Contrastive Conditional Adversarial Autoencoder With Class-Specific Forces for Imbalanced Open-Set Fault Detection
In real-world industrial scenarios, fault detection faces the widely recognized challenge of data imbalance, which not only refers to the scarcity of fault data but also includes the imbalance in healthy data. This article is concerned with imbalanced open-set fault detection (IOSFD), a practical ye...
Saved in:
| Published in: | IEEE transactions on automation science and engineering Vol. 22; pp. 15757 - 15767 |
|---|---|
| Main Authors: | , , , , , , |
| Format: | Journal Article |
| Language: | English |
| Published: |
IEEE
2025
|
| Subjects: | |
| ISSN: | 1545-5955, 1558-3783 |
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Abstract | In real-world industrial scenarios, fault detection faces the widely recognized challenge of data imbalance, which not only refers to the scarcity of fault data but also includes the imbalance in healthy data. This article is concerned with imbalanced open-set fault detection (IOSFD), a practical yet challenging scenario in industrial applications where multiple healthy operating conditions and multiple fault types are imbalanced. In this article, we propose a new contrastive conditional adversarial autoencoder for IOSFD. It constructs an end-to-end unified model based on multi-class known healthy and faulty data to address the reliance of traditional methods on fault samples, while optimizing with class-specific weighted forces to ensure equal attention to imbalanced known classes. Input and feature reconstruction conditioned on operating modes are utilized to learn a compact decision plane and achieve both unknown fault detection and known data classification. Significantly, we formulate the optimization objective of conditional reconstruction based on contrastive learning and introduce adversarial training to further enhance the model's performance. The effectiveness of the proposed method is validated through real-world pipeline leak detection and Tennessee-Eastman multi-fault detection. Note to Practitioners-Existing data-driven fault detection methods often assume a balanced training dataset, where the quantities of data for various fault types and healthy conditions are roughly equal. However, in real-world industrial settings, fault data is frequently scarce, and the amount of data for different healthy operating conditions can vary significantly. This imbalance can lead to biased models that do not perform well in practice. To address this challenge and leverage the available data effectively, this paper develops a contrastive conditional adversarial autoencoder. The proposed method directly models multiple classes of known fault and healthy data to identify unknown faults, eliminating the need for unknown fault samples in model training. Additionally, the model employs class-specific weighted forces to ensure that all categories, regardless of their frequency in the training set, are given equal attention during learning. This helps to mitigate the effects of data imbalance, ensuring that the model learns from both the abundant healthy data and the sparse known fault data. The proposed approach has been validated through real-world applications, including pipeline leak detection and the Tennessee-Eastman multi-fault detection task. These case studies demonstrate the effectiveness of the method in practical settings. This technique offers a robust solution for industries where fault data is scarce and operating conditions are varied, making it a valuable tool for practitioners working with imbalanced datasets. |
