Smart Structures and Systems

Techno-Press (테크노프레스)
권호 표지
DOI QR코드

DOI QR Code

Structural health monitoring response reconstruction based on UAGAN under structural condition variations with few-shot learning

  • Jun, Li (Centre for Infrastructure Monitoring and Protection, School of Civil and Mechanical Engineering, Curtin University) ;
  • Zhengyan, He (School of Civil Engineering, Guangzhou University) ;
  • Gao, Fan (School of Civil Engineering, Guangzhou University)
  • Received : 2022.06.24
  • Accepted : 2022.11.03
  • Published : 2022.12.25
⟨ Previous Next ⟩

Abstract

Inevitable response loss under complex operational conditions significantly affects the integrity and quality of measured data, leading the structural health monitoring (SHM) ineffective. To remedy the impact of data loss, a common way is to transfer the recorded response of available measure point to where the data loss occurred by establishing the response mapping from measured data. However, the current research has yet addressed the structural condition changes afterward and response mapping learning from a small sample. So, this paper proposes a novel data driven structural response reconstruction method based on a sophisticated designed generating adversarial network (UAGAN). Advanced deep learning techniques including U-shaped dense blocks, self-attention and a customized loss function are specialized and embedded in UAGAN to improve the universal and representative features extraction and generalized responses mapping establishment. In numerical validation, UAGAN efficiently and accurately captures the distinguished features of structural response from only 40 training samples of the intact structure. Besides, the established response mapping is universal, which effectively reconstructs responses of the structure suffered up to 10% random stiffness reduction or structural damage. In the experimental validation, UAGAN is trained with ambient response and applied to reconstruct response measured under earthquake. The reconstruction losses of response in the time and frequency domains reached 16% and 17%, that is better than the previous research, demonstrating the leading performance of the sophisticated designed network. In addition, the identified modal parameters from reconstructed and the corresponding true responses are highly consistent indicates that the proposed UAGAN is very potential to be applied to practical civil engineering.

Keywords

Acknowledgement

The support from the National Natural Science Foundation of China project No. 52178279 and Guangzhou Basic and Applied Basic Research Foundation project, is greatly appreciated.

