SAE-FSC A Siamese Attention Encoder–Based Few-Shot Cross-Domain Fault Diagnosis Framework for Bearings
##plugins.themes.bootstrap3.article.main##
##plugins.themes.bootstrap3.article.sidebar##
Abstract
Rolling element bearings are critical components in rotating machinery, where failures can cause severe downtime and safety risks. Existing fault diagnosis methods are predominantly supervised, requiring large amounts of labeled data across multiple operating conditions. However, in realistic industrial scenarios, such labeled datasets are scarce, and models trained on one regime often fail to generalize to others. To overcome this cross-domain generalization challenge, we propose a Siamese Attention Encoder–based few-shot cross-domain fault diagnosis (SAE-FSC) framework. The key novelty of this work lies in an attentionaugmented Siamese encoder that extracts highly discriminative and transferable time-series features, coupled with a composite objective function that jointly optimizes supervised cross-entropy, pairwise binary cross-entropy, and domain adversarial loss. This combination enforces intra-class domain invariant feature learning across multiple operating conditions. Extensive experiments on the Case Western Reserve University (CWRU) dataset under leave-one-fault-out (LOFO) and leave-two-fault-out (LTFO) protocols demonstrate robust generalization across unseen fault types, load conditions, and fault severities, achieving a prediction accuracy of 87% for 5 shot learning.
##plugins.themes.bootstrap3.article.details##
Cross-domain Adaptation, Few-shot Learning, Fault Diagnosis, Siamese Encoder
Chen, Z., Li, C., & Sanchez, V. (2017). Multisensor feature fusion for bearing fault diagnosis using sparse autoencoder and deep belief network. IEEE Transactions on Instrumentation and Measurement, 66(7), 1693–1702.
Chen, Z., Li, C., & Sanchez, V. (2020). Deep neural networks for vibration-based fault diagnosis of rotating machinery: A review. Mechanical Systems and Signal Processing, 133, 106983.
Finn, C., Abbeel, P., & Levine, S. (2017). Model-agnostic meta-learning for fast adaptation of deep networks. In International conference on machine learning (icml) (pp. 1126–1135).
Ganin, Y., & Lempitsky, V. (2016). Domain-adversarial training of neural networks. In Advances in neural information processing systems (neurips) (Vol. 29, pp. 2096–2104).
Jardine, A. K., Lin, D., & Banjevic, D. (2006). A review on machinery diagnostics and prognostics implementing condition-based maintenance. Mechanical Systems and Signal Processing, 20(7), 1483–1510.
Kankar, P. K., Sharma, S. C., & Harsha, S. P. (2011). Rolling element bearing fault diagnosis using wavelet transform. Neurocomputing, 74(10), 1638–1645.
Koch, G., Zemel, R., & Salakhutdinov, R. (2015). Siamese neural networks for one-shot image recognition. In Icml deep learning workshop.
Lee, H., Kim, J., & Park, S. (2023). Meta-learning for crossdomain fault diagnosis with limited labeled data. Mechanical Systems and Signal Processing, 185, 109729.
Long, M., Cao, Y., Wang, J., & Jordan, M. I. (2015). Learning transferable features with deep adaptation networks. In Proceedings of the 32nd international conference on machine learning (icml) (pp. 97–105).
Ma, J., Yu, S., & Cheng, W. (2022). Composite fault diagnosis of rolling bearing based on chaotic honey badger algorithm optimizing vmd and elm. Machines, 10(6), 469.
Mishra, R. K., Choudhary, A., Mohanty, A. R., & Fatima, S. (2021). Multi-domain bearing fault diagnosis using support vector machine. In 2021 ieee 4th international conference on computing, power and communication technologies (gucon) (pp. 1–6).
Ocak, H., & Loparo, K. A. (2005). Hmm-based fault detection and diagnosis scheme for rolling element bearings. Journal of Vibration and Acoustics, 127(4), 299–306.
Pei, Z., Cao, Z., Long, M., & Wang, J. (2018). Multiadversarial domain adaptation. In Aaai conference on artificial intelligence (Vol. 32).
Randall, R. B., & Antoni, J. (2011). Rolling element bearing diagnostics—a tutorial. Mechanical Systems and Signal Processing, 25(2), 485–520.
Shao, H., Jiang, H., Lin, Y., & Li, X. (2017). A novel deep autoencoder feature learning method for rotating machinery fault diagnosis. Mechanical Systems and Signal Processing, 95, 187–204.
Shao, H., Xia, M., Wan, J., Zhang, W., & Li, X. (2021). A novel multiple source domain adaptation method for intelligent fault diagnosis. IEEE Transactions on Industrial Electronics, 68(5), 4374–4384.
Shen, T., Liu, X., Zhang, J., & Zhao, W. (2023). Few-shot learning-based intelligent fault diagnosis for rotating machinery: A survey. Mechanical Systems and Signal Processing, 181, 109537.
Snell, J., Swersky, K., & Zemel, R. S. (2017). Prototypical networks for few-shot learning. In Advances in neural information processing systems (neurips) (Vol. 30, pp. 4077–4087).
Sun, B., & Saenko, K. (2016). Deep coral: Correlation alignment for deep domain adaptation. In European conference on computer vision (eccv) workshops (pp. 443–450). Springer.
Tian, J., Morillo, C., Azarian, M. H., & Pecht, M. (2015). Motor bearing fault detection using spectral kurtosisbased feature extraction coupled with k-nearest neighbor distance analysis. IEEE Transactions on Industrial Electronics, 63(3), 1793–1803.
Wang, Y., Peter, W. T., Tang, B., Qin, Y., Deng, L., Huang, T., & Xu, G. (2019). Order spectrogram visualization for rolling bearing fault detection under speed variation conditions. Mechanical Systems and Signal Processing, 122, 580–596.
Yassine, T., Bengherbia, B., Benyezza, H., & Ould, Z. M. (2022). Bearing faults diagnosis using envelope analysis and 1d convolutional neural network. AINTELIA Science Notes Journal, 1(1).
Zhang, W., Li, C., Ding, X., & Li, Z. (2019). Few-shot bearing fault diagnosis with siamese networks. IEEE Transactions on Instrumentation and Measurement, 68(6), 1503–1512.
Zhang, W., Li, C., Ding, X., & Li, Z. (2022). Graph convolutional networks for intelligent fault diagnosis of machines with limited data. IEEE Transactions on Industrial Electronics, 69(1), 139–149.
Zhu, K., Xu, Y., & Wang, X. (2023). Domain generalization in intelligent fault diagnosis: A comprehensive review. Mechanical Systems and Signal Processing, 187,109947.
This work is licensed under a Creative Commons Attribution 3.0 Unported License.