|---|---|
| AbstractList | In real-world industrial scenarios, fault detection faces the widely recognized challenge of data imbalance, which not only refers to the scarcity of fault data but also includes the imbalance in healthy data. This article is concerned with imbalanced open-set fault detection (IOSFD), a practical yet challenging scenario in industrial applications where multiple healthy operating conditions and multiple fault types are imbalanced. In this article, we propose a new contrastive conditional adversarial autoencoder for IOSFD. It constructs an end-to-end unified model based on multi-class known healthy and faulty data to address the reliance of traditional methods on fault samples, while optimizing with class-specific weighted forces to ensure equal attention to imbalanced known classes. Input and feature reconstruction conditioned on operating modes are utilized to learn a compact decision plane and achieve both unknown fault detection and known data classification. Significantly, we formulate the optimization objective of conditional reconstruction based on contrastive learning and introduce adversarial training to further enhance the model's performance. The effectiveness of the proposed method is validated through real-world pipeline leak detection and Tennessee-Eastman multi-fault detection. Note to Practitioners-Existing data-driven fault detection methods often assume a balanced training dataset, where the quantities of data for various fault types and healthy conditions are roughly equal. However, in real-world industrial settings, fault data is frequently scarce, and the amount of data for different healthy operating conditions can vary significantly. This imbalance can lead to biased models that do not perform well in practice. To address this challenge and leverage the available data effectively, this paper develops a contrastive conditional adversarial autoencoder. The proposed method directly models multiple classes of known fault and healthy data to identify unknown faults, eliminating the need for unknown fault samples in model training. Additionally, the model employs class-specific weighted forces to ensure that all categories, regardless of their frequency in the training set, are given equal attention during learning. This helps to mitigate the effects of data imbalance, ensuring that the model learns from both the abundant healthy data and the sparse known fault data. The proposed approach has been validated through real-world applications, including pipeline leak detection and the Tennessee-Eastman multi-fault detection task. These case studies demonstrate the effectiveness of the method in practical settings. This technique offers a robust solution for industries where fault data is scarce and operating conditions are varied, making it a valuable tool for practitioners working with imbalanced datasets. |
| Author | Li, Zheng Pang, Zhibo Meng, Tao Liu, Tong Chen, Zhansheng Zhang, Hao Zuo, Zhonglin |