References

  1. Bao, Y. and Li, H. (2021), "Machine learning paradigm for structural health monitoring", Struct. Health Monitor., 20(4), 1353-1372. https://doi.org/10.1177/1475921720972416
  2. Bao, Y., Li, H., Sun, X., Yu, Y. and Ou, J. (2013), "Compressive sampling-based data loss recovery for wireless sensor networks used in civil structural health monitoring", Struct. Health Monitor., 12(1), 78-95. https://doi.org/10.1177/1475921712462936
  3. Bao, J., Chen, D., Wen, F., Li, H. and Hua, G. (2017), "CVAE-GAN: fine-grained image generation through asymmetric training", Proceedings of 2017 IEEE International Conference on Computer Vision (ICCV), Venice, Italy, October.
  4. Cross, E.J., Koo, K.Y., Brownjohn, J.M.W. and Worden, K. (2013), "Long-term monitoring and data analysis of the Tamar Bridge", Mech. Syst. Signal Process., 35(1-2), 16-34. https://doi.org/10.1016/j.ymssp.2012.08.026
  5. Fan, G., Li, J. and Hao, H. (2019), "Lost data recovery for structural health monitoring based on convolutional neural networks", Struct. Control Health Monitor., 26(10), e2433. https://doi.org/10.1002/stc.2433
  6. Fan, G., Li, J. and Hao, H. (2020), "Vibration signal denoising for structural health monitoring by residual convolutional neural networks", Measurement, 157, p. 107651. https://doi.org/10.1016/j.measurement.2020.107651
  7. Fan, G., Li, J. and Hao, H. (2021a), "Dynamic response reconstruction for structural health monitoring using densely connected convolutional networks", Struct. Health Monitor., 20(4), 1373-1391. https://doi.org/10.1177/1475921720916881
  8. Fan, G., Li, J., Hao, H. and Xin, Y. (2021b), "Data driven structural dynamic response reconstruction using segment based generative adversarial networks", Eng. Struct., 234, p. 111970. https://doi.org/10.1016/j.engstruct.2021.111970
  9. Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A. and Bengio, Y. (2020), "Generative adversarial networks", Commun. ACM, 63(11), 139-144. https://doi.org/10.1145/3422622
  10. He, J., Zhou, Y., Guan, X., Zhang, W., Zhang, W. and Liu, Y. (2016), "Time domain strain/stress reconstruction based on empirical mode decomposition: numerical study and experimental validation", Sensors, 16(8), 22. https://doi.org/10.3390/s16081290
  11. Huang, Y., Beck, J.L., Wu, S. and Li, H. (2014), "Robust Bayesian compressive sensing for signals in structural health monitoring", Comput.-Aided Civil Infrastr. Eng., 29(3), 160-179. https://doi.org/10.1111/mice.12051
  12. Jeong, S., Ferguson, M., Hou, R., Lynch, J.P., Sohn, H. and Law, K.H. (2019), "Sensor data reconstruction using bidirectional recurrent neural network with application to bridge monitoring", Adv. Eng. Inform., 42, p. 100991. https://doi.org/10.1016/j.aei.2019.100991
  13. Jiang, K., Han, Q., Du, X. and Ni, P. (2021a), "Structural dynamic response reconstruction and virtual sensing using a sequence to sequence modeling with attention mechanism", Automat. Constr., 131, p. 103895. https://doi.org/10.1016/j.autcon.2021.103895
  14. Jiang, H., Wan, C., Yang, K., Ding, Y. and Xue, S. (2021b), "Continuous missing data imputation with incomplete dataset by generative adversarial networks-based unsupervised learning for long-term bridge health monitoring", Struct. Health Monitor., 21(3), 1093-1109. https://doi.org/10.1177/14759217211021942
  15. Kalman, R.E. (1960), "A new approach to linear filtering and prediction problems", J. Basic Eng., 82(1), 35-45. https://doi.org/10.1115/1.3662552
  16. Klikowicz, P., Salamak, M. and Poprawa, G. (2016), "Structural health monitoring of urban structures", In: World Multidisciplinary Civil Engineering - Architecture - Urban Planning Symposium (WMCAUS), Prague, Czech, August.
  17. Kullaa, J. (2013), "Detection, identification, and quantification of sensor fault in a sensor network", Mech. Syst. Signal Process., 40(1), 208-221. https://doi.org/10.1016/j.ymssp.2013.05.007
  18. Law, S.S., Li, J. and Ding, Y. (2011), "Structural response reconstruction with transmissibility concept in frequency domain", Mech. Syst. Signal Process., 25(3), 952-968. https://doi.org/10.1016/j.ymssp.2010.10.001
  19. Lei, X., Sun, L. and Xia, Y. (2021), "Lost data reconstruction for structural health monitoring using deep convolutional generative adversarial networks", Struct. Health Monitor., 20(4), 2069-2087. https://doi.org/10.1177/1475921720959226
  20. Li, J. and Hao, H. (2014), "Substructure damage identification based on wavelet-domain response reconstruction", Eng. Struct., 13(4), 389-405. https://doi.org/10.1177/1475921714532991
  21. Li, J., Law, S.S. and Ding, Y. (2012), "Substructure damage identification based on response reconstruction in frequency domain and model updating", Eng. Struct., 41, 270-284. https://doi.org/10.1016/j.engstruct.2012.03.035
  22. Li, J., Law, S.S. and Ding, Y. (2013), "Damage detection of a substructure based on response reconstruction in frequency domain", Key Eng. Mater., 569, 823-830. https://doi.org/10.4028/www.scientific.net/KEM.569-570.823
  23. Li, J., Hao, H., Fan, G., Ni, P., Wang, X., Wu, C., Lee, J.M. and Jung, K.H. (2017), "Numerical and experimental verifications on damping identification with model updating and vibration monitoring data", Smart Struct. Syst., Int. J., 20(2), 127-137. https://doi.org/10.12989/sss.2017.20.2.127
  24. Li, Y., Ni, P., Sun, L. and Zhu, W. (2022), "A convolutional neural network-based full-field response reconstruction framework with multitype inputs and outputs", Struct. Control Health Monitor., p. e2961. https://doi.org/10.1002/stc.2961
