust data-...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/182763 |
| _version_ | 1826133446112575488 |
|---|---|
| author | Xia, Yan Xu, Yan |
| author2 | School of Electrical and Electronic Engineering |
| author_facet | School of Electrical and Electronic Engineering Xia, Yan Xu, Yan |
| author_sort | Xia, Yan |
| collection | NTU |
| description | Machine learning (ML) techniques have shown great potential for power converter fault diagnosis. However, the data measured by the diagnostic processor may be corrupted in real-world applications, which would degrade the performance of ML-based diagnostic models. This article proposed a robust data-driven method for power switch open-circuit fault diagnosis under low-quality data issues with missing values, outliers, noises. At offline stage, a robust subspace matrix is first trained and adopted to recover the corrupted data from missing data and outliers. Then, the recovered data is further denoized through joint sparse coding and transform learning, where a transform weight matrix can be obtained. By using the processed data as the input, a random vector functional link network is trained to generate the diagnostic model. At online stage, real-time current signals are firstly recovered\denoized based on the trained matrixes and then sent to the trained diagnostic model to generate the diagnostic result. Simulation and real-time tests have demonstrated the high accuracy and strong robustness of the proposed method under various scenarios. |
| first_indexed | 2025-03-09T16:09:25Z |
| format | Journal Article |
| id | ntu-10356/182763 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2025-03-09T16:09:25Z |
| publishDate | 2025 |
| record_format | dspace |
| spelling | ntu-10356/1827632025-02-24T07:11:54Z A robust data-driven method for open-circuit fault diagnosis of power switches in three-phase inverters with low-quality data Xia, Yan Xu, Yan School of Electrical and Electronic Engineering Centre for Power Engineering Engineering Low-quality data Open-circuit fault diagnosis Machine learning (ML) techniques have shown great potential for power converter fault diagnosis. However, the data measured by the diagnostic processor may be corrupted in real-world applications, which would degrade the performance of ML-based diagnostic models. This article proposed a robust data-driven method for power switch open-circuit fault diagnosis under low-quality data issues with missing values, outliers, noises. At offline stage, a robust subspace matrix is first trained and adopted to recover the corrupted data from missing data and outliers. Then, the recovered data is further denoized through joint sparse coding and transform learning, where a transform weight matrix can be obtained. By using the processed data as the input, a random vector functional link network is trained to generate the diagnostic model. At online stage, real-time current signals are firstly recovered\denoized based on the trained matrixes and then sent to the trained diagnostic model to generate the diagnostic result. Simulation and real-time tests have demonstrated the high accuracy and strong robustness of the proposed method under various scenarios. Ministry of Education (MOE) This work was supported by the Ministry of Education, Republic of Singapore, under Grant AcRF TIER-1 RT9/22. 2025-02-24T07:11:54Z 2025-02-24T07:11:54Z 2025 Journal Article Xia, Y. & Xu, Y. (2025). A robust data-driven method for open-circuit fault diagnosis of power switches in three-phase inverters with low-quality data. IEEE Transactions On Power Electronics, 40(4), 5949-5958. https://dx.doi.org/10.1109/TPEL.2024.3494858 0885-8993 https://hdl.handle.net/10356/182763 10.1109/TPEL.2024.3494858 2-s2.0-85209725152 4 40 5949 5958 en RT9/22 IEEE Transactions on Power Electronics © 2024 IEEE. All rights reserved. |
| spellingShingle | Engineering Low-quality data Open-circuit fault diagnosis Xia, Yan Xu, Yan A robust data-driven method for open-circuit fault diagnosis of power switches in three-phase inverters with low-quality data |
| title | A robust data-driven method for open-circuit fault diagnosis of power switches in three-phase inverters with low-quality data |
| title_full | A robust data-driven method for open-circuit fault diagnosis of power switches in three-phase inverters with low-quality data |
| title_fullStr | A robust data-driven method for open-circuit fault diagnosis of power switches in three-phase inverters with low-quality data |
| title_full_unstemmed | A robust data-driven method for open-circuit fault diagnosis of power switches in three-phase inverters with low-quality data |
| title_short | A robust data-driven method for open-circuit fault diagnosis of power switches in three-phase inverters with low-quality data |
| title_sort | robust data driven method for open circuit fault diagnosis of power switches in three phase inverters with low quality data |
| topic | Engineering Low-quality data Open-circuit fault diagnosis |
| url | https://hdl.handle.net/10356/182763 |
| work_keys_str_mv | AT xiayan arobustdatadrivenmethodforopencircuitfaultdiagnosisofpowerswitchesinthreephaseinverterswithlowqualitydata AT xuyan arobustdatadrivenmethodforopencircuitfaultdiagnosisofpowerswitchesinthreephaseinverterswithlowqualitydata AT xiayan robustdatadrivenmethodforopencircuitfaultdiagnosisofpowerswitchesinthreephaseinverterswithlowqualitydata AT xuyan robustdatadrivenmethodforopencircuitfaultdiagnosisofpowerswitchesinthreephaseinverterswithlowqualitydata |