A Novel Method for Adapting Synchronous Motor Diagnostic Models in the Absence of Target-domain Fault Data
DOI:
https://doi.org/10.14529/cmse260306Keywords:
FD-GAP, gap loss, domain adaptation, one-class learning, inter-turn short circuits, synchronous motors, stator current-based diagnostics, Fr´echet distance, neural network ensemble, unstable model filteringAbstract
Among the approaches to fault diagnosis of rotating machinery under domain shift, the equipment commissioning scenario remains insufficiently studied, where only normal-state data can be collected and examples of target faults are completely unavailable. Conventional unsupervised domain adaptation methods are not directly applicable in this setting because target fault samples required for clustering are absent, while one-class anomaly detection methods ignore valuable knowledge about faults available in the source domain. Approaches based on digital twins or generative adversarial networks require computationally expensive physical modeling and may suffer from discrepancies between synthetic and real data. We propose a novel FD-GAP domain adaptation approach for the setting in which the target domain contains only normal operating data. Unlike previous approaches limited to distribution alignment, the proposed method transfers the geometric boundary of the fault class from the source domain to the target domain using the Fréchet distance and gap loss to push fault states away from the normal-state region. To compensate for stochastic model instability caused by different random initializations, a novel disagreement-based ensemble filtering mechanism is introduced. The mechanism operates exclusively on normal-state data from the target equipment and improves deployment reliability while maintaining a prescribed false-positive rate.
References
Li W., Huang R., Li J., Liao Y., et al. A perspective survey on deep transfer learning for fault diagnosis in industrial scenarios: Theories, applications and challenges. Mechanical Systems and Signal Processing. 2022. Vol. 167. DOI: 10.1016/j.ymssp.2021.108487.
Sullivan G., Pugh R., Melendez A.P., Hunt W.D. Operations & Maintenance Best Practices - A Guide to Achieving Operational Efficiency (Release 3): tech. rep. / Pacific Northwest National Laboratory (PNNL), Richland, WA (United States). Richland, WA, United States, Aug. 2010. DOI: 10.2172/1034595.
Khan S., Yairi T. A review on the application of deep learning in system health management. Mechanical Systems and Signal Processing. 2018. Vol. 107. P. 241–265. DOI: 10.1016/j.ymssp.2017.11.024.
Li C., Zhang S., Qin Y., Estupinan E. A systematic review of deep transfer learning for machinery fault diagnosis. Neurocomputing. 2020. Vol. 407. P. 121–135. DOI: 10.1016/j.neucom.2020.04.045.
Fink O., Wang Q., Svensén M., Dersin P., et al. Potential, challenges and future directions for deep learning in prognostics and health management applications. Engineering Applications of Artificial Intelligence. 2020. Vol. 92. DOI: 10.1016/j.engappai.2020.103678.
Gao T., Yang J., Tang Q. A multi-source domain information fusion network for rotating machinery fault diagnosis under variable operating conditions. Information Fusion. 2024. Vol. 106. DOI: 10.1016/j.inffus.2024.102278.
Jia N., Huang W., Ding C., Wang J., et al. Physics-informed unsupervised domain adaptation framework for cross-machine bearing fault diagnosis. Advanced Engineering Informatics. 2024. Vol. 62. DOI: 10.1016/j.aei.2024.102774.
Pan X., Chen H., Wang W., Su X. Adversarial domain adaptation based on contrastive learning for bearings fault diagnosis. Simulation Modelling Practice and Theory. 2025. Vol. 139. DOI: 10.1016/j.simpat.2024.103058.
Pang B., Liu Q., Sun Z., Xu Z., et al. Time-frequency supervised contrastive learning via pseudo-labeling: An unsupervised domain adaptation network for rolling bearing fault diagnosis under time-varying speeds. Advanced Engineering Informatics. 2024. Vol. 59. DOI: 10.1016/j.aei.2023.102304.
Lin F., Guo C., Zhao Z., Zhang X., et al. A Universal Domain Adaptation Method With Cluster Matching for Machinery Fault Diagnosis. IEEE Internet of Things Journal. 2025. Vol. 12, no. 6. P. 7487–7502. DOI: 10.1109/JIOT.2024.3496928.
Liu Y., Deng A., Chen G., Shi Y., et al. Universal domain adaptation in rotating machinery fault diagnosis: A self-supervised orthogonal clustering approach. Reliability Engineering & System Safety. 2025. Vol. 257. DOI: 10.1016/j.ress.2025.110828.
Tian J., Han D., Karimi H.R., Zhang Y., et al. A universal multi-source domain adaptation method with unsupervised clustering for mechanical fault diagnosis under incomplete data. Neural Networks. 2024. Vol. 173. DOI: 10.1016/j.neunet.2024.106167.