  25. Lin, Y.Z., Nie, Z.H. and Ma, H.W. (2017), "Structural damage detection with automatic feature-extraction through deep learning", Comput.-Aided Civil Infrastr. Eng., 32(12), 1025-1046. https://doi.org/10.1111/mice.12313
  26. Lin, Z., Li, M., Zheng, Z., Cheng, Y. and Yuan, C. (2020), "Self-attention convlstm for spatiotemporal prediction", Proceedings of the AAAI Conference on Artificial Intelligence, New York, NY, USA, February.
  27. Luong, M.T., Pham, H. and Manning, C.D. (2015), "Effective approaches to attention-based neural machine translation", arXiv preprint arXiv:1508.04025. https://doi.org/10.48550/arXiv.1508.04025
  28. Nagarajaiah, S. and Erazo, K. (2016), "Structural monitoring and identification of civil infrastructure in the United States", Struct. Monitor. Maint., Int. J., 3(1), 51-69. https://doi.org/10.12989/smm.2016.3.1.051
  29. Ni, P., Li, J., Hao, H. and Xia, Y. (2018), "Stochastic dynamic analysis of marine risers considering Gaussian system uncertainties", J. Sound Vib., 416, 224-243. https://doi.org/10.1016/j.jsv.2017.11.049
  30. Ni, P., Li, J., Hao, H., Xia, Y. and Du, X. (2019a), "Stochastic dynamic analysis of marine risers considering fluid-structure interaction and system uncertainties", Eng. Struct., 198, 14. https://doi.org/10.1016/j.engstruct.2019.109507
  31. Ni, P., Xia, Y., Li, J., Hao, H., Bi, K. and Zuo, H. (2019b), "Multi-scale stochastic dynamic response analysis of offshore risers with lognormal uncertainties", Ocean Eng., 189, p. 106333. https://doi.org/10.1016/j.oceaneng.2019.106333
  32. Niu, Y., Fritzen, C.P., Jung, H., Buethe, I., Ni, Y.Q. and Wang, Y.W. (2015), "Online simultaneous reconstruction of wind load and structural responses-Theory and application to Canton Tower", Comput.-Aided Civil Infrastr. Eng., 30(8), 666-681. https://doi.org/10.1111/mice.12134
  33. Oh, B.K., Glisic, B., Kim, Y. and Park, H.S. (2020), "Convolutional neural network-based data recovery method for structural health monitoring", Struct. Health Monitor., 19(6), 1821-1838. https://doi.org/10.1177/1475921719897571
  34. Petersen, O.W., Oiseth, O., Nord, T.S. and Lourens, E. (2018), "Estimation of the full-field dynamic response of a floating bridge using Kalman-type filtering algorithms", Mech. Syst. Signal Process., 107, 12-28. https://doi.org/10.1016/j.ymssp.2018.01.022
  35. Ronneberger, O., Fischer, P. and Brox, T. (2015), "U-net: Convolutional networks for biomedical image segmentation", International Conference on Medical Image Computing and Computer-Assisted Intervention, November.
  36. Sun, L., Shang, Z., Xia, Y., Bhowmick, S. and Nagarajaiah, S. (2020), "Review of bridge structural health monitoring aided by big data and artificial intelligence: From condition assessment to damage detection", J. Struct. Eng., 146(5), 04020073. https://doi.org/10.1061/(asce)st.1943-541x.0002535
  37. Thadikemalla, V.S.G. and Gandhi, A.S. (2018), "A data loss recovery technique using compressive sensing for structural health monitoring applications", KSCE J. Civil Eng., 22(12), 5084-5093. https://doi.org/10.1007/s12205-017-2070-z
  38. Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A.N., Kaiser, L. and Polosukhin, I. (2017), "Attention is all you need", Proceedings of the 31st International Conference on Neural Information Processing Systems, Long Beach, CA, USA, December.
  39. Wan, Z., Wang, T., Huang, Q. and Li, L. (2014), "Structural response reconstruction for non-proportionally damped systems in the presence of closely spaced modes", J. Vibroeng., 16(8), 3740-3758. https://www.extrica.com/article/15202
  40. Xu, T., Zhang, P., Huang, Q., Zhang, H., Gan, Z., Huang, X. and He, X. (2018), "Attngan: Fine-grained text to image generation with attentional generative adversarial networks", Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA, June.
  41. Yi, T.H., Li, H.N. and Gu, M. (2013), "Recent research and applications of GPS-based monitoring technology for high-rise structures", Struct. Control Health Monitor., 20(5), 649-670. https://doi.org/10.1002/stc.1501
  42. Zhang, Y. and Lei, Y. (2021), "Data anomaly detection of bridge structures using convolutional neural network based on structural vibration signals", Symmetry, 13(7), p. 1186. https://doi.org/10.3390/sym13071186
  43. Zhang, X.H. and Wu, Z.B. (2019), "Dual-type structural response reconstruction based on moving-window Kalman filter with unknown measurement noise", J. Aerosp. Eng., 32(4), 14. https://doi.org/10.1061/(asce)as.1943-5525.0001016
  44. Zhang, C.D. and Xu, Y.L. (2016), "Structural damage identification via multi-type sensors and response reconstruction", Struct. Health Monitor., 15(6), 715-729. https://doi.org/10.1177/1475921716659787
  45. Zhang, L., Ji, Y., Lin, X. and Liu, C. (2017), "Style transfer for anime sketches with enhanced residual u-net and auxiliary classifier gan", In: 2017 4th IAPR Asian Conference on Pattern Recognition (ACPR), Nanjing, China, November.
  46. Zhang, X.H., Zhu, Z., Yuan, G.K. and Zhu, S. (2021), "Adaptive Mode Selection Integrating Kalman Filter for Dynamic Response Reconstruction", J. Sound Vib., 515, 18. https://doi.org/10.1016/j.jsv.2021.116497
  47. Zhou, H., Zhang, S., Peng, J., Zhang, S., Li, J., Xiong, H. and Zhang, W. (2021), "Informer: Beyond efficient transformer for long sequence time-series forecasting", Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 35, No. 12, February.
  48. Zhu, S., Zhang, X.H., Xu, Y.L. and Zhan, S. (2013), "Multi-type sensor placement for multi-scale response reconstruction", Adv. Struct. Eng., 16(10), 1779-1797. https://doi.org/10.1260/1369-4332.16.10.1779