| Author_xml | – sequence: 1 givenname: Zhonglin orcidid: 0000-0001-6283-1668 surname: Zuo fullname: Zuo, Zhonglin organization: College of Control Science and Engineering, Zhejiang University, Hangzhou, China – sequence: 2 givenname: Tao orcidid: 0000-0001-7871-0457 surname: Meng fullname: Meng, Tao email: mengtao@zju.edu.cn organization: School of Aeronautics and Astronautics, Zhejiang University, Hangzhou, China – sequence: 3 givenname: Zheng orcidid: 0000-0001-6953-6768 surname: Li fullname: Li, Zheng organization: School of Automation, Chongqing University, Chongqing, China – sequence: 4 givenname: Hao orcidid: 0000-0002-4036-7961 surname: Zhang fullname: Zhang, Hao organization: School of Automation, Chongqing University, Chongqing, China – sequence: 5 givenname: Tong surname: Liu fullname: Liu, Tong email: t.liu@sheffield.ac.uk organization: Department of Computer Science, The University of Sheffield, Sheffield, U.K – sequence: 6 givenname: Zhansheng surname: Chen fullname: Chen, Zhansheng organization: Shanghai Aerospace Space Technology Company Ltd., Shanghai, China – sequence: 7 givenname: Zhibo orcidid: 0000-0002-7474-4294 surname: Pang fullname: Pang, Zhibo organization: Department of Automation Technology, ABB Corporate Research Sweden, Västerås, Sweden |
| BookMark | eNpFkNFqwjAUhsNwMHV7gMEu8gJ1J02TppfS6SYIXujYZUnTU5ZRG0misLdfi8Kuznfg_D-cb0YmveuRkGcGC8ageD0s96tFCqlYcJED5PkdmTIhVMJzxScjZyIRhRAPZBbCD0CaqQKmxJeuj16HaC9IB25stK7XHV02F_RBezvyOTrsjWvQ0y8bv2nZ6RCS_QmNba2ha-cNBto6TzfHWne6N9jQ3Qn7ZI-RrvW5i_QNI5qx_JHct7oL-HSbc_K5Xh3Kj2S7e9-Uy21i0hRikhpR8yIXUspCMTVsRY3cSMzAFAKarMk4MlEDb6XG4VU0tcy1UVC3eYuKzwm79hrvQvDYVidvj9r_VgyqUVo1SqtGadVN2pB5uWYsIv7fM4AMJOd_251saQ |
| CODEN | ITASC7 |
| Cites_doi | 10.1016/j.compind.2024.104102 10.1109/TIE.2023.3294645 10.1007/978-3-030-58621-8_45 10.1109/CVPR.2019.00241 10.1109/TII.2022.3149935 10.1109/TII.2023.3275696 10.1016/0098-1354(93)80018-I 10.1109/TII.2020.3046566 10.1109/CVPR.2017.463 10.1109/TNNLS.2020.2980749 10.5555/3524938.3525087 10.1109/TIM.2022.3196436 10.1016/j.patcog.2020.107706 10.1109/CVPR.2019.00949 10.1016/j.ress.2021.108126 10.1109/TASE.2019.2957232 10.1109/TPAMI.2022.3200384 10.1109/TASE.2022.3149591 10.1109/TMECH.2022.3179289 10.5220/0007364500002108 10.1109/TR.2022.3214519 10.1109/TCYB.2025.3531494 10.1145/1390156.1390294 10.1145/3097983.3098052 10.1109/TASE.2021.3138925 10.1109/TPAMI.2014.2321392 10.1007/978-3-030-20893-6_39 10.1109/TASE.2023.3322156 10.1109/TII.2020.2988208 10.1109/TMECH.2021.3058147 10.1109/TIE.2022.3194654 10.1109/tase.2024.3417208 10.1109/TII.2021.3112504 |
| ContentType | Journal Article |
| DBID | 97E RIA RIE AAYXX CITATION |
| DOI | 10.1109/TASE.2025.3570077 |
| DatabaseName | IEEE All-Society Periodicals Package (ASPP) 2005–Present IEEE All-Society Periodicals Package (ASPP) 1998–Present IEEE Electronic Library (IEL) CrossRef |
| DatabaseTitle | CrossRef |
| DatabaseTitleList | |
| Database_xml | – sequence: 1 dbid: RIE name: IEEE Electronic Library (IEL) url: https://ieeexplore.ieee.org/ sourceTypes: Publisher |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Engineering |
| EISSN | 1558-3783 |
| EndPage | 15767 |
| ExternalDocumentID | 10_1109_TASE_2025_3570077 11004063 |
| Genre | orig-research |
| GrantInformation_xml | – fundername: Zhejiang Provincial Outstanding Youth Science Foundation of China grantid: LR24F030001 funderid: 10.13039/501100017623 |
| GroupedDBID | -~X 0R~ 29I 4.4 5GY 5VS 6IK 97E AAJGR AARMG AASAJ AAWTH ABAZT ABQJQ ABVLG ACGFO ACGFS ACIWK AENEX AETIX AGQYO AGSQL AHBIQ AIBXA AKJIK AKQYR ALMA_UNASSIGNED_HOLDINGS ATWAV BEFXN BFFAM BGNUA BKEBE BPEOZ CS3 DU5 EBS EJD F5P HZ~ H~9 IFIPE IPLJI JAVBF LAI M43 O9- OCL PQQKQ RIA RIE RNS AAYXX CITATION |
