Cost-Sensitive Extremely Randomized Trees Algorithm for Online Fault Detection of Wind Turbine Generators
The number of normal samples of wind turbine generators is much larger than the number of fault samples. To solve the problem of imbalanced classification in wind turbine generator fault detection, a cost-sensitive extremely randomized trees (CS-ERT) algorithm is proposed in this paper, in which the...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2021-05-01
|
| Series: | Frontiers in Energy Research |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fenrg.2021.686616/full |
| _version_ | 1818606828307611648 |
|---|---|
| author | Mingzhu Tang Yutao Chen Huawei Wu Qi Zhao Wen Long Victor S. Sheng Jiabiao Yi |
| author_facet | Mingzhu Tang Yutao Chen Huawei Wu Qi Zhao Wen Long Victor S. Sheng Jiabiao Yi |
| author_sort | Mingzhu Tang |
| collection | DOAJ |
| description | The number of normal samples of wind turbine generators is much larger than the number of fault samples. To solve the problem of imbalanced classification in wind turbine generator fault detection, a cost-sensitive extremely randomized trees (CS-ERT) algorithm is proposed in this paper, in which the cost-sensitive learning method is introduced into an extremely randomized trees (ERT) algorithm. Based on the classification misclassification cost and class distribution, the misclassification cost gain (MCG) is proposed as the score measure of the CS-ERT model growth process to improve the classification accuracy of minority classes. The Hilbert-Schmidt independence criterion lasso (HSICLasso) feature selection method is used to select strongly correlated non-redundant features of doubly-fed wind turbine generators. The effectiveness of the method was verified by experiments on four different failure datasets of wind turbine generators. The experiment results show that average missing detection rate, average misclassification cost and gMean of the improved algorithm better than those of the ERT algorithm. In addition, compared with the CSForest, AdaCost and MetaCost methods, the proposed method has better real-time fault detection performance. |
| first_indexed | 2024-12-16T14:17:03Z |
| format | Article |
| id | doaj.art-d49bc9dfa9ee41c4bdfa206013d1d186 |
| institution | Directory Open Access Journal |
| issn | 2296-598X |
| language | English |
| last_indexed | 2024-12-16T14:17:03Z |
| publishDate | 2021-05-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Energy Research |
| spelling | doaj.art-d49bc9dfa9ee41c4bdfa206013d1d1862022-12-21T22:28:34ZengFrontiers Media S.A.Frontiers in Energy Research2296-598X2021-05-01910.3389/fenrg.2021.686616686616Cost-Sensitive Extremely Randomized Trees Algorithm for Online Fault Detection of Wind Turbine GeneratorsMingzhu Tang0Yutao Chen1Huawei Wu2Qi Zhao3Wen Long4Victor S. Sheng5Jiabiao Yi6School of Energy and Power Engineering, Changsha University of Science and Technology, Changsha, ChinaSchool of Energy and Power Engineering, Changsha University of Science and Technology, Changsha, ChinaHubei Key Laboratory of Power System Design and Test for Electrical Vehicle, Hubei University of Arts and Science, Xiangyang, ChinaSchool of Energy and Power Engineering, Changsha University of Science and Technology, Changsha, ChinaGuizhou Key Laboratory of Economics System Simulation, Guizhou University of Finance and Economics, Guiyang, ChinaComputer Science Department, Texas Tech University, Lubbock, TX, United StatesSchool of Energy and Power Engineering, Changsha University of Science and Technology, Changsha, ChinaThe number of normal samples of wind turbine generators is much larger than the number of fault samples. To solve the problem of imbalanced classification in wind turbine generator fault detection, a cost-sensitive extremely randomized trees (CS-ERT) algorithm is proposed in this paper, in which the cost-sensitive learning method is introduced into an extremely randomized trees (ERT) algorithm. Based on the classification misclassification cost and class distribution, the misclassification cost gain (MCG) is proposed as the score measure of the CS-ERT model growth process to improve the classification accuracy of minority classes. The Hilbert-Schmidt independence criterion lasso (HSICLasso) feature selection method is used to select strongly correlated non-redundant features of doubly-fed wind turbine generators. The effectiveness of the method was verified by experiments on four different failure datasets of wind turbine generators. The experiment results show that average missing detection rate, average misclassification cost and gMean of the improved algorithm better than those of the ERT algorithm. In addition, compared with the CSForest, AdaCost and MetaCost methods, the proposed method has better real-time fault detection performance.https://www.frontiersin.org/articles/10.3389/fenrg.2021.686616/fullfault detectionfault diagnosiscost-sensitive learningextremely randomized treesclass imbalancewind turbine generator |
| spellingShingle | Mingzhu Tang Yutao Chen Huawei Wu Qi Zhao Wen Long Victor S. Sheng Jiabiao Yi Cost-Sensitive Extremely Randomized Trees Algorithm for Online Fault Detection of Wind Turbine Generators Frontiers in Energy Research fault detection fault diagnosis cost-sensitive learning extremely randomized trees class imbalance wind turbine generator |
| title | Cost-Sensitive Extremely Randomized Trees Algorithm for Online Fault Detection of Wind Turbine Generators |
| title_full | Cost-Sensitive Extremely Randomized Trees Algorithm for Online Fault Detection of Wind Turbine Generators |
| title_fullStr | Cost-Sensitive Extremely Randomized Trees Algorithm for Online Fault Detection of Wind Turbine Generators |
| title_full_unstemmed | Cost-Sensitive Extremely Randomized Trees Algorithm for Online Fault Detection of Wind Turbine Generators |
| title_short | Cost-Sensitive Extremely Randomized Trees Algorithm for Online Fault Detection of Wind Turbine Generators |
| title_sort | cost sensitive extremely randomized trees algorithm for online fault detection of wind turbine generators |
| topic | fault detection fault diagnosis cost-sensitive learning extremely randomized trees class imbalance wind turbine generator |
| url | https://www.frontiersin.org/articles/10.3389/fenrg.2021.686616/full |
| work_keys_str_mv | AT mingzhutang costsensitiveextremelyrandomizedtreesalgorithmforonlinefaultdetectionofwindturbinegenerators AT yutaochen costsensitiveextremelyrandomizedtreesalgorithmforonlinefaultdetectionofwindturbinegenerators AT huaweiwu costsensitiveextremelyrandomizedtreesalgorithmforonlinefaultdetectionofwindturbinegenerators AT qizhao costsensitiveextremelyrandomizedtreesalgorithmforonlinefaultdetectionofwindturbinegenerators AT wenlong costsensitiveextremelyrandomizedtreesalgorithmforonlinefaultdetectionofwindturbinegenerators AT victorssheng costsensitiveextremelyrandomizedtreesalgorithmforonlinefaultdetectionofwindturbinegenerators AT jiabiaoyi costsensitiveextremelyrandomizedtreesalgorithmforonlinefaultdetectionofwindturbinegenerators |