A novel method for power transformer fault diagnosis considering imbalanced data samples
IntroductionMachine learning-based power transformer fault diagnosis methods often grapple with the challenge of imbalanced fault case distributions across different categories, potentially degrading diagnostic accuracy. To address this issue and enhance the accuracy and operational efficiency of po...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2025-01-01
|
| Series: | Frontiers in Energy Research |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fenrg.2024.1500548/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841558768736796672 |
|---|---|
| author | Jun Chen Yong Wang Lingming Kong Yilong Chen Mianzhi Chen Qian Cai Gehao Sheng |
| author_facet | Jun Chen Yong Wang Lingming Kong Yilong Chen Mianzhi Chen Qian Cai Gehao Sheng |
| author_sort | Jun Chen |
| collection | DOAJ |
| description | IntroductionMachine learning-based power transformer fault diagnosis methods often grapple with the challenge of imbalanced fault case distributions across different categories, potentially degrading diagnostic accuracy. To address this issue and enhance the accuracy and operational efficiency of power transformer fault diagnosis models, this paper presents a novel fault diagnosis model that integrates Neighborhood Component Analysis (NCA) and k-Nearest Neighbor (KNN) learning, with the incorporation of correction factors.MethodsThe methodology begins by introducing a correction factor into the objective function of the NCA algorithm to reduce the impact of sample imbalance on model training. We derive a sample parameter correlation quantization matrix from oil chromatography fault data using association rules, which serves as the initial value for the NCA algorithm’s training metric matrix. The metric matrix obtained from training is then applied to perform a mapping transformation on the input data for the KNN classifier, thereby reducing the distance between similar samples and enhancing KNN classification performance. Hyperparameter tuning is achieved through the Bayesian optimization algorithm to identify the model parameter set that maximizes test set accuracy.ResultsAnalysis of the transformer fault case library reveals that the model proposed in this paper reduces diagnostic time by nearly half compared to traditional machine learning diagnosis models. Additionally, the accuracy for minority sample classes is improved by at least 15% compared to other models.DiscussionThe integration of NCA and KNN with correction factors not only mitigates the effects of sample imbalance but also significantly enhances the operational efficiency and diagnostic accuracy of power transformer fault diagnosis. The proposed model’s performance improvements highlight the potential of this approach for practical applications in the field of power transformer maintenance and diagnostics. |
| format | Article |
| id | doaj-art-56881a632afc48e583d29c9b036bb4a2 |
| institution | Kabale University |
| issn | 2296-598X |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Energy Research |
| spelling | doaj-art-56881a632afc48e583d29c9b036bb4a22025-01-06T06:58:52ZengFrontiers Media S.A.Frontiers in Energy Research2296-598X2025-01-011210.3389/fenrg.2024.15005481500548A novel method for power transformer fault diagnosis considering imbalanced data samplesJun Chen0Yong Wang1Lingming Kong2Yilong Chen3Mianzhi Chen4Qian Cai5Gehao Sheng6Guangzhou Power Supply Bureau of Guangdong Power Grid Co. Ltd., Guangzhou, Guangdong, ChinaGuangzhou Power Supply Bureau of Guangdong Power Grid Co. Ltd., Guangzhou, Guangdong, ChinaGuangzhou Power Supply Bureau of Guangdong Power Grid Co. Ltd., Guangzhou, Guangdong, ChinaGuangzhou Power Supply Bureau of Guangdong Power Grid Co. Ltd., Guangzhou, Guangdong, ChinaGuangzhou Power Supply Bureau of Guangdong Power Grid Co. Ltd., Guangzhou, Guangdong, ChinaDepartment of Electrical Engineering, Shanghai Jiao Tong University, Shanghai, ChinaDepartment of Electrical Engineering, Shanghai Jiao Tong University, Shanghai, ChinaIntroductionMachine learning-based power transformer fault diagnosis methods often grapple with the challenge of imbalanced fault case distributions across different categories, potentially degrading diagnostic accuracy. To address this issue and enhance