| ID | FETCH-LOGICAL-c220t-2c5b39756669818c5b9be3c6e40c950d4d43e15b03f6ae558ecb67ac80bf7fe83 |
| IEDL.DBID | RIE |
| ISICitedReferencesCount | 0 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001499531700016&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 1545-5955 |
| IngestDate | Sat Nov 29 07:47:41 EST 2025 Wed Aug 27 01:49:20 EDT 2025 |
| IsPeerReviewed | false |
| IsScholarly | true |
| Language | English |
| License | https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html https://doi.org/10.15223/policy-029 https://doi.org/10.15223/policy-037 |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-c220t-2c5b39756669818c5b9be3c6e40c950d4d43e15b03f6ae558ecb67ac80bf7fe83 |
| ORCID | 0000-0002-4036-7961 0000-0002-7474-4294 0000-0001-6283-1668 0000-0001-6953-6768 0000-0001-7871-0457 |
| PageCount | 11 |
| ParticipantIDs | crossref_primary_10_1109_TASE_2025_3570077 ieee_primary_11004063 |
| PublicationCentury | 2000 |
| PublicationDate | 20250000 2025-00-00 |
| PublicationDateYYYYMMDD | 2025-01-01 |
| PublicationDate_xml | – year: 2025 text: 20250000 |
| PublicationDecade | 2020 |
| PublicationTitle | IEEE transactions on automation science and engineering |
| PublicationTitleAbbrev | TASE |
| PublicationYear | 2025 |
| Publisher | IEEE |
| Publisher_xml | – name: IEEE |
| References | ref13 ref35 ref12 ref34 ref15 ref37 ref14 ref31 ref30 ref11 ref33 ref10 ref2 ref1 ref17 ref16 Rifai (ref25) ref18 Ruff (ref19) ref24 ref23 ref26 ref20 ref22 ref21 van den Oord (ref32) 2018 ref28 ref27 ref29 ref8 ref7 ref9 ref4 ref3 ref6 ref5 Wang (ref36); 119 |
| References_xml | – ident: ref8 doi: 10.1016/j.compind.2024.104102 – ident: ref29 doi: 10.1109/TIE.2023.3294645 – ident: ref33 doi: 10.1007/978-3-030-58621-8_45 – ident: ref11 doi: 10.1109/CVPR.2019.00241 – ident: ref16 doi: 10.1109/TII.2022.3149935 – ident: ref7 doi: 10.1109/TII.2023.3275696 – ident: ref37 doi: 10.1016/0098-1354(93)80018-I – ident: ref13 doi: 10.1109/TII.2020.3046566 – ident: ref30 doi: 10.1109/CVPR.2017.463 – ident: ref26 doi: 10.1109/TNNLS.2020.2980749 – ident: ref31 doi: 10.5555/3524938.3525087 – year: 2018 ident: ref32 article-title: Representation learning with contrastive predictive coding publication-title: arXiv:1807.03748 – ident: ref21 doi: 10.1109/TIM.2022.3196436 – ident: ref24 doi: 10.1016/j.patcog.2020.107706 – volume: 119 start-page: 9929 volume-title: Proc. 37th Int. Conf. Mach. Learn. ident: ref36 article-title: Understanding contrastive representation learning through alignment and uniformity on the hypersphere – ident: ref15 doi: 10.1109/CVPR.2019.00949 – ident: ref34 doi: 10.1016/j.ress.2021.108126 – ident: ref1 doi: 10.1109/TASE.2019.2957232 – ident: ref10 doi: 10.1109/TPAMI.2022.3200384 – ident: ref3 doi: 10.1109/TASE.2022.3149591 – ident: ref35 doi: 10.1109/TMECH.2022.3179289 – ident: ref28 doi: 10.5220/0007364500002108 – ident: ref14 doi: 10.1109/TR.2022.3214519 – ident: ref17 doi: 10.1109/TCYB.2025.3531494 – ident: ref22 doi: 10.1145/1390156.1390294 – ident: ref23 doi: 10.1145/3097983.3098052 – ident: ref4 doi: 10.1109/TASE.2021.3138925 – ident: ref9 doi: 10.1109/TPAMI.2014.2321392 – start-page: 833 volume-title: Proc. 28th Int. Conf. Mach. Learn. (ICML) ident: ref25 article-title: Contractive auto-encoders: Explicit invariance during feature extraction – ident: ref18 doi: 10.1007/978-3-030-20893-6_39 – ident: ref20 doi: 10.1109/TASE.2023.3322156 – ident: ref6 doi: 10.1109/TII.2020.2988208 – start-page: 4393 volume-title: Proc. Int. Conf. Mach. Learn. ident: ref19 article-title: Deep one-class classification – ident: ref27 doi: 10.1109/TMECH.2021.3058147 – ident: ref2 doi: 10.1109/TIE.2022.3194654 – ident: ref12 doi: 10.1109/tase.2024.3417208 – ident: ref5 doi: 10.1109/TII.2021.3112504 |
