Abstract
Transfer learning methods have received abundant attention and extensively utilized in cross-domain fault diagnosis, which suppose that the label sets in the source and target domains are coincident. However, the open set domain adaptation problem which include new fault modes in the target domain is not well solved. To address the problem, an unknown-class recognition adversarial network (UCRAN) is proposed for the cross-domain fault diagnosis. Specifically, a three-dimensional discriminator is designed to conduct domain-invariant learning on the source domain, target known domain and target unknown domain. Then, an entropy minimization is introduced to determine the decision boundaries. Finally, a posteriori inference method is developed to calculate the open set recognition weight, which are used to adaptively weigh the importance between known class and unknown class. The effectiveness and practicability of the proposed UCRAN is validated by a series of experiments. The experimental results show that compared to other existing methods, the proposed UCRAN realizes better diagnosis performance in different domain transfer task.
Similar content being viewed by others
Data availability
The data analyzed in this study is partly from https://mb.uni-paderborn.de/kat/forschung/datacenter/bearing-datacenter/ and partly available on request from the corresponding author upon reasonable request.
Abbreviations
- UCRAN:
-
Unknown-class recognition adversarial network
- DA:
-
Domain adaptation
- CSDA:
-
Closed set domain adaptation
- OSDA:
-
Open set domain adaptation
- SD:
-
Source domain
- TDK:
-
Target domain-known
- TDU:
-
Target domain-unknown
- \({D}_{s}\) :
-
Source domain datasets
- \({D}_{t}\) :
-
Target domain datasets
- \({\mathcal{C}}_{s}\) :
-
The set of source domain labels
- \({\mathcal{C}}_{t}\) :
-
The set of target domain labels
- \(\mathcal{C}\) :
-
The known labels
- \(\overline{{\mathcal{C}}}_{t}\) :
-
The target-unknown labels
- \({f}_{{\theta }_{G}}\) :
-
Feature extraction module
- \({f}_{{\theta }_{D}}\) :
-
Domain discriminator
- \({f}_{{\theta }_{E}}\) :
-
Unknown recognizer
- \({f}_{{\theta }_{C}}\) :
-
Domain classifier
- \({WT}_{f}\) :
-
Operation of Continuous wavelet transform
- \({L}_{ce}\) :
-
Cross-entropy loss function
- \({D}_{SD}\) :
-
Domain discriminator for source domain
- \({D}_{TDK}\) :
-
Domain discriminator for target known domain
- \({D}_{TDU}\) :
-
Domain discriminator for target unknown domain
- \({L}_{D}^{s}\) :
-
The source domain discrimination loss function
- \({L}_{D}^{t}\) :
-
The target domain discrimination loss function
- \({w}_{x}\) :
-
Open set recognition weight
- \({L}_{D}^{TDK}\) :
-
The target domain-known domain discrimination loss function
- \({L}_{D}^{TDU}\) :
-
The target domain-unknown domain discrimination loss function
- \({L}_{D}\) :
-
Overall domain discrimination loss of
- \({L}_{G}\) :
-
Overall generator loss
- \({L}_{e}^{s}\) :
-
The loss function of source domain
- \({L}_{ent}^{t}\) :
-
The entropy loss
References
Ainapure, A., Li, X., Singh, J., Yang, Q., & Lee, J. (2020). Deep learning-based cross-machine health identification method for vacuum pumps with domain adaptation. Procedia Manufacturing, 48, 1088–1093. https://doi.org/10.1016/j.promfg.2020.05.149
Cao, X. C., Chen, B. Q., & Zeng, N. Y. (2020a). A deep domain adaption model with multi-task networks for planetary gearbox fault diagnosis. Neurocomputing, 409, 173–190. https://doi.org/10.1016/j.neucom.2020.05.064