the accuracy and operational efficiency of power transformer fault diagnosis models, this paper presents a novel fault diagnosis model that integrates Neighborhood Component Analysis (NCA) and k-Nearest Neighbor (KNN) learning, with the incorporation of correction factors.MethodsThe methodology begins by introducing a correction factor into the objective function of the NCA algorithm to reduce the impact of sample imbalance on model training. We derive a sample parameter correlation quantization matrix from oil chromatography fault data using association rules, which serves as the initial value for the NCA algorithm’s training metric matrix. The metric matrix obtained from training is then applied to perform a mapping transformation on the input data for the KNN classifier, thereby reducing the distance between similar samples and enhancing KNN classification performance. Hyperparameter tuning is achieved through the Bayesian optimization algorithm to identify the model parameter set that maximizes test set accuracy.ResultsAnalysis of the transformer fault case library reveals that the model proposed in this paper reduces diagnostic time by nearly half compared to traditional machine learning diagnosis models. Additionally, the accuracy for minority sample classes is improved by at least 15% compared to other models.DiscussionThe integration of NCA and KNN with correction factors not only mitigates the effects of sample imbalance but also significantly enhances the operational efficiency and diagnostic accuracy of power transformer fault diagnosis. The proposed model’s performance improvements highlight the potential of this approach for practical applications in the field of power transformer maintenance and diagnostics.https://www.frontiersin.org/articles/10.3389/fenrg.2024.1500548/fullfault diagnosistransformerk-nearest neighbor learningmachine learningimbalanced samples |
| spellingShingle | Jun Chen Yong Wang Lingming Kong Yilong Chen Mianzhi Chen Qian Cai Gehao Sheng A novel method for power transformer fault diagnosis considering imbalanced data samples Frontiers in Energy Research fault diagnosis transformer k-nearest neighbor learning machine learning imbalanced samples |
| title | A novel method for power transformer fault diagnosis considering imbalanced data samples |
| title_full | A novel method for power transformer fault diagnosis considering imbalanced data samples |
| title_fullStr | A novel method for power transformer fault diagnosis considering imbalanced data samples |
| title_full_unstemmed | A novel method for power transformer fault diagnosis considering imbalanced data samples |
| title_short | A novel method for power transformer fault diagnosis considering imbalanced data samples |
| title_sort | novel method for power transformer fault diagnosis considering imbalanced data samples |
| topic | fault diagnosis transformer k-nearest neighbor learning machine learning imbalanced samples |
| url | https://www.frontiersin.org/articles/10.3389/fenrg.2024.1500548/full |
| work_keys_str_mv | AT junchen anovelmethodforpowertransformerfaultdiagnosisconsideringimbalanceddatasamples AT yongwang anovelmethodforpowertransformerfaultdiagnosisconsideringimbalanceddatasamples AT lingmingkong anovelmethodforpowertransformerfaultdiagnosisconsideringimbalanceddatasamples AT yilongchen anovelmethodforpowertransformerfaultdiagnosisconsideringimbalanceddatasamples AT mianzhichen anovelmethodforpowertransformerfaultdiagnosisconsideringimbalanceddatasamples AT qiancai anovelmethodforpowertransformerfaultdiagnosisconsideringimbalanceddatasamples AT gehaosheng anovelmethodforpowertransformerfaultdiagnosisconsideringimbalanceddatasamples AT junchen novelmethodforpowertransformerfaultdiagnosisconsideringimbalanceddatasamples AT yongwang novelmethodforpowertransformerfaultdiagnosisconsideringimbalanceddatasamples AT lingmingkong novelmethodforpowertransformerfaultdiagnosisconsideringimbalanceddatasamples AT yilongchen novelmethodforpowertransformerfaultdiagnosisconsideringimbalanceddatasamples AT mianzhichen novelmethodforpowertransformerfaultdiagnosisconsideringimbalanceddatasamples AT qiancai novelmethodforpowertransformerfaultdiagnosisconsideringimbalanceddatasamples AT gehaosheng novelmethodforpowertransformerfaultdiagnosisconsideringimbalanceddatasamples |