View in EDS

An optimized machine learning framework for diagnosing demagnetization faults in PMSG using FEM, feature engineering, and experimental validation.

Saved in:
Bibliographic Details
Title: An optimized machine learning framework for diagnosing demagnetization faults in PMSG using FEM, feature engineering, and experimental validation.
Authors: Shahbaz, Nadeem1 (AUTHOR), Chen, Yu1 (AUTHOR) chenyu@xjtu.edu.cn, Liang, Feng1 (AUTHOR), Siyu, Du1 (AUTHOR), Zhang, Sichao1 (AUTHOR), Zhao, Shouwang1 (AUTHOR), Wang, Shuang1 (AUTHOR), Min, Zhang2 (AUTHOR), Ma, Yong3 (AUTHOR), Li, Chong3 (AUTHOR), Deng, Wei3 (AUTHOR), Zhao, Yong3 (AUTHOR)
Source: Electrical Engineering. Mar2026, Vol. 108 Issue 3, p1-15. 15p.
Subject Terms: *Fault diagnosis, *Demagnetization, *Wind power plants, *Finite element method, *Permanent magnet generators, *Machine learning, *Feature extraction, *Model validation
Abstract: The reliability and efficiency of Permanent Magnet Synchronous Generators (PMSG) are crucial for the long-term performance and fault-free operation of wind power systems, where condition monitoring and early fault detection are essential. This study aims to improve the accuracy of the fault detection model and reduce processing time. To achieve these goals, a robust fault diagnosis framework is developed that combines the Finite Element Method (FEM), feature engineering, and experimental validation to detect various levels of demagnetization in PMSG. Four fault conditions are examined: healthy, 50%, 75%, and 100% unipolar demagnetization using stray flux and current signals. Features are extracted using the Discrete Wavelet Transform (DWT), ranked with the Kruskal–Wallis filter, and optimized through Feature Dimension Coordination (FDC). Three fundamental machine learning (ML) algorithms (KNN, SVM, and Ensemble), including seventeen sub-classifiers, are evaluated for fault classification. Results show that the Cosine KNN and six features (Range, L2 Norm, L1 Norm, Standard Deviation, Median, and Mean Absolute Deviation) are the most effective for classifying faults using flux signals. The feature engineering approach improves processing time (from 1.9 to 0.6 s) and accuracy (from 95 to 100%). The method was validated with experimental data, confirming its robustness for wind energy systems. [ABSTRACT FROM AUTHOR]
Database: Energy & Power Source
FullText Text:
  Availability: 0
Header DbId: enr
DbLabel: Energy & Power Source
An: 191606167
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: An optimized machine learning framework for diagnosing demagnetization faults in PMSG using FEM, feature engineering, and experimental validation.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Shahbaz%2C+Nadeem%22">Shahbaz, Nadeem</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Chen%2C+Yu%22">Chen, Yu</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> chenyu@xjtu.edu.cn</i><br /><searchLink fieldCode="AR" term="%22Liang%2C+Feng%22">Liang, Feng</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Siyu%2C+Du%22">Siyu, Du</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Zhang%2C+Sichao%22">Zhang, Sichao</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Zhao%2C+Shouwang%22">Zhao, Shouwang</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Wang%2C+Shuang%22">Wang, Shuang</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Min%2C+Zhang%22">Min, Zhang</searchLink><relatesTo>2</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Ma%2C+Yong%22">Ma, Yong</searchLink><relatesTo>3</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Li%2C+Chong%22">Li, Chong</searchLink><relatesTo>3</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Deng%2C+Wei%22">Deng, Wei</searchLink><relatesTo>3</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Zhao%2C+Yong%22">Zhao, Yong</searchLink><relatesTo>3</relatesTo> (AUTHOR)
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22Electrical+Engineering%22">Electrical Engineering</searchLink>. Mar2026, Vol. 108 Issue 3, p1-15. 15p.