Chen, X. K., Shao, H. D., Xiao, Y. M., Yan, S., Cai, B. P., & Liu, B. (2023a). Collaborative fault diagnosis of rotating machinery via dual adversarial guided unsupervised multi-domain adaptation network. Mechanical Systems and Signal Processing, 198, 110427. https://doi.org/10.1016/j.ymssp.2023.110427
Chen, X. H., Yang, R., Xue, Y. H., Huang, M. J., Ferrero, R., & Wang, Z. D. (2023b). Deep transfer learning for bearing fault diagnosis: A systematic review since 2016. IEEE Transactions on Instrumentation and Measurement, 72, 3508221. https://doi.org/10.1109/TIM.2023.3244237
Chen, Y. Y., Zhang, D., Zhu, K. P., & Yan, R. Q. (2023c). An adaptive activation transfer learning approach for fault diagnosis. IEEE/ASME Transactions on Mechatronics, 28, 2645–2656. https://doi.org/10.1109/TMECH.2023.3243533
Chen, Z., Liao, Y., Li, J., Huang, R., Xu, L., Jin, G., & Li, W. (2022a). A multi-source weighted deep transfer network for open-set fault diagnosis of rotary machinery. IEEE Transactions on Cybernetics. https://doi.org/10.1109/TCYB.2022.3195355
Chen, Z., Wu, J., Deng, C., Wang, C., & Wang, Y. (2022b). Residual deep subdomain adaptation network: A new method for intelligent fault diagnosis of bearings across multiple domains. Mechanism and Machine Theory, 169, 104635. https://doi.org/10.1016/j.mechmachtheory.2021.104635
Chen, Z. Y., Li, Z. R., Wu, J., Deng, C., & Dai, W. (2022c). Deep residual shrinkage relation network for anomaly detection of rotating machines. Journal of Manufacturing Systems, 65, 579–590. https://doi.org/10.1016/j.jmsy.2022.10.014
Ganin, Y., Ustinova, E., Ajakan, H., Germain, P., Larochelle, H., Laviolette, F., Marchand, M., & Lempitsky, V. (2016). Domain-adversarial training of neural networks. Journal of Machine Learning Research, 17, 1–35. https://doi.org/10.1007/978-3-319-58347-1_10
Guo, L., Yu, Y. X., Liu, Y. K., Gao, H. L., & Chen, T. (2022). Reconstruction domain adaptation transfer network for partial transfer learning of machinery fault diagnostics. IEEE Transactions on Instrumentation and Measurement, 71, 2502710. https://doi.org/10.1109/TIM.2021.3129213
Guo, Q., Li, J., Zhou, F., Li, G., & Lin, J. (2023b). An open-set fault diagnosis framework for MMCs based on optimized temporal convolutional network. Applied Soft Computing, 133, 109959. https://doi.org/10.1016/j.asoc.2022.109959
Guo, Y., Zhang, J. D., Sun, B., & Wang, Y. K. (2023a). Adversarial deep transfer learning in fault diagnosis: progress, challenges, and future prospects. Sensors, 23, 7263. https://doi.org/10.3390/s23167263
Hu, W. F., Shao, J. Y., Jiao, Q., Wang, C. X., Cheng, J., Liu, Z. Y., & Tan, J. R. (2022). A new differentiable architecture search method for optimizing convolutional neural networks in the digital twin of intelligent robotic grasping. Journal of Intelligent Manufacturing, 34, 2943–2961. https://doi.org/10.1007/s10845-022-01971-8
Jang, J., Na, B., Shin, D. H., Ji, M., Song, K., & Moon, I.-C. (2022). Unknown-aware domain adversarial learning for open-set domain adaptation. Advances in Neural Information Processing Systems, 35, 16755–16767. https://doi.org/10.48550/arXiv.2206.07551
Jiang, W., Wu, J., Zhu, H., Li, X., & Gao, L. (2023). Paired ensemble and group knowledge measurement for health evaluation of wind turbine gearbox under compound fault scenarios. Journal of Manufacturing Systems, 70, 382–394. https://doi.org/10.1016/j.jmsy.2023.08.004
