DOI QR코드

DOI QR Code

Damage localization and quantification of a truss bridge using PCA and convolutional neural network

  • Jiajia, Hao (School of Civil and Environmental Engineering, Faculty of Engineering and Information Technology, University of Technology Sydney) ;
  • Xinqun, Zhu (School of Civil and Environmental Engineering, Faculty of Engineering and Information Technology, University of Technology Sydney) ;
  • Yang, Yu (School of Civil and Environmental Engineering, Faculty of Engineering and Information Technology, University of Technology Sydney) ;
  • Chunwei, Zhang (Multidisciplinary Center for Infrastructure Engineering, Shenyang University of Technology) ;
  • Jianchun, Li (School of Civil and Environmental Engineering, Faculty of Engineering and Information Technology, University of Technology Sydney)
  • 투고 : 2022.03.28
  • 심사 : 2022.09.02
  • 발행 : 2022.12.25

초록

Deep learning algorithms for Structural Health Monitoring (SHM) have been extracting the interest of researchers and engineers. These algorithms commonly used loss functions and evaluation indices like the mean square error (MSE) which were not originally designed for SHM problems. An updated loss function which was specifically constructed for deep-learning-based structural damage detection problems has been proposed in this study. By tuning the coefficients of the loss function, the weights for damage localization and quantification can be adapted to the real situation and the deep learning network can avoid unnecessary iterations on damage localization and focus on the damage severity identification. To prove efficiency of the proposed method, structural damage detection using convolutional neural networks (CNNs) was conducted on a truss bridge model. Results showed that the validation curve with the updated loss function converged faster than the traditional MSE. Data augmentation was conducted to improve the anti-noise ability of the proposed method. For reducing the training time, the normalized modal strain energy change (NMSEC) was extracted, and the principal component analysis (PCA) was adopted for dimension reduction. The results showed that the training time was reduced by 90% and the damage identification accuracy could also have a slight increase. Furthermore, the effect of different modes and elements on the training dataset was also analyzed. The proposed method could greatly improve the performance for structural damage detection on both the training time and detection accuracy.

키워드

과제정보

This research is financially supported by the Ministry of Science and Technology of China (Grant No. 2019YFE0112400).