Jiao J., Li H., Zhang T., Lin J. Source-Free Adaptation Diagnosis for Rotating Machinery. IEEE Transactions on Industrial Informatics. 2023. Vol. 19, no. 9. P. 9586–9595. DOI: 10.1109/TII.2022.3231414.
Li Q., Tang B., Deng L., Yang Q., et al. Adaptive centroid prototype-based domain adaptation for fault diagnosis of rotating machinery without source data. Reliability Engineering & System Safety. 2024. Vol. 251. DOI: 10.1016/j.ress.2024.110393.
Liu Z., Zhao K., Shao H., Chen J., et al. Reinforced fuzzy domain adaptation: Revolutionizing data-unaccessible rotating machinery fault diagnosis across multiple domains. Expert Systems with Applications. 2024. Vol. 252. DOI: 10.1016/j.eswa.2024.124094.
Yu A., Cai B., Wu Q., Martínez García M., et al. Source-free domain adaptation method for fault diagnosis of rotation machinery under partial information. Reliability Engineering & System Safety. 2024. Vol. 248. DOI: 10.1016/j.ress.2024.110181.
Lin Y., Wang Y., Zhang M., Zhao M. A robust source-free unsupervised domain adaptation method based on uncertainty measure and adaptive calibration for rotating machinery fault diagnosis. Reliability Engineering & System Safety. 2025. Vol. 253. DOI: 10.1016/j.ress.2024.110516.
Jeong H., Kim S., Seo D., Kwon J. Source-Free Domain Adaptation Framework for Rotary Machine Fault Diagnosis. Sensors. 2025. Vol. 25, no. 14. DOI: 10.3390/s25144383.
Zhang Y., Ren Z., Feng K., Yu K., et al. Universal source-free domain adaptation method for cross-domain fault diagnosis of machines. Mechanical Systems and Signal Processing. 2023. Vol. 191. DOI: 10.1016/j.ymssp.2023.110159.
Chen L., Li Q., Shen C., Zhu J., et al. Adversarial Domain-Invariant Generalization: A Generic Domain-Regressive Framework for Bearing Fault Diagnosis Under Unseen Conditions. IEEE Transactions on Industrial Informatics. 2022. Vol. 18, no. 3. P. 1790–1800. DOI: 10.1109/TII.2021.3078712.
Han T., Li Y.-F., Qian M. A Hybrid Generalization Network for Intelligent Fault Diagnosis of Rotating Machinery Under Unseen Working Conditions. IEEE Transactions on Instrumentation and Measurement. 2021. Vol. 70. P. 1–11. DOI: 10.1109/TIM.2021.3088489.
Li X., Zhang W., Ma H., Luo Z., et al. Domain generalization in rotating machinery fault diagnostics using deep neural networks. Neurocomputing. 2020. Vol. 403. P. 409–420. DOI: 10.1016/j.neucom.2020.05.014.
Fan Z., Xu Q., Jiang C., Ding S.X. Deep Mixed Domain Generalization Network for Intelligent Fault Diagnosis Under Unseen Conditions. IEEE Transactions on Industrial Electronics. 2024. Vol. 71, no. 1. P. 965–974. DOI: 10.1109/TIE.2023.3243293.
Shi Y., Deng A., Deng M., Xu M., et al. Domain augmentation generalization network for real-time fault diagnosis under unseen working conditions. Reliability Engineering & System Safety. 2023. Vol. 235. DOI: 10.1016/j.ress.2023.109188.
Gao T., Yang J., Wang W., Fan X. A domain feature decoupling network for rotating machinery fault diagnosis under unseen operating conditions. Reliability Engineering & System Safety. 2024. Vol. 252. DOI: 10.1016/j.ress.2024.110449.
Wang R., Huang W., Lu Y., Zhang X., et al. A novel domain generalization network with multidomain specific auxiliary classifiers for machinery fault diagnosis under unseen working conditions. Reliability Engineering & System Safety. 2023. Vol. 238. DOI: 10.1016/j.ress.2023.109463.
Zhao C., Shen W. A domain generalization network combing invariance and specificity towards real-time intelligent fault diagnosis. Mechanical Systems and Signal Processing. 2022. Vol. 173. DOI: 10.1016/j.ymssp.2022.108990.
Qian Q., Zhou J., Qin Y. Relationship Transfer Domain Generalization Network for Rotating Machinery Fault Diagnosis Under Different Working Conditions. IEEE Transactions on Industrial Informatics. 2023. Vol. 19, no. 9. P. 9898–9908. DOI: 10.1109/TII.2022.3232842.
Liu X., Sun W., Li H., Li Q., et al. Unknown working condition fault diagnosis of rotate machine without training sample based on local fault semantic attribute. Advanced Engineering Informatics. 2024. Vol. 61. DOI: 10.1016/j.aei.2024.102515.
Liu C., Gryllias K. Simulation-Driven Domain Adaptation for Rolling Element Bearing Fault Diagnosis. IEEE Transactions on Industrial Informatics. 2022. Vol. 18, no. 9. P. 5760–5770. DOI: 10.1109/TII.2021.3103412.