Kim, H., Lee, H. J., Kim, S., & Kim, S. W. (2023). Attention recurrent neural network-based severity estimation method for early-stage fault diagnosis in robot harness cable. Sensors, 23, 5299. https://doi.org/10.3390/s23115299
Lessmeier, C., Kimotho, J. K., Zimmer, D., & Sextro, W. (2016). Condition monitoring of bearing damage in electromechanical drive systems by using motor current signals of electric motors: A benchmark data set for data-driven classification. In PHM Society European Conference (pp. 5–8). https://doi.org/10.36001/PHME.2016.V3I1.1577
Li, G.R., Kang, G.L., Zhu, Y., Wei, Y.C., Yang, Y., & IEEE Comp, S.O.C. (2021). Domain consensus clustering for universal domain adaptation. In 2021 IEEE/CVF conference on computer vision and pattern recognition, CVPR 2021 (pp. 9752–9761).https://doi.org/10.1109/CVPR46437.2021.00963
Li, X., Zhang, W., & Ding, Q. (2018). A robust intelligent fault diagnosis method for rolling element bearings based on deep distance metric learning. Neurocomputing, 310, 77–95. https://doi.org/10.1016/j.neucom.2018.05.021
Li, X., Zhang, W., Ma, H., Luo, Z., & Li, X. (2020). Partial transfer learning in machinery cross-domain fault diagnostics using class-weighted adversarial networks. Neural Networks, 129, 313–322. https://doi.org/10.1016/j.neunet.2020.06.014
Li, X. Y., Yuan, P., Wang, X. K., Li, D. Y., Xie, Z. J., & Kong, X. W. (2023). An unsupervised transfer learning bearing fault diagnosis method based on depthwise separable convolution. Measurement Science and Technology, 34, 095401. https://doi.org/10.1088/1361-6501/acda55
Liu, H., Cao, Z., Long, M., Wang, J., & Yang, Q. (2020). Separate to adapt: open set domain adaptation via progressive separation. In: 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (pp. 2922–2931). https://doi.org/10.1109/CVPR.2019.00304
Liu, G. K., Shen, W. M., Gao, L., & Kusiak, A. (2022a). Knowledge transfer in fault diagnosis of rotary machines. IET Collaborative Intelligent Manufacturing, 4, 17–34. https://doi.org/10.1049/cim2.12047
Liu, Z., Chen, G., Li, Z., Kang, Y., Qu, S., & Jiang, C. (2022b). PSDC: A prototype-based shared-dummy classifier model for open-set domain adaptation. IEEE Transactions on Cybernetics, 53, 1–14. https://doi.org/10.1109/tcyb.2022.3228301
Lu, N., Cui, Z. Y., Hu, H. Y., & Yin, T. (2023b). Multi-view and multi-level network for fault diagnosis accommodating feature transferability. Expert Systems with Applications, 213, 119057. https://doi.org/10.1016/j.eswa.2022.119057
Lu, S. L., Lu, J. F., An, K., Wang, X. X., & He, Q. B. (2023a). Edge computing on IoT for machine signal processing and fault diagnosis: A review. IEEE Internet Things, 10, 11093–11116. https://doi.org/10.1109/JIOT.2023.3239944
Ma, X. B., Yan, B. X., Wang, H., & Liao, H. T. (2023b). Decision-level machinery fault prognosis using N-BEATS-based degradation feature prediction and reconstruction. Mechanical Systems and Signal Processing. https://doi.org/10.1016/j.ymssp.2023.110435
Ma, Y. C., Liu, Y. Q., Yang, Z. L., Cheng, M., & Meng, H. (2023a). Deep adversarial transfer neural network for fault diagnosis of wind turbine gearbox. International Journal of Green Energy, 20, 1750–1762. https://doi.org/10.1080/15435075.2023.2194375
Nie, X. Y., & Xie, G. (2021). A novel normalized recurrent neural network for fault diagnosis with noisy labels. Journal of Intelligent Manufacturing, 32, 1271–1288. https://doi.org/10.1007/s10845-020-01608-8
