A Review of Deep Learning-Based Intelligent Fault Diagnosis Methods for Rotating Machinery under Small-Sample Conditions
Main Article Content
Keywords
rotating machinery, fault analysis, small sample, fault diagnosis, meta-learning, transfer learning, domain generalization, data augmentation, self-supervised learning
Abstract
Rotating machinery occupies a central position in modern industrial equipment, and accurate identification of its operating condition is of great significance for ensuring production safety. However, in real-world operating conditions, the high cost of acquiring fault samples, complex service environments, and other difficulties result in extremely scarce available training samples, posing severe challenges to traditional deep learning methods that rely on big data. Therefore, fault diagnosis technology under small-sample scenarios has become a frontier hotspot in both academia and engineering. This paper reviews the research progress on deep learning-based intelligent fault diagnosis methods for rotating machinery under small-sample conditions. It elaborates on the core ideas and current application status of key approaches, including meta-learning, transfer learning, domain generalization, data augmentation, and self-supervised learning, with the aim of providing useful references for subsequent research in this field.
References
- [1] Li, L. F. (2015). Research on fault identification and intelligent diagnosis of rotating machinery [Master’s thesis, China Jiliang University]. CNKI. https://kns.cnki.net/kcms2/article/abstract?v=zO3wb1M9eky5mQii3rkQ2igyGY_4h3-KZuju03hs8f0Nw89sg-rd8FYZdVpJRzsWVRNKE1EJjNViaGq86VerDe7k3x35Vr5gw5iTo7v4De_AAiHgkeZmirYNSI6Zn9DN6IrSU3ZbsZh-250aeznrHB0ckFA9jenaFLsxgKn05YfrjpZgHZn2Uw==&uniplatform=NZKPT&language=CHS.
- [2] Xia, P. C., Huang, Y. X., Wang, Y. X., Liu, C. L., & Liu, J. (2023). Application of discriminative meta-learning method based on data augmentation in cross-device few-shot fault diagnosis. Scientia Sinica Technologica, 53(11), 2003-2004.
- [3] Zheng, Q. M., Wang, X. M., & Jiang, L. (2025). Research on fault diagnosis method of point machine based on unlabeled self-supervised representation learning. Journal of Railway Science and Engineering, 22(01), 404-415. https://doi.org/10.19713/j.cnki.43-1423/u.T20240574.
- [4] Wu, K., Wu, J., Shu, Q. M., Shen, W. M., & Song, W. B. (2025). Review of deep learning-based small-sample fault diagnosis methods for rotating machinery. Chinese Journal of Ship Research, 20(02), 3-19. https://doi.org/10.19693/j.issn.1673-3185.04175.
- [5] Chen, J. X. (2025). Few-shot bearing fault diagnosis based on time-frequency information self-supervised contrastive learning and fusion enhancement [Master’s thesis, Yanshan University]. https://doi.org/10.27440/d.cnki.gysdu.2025.000938.
- [6] Zhang, R. (2023). Research on fault diagnosis and health assessment of rotating machinery based on meta-learning [Master’s thesis, Shijiazhuang Tiedao University]. https://doi.org/10.27334/d.cnki.gstdy.2023.000623.
- [7] Song, Y. M. (2024). Research on rolling bearing fault diagnosis method based on self-supervised learning [Master’s thesis, Dalian University of Technology]. https://doi.org/10.26991/d.cnki.gdllu.2024.005073.
Article Sidebar
This work is licensed under a Creative Commons Attribution 4.0 International License.