Smart Structures and Systems

Techno-Press (테크노프레스)
권호 표지
DOI QR코드

DOI QR Code

Dual-loss CNN: A separability-enhanced network for current-based fault diagnosis of rolling bearings

  • Lingli Cui (Key Laboratory of Advanced Manufacturing Technology, Beijing University of Technology) ;
  • Gang Wang (Key Laboratory of Advanced Manufacturing Technology, Beijing University of Technology) ;
  • Dongdong Liu (Key Laboratory of Advanced Manufacturing Technology, Beijing University of Technology) ;
  • Jiawei Xiang (College of Mechanical and Electrical Engineering, Wenzhou University) ;
  • Huaqing Wang (College of Mechanical and Electrical Engineering, Beijing University of Chemical Technology)
  • Received : 2023.04.03
  • Accepted : 2024.03.18
  • Published : 2024.04.25
⟨ Previous Next ⟩

Abstract

Current-based mechanical fault diagnosis is more convenient and low cost since additional sensors are not required. However, it is still challenging to achieve this goal due to the weak fault information in current signals. In this paper, a dual-loss convolutional neural network (DLCNN) is proposed to implement the intelligent bearing fault diagnosis via current signals. First, a novel similarity loss (SimL) function is developed, which is expected to maximize the intra-class similarity and minimize the inter-class similarity in the model optimization operation. In the loss function, a weight parameter is further introduced to achieve a balance and leverage the performance of SimL function. Second, the DLCNN model is constructed using the presented SimL and the cross-entropy loss. Finally, the two-phase current signals are fused and then fed into the DLCNN to provide more fault information. The proposed DLCNN is tested by experiment data, and the results confirm that the DLCNN achieves higher accuracy compared to the conventional CNN. Meanwhile, the feature visualization presents that the samples of different classes are separated well.

Keywords

Acknowledgement

This work is supported by the National Natural Science Foundation of China under Grant No. 52075008 and Grant 52305086.

