Diagnostics for Mechanical Systems with Unknown Fault Modes: A Novel Open Set Recognition Approach
##plugins.themes.bootstrap3.article.main##
##plugins.themes.bootstrap3.article.sidebar##
Abstract
A common challenge in condition-based maintenance is that not all fault modes of the system are known from historical data, particularly in systems with evolving operating conditions, or are newly developed. Conventional data-driven diagnostic methods typically rely on a closed-set assumption, where all possible fault modes are represented during training. As a result, previously unseen fault modes are often incorrectly assigned to known ones with high confidence, potentially leading to ineffective or even risky maintenance decisions. To address this limitation, this paper proposes an open-set diagnostic approach that integrates supervised contrastive learning with a simplified Hopfield energy score. An encoder is trained using a supervised contrastive loss function to obtain well-separated embeddings of known system states. During inference, the alignment between a test observation and the learned state prototypes is quantified using the simplified Hopfield energy score. Observations with low similarity to known states are identified as unknown through thresholding. Experimental results on a benchmark dataset demonstrate that the proposed method effectively distinguishes unknown states while maintaining an accurate classification of known states, achieving competitive performance compared to established baselines. By explicitly identifying unknown states, the proposed approach enables more reliable and risk-aware maintenance decisions, particularly in safety-critical applications.
How to Cite
##plugins.themes.bootstrap3.article.details##
Fault Diagnostics, Open Set Recognition
Fink, O., Wang, Q., Svensen, M., Dersin, P., Lee, W.-J., & Ducoffe, M. (2020). Potential, challenges and future directions for deep learning in prognostics and health management applications. Engineering Applications of Artificial Intelligence, 92, 103678.
Gavarini, G., Stucchi, D., Ruospo, A., Boracchi, G., & Sanchez, E. (2022). Open-set recognition: An inexpensive strategy to increase DNN reliability. In 2022 IEEE 28th International Symposium on On-Line Testing and Robust System Design (IOLTS) (pp. 1–7).
He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR).
Hendrycks, D., & Gimpel, K. (2016). A baseline for detecting misclassified and out-of-distribution examples in neural networks. arXiv preprint arXiv:1610.02136.
Hu, Y., Miao, X., Si, Y., Pan, E., & Zio, E. (2022). Prognostics and health management: A review from the perspectives of design, development and decision. Reliability Engineering & System Safety, 217, 108063.
Kemmerzell, N., Schreiner, A., Khalid, H., Schalk, M., & Bordoli, L. (2025). Towards a better understanding of evaluating trustworthiness in AI systems. ACM Computing Surveys, 57(9), 1–38.
Khosla, P., Teterwak, P., Wang, C., Sarna, A., Tian, Y., Isola, P., ... Krishnan, D. (2020). Supervised contrastive learning. Advances in Neural Information Processing Systems, 33, 18661–18673.
Lai, C., Baraldi, P., & Zio, E. (2024). Physics-informed deep autoencoder for fault detection in new-design systems. Mechanical Systems and Signal Processing, 215, 111420.
Lee, K., Lee, K., Lee, H., & Shin, J. (2018). A simple unified framework for detecting out-of-distribution samples and adversarial attacks. Advances in Neural Information Processing Systems, 31.
Li, J., Yue, K., Chen, Z., Xia, J., Li, W., & Zhang, X. (2024). An uncertainty-aware continual learning framework for fault diagnosis of rotating machinery with homogeneous-heterogeneous faults. IEEE Transactions on Automation Science and Engineering.
Li, X., Hu, Y., Li, M., & Zheng, J. (2020). Fault diagnostics between different types of components: A transfer learning approach. Applied Soft Computing, 86, 105950.
Li, X., Hu, Y., Zheng, J., Li, M., & Ma, W. (2021). Central moment discrepancy-based domain adaptation for intelligent bearing fault diagnosis. Neurocomputing, 429, 12–24.
Liang, S., Li, Y., & Srikant, R. (2017). Enhancing the reliability of out-of-distribution image detection in neural networks. arXiv preprint arXiv:1706.02690.
Lin, F., Ren, J., Zhao, Z., Zhang, X., & Chen, X. (2025). An uncertainty-guided contrastive learning method for OOD detection in trustworthy fault diagnosis. Reliability Engineering & System Safety, 111967.
Łuczak, D. (2024). Machine fault diagnosis through vibration analysis: Continuous wavelet transform with complex Morlet wavelet and time–frequency RGB image recognition via convolutional neural network. Electronics, 13(2), 452.
Peng, P., Lu, J., Xie, T., Tao, S., Wang, H., & Zhang, H. (2022). Open-set fault diagnosis via supervised contrastive learning with negative out-of-distribution data augmentation. IEEE Transactions on Industrial Informatics, 19(3), 2463–2473.
Ramsauer, H., Schäfl, B., Lehner, J., Seidl, P., Widrich, M., Adler, T., ... others. (2020). Hopfield networks is all you need. arXiv preprint arXiv:2008.02217.
Rombach, K., Michau, G., & Fink, O. (2023). Controlled generation of unseen faults for partial and open-partial domain adaptation. Reliability Engineering & System Safety, 230, 108857.
Smith, W. A., & Randall, R. B. (2015). Rolling element bearing diagnostics using the Case Western Reserve University data: A benchmark study. Mechanical Systems and Signal Processing, 64, 100–131.
Sun, Y., Ming, Y., Zhu, X., & Li, Y. (2022). Out-of-distribution detection with deep nearest neighbors. In International Conference on Machine Learning (pp. 20827–20840).
Wu, C., Gao, F., & Zhang, R. (2026). An out-of-distribution fault detection framework using deep global feature modeling and extended logit fusion for industrial processes. Process Safety and Environmental Protection, 108778.
Xie, W., Han, T., Pei, Z., & Xie, M. (2023). A unified out-of-distribution detection framework for trustworthy prognostics and health management in renewable energy systems. Engineering Applications of Artificial Intelligence, 125, 106707.
Yang, J., Zhou, K., Li, Y., & Liu, Z. (2024). Generalized out-of-distribution detection: A survey. International Journal of Computer Vision, 132(12), 5635–5662.
Yoon, J., Kim, D., & Kim, D. (2022). SDBOSR: Separable decision boundary-based open-set recognition for manufacturing equipment fault classification. In 2022 IEEE International Conference on Prognostics and Health Management (ICPHM) (pp. 121–126).
Zhang, J., Fu, Q., Chen, X., Du, L., Li, Z., Wang, G., ... others. (2022). Out-of-distribution detection based on in-distribution data patterns memorization with modern Hopfield energy. In The Eleventh International Conference on Learning Representations.
Zhang, Y., Ding, J., Li, Y., Ren, Z., & Feng, K. (2024). Multi-modal data cross-domain fusion network for gearbox fault diagnosis under variable operating conditions. Engineering Applications of Artificial Intelligence, 133, 108236.
Zhu, J., Chen, N., & Peng, W. (2018). Estimation of bearing remaining useful life based on multiscale convolutional neural network. IEEE Transactions on Industrial Electronics, 66(4), 3208–3216.
Zohrabi, R., Hasani, H., Baghshah, M., Rohrbach, A., Rohrbach, M., & Rohban, M. H. (2026). Spurious-aware prototype refinement for reliable out-of-distribution detection. Advances in Neural Information Processing Systems, 38, 44545–44589.
Zonta, T., Da Costa, C. A., da Rosa Righi, R., De Lima, M. J., Da Trindade, E. S., & Li, G. P. (2020). Predictive maintenance in Industry 4.0: A systematic literature review. Computers & Industrial Engineering, 150, 106889.
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.