Evaluation of Input Presentation in Transfer Learning for Bearing Fault Detection
##plugins.themes.bootstrap3.article.main##
##plugins.themes.bootstrap3.article.sidebar##
Abstract
Transfer learning is a promising technique to overcome data insufficiency, a noticeable barrier to real-world application of intelligent maintenance solutions. Although the importance of data preparation and preprocessing is widely acknowledged, no study in particular has investigated the effect of vibration data input presentations on transferability capabilities. This study aims to fill in this gap by conducting experiments across three benchmark datasets to evaluate direct transfer, catastrophic forgetting and data efficiency for bearing fault classification. Moreover, we explore the opportunity to employ source model pseudo-labeling to reduce the need for data labeled by human experts. Our findings show that not only does the choice of preprocessing pipeline significantly affect target-set performance, but also that the vulnerability to catastrophic forgetting varies accordingly. Thus, we conclude that finding the right data processing routine is also a key component in achieving supreme transfer learning performance and, indeed, it deserves more attention. The code base of this study is open-sourced and made publicly available to support reproducibility, transparency, and further research.
How to Cite
##plugins.themes.bootstrap3.article.details##
Transfer Learning, Input Representation, Vibration Analysis, Intelligent Fault Detection, Bearing Fault Detection
Berenji, A., Nowaczyk, S., & Taghiyarrenani, Z. (2023). Data-centric perspective on explainability versus performance trade-off. In International Symposium on Intelligent Data Analysis (pp. 42–54).
Berenji, A., & Taghiyarrenani, Z. (2022). An analysis of vibrations and currents for broken rotor bar detection in three-phase induction motors. In PHM Society European Conference (Vol. 7, pp. 43–48).
Carvalho, T. P., Soares, F. A., Vita, R., Francisco, R. d. P., Basto, J. P., & Alcalá, S. G. (2019). A systematic literature review of machine learning methods applied to predictive maintenance. Computers & Industrial Engineering, 137, 106024.
Li, M., He, S., Chen, J., Feng, Y., & Xie, J. (2025). Label-smoothing dynamic decoupling augmented network for intelligent fault diagnosis under imbalanced data distribution with noisy labels. Measurement, 118664.
Liu, R., Yang, B., Zio, E., & Chen, X. (2018). Artificial intelligence for fault diagnosis of rotating machinery: A review. Mechanical Systems and Signal Processing, 108, 33–47.
Lu, T., Yu, F., Han, B., & Wang, J. (2020). A generic intelligent bearing fault diagnosis system using convolutional neural networks with transfer learning. IEEE Access, 8, 164807–164814.
Lu, Y., Xu, Z., & He, W. (2023). Rethinking label refurbishment: Model robustness under label noise. In Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 37, pp. 15000–15008).
Misbah, I., Lee, C. K., & Keung, K. L. (2024). Fault diagnosis in rotating machines based on transfer learning: Literature review. Knowledge-Based Systems, 283, 111158.
Reed, S., Lee, H., Anguelov, D., Szegedy, C., Erhan, D., & Rabinovich, A. (2014). Training deep neural networks on noisy labels with bootstrapping. arXiv preprint arXiv:1412.6596.
Shao, S., McAleer, S., Yan, R., & Baldi, P. (2018). Highly accurate machine fault diagnosis using deep transfer learning. IEEE Transactions on Industrial Informatics, 15(4), 2446–2455.
Sukhbaatar, S., & Fergus, R. (2014). Learning from noisy labels with deep neural networks. arXiv preprint arXiv:1406.2080.
Thomas, D., & Weiss, B. (2021). Maintenance costs and advanced maintenance techniques in manufacturing machinery: Survey and analysis. International Journal of Prognostics and Health Management, 12(1). doi: 10.36001/ijphm.2021.v12i1.2934
Wang, S., Tian, J., Zhang, J., Yang, H., Yuan, X., & Liang, P. (2026). A dual model joint learning framework for intelligent fault diagnosis of rotating machinery under noisy labels. Advanced Engineering Informatics, 69, 104024.
Zhao, B., Zhang, X., Zhan, Z., & Pang, S. (2020). Deep multi-scale convolutional transfer learning network: A novel method for intelligent fault diagnosis of rolling bearings under variable working conditions and domains. Neurocomputing, 407, 24–38.
Zhao, Y., & Xu, J. (2024). Bearing fault diagnosis based on data missing and feature shift suppression strategy. Proceedings of the Institution of Mechanical Engineers, Part I: Journal of Systems and Control Engineering, 238(7), 1193–1205.
Zhao, Z., Alzubaidi, L., Zhang, J., Duan, Y., & Gu, Y. (2024). A comparison review of transfer learning and self-supervised learning: Definitions, applications, advantages and limitations. Expert Systems with Applications, 242, 122807.
Zhong, J., Yang, Y., Mao, H., Qin, A., Li, X., & Tang, W. (2024). Contrastive regularization-guided label refurbishment for fault diagnosis under label noise. Advanced Engineering Informatics, 61, 102478.
This work is licensed under a Creative Commons Attribution 3.0 Unported License.
The Prognostic and Health Management Society advocates open-access to scientific data and uses a Creative Commons license for publishing and distributing any papers. A Creative Commons license does not relinquish the author’s copyright; rather it allows them to share some of their rights with any member of the public under certain conditions whilst enjoying full legal protection. By submitting an article to the International Conference of the Prognostics and Health Management Society, the authors agree to be bound by the associated terms and conditions including the following:
As the author, you retain the copyright to your Work. By submitting your Work, you are granting anybody the right to copy, distribute and transmit your Work and to adapt your Work with proper attribution under the terms of the Creative Commons Attribution 3.0 United States license. You assign rights to the Prognostics and Health Management Society to publish and disseminate your Work through electronic and print media if it is accepted for publication. A license note citing the Creative Commons Attribution 3.0 United States License as shown below needs to be placed in the footnote on the first page of the article.
First Author et al. This is an open-access article distributed under the terms of the Creative Commons Attribution 3.0 United States License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.