Deep Ensemble Model for Unknown Partial Discharge Diagnosis in Gas-Insulated Switchgears Using Convolutional Neural Networks
Deep neural networks (DNNs) are widely used for fault classification using partial discharges (PDs) to evaluate various electrical apparatuses and achieve high classification accuracy pertaining to trained PD faults. However, there is a risk of false alarm in the case of untrained PD faults because...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2021-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9444415/ |
| _version_ | 1819010343785988096 |
|---|---|
| author | Vo-Nguyen Tuyet-Doan Ha-Anh Pho Byeongho Lee Yong-Hwa Kim |
| author_facet | Vo-Nguyen Tuyet-Doan Ha-Anh Pho Byeongho Lee Yong-Hwa Kim |
| author_sort | Vo-Nguyen Tuyet-Doan |
| collection | DOAJ |
| description | Deep neural networks (DNNs) are widely used for fault classification using partial discharges (PDs) to evaluate various electrical apparatuses and achieve high classification accuracy pertaining to trained PD faults. However, there is a risk of false alarm in the case of untrained PD faults because it is difficult for DNNs to predict data that were not included in the training process. In this paper, we research classification problems of unknown classes using PDs in gas-insulated switchgears (GISs) and propose a deep ensemble model to obtain the confidence of output probability and determine thresholds to detect unknown fault classes. The proposed model was verified by real-world phase-resolved PD (PRPD) experiments using online ultra-high frequency (UHF) PD measurement systems. The experimental results show that the proposed model achieves better unknown detection performance for the untrained PD faults and retains the classification performance for the trained PD faults. |
| first_indexed | 2024-12-21T01:10:46Z |
| format | Article |
| id | doaj.art-25baff4c346640c7b80eed21073867cb |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-21T01:10:46Z |
| publishDate | 2021-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-25baff4c346640c7b80eed21073867cb2022-12-21T19:20:55ZengIEEEIEEE Access2169-35362021-01-019805248053410.1109/ACCESS.2021.30849509444415Deep Ensemble Model for Unknown Partial Discharge Diagnosis in Gas-Insulated Switchgears Using Convolutional Neural NetworksVo-Nguyen Tuyet-Doan0https://orcid.org/0000-0002-2156-4708Ha-Anh Pho1https://orcid.org/0000-0001-9444-0020Byeongho Lee2https://orcid.org/0000-0001-8999-0861Yong-Hwa Kim3https://orcid.org/0000-0003-2183-5085Department of Electronic Engineering, Myongji University, Yongin, South KoreaDepartment of Electronic Engineering, Myongji University, Yongin, South KoreaDepartment of Electrical and Computer Engineering, Institute of New Media and Communications (INMC), Seoul National University (SNU), Seoul, South KoreaDepartment of Data Science, Korea National University of Transportation, Uiwang, South KoreaDeep neural networks (DNNs) are widely used for fault classification using partial discharges (PDs) to evaluate various electrical apparatuses and achieve high classification accuracy pertaining to trained PD faults. However, there is a risk of false alarm in the case of untrained PD faults because it is difficult for DNNs to predict data that were not included in the training process. In this paper, we research classification problems of unknown classes using PDs in gas-insulated switchgears (GISs) and propose a deep ensemble model to obtain the confidence of output probability and determine thresholds to detect unknown fault classes. The proposed model was verified by real-world phase-resolved PD (PRPD) experiments using online ultra-high frequency (UHF) PD measurement systems. The experimental results show that the proposed model achieves better unknown detection performance for the untrained PD faults and retains the classification performance for the trained PD faults.https://ieeexplore.ieee.org/document/9444415/Fault diagnosisconvolutional neural network (CNN)ensemble modelpartial discharges (PDs)gas-insulated switchgear (GIS) |
| spellingShingle | Vo-Nguyen Tuyet-Doan Ha-Anh Pho Byeongho Lee Yong-Hwa Kim Deep Ensemble Model for Unknown Partial Discharge Diagnosis in Gas-Insulated Switchgears Using Convolutional Neural Networks IEEE Access Fault diagnosis convolutional neural network (CNN) ensemble model partial discharges (PDs) gas-insulated switchgear (GIS) |
| title | Deep Ensemble Model for Unknown Partial Discharge Diagnosis in Gas-Insulated Switchgears Using Convolutional Neural Networks |
| title_full | Deep Ensemble Model for Unknown Partial Discharge Diagnosis in Gas-Insulated Switchgears Using Convolutional Neural Networks |
| title_fullStr | Deep Ensemble Model for Unknown Partial Discharge Diagnosis in Gas-Insulated Switchgears Using Convolutional Neural Networks |
| title_full_unstemmed | Deep Ensemble Model for Unknown Partial Discharge Diagnosis in Gas-Insulated Switchgears Using Convolutional Neural Networks |
| title_short | Deep Ensemble Model for Unknown Partial Discharge Diagnosis in Gas-Insulated Switchgears Using Convolutional Neural Networks |
| title_sort | deep ensemble model for unknown partial discharge diagnosis in gas insulated switchgears using convolutional neural networks |
| topic | Fault diagnosis convolutional neural network (CNN) ensemble model partial discharges (PDs) gas-insulated switchgear (GIS) |
| url | https://ieeexplore.ieee.org/document/9444415/ |
| work_keys_str_mv | AT vonguyentuyetdoan deepensemblemodelforunknownpartialdischargediagnosisingasinsulatedswitchgearsusingconvolutionalneuralnetworks AT haanhpho deepensemblemodelforunknownpartialdischargediagnosisingasinsulatedswitchgearsusingconvolutionalneuralnetworks AT byeongholee deepensemblemodelforunknownpartialdischargediagnosisingasinsulatedswitchgearsusingconvolutionalneuralnetworks AT yonghwakim deepensemblemodelforunknownpartialdischargediagnosisingasinsulatedswitchgearsusingconvolutionalneuralnetworks |