References

  1. Bae, J., Jung, W. and Park, Y.H. (2022), "Normal data based rotating machine anomaly detection using CNN with self-labeling", Smart Struct. Syst., Int. J., 29(6), 757-766. https://doi.org/10.12989/sss.2022.29.6.757
  2. Cui, L. Li, W., Wang, X., Zhao, D. and Wang, H. (2022), "Comprehensive remaining useful life prediction for rolling element bearings based on time-varying particle filtering", IEEE Trans. Instrum. Meas., 71, 1-10. https://doi.org/10.1109/TIM.2022.3163167
  3. Elbouchikhi, E., Choqueuse, V., Auger, F. and Benbouzid, M.E.H. (2017), "Motor current signal analysis based on a matched subspace detector", IEEE Trans. Instrum. Meas., 66(12), 3260-3270. https://doi.org/10.1109/TIM.2017.2749858
  4. Gao, Y., Liu, X. and Xiang, J. (2020), "FEM simulation-based generative adversarial networks to detect bearing faults", IEEE Trans. Ind. Inform., 16(7), 4961-4971. https://doi.org/10.1109/TII.2020.2968370
  5. Guan, Y. and Feng, Z. (2021), "Adaptive linear chirplet transform for analyzing signals with crossing frequency trajectories", IEEE Trans. Ind. Electron., 69(8), 8396-8410. https://doi.org/10.1109/TIE.2021.3097605
  6. Herold, T., Mboo, C.P. and Hameyer, K. (2013), "Evaluation of the use of an electrical drive as a sensor for the detection of bearing damage", Proc. Conference on Acoustics, AIA-DAGA., pp. 1538-1540.
  7. Hoang, D.T. and Kang, H.J. (2020), "A motor current signal-based bearing fault diagnosis using deep learning and information fusion", IEEE Trans. Instrum. Meas., 69(6), 3325-3333. https://doi.org/10.1109/TIM.2019.2933119
  8. Hsueh, Y.M., Ittangihal, V.R., Wu, W.B., Chang, H.C. and Kuo, C.C. (2019), "Fault diagnosis system for induction motors by CNN using empirical wavelet transform", Symmetry, 11(10), 1212. https://doi.org/10.3390/sym11101212 [EWT]
  9. Kim, I., Lee, Y., Oh, J. and Kim, N. (2022), "Fault detection and classification of permanent magnet synchronous machine using signal injection", Smart Struct. Syst., Int. J., 29(6), 785-790. https://doi.org/10.12989/sss.2022.29.6.785
  10. Kingma, D.P. and Ba, J. (2014), "Adam: A method for stochastic optimization", arXiv preprint. https://doi.org/10.48550/arXiv.1412.6980
  11. Kumar, A., Vashishtha, G., Gandhi, C.P., Zhou, Y., Glowacz, A. and Xiang, J. (2021), "Novel convolutional neural network (NCNN) for the diagnosis of bearing defects in rotary machinery", IEEE Trans. Instrum. Meas., 70, 1-10. https://doi.org/10.1109/TIM.2021.3055802
  12. Lecun, Y., Bottou, L., Bengio, Y. and Haffner, P. (1998), "Gradient-based learning applied to document recognition", Proc. IEEE, 86(11), 2278-2324. https://doi.org/10.1109/5.726791
  13. Leite, V.C.M.N., da Silva, J.G.B., Torres, G.L., Veloso, G.F.C., da Silva, L.E.B., Bonaldi, E.L. and de Oliveira, L.E.D.L. (2017), "Bearing fault detection in induction machine using squared envelope analysis of stator current", Bearing Technology; Rijeka, Croatia: InTech.
  14. Lessmeier, C., Kimotho, J.K., Zimmer, D. and 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", Proc. Eur. Conf. Prognostics Health Manage. Soc., 3(1), 5-8. https://doi.org/10.36001/phme.2016.v3i1.1577
  15. Li, Y., Dai, W. and Zhang, W. (2020), "Bearing Fault Feature Selection Method Based on Weighted Multidimensional Feature Fusion", IEEE Access, 8, 19008-19025. https://doi.org/10.1109/ACCESS.2020.2967537
  16. Liu, D., Cui, L., Cheng, W., Zhao, D. and Wen, W. (2022a), "Rolling bearing fault severity recognition via data mining integrated with convolutional neural network", IEEE Sens. J., 22(6), 5768-5777. https://doi.org/10.1109/JSEN.2022.3146151
  17. Liu, X., Liu, S., Xiang, J., Sun, R. and Wei, Y. (2022b), "An ensemble and shared selective adversarial network for partial domain fault diagnosis of machinery", Eng. Appl. Artif. Intel., 113, 104906. https://doi.org/10.1016/j.engappai.2022.104906
  18. Liu, D., Cui, L. and Cheng, W. (2023), "Fault diagnosis of wind turbines under nonstationary conditions based on a novel tacho-less generalized demodulation", Renew. Energy, 206, 645-657. https://doi.org/10.1016/j.renene.2023.01.056
  19. Randall, R.B. and Antoni, J. (2011), "Rolling element bearing diagnostics-A tutorial", Mech. Syst. signal Process., 25(2), 485-520. https://doi.org/10.1016/j.ymssp.2010.07.017
  20. Sadoughi, M., Downey, A., Bunge, G., Ranawat, A., Hu, C. and Laflamme, S. (2018), "A deep learning-based approach for fault diagnosis of roller element bearings", In: 2018 Annual Conference of the Prognostics and Health Management Society, 10(1), 1-7.
  21. Shao, H., Jiang, H., Zhang, H. and Liang, T. (2018), "Electric locomotive bearing fault diagnosis using a novel convolutional deep belief network", IEEE Trans. Ind. Electron., 65(3), 2727-2736. https://doi.org/10.1109/TIE.2017.2745473
  22. Shao, H., Xia, M., Wan, J. and Silva, C.W. (2022), "Modified stacked autoencoder using adaptive Morlet wavelet for intelligent fault diagnosis of rotating machinery", IEEE-ASME Trans. Mech., 27(1), 24-33. https://doi.org/10.1109/TMECH.2021.3058061
  23. Shen, C., Wang, D., Liu, Y., Kong, F. and Tse, P.W. (2014), "Recognition of rolling bearing fault patterns and sizes based on two-layer support vector regression machines", Smart Struct. Syst., Int. J., 13(3), 453-471. https://doi.org/10.12989/sss.2014.13.3.453
  24. Singh, S., Kumar, A. and Kumar, N. (2014), "Motor current signature analysis for bearing fault detection in mechanical systems", Procedia Mater. Sci., 6, 171-177. https://doi.org/10.1016/j.mspro.2014.07.021
  25. Toma, R.N., Prosvirin, A.E. and Kim, J.M. (2020), "Bearing fault diagnosis of induction motors using a genetic algorithm and machine learning classifiers", Sensors, 20(7), 1884. https://doi.org/10.3390/s20071884
  26. Van, V.M.L. and Hinton, G. (2008), "Visualizing data using t-SNE", J. Mach. Learn. Res., 9, 2579-2605.
  27. Wang, G. and Xiang, J. (2021), "Railway sleeper crack recognition based on edge detection and CNN", Smart Struct. Syst., Int. J., 28(6), 779-789. https://doi.org/10.12989/sss.2021.28.6.779
  28. Wang, H., Xu, J., Yan, R., Sun, C. and Chen, X. (2019), "Intelligent bearing fault diagnosis using multi-head attention-based CNN", Procedia Manuf., 49, 112-118. https://doi.org/10.1016/j.promfg.2020.07.005
  29. Wang, H., Xu, J., Yan, R. and Gao, R.X. (2020), "A new intelligent bearing fault diagnosis method using SDP representation and SE-CNN", IEEE Trans. Instrum. Meas., 69(5), 2377-2389. https://doi.org/10.1109/TIM.2019.2956332