Oliveira, A., Zemouri, R., Cambron, P., & Tahan, A. (2023). Early detection and diagnosis of wind turbine abnormal conditions using an interpretable supervised variational autoencoder model. Energies, 16, 4544. https://doi.org/10.3390/en16124544
Qu, S., Zou, T., Roehrbein, F., Lu, C., Chen, G., Tao, D., & Jiang, C. (2023). Upcycling models under domain and category shift. In IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (pp. 20019–20028). https://doi.org/10.1109/CVPR52729.2023.01917
She, D. M., Chen, J., Yan, X. A., Zhao, X. L., & Pecht, M. (2023). Diversity maximization-based transfer diagnosis approach of rotating machinery. Structural Health Monitoring, 23, 410–420. https://doi.org/10.1177/14759217231164921
Tian, J., Han, D., Karimi, H. R., Zhang, Y., & Shi, P. (2023). Deep learning-based open set multi-source domain adaptation with complementary transferability metric for mechanical fault diagnosis. Neural Networks, 162, 69–82. https://doi.org/10.1016/j.neunet.2023.02.025
Wang, C., Tian, B., Zirui, L., Xiaoqi, W., & Jun, W. (2023a). Self-attention and subdomain adaptive adversarial network for bearing fault diagnosis under varying operation conditions. Chinese Journal of Ship Research, 18, 260–268. https://doi.org/10.19693/j.issn.1673-3185.03092
Wang, Q. B., Xu, Y. B., Yang, S. K., Chang, J. T., Zhang, J. G., & Kong, X. G. (2023b). A domain adaptation method for bearing fault diagnosis using multiple incomplete source data. Journal of Intelligent Manufacturing, 35, 777–791. https://doi.org/10.1007/s10845-023-02075-7
Wen, L., Li, X. Y., Gao, L., & Zhang, Y. Y. (2018). A new convolutional neural network-based data-driven fault diagnosis method. IEEE Transactions on Industrial Electronics, 65, 5990–5998. https://doi.org/10.1109/TIE.2017.2774777
Wu, Z. H., Jiang, H. K., Liu, S. W., Liu, Y. P., & Yang, W. F. (2023). Conditional distribution-guided adversarial transfer learning network with multi-source domains for rolling bearing fault diagnosis. Advanced Engineering Informatics, 56, 101993. https://doi.org/10.1016/j.aei.2023.101993
Xiao, Y. M., Shao, H. D., Feng, M. J., Han, T., Wan, J. F., & Liu, B. (2023). Towards trustworthy rotating machinery fault diagnosis via attention uncertainty in transformer. Journal of Machine Learning Research, 70, 186–201. https://doi.org/10.1016/j.jmsy.2023.07.012
Yan, X. A., She, D. M., & Xu, Y. D. (2023). Deep order-wavelet convolutional variational autoencoder for fault identification of rolling bearing under fluctuating speed conditions. Expert Systems with Applications, 216, 119479. https://doi.org/10.1016/j.eswa.2022.119479
Yang, B., Lei, Y. G., Li, X., & Roberts, C. (2023). Deep targeted transfer learning along designable adaptation trajectory for fault diagnosis across different machines. IEEE Transactions on Industrial Electronics, 70, 9463–9473. https://doi.org/10.1109/TIE.2022.3212415
Yang, G., Tang, H., Zhong, Z., Ding, M., Shao, L., Sebe, N., & Ricci, E. (2021). Transformer-based source-free domain adaptation. In CVPR. arXiv:2105.14138
Yu, X., Zhao, Z., Zhang, X., Chen, X., & Cai, J. (2023). Statistical identification guided open-set domain adaptation in fault diagnosis. Reliability Engineering and System Safety. https://doi.org/10.1016/j.ress.2022.109047
Yu, X., Zhao, Z., Zhang, X., Zhang, Q., Liu, Y., Sun, C., & Chen, X. (2022). Deep-learning-based open set fault diagnosis by extreme value theory. IEEE Transactions on Industrial Informatics, 18, 185–196. https://doi.org/10.1109/TII.2021.3070324
