Multi-Class Gearbox Fault Diagnosis via Pre-Trained Model-based Domain Adaptation with Healthy-Only Target Data
##plugins.themes.bootstrap3.article.main##
##plugins.themes.bootstrap3.article.sidebar##
Abstract
Accurate gearbox fault diagnosis under varying operational speeds is critical for industrial predictive maintenance. A significant challenge is domain shift, where models trained under one condition fail to generalize to another, especially when only healthy data from the target domain is available for training. This study proposes a novel domain adaptation framework, CDANet, that directly leverages raw sensor data to perform multi-class fault classification without manual feature engineering. The model combines a lightweight CNNbased temporal feature extractor with a frozen DistilBERT encoder to capture transferable, domain-invariant representations, combined with a maximum mean discrepancy loss to align the feature distributions between the source and target domains using only healthy samples. Experimental results demonstrate that our proposed model significantly outperforms conventional deep learning approaches, achieving high classification accuracy across six domain adaptation tasks. This work validates the effectiveness of applying pre-trained models in domain adaptation for gearbox fault diagnosis under real-world domain shift constraints.
##plugins.themes.bootstrap3.article.details##
Fault Diagnosis, Domain Adaptation
Huang, C., Bu, S., Lee, H. H., Chan, C. H., Kong, S. W., & Yung, W. K. (2024). Prognostics and health management for predictive maintenance: A review. Journal of Manufacturing Systems, 75, 78–101.
Ji, D.-Y., Minami, T., & Lee, J. (2025). Intelligent fault diagnosis for planetary gearboxes using a hybrid cnn-transformer model with skip connections. In 2025 ieee international conference on prognostics and health management (icphm) (p. 1-6). doi:10.1109/ICPHM65385.2025.11062027
Ji, D.-Y., Sumiya, M., Kamaji, Y., Matsukura, S., Li, W., & Lee, J. (2025). Improving machine calibration performance through systematic feature design in semiconductor manufacturing. In 2025 36th annual semi advanced semiconductor manufacturing conference (asmc) (pp. 1–6).
Ji, D.-Y., Wang, B., Inoue, H., & Kanemaru, M. (2025). Motor fault detection with a hybrid physics-based and data-driven method. In 2025 ieee international electric machines drives conference (iemdc) (p. 1178-1183). doi: 10.1109/IEMDC60492.2025.11061000
Khanal, S., Huang, H.-Z., Huang, C.-G., Dahal, A., Huang, T., & Niazi, S. G. (2025). Domain-specific dual network with unsupervised domain adaptation for transfer fault prognosis across machines using multiple source domains. IEEE Transactions on Instrumentation and Measurement.
Kumar, S., Singh, O. P., Kumar, V., & Sarangi, S. (2025). Ml-based bevel gearbox fault diagnosis: An extensive time domain feature extraction approach with limited data. International Journal of Non-Linear Mechanics, 170, 105003.
Lou, Y., Kumar, A., & Xiang, J. (2022). Machinery fault diagnosis based on domain adaptation to bridge the gap between simulation and measured signals. IEEE Transactions on Instrumentation and Measurement, 71, 1–9.
Sanh, V., Debut, L., Chaumond, J., & Wolf, T. (2019). Distilbert, a distilled version of bert: smaller, faster, cheaper and lighter. arXiv preprint arXiv:1910.01108.
Shao, Z., Jiang, H., Zhang, X., Zhou, J., & Huang, W. (2025). Pll-wcan: Pseudo-label progressive learning guided wavelet class-aware adaptive network for gearbox cross-domain fault diagnosis. Mechanical Systems and Signal Processing, 230, 112624.
Shin, C.-Y., Park, J.-T., Baek, U.-J., & Kim, M.-S. (2023). A feasible and explainable network traffic classifier uti-lizing distilbert. IEEE Access, 11, 70216–70237.
Soualhi, A., Hawwari, Y., Medjaher, K., Clerc, G., Hubert, R., & Guillet, F. (2018). Phm survey: Implementation of signal processing methods for monitoring bearings and gearboxes. International Journal of Prognostics and Health Management, 9(2), 0.
Wang, Q., Taal, C., & Fink, O. (2021). Integrating expert knowledge with domain adaptation for unsupervised fault diagnosis. IEEE Transactions on Instrumentation and Measurement, 71, 1–12.
Yao, S., Kang, Q., Zhou, M., Rawa, M. J., & Abusorrah, A. (2023). A survey of transfer learning for machinery diagnostics and prognostics. Artificial Intelligence Review, 56(4), 2871–2922.
Ye, M., Yan, X., Jiang, D., & Chen, N. (2025). A multibranch attention coupled convolutional domain adaptation network for bearing intelligent fault recognition under unlabeled sample scenarios. Applied Soft Computing, 174, 113053.
Zhang, X., Li, Z., & Wang, J. (2025). Joint domain-adaptive transformer model for bearing remaining useful life prediction across different domains. Engineering Applications of Artificial Intelligence, 159, 111615.
Zhu, P., Deng, L., Tang, B., Yang, Q., & Li, Q. (2025). Digital twin-enabled entropy regularized wavelet attention domain adaptation network for gearboxes fault diagnosis without fault data. Advanced Engineering Informatics, 64, 103055.
This work is licensed under a Creative Commons Attribution 3.0 Unported License.