| SSID | ssj0024890 |
| Score | 2.389352 |
| Snippet | In real-world industrial scenarios, fault detection faces the widely recognized challenge of data imbalance, which not only refers to the scarcity of fault... |
| SourceID | crossref ieee |
| SourceType | Index Database Publisher |
| StartPage | 15757 |
| SubjectTerms | Adaptation models Autoencoders Automation class-specific weighting conditional autoencoder Contrastive learning Data models Fault detection Fault diagnosis Feature extraction imbalanced data Open-set fault detection Representation learning Training |
| Title | Contrastive Conditional Adversarial Autoencoder With Class-Specific Forces for Imbalanced Open-Set Fault Detection |
| URI | https://ieeexplore.ieee.org/document/11004063 |
| Volume | 22 |
| WOSCitedRecordID | wos001499531700016&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| journalDatabaseRights | – providerCode: PRVIEE databaseName: IEEE Electronic Library (IEL) customDbUrl: eissn: 1558-3783 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0024890 issn: 1545-5955 databaseCode: RIE dateStart: 20040101 isFulltext: true titleUrlDefault: https://ieeexplore.ieee.org/ providerName: IEEE |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3NS8MwFA9ueNCDnxPnFzl4ErJlbdM0x6EWBRnCJu5W8vGKBe2k6_z7TdLK5sGDt6S0obwX3vvl5b3fQ-iaa6YgkiHJleYkyhNKJNURiTmVMjAszo0ncX3ik0kyn4vntljd18IAgE8-g4Eb-rt8s9ArFyobenoz61M7qMN53BRrrYn1Eh9QcZCAMMFYe4U5omI4G0_v7VEwYIPQ0blz_ssJbXRV8U4l3f_n7xygvRY94nGj7kO0BeUR2t3gFDxGleObquTSmTFsx6Zoon3Yt15eSrfh8HhVLxyDpYEKvxb1G_a9MYlvRp8XGqeLytoPbAEtfvxQLvtRg8Eu-YRMocapXL3X-A5qn8dV9tBLej-7fSBtYwWig4DWJLAKsjjEIrlYWIdtZ0JBqGOIqBaMmshEIYyYomEeS2AsAa1iLnVCVc5zSMIT1C0XJZwiLJQFREYxYXFIpFUgdRAbDXYRZY1FmPfRzY-ks8-GPyPz5w4qMqeWzKkla9XSRz0n5fWLrYDP_nh-jnbc501E5AJ162oFl2hbf9XFsrry2-MbP866LA |
| linkProvider | IEEE |
| linkToHtml | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LSwMxEA5aBfXgs2J95uBJSJvuJrubY9GWFmsRrNjbkscsFrSV7dbfb5JdsR48eMuGJYSZMPNlMvMNQtex5gqYDEmmdExYllAiqWYkiqmUgeFRZjyJ6zAejZLJRDxWxeq-FgYAfPIZNN3Qv-WbuV66UFnL05tZn7qONjhjAS3LtX6o9RIfUnGggHDBefWI2aaiNe48de1lMODN0BG6x_EvN7TSV8W7ld7ePze0j3Yr_Ig7pcIP0BrMDtHOCqvgEcod41QuF86QYTs20zLeh33z5YV0Rw53lsXccVgayPHLtHjFvjsm8e3os6nGvXluLQi2kBYP3pXLf9RgsEs_IU9Q4J5cvhX4DgqfyTWro-ded3zbJ1VrBaKDgBYksCqySMRiuUhYl22_hIJQR8CoFpwaZlgIba5omEUSOE9AqyiWOqEqizNIwmNUm81ncIKwUBYSGcWFRSJMq0DqIDIa7CLKmoswa6Cbb0mnHyWDRupvHlSkTi2pU0taqaWB6k7KPz9WAj79Y_4KbfXHD8N0OBjdn6Ftt1QZHzlHtSJfwgXa1J_FdJFf-qPyBRtZvXM |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Contrastive+Conditional+Adversarial+Autoencoder+With+Class-Specific+Forces+for+Imbalanced+Open-Set+Fault+Detection&rft.jtitle=IEEE+transactions+on+automation+science+and+engineering&rft.au=Zuo%2C+Zhonglin&rft.au=Meng%2C+Tao&rft.au=Li%2C+Zheng&rft.au=Zhang%2C+Hao&rft.date=2025&rft.issn=1545-5955&rft.eissn=1558-3783&rft.volume=22&rft.spage=15757&rft.epage=15767&rft_id=info:doi/10.1109%2FTASE.2025.3570077&rft.externalDBID=n%2Fa&rft.externalDocID=10_1109_TASE_2025_3570077 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1545-5955&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1545-5955&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1545-5955&client=summon |