Learning Power Systems Waveform Incipient Patterns Through Few-Shot Meta-Learning
Incipient faults (IFs) are abnormal states before the permanent failure of power equipment. IFs are typically transient and generally do not trigger the operation of relay protection devices. This leads the difficulty in capturing IF data from waveform monitoring or recording devices. However, tradi...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Open Access Journal of Power and Energy |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10713429/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832592865963802624 |
|---|---|
| author | Lixian Shi Qiushi Cui Yang Weng Yigong Zhang Shilong Chen Jian Li Wenyuan Li |
| author_facet | Lixian Shi Qiushi Cui Yang Weng Yigong Zhang Shilong Chen Jian Li Wenyuan Li |
| author_sort | Lixian Shi |
| collection | DOAJ |
| description | Incipient faults (IFs) are abnormal states before the permanent failure of power equipment. IFs are typically transient and generally do not trigger the operation of relay protection devices. This leads the difficulty in capturing IF data from waveform monitoring or recording devices. However, traditional detection methods cannot achieve satisfactory performance when faced with limited data. Besides, some signal analysis methods based on waveform conversion to images cannot obtain understandable image data and cannot analyze both current and voltage signals simultaneously. To resolve these problems, a few-shot meta-learning framework for incipient fault detection (FSMLF-IFD) is proposed in this paper. For better data processing, a waveform image conversion strategy is proposed to convert waveforms into understandable images from the time domain perspective. Then, an adaptive image fusion strategy is developed to concurrently analyze voltage and current images. Next, at the meta-training stage, an adaptability-enhancing weighting initialization strategy is constructed to address the data differences between the meta-training stage and IF detection stage. Finally, an IF detection model based on convolutional neural networks (CNNs) is obtained through the fine-tuning process. In the numerical results, the IF detection and classification accuracy of FSMLF-IFD reached 0.9720 and 0.9840 based on simulation and field IF data, which validates the effectiveness of the proposed method. |
| format | Article |
| id | doaj-art-70389b81df6648f4b3ef8f2eed2bb07a |
| institution | Kabale University |
| issn | 2687-7910 |
| language | English |
| publishDate | 2024-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Open Access Journal of Power and Energy |
| spelling | doaj-art-70389b81df6648f4b3ef8f2eed2bb07a2025-01-21T00:03:08ZengIEEEIEEE Open Access Journal of Power and Energy2687-79102024-01-011153254510.1109/OAJPE.2024.347763010713429Learning Power Systems Waveform Incipient Patterns Through Few-Shot Meta-LearningLixian Shi0https://orcid.org/0000-0002-1846-2399Qiushi Cui1https://orcid.org/0000-0002-3471-3936Yang Weng2https://orcid.org/0000-0002-5267-1303Yigong Zhang3https://orcid.org/0000-0002-6412-6599Shilong Chen4Jian Li5Wenyuan Li6https://orcid.org/0000-0002-5113-7280School of Electrical Engineering, Chongqing University, Chongqing, ChinaSchool of Electrical Engineering, Chongqing University, Chongqing, ChinaDepartment of Electrical, Computer and Energy Engineering, Arizona State University, Tempe, AZ, USASchool of Electrical Engineering, Chongqing University, Chongqing, ChinaFaculty of Electric Power Engineering, Kunming University of Science and Technology, Kunming, ChinaSchool of Electrical Engineering, Chongqing University, Chongqing, ChinaSchool of Electrical Engineering, Chongqing University, Chongqing, ChinaIncipient faults (IFs) are abnormal states before the permanent failure of power equipment. IFs are typically transient and generally do not trigger the operation of relay protection devices. This leads the difficulty in capturing IF data from waveform monitoring or recording devices. However, traditional detection methods cannot achieve satisfactory performance when faced with limited data. Besides, some signal analysis methods based on waveform conversion to images cannot obtain understandable image data and cannot analyze both current and voltage signals simultaneously. To resolve these problems, a few-shot meta-learning framework for incipient fault detection (FSMLF-IFD) is proposed in this paper. For better data processing, a waveform image conversion strategy is proposed to convert waveforms into understandable images from the time domain perspective. Then, an adaptive image fusion strategy is developed to concurrently analyze voltage and current images. Next, at the meta-training stage, an adaptability-enhancing weighting initialization strategy is constructed to address the data differences between the meta-training stage and IF detection stage. Finally, an IF detection model based on convolutional neural networks (CNNs) is obtained through the fine-tuning process. In the numerical results, the IF detection and classification accuracy of FSMLF-IFD reached 0.9720 and 0.9840 based on simulation and field IF data, which validates the effectiveness of the proposed method.https://ieeexplore.ieee.org/document/10713429/Incipient fault detectionpower qualitydata scarcitywaveform imagefew-shot learningmeta-learning |
| spellingShingle | Lixian Shi Qiushi Cui Yang Weng Yigong Zhang Shilong Chen Jian Li Wenyuan Li Learning Power Systems Waveform Incipient Patterns Through Few-Shot Meta-Learning IEEE Open Access Journal of Power and Energy Incipient fault detection power quality data scarcity waveform image few-shot learning meta-learning |
| title | Learning Power Systems Waveform Incipient Patterns Through Few-Shot Meta-Learning |
| title_full | Learning Power Systems Waveform Incipient Patterns Through Few-Shot Meta-Learning |
| title_fullStr | Learning Power Systems Waveform Incipient Patterns Through Few-Shot Meta-Learning |
| title_full_unstemmed | Learning Power Systems Waveform Incipient Patterns Through Few-Shot Meta-Learning |
| title_short | Learning Power Systems Waveform Incipient Patterns Through Few-Shot Meta-Learning |
| title_sort | learning power systems waveform incipient patterns through few shot meta learning |
| topic | Incipient fault detection power quality data scarcity waveform image few-shot learning meta-learning |
| url | https://ieeexplore.ieee.org/document/10713429/ |
| work_keys_str_mv | AT lixianshi learningpowersystemswaveformincipientpatternsthroughfewshotmetalearning AT qiushicui learningpowersystemswaveformincipientpatternsthroughfewshotmetalearning AT yangweng learningpowersystemswaveformincipientpatternsthroughfewshotmetalearning AT yigongzhang learningpowersystemswaveformincipientpatternsthroughfewshotmetalearning AT shilongchen learningpowersystemswaveformincipientpatternsthroughfewshotmetalearning AT jianli learningpowersystemswaveformincipientpatternsthroughfewshotmetalearning AT wenyuanli learningpowersystemswaveformincipientpatternsthroughfewshotmetalearning |