Zhang, P. B., Chen, R. X., Xu, X. Y., Yang, L. X., & Ran, M. Y. (2023a). Recent progress and prospective evaluation of fault diagnosis strategies for electrified drive powertrains: A comprehensive review. Measurement, 222, 113711. https://doi.org/10.1016/j.measurement.2023.113711
Zhang, S. Y., Su, L., Gu, J. F., Li, K., Zhou, L., & Pecht, M. (2023b). Rotating machinery fault detection and diagnosis based on deep domain adaptation: A survey. Chinese J Aeronaut, 36, 45–74. https://doi.org/10.1016/j.cja.2021.10.006
Zhang, X., Wang, H. F., Wu, B., Zhou, Q., & Hu, Y. M. (2023c). A novel data-driven method based on sample reliability assessment and improved CNN for machinery fault diagnosis with non-ideal data. Journal of Intelligent Manufacturing, 34, 2449–2462. https://doi.org/10.1007/s10845-022-01944-x
Zhang, Y., Ji, J. C., Ren, Z., Ni, Q., & Wen, B. (2023d). Multi-sensor open-set cross-domain intelligent diagnostics for rotating machinery under variable operating conditions. Mechanical Systems and Signal Processing, 191, 110172. https://doi.org/10.1016/j.ymssp.2023.110172
Zhang, Z. W., Nie, G. C., Shao, M. Y., Li, L., Zhou, J. L., & Shao, S. J. (2023e). Multi-sample-distances-fusion- and generalized-Pareto-distribution-based open-set fault diagnosis of rolling bearing. Nonlinear Dynamics, 111, 11407–11428. https://doi.org/10.1007/s11071-023-08441-7
Zheng, X. R., He, Z. W., Nie, J. H., Li, P., Dong, Z. K., & Gao, M. Y. (2024). A progressive multi-source domain adaptation method for bearing fault diagnosis. Applied Acoustics, 216, 109797. https://doi.org/10.1016/j.apacoust.2023.109797
Zhou, H. S., Ouyang, D. T., Tian, X. L., & Zhang, L. M. (2023). DiagDO: An efficient model based diagnosis approach with multiple observations. Frontiers of Computer Science-Chinese, 17, 176407. https://doi.org/10.1007/s11704-022-2261-8
Zhu, J., Huang, C.-G., Shen, C., & Shen, Y. (2022). Cross-domain open-set machinery fault diagnosis based on adversarial network with multiple auxiliary classifiers. IEEE Transactions on Industrial Informatics, 18, 8077–8086. https://doi.org/10.1109/TII.2021.3138558
Acknowledgements
This research is supported in part by the National Natural Science Foundation of China under the Grant No. 51875225, and in part by Hubei Provincial Natural Science Foundation for Innovation Groups under Grant No. 2021CFA026.
Author information
Authors and Affiliations
Contributions
Ke Wu: Writing-Original draft preparation, Methodology, Formal analysis. Wei Xu: Software, Validation, Visualization, Investigation. Qiming Shu: Project administration. Wenjun Zhang: Conceptualization, Methodology. Xiaolong Cui: Editing, Supervision. Jun Wu: Writing-Reviewing and Editing, Conceptualization, Methodology, Supervision, Funding acquisition.
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that this manuscript is original, has not been published before, and is not currently being considered for publication elsewhere. The authors have no relevant financial or non-financial interests to disclose. All authors certify that they are not affiliated with any organization or entity that has a financial or non-financial interest in the subject matter of this manuscript. The authors declare that they have no conflicts of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Wu, K., Xu, W., Shu, Q. et al. Unknown-class recognition adversarial network for open set domain adaptation fault diagnosis of rotating machinery. J Intell Manuf (2024). https://doi.org/10.1007/s10845-024-02395-2
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s10845-024-02395-2