Lou Y., Kumar A., Xiang J. Machinery Fault Diagnosis Based on Domain Adaptation to Bridge the Gap Between Simulation and Measured Signals. IEEE Transactions on Instrumentation and Measurement. 2022. Vol. 71. P. 1–9. DOI: 10.1109/TIM.2022.3180416.
Xiao Y., Shao H., Yan S., Wang J., et al. Domain generalization for rotating machinery fault diagnosis: A survey. Advanced Engineering Informatics. 2025. Vol. 64. DOI: 10.1016/j.aei.2024.103063.
Zhang Y., Ji J., Ren Z., Ni Q., et al. Digital twin-driven partial domain adaptation network for intelligent fault diagnosis of rolling bearing. Reliability Engineering & System Safety. 2023. Vol. 234. DOI: 10.1016/j.ress.2023.109186.
Azari M.S., Santini S., Edrisi F., Flammini F. Self-adaptive fault diagnosis for unseen working conditions based on digital twins and domain generalization. Reliability Engineering & System Safety. 2025. Vol. 254. DOI: 10.1016/j.ress.2024.110560.
Li Y., Wang N., Shi J., Hou X., et al. Adaptive Batch Normalization for practical domain adaptation. Pattern Recognition. 2018. Vol. 80. P. 109–117. DOI: 10.1016/j.patcog.2018.03.005.
Dowson D.C., Landau B.V. The Fréchet distance between multivariate normal distributions. Journal of Multivariate Analysis. 1982. Vol. 12, no. 3. P. 450–455. DOI: 10.1016/0047-259X(82)90077-X.
Ledoit O., Wolf M. A well-conditioned estimator for large-dimensional covariance matrices. Journal of Multivariate Analysis. 2004. Vol. 88, no. 2. P. 365–411. DOI: 10.1016/S0047-259X(03)00096-4.
Higham N.J. Functions of Matrices: Theory and Computation. Philadelphia: Society for Industrial, Applied Mathematics, 2008. Other Titles in Applied Mathematics. DOI: 10.1137/1.9780898717778.
Ionescu C., Vantzos O., Sminchisescu C. Matrix Backpropagation for Deep Networks with Structured Layers. 2015 IEEE International Conference on Computer Vision (ICCV), Santiago, Chile, 13–16 December 2015. IEEE, 2015. P. 2965–2973. DOI: 10.1109/ICCV.2015.339.
Chu N., Wang L., Yu L., He C., et al. An Adaptive Autogram Approach Based on a CFAR Detector for Incipient Cavitation Detection. Sensors. 2020. Vol. 20, no. 8. DOI: 10.3390/s20082303.
Cheng J., Yang D.,Wang S. Unsupervised Learning Applications for Predictive Maintenance in Consumer Electronics and Intelligent Energy Networks. IEEE Transactions on Consumer Electronics. 2025. Vol. 71, no. 3. P. 8347–8354. DOI: 10.1109/TCE.2025.3573649.
Lakshminarayanan B., Pritzel A., Blundell C. Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles. Advances in Neural Information Processing Systems, Long Beach, California, USA, 4–9 December 2017. Vol. 30. 2017. P. 6402–6413. URL: https://papers.nips.cc/paper_files/paper/2017/hash/9ef2ed4b7fd2c810847ffa5fa85bce38-Abstract.html.
Tifrea A., Stavarache E., Yang F. Semi-supervised novelty detection using ensembles with regularized disagreement. Proceedings of the Thirty-Eighth Conference on Uncertainty in Artificial Intelligence, Eindhoven, The Netherlands, 1–5 August 2022. Vol. 180 / ed. by J. Cussens, K. Zhang. PMLR, 2022. P. 1939–1948. Proceedings of Machine Learning Research. URL: https://proceedings.mlr.press/v180/tifrea22a.html.
Jung W., Kim S.-H., Yun S.-H., Bae J., et al. Vibration, acoustic, temperature, and motor current dataset of rotating machine under varying operating conditions for fault diagnosis. Data in Brief. 2023. Vol. 48. DOI: 10.1016/j.dib.2023.109049.
Barandas M., Folgado D., Fernandes L., Santos S., et al. TSFEL: Time Series Feature Extraction Library. SoftwareX. 2020. Vol. 11. DOI: 10.1016/j.softx.2020.100456.
Shestakov A.L., Galyshev D.V., Eremeeva V.A., Ibryaeva O.L. Domain-Adaptive and Multifactor Invariant Diagnosis of Inter-Turn Short Circuits in Permanent Magnet Synchronous Motors. 2026 28th International Conference on Digital Signal Processing and its Applications (DSPA), Moscow, Russian Federation, 25–27 March 2026. IEEE, 2026. P. 1–6. DOI: 10.1109/DSPA69176.2026.11476735.