– Name: Subject
  Label: Subject Terms
  Group: Su
  Data: *<searchLink fieldCode="DE" term="%22Fault+diagnosis%22">Fault diagnosis</searchLink><br />*<searchLink fieldCode="DE" term="%22Demagnetization%22">Demagnetization</searchLink><br />*<searchLink fieldCode="DE" term="%22Wind+power+plants%22">Wind power plants</searchLink><br />*<searchLink fieldCode="DE" term="%22Finite+element+method%22">Finite element method</searchLink><br />*<searchLink fieldCode="DE" term="%22Permanent+magnet+generators%22">Permanent magnet generators</searchLink><br />*<searchLink fieldCode="DE" term="%22Machine+learning%22">Machine learning</searchLink><br />*<searchLink fieldCode="DE" term="%22Feature+extraction%22">Feature extraction</searchLink><br />*<searchLink fieldCode="DE" term="%22Model+validation%22">Model validation</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: The reliability and efficiency of Permanent Magnet Synchronous Generators (PMSG) are crucial for the long-term performance and fault-free operation of wind power systems, where condition monitoring and early fault detection are essential. This study aims to improve the accuracy of the fault detection model and reduce processing time. To achieve these goals, a robust fault diagnosis framework is developed that combines the Finite Element Method (FEM), feature engineering, and experimental validation to detect various levels of demagnetization in PMSG. Four fault conditions are examined: healthy, 50%, 75%, and 100% unipolar demagnetization using stray flux and current signals. Features are extracted using the Discrete Wavelet Transform (DWT), ranked with the Kruskal–Wallis filter, and optimized through Feature Dimension Coordination (FDC). Three fundamental machine learning (ML) algorithms (KNN, SVM, and Ensemble), including seventeen sub-classifiers, are evaluated for fault classification. Results show that the Cosine KNN and six features (Range, L2 Norm, L1 Norm, Standard Deviation, Median, and Mean Absolute Deviation) are the most effective for classifying faults using flux signals. The feature engineering approach improves processing time (from 1.9 to 0.6 s) and accuracy (from 95 to 100%). The method was validated with experimental data, confirming its robustness for wind energy systems. [ABSTRACT FROM AUTHOR]
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=enr&AN=191606167
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1007/s00202-025-03485-x
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 15
        StartPage: 1
    Subjects:
      – SubjectFull: Fault diagnosis
        Type: general
      – SubjectFull: Demagnetization
        Type: general
      – SubjectFull: Wind power plants
        Type: general
      – SubjectFull: Finite element method
        Type: general
      – SubjectFull: Permanent magnet generators
        Type: general
      – SubjectFull: Machine learning
        Type: general
      – SubjectFull: Feature extraction
        Type: general
      – SubjectFull: Model validation
        Type: general
    Titles:
      – TitleFull: An optimized machine learning framework for diagnosing demagnetization faults in PMSG using FEM, feature engineering, and experimental validation.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Shahbaz, Nadeem
      – PersonEntity:
          Name:
            NameFull: Chen, Yu
      – PersonEntity:
          Name:
            NameFull: Liang, Feng
      – PersonEntity:
          Name:
            NameFull: Siyu, Du
      – PersonEntity:
          Name:
            NameFull: Zhang, Sichao
      – PersonEntity:
          Name:
            NameFull: Zhao, Shouwang
      – PersonEntity:
          Name:
            NameFull: Wang, Shuang
      – PersonEntity:
          Name:
            NameFull: Min, Zhang
      – PersonEntity:
          Name:
            NameFull: Ma, Yong
      – PersonEntity:
          Name:
            NameFull: Li, Chong
      – PersonEntity:
          Name:
            NameFull: Deng, Wei
      – PersonEntity:
          Name:
            NameFull: Zhao, Yong
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 03
              Text: Mar2026
              Type: published
              Y: 2026
          Identifiers:
            – Type: issn-print
              Value: 09487921
          Numbering:
            – Type: volume
              Value: 108
            – Type: issue
              Value: 3
          Titles:
            – TitleFull: Electrical Engineering
              Type: main
ResultId 1