Power Transformer Health Index Using Cost-Sensitive Learning to Consider the Impact of Misclassification
Power companies employ comprehensive evaluations to manage facilities, such as transformers, lines, and generators, assessing their status and calculating the probability of failure. Within this asset management framework, a key metric that indicates the condition of facilities is the health index....
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10798105/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841557024462077952 |
|---|---|
| author | Junsoo Che Gihun Park Jeongsik Oh Su-Han Pyo Byeonghyeon An Taesik Park |
| author_facet | Junsoo Che Gihun Park Jeongsik Oh Su-Han Pyo Byeonghyeon An Taesik Park |
| author_sort | Junsoo Che |
| collection | DOAJ |
| description | Power companies employ comprehensive evaluations to manage facilities, such as transformers, lines, and generators, assessing their status and calculating the probability of failure. Within this asset management framework, a key metric that indicates the condition of facilities is the health index. To develop an optimal investment strategy for facility maintenance and management, a method for accurately calculating the health index is essential. Recent studies have explored the application of machine learning to improve accuracy. However, these studies ignore the risk of misclassification, which can result in significant financial losses when predicting the health index of power transformers. Therefore, this study proposes a health index calculation algorithm that incorporates cost-sensitive learning to assess the impact of each misclassification scenario, construct a cost matrix, and minimize the cost through learning. Input parameters are selected on the basis of the power transformer’s structure, failure mode effect analysis, failure diagnosis techniques, and historical failure cases. During the learning process, the relationship between actual and predicted classes is analyzed, and the risk associated with each misclassification scenario is quantified to describe the cost matrix. The effectiveness of the proposed health index calculation algorithm is validated using in-service power transformer inspection data, demonstrating a reduction in the risk of specific misclassification scenarios. |
| format | Article |
| id | doaj-art-8246712e25cd4d31a5945fac7629a759 |
| institution | Kabale University |
| issn | 2169-3536 |
| language | English |
| publishDate | 2024-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-8246712e25cd4d31a5945fac7629a7592025-01-07T00:01:55ZengIEEEIEEE Access2169-35362024-01-011219179019180710.1109/ACCESS.2024.351678510798105Power Transformer Health Index Using Cost-Sensitive Learning to Consider the Impact of MisclassificationJunsoo Che0https://orcid.org/0000-0002-2491-1832Gihun Park1https://orcid.org/0009-0009-1860-9811Jeongsik Oh2https://orcid.org/0009-0001-4506-2673Su-Han Pyo3https://orcid.org/0000-0003-1119-4755Byeonghyeon An4https://orcid.org/0000-0003-0006-3880Taesik Park5https://orcid.org/0000-0003-2372-2332Department of Electrical Engineering, Mokpo National University, Muan, Republic of KoreaAsset Management Project, KEPCO Research Institute, Daejeon, Republic of KoreaDepartment of Electrical Engineering, Mokpo National University, Muan, Republic of KoreaPower Conversion Research Center, Korea Electrotechnology Research Institute, Gwangju, Republic of KoreaDepartment of Electrical Engineering, Mokpo National University, Muan, Republic of KoreaDepartment of Electrical Engineering, Mokpo National University, Muan, Republic of KoreaPower companies employ comprehensive evaluations to manage facilities, such as transformers, lines, and generators, assessing their status and calculating the probability of failure. Within this asset management framework, a key metric that indicates the condition of facilities is the health index. To develop an optimal investment strategy for facility maintenance and management, a method for accurately calculating the health index is essential. Recent studies have explored the application of machine learning to improve accuracy. However, these studies ignore the risk of misclassification, which can result in significant financial losses when predicting the health index of power transformers. Therefore, this study proposes a health index calculation algorithm that incorporates cost-sensitive learning to assess the impact of each misclassification scenario, construct a cost matrix, and minimize the cost through learning. Input parameters are selected on the basis of the power transformer’s structure, failure mode effect analysis, failure diagnosis techniques, and historical failure cases. During the learning process, the relationship between actual and predicted classes is analyzed, and the risk associated with each misclassification scenario is quantified to describe the cost matrix. The effectiveness of the proposed health index calculation algorithm is validated using in-service power transformer inspection data, demonstrating a reduction in the risk of specific misclassification scenarios.https://ieeexplore.ieee.org/document/10798105/Health indexasset managementpower transformercost-sensitive learning |
| spellingShingle | Junsoo Che Gihun Park Jeongsik Oh Su-Han Pyo Byeonghyeon An Taesik Park Power Transformer Health Index Using Cost-Sensitive Learning to Consider the Impact of Misclassification IEEE Access Health index asset management power transformer cost-sensitive learning |
| title | Power Transformer Health Index Using Cost-Sensitive Learning to Consider the Impact of Misclassification |
| title_full | Power Transformer Health Index Using Cost-Sensitive Learning to Consider the Impact of Misclassification |
| title_fullStr | Power Transformer Health Index Using Cost-Sensitive Learning to Consider the Impact of Misclassification |
| title_full_unstemmed | Power Transformer Health Index Using Cost-Sensitive Learning to Consider the Impact of Misclassification |
| title_short | Power Transformer Health Index Using Cost-Sensitive Learning to Consider the Impact of Misclassification |
| title_sort | power transformer health index using cost sensitive learning to consider the impact of misclassification |
| topic | Health index asset management power transformer cost-sensitive learning |
| url | https://ieeexplore.ieee.org/document/10798105/ |
| work_keys_str_mv | AT junsooche powertransformerhealthindexusingcostsensitivelearningtoconsidertheimpactofmisclassification AT gihunpark powertransformerhealthindexusingcostsensitivelearningtoconsidertheimpactofmisclassification AT jeongsikoh powertransformerhealthindexusingcostsensitivelearningtoconsidertheimpactofmisclassification AT suhanpyo powertransformerhealthindexusingcostsensitivelearningtoconsidertheimpactofmisclassification AT byeonghyeonan powertransformerhealthindexusingcostsensitivelearningtoconsidertheimpactofmisclassification AT taesikpark powertransformerhealthindexusingcostsensitivelearningtoconsidertheimpactofmisclassification |