참고문헌

  1. Abdeljaber, O., Avci, O., Kiranyaz, M.S., Gabbouj, M. and Inman, D.J. (2017), "Real-time vibration-based structural damage detection using one-dimensional convolutional neural networks", J. Sound Vib., 388, 154-170. https://doi.org/10.1016/j.jsv.2016.10.043
  2. Abdeljaber, O., Avci, O., Kiranyaz, M.S., Boashash, B., Sodano, H., and Inman, D.J. (2018), "1-D CNNs for structural damage detection: Verification on a structural health monitoring benchmark data", Neurocomputing, 275, 1308-1317. https://doi.org/10.1016/j.neucom.2017.09.069
  3. Aswal, N., Sen, S. and Mevel, L. (2021), "Estimation of local failure in tensegrity using Interacting Particle-Ensemble Kalman Filter", Mech. Syst. Signal Process., 160, 107824. https://doi.org/10.1016/j.ymssp.2021.107824
  4. Cao, M., Radzienski, M., Xu, W. and Ostachowicz, W. (2014), "Identification of multiple damage in beams based on robust curvature mode shapes", Mech. Syst. Signal Process, 46(2), 468-480. https://doi.org/10.1016/j.ymssp.2014.01.004
  5. Cha, Y.-J., Choi, C. and Buyukozturk, O. (2017), "Deep learning-based crack damage detection using convolutional neural networks", Comput.-Aided Civil Infrastr. Eng., 32(5), 361-378. https://doi.org/10.1111/mice.12263
  6. Huynh, T.-C., Park, J.-H., Jung, H.-J. and Kim, J.-T. (2019), "Quasi-autonomous bolt-loosening detection method using vision-based deep learning and image processing", Automat. Constr., 105, 102844. https://doi.org/10.1016/j.autcon.2019.102844
  7. Islam, M.M. and Kim, J.-H. (2019), "Vision-based autonomous crack detection of concrete structures using a fully convolutional encoder-decoder network", Sensors, 19(19), 4251. https://doi.org/10.3390/s19194251
  8. Kim, C.-W., Zhang, Y., Wang, Z.R., Oshima, Y. and Morita, T. (2018), "Long-term bridge health monitoring and performance assessment based on a Bayesian approach", Struct. Infrastr. Eng., 14(7), 883-894. https://doi.org/10.1080/15732479.2018.1436572
  9. Kordestani, H., Xiang, Y.-Q. and Ye, X.-W. (2018), "Output-only damage detection of steel beam using moving average filter", Shock and Vibration, 2018, 1-13. https://doi.org/10.1155/2018/2067680
  10. Kordestani, H., Zhang, C.W., Masri, S.F. and Shadabfar, M. (2021), "An empirical time-domain trend line-based bridge signal decomposing algorithm using Savitzky-Golay filter", Struct. Control Health Monitor., 28(7), e2750. https://doi.org/10.1002/stc.2750
  11. Lee, S., Park, S., Kim, T., Lieu, Q.X. and Lee, J. (2021), "Damage quantification in truss structures by limited sensor-based surrogate model", Appl. Acoust., 172, 107547. https://doi.org/10.1016/j.apacoust.2020.107547
  12. Lin, Y.Z., Ni, 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
  13. Maaten, L. and Hinton, G. (2008), "Visualizing data using t-SNE", J. Mach. Learn. Res., 9(86), 2579-2605.
  14. Nick, H. and Aziminejad, A. (2021), "Vibration-based damage identification in steel girder bridges using artificial neural network under noisy conditions", J. Nondestr. Eval., 40(1). https://doi.org/10.1007/s10921-020-00744-8
  15. Paz, M. and Kim, Y.H. (2019), Structural Dynamics: Theory and Computation, Cham: Springer International Publishing. https://doi.org/10.1007/978-3-319-94743-3
  16. Santos, F.L.M. dos, Peeters, B., Van der Auweraer, H., Goes, L.C.S. and Desmet, W. (2016), "Vibration-based damage detection for a composite helicopter main rotor blade", Case Stud. Mech. Syst. Signal Process., 3, 22-27. https://doi.org/10.1016/j.csmssp.2016.01.001
  17. Sharma, S. and Sen, S. (2021), "Bridge damage detection in presence of varying temperature using two-step neural network approach", J. Bridge Eng., 26(6). https://doi.org/10.1061/(ASCE)BE.1943-5592.0001708
  18. Spencer, B.F., Hoskere, V. and Narazaki, Y. (2019), "Advances in computer vision-based civil infrastructure inspection and monitoring", Eng., 5(2), 199-222. https://doi.org/10.1016/j.eng.2018.11.030
  19. Sun, L.M., Shang, Z.Q., 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
  20. Talebpour, M.H., Goudarzi, Y. and Sharifnezhad, M. (2020), "Clustering elements of truss structures for damage identification by CBO", Periodica Polytechnica Civil Engineering, October. https://doi.org/10.3311/PPci.16636
  21. Teng, S., Chen, G.F., Gong, P.P., Liu, G. and Cui, F.S. (2020), "Structural damage detection using convolutional neural networks combining strain energy and dynamic response", Meccanica, 55(4), 945-959. https://doi.org/10.1007/s11012-019-01052-w
  22. Wang, S.Q. and Xu, M.Q. (2019), "Modal strain energy-based structural damage identification: a review and comparative study", Struct. Eng. Int., 29(2), 234-248. https://doi.org/10.1080/10168664.2018.1507607
  23. Wang, R.H., Chencho, An, S., Li, J., Li, L., Hao, H. and Liu, W.Q. (2020), "Deep residual network framework for structural health monitoring", Struct. Health Monitor., 20(4), 147592172091837. https://doi.org/10.1177/1475921720918378
  24. Xu, J., Hao, J.J., Li, H.N., Luo, M.Z., Guo, W. and Li, W.J. (2017), "Experimental damage identification of a model reticulated shell", Appl. Sci., 7(4), 362. https://doi.org/10.3390/app7040362
  25. Yu, Y., Wang, C.Y., Gu, X.Y. and Li, J.C. (2019), "A novel deep learning-based method for damage identification of smart building structures", Struct. Health Monitor., 18(1), 143-163. https://doi.org/10.1177/1475921718804132
  26. Yu, Y., Rashidi, M., Samali, B., Mohammadi, M., Nguyen, T.N. and Zhou, X.X. (2022), "Crack detection of concrete structures using deep convolutional neural networks optimized by enhanced chicken swarm algorithm", Struct. Health Monitor., p. 14759217211053546. https://doi.org/10.1177/14759217211053546
  27. Zhou, Z., Wegner, L.D. and Sparling, B.F. (2021), "Data quality indicators for vibration-based damage detection and localization", Eng. Struct., 230, 111703. https://doi.org/10.1016/j.engstruct.2020.111703