Smart Structures and Systems

Techno-Press (테크노프레스)
권호 표지
DOI QR코드

DOI QR Code

Hybrid machine learning with mode shape assessment for damage identification of plates

  • Pei Yi Siow (Department of Mechanical Engineering, Faculty of Engineering, Universiti Malaya) ;
  • Zhi Chao Ong (Department of Mechanical Engineering, Faculty of Engineering, Universiti Malaya) ;
  • Shin Yee Khoo (Department of Mechanical Engineering, Faculty of Engineering, Universiti Malaya) ;
  • Kok-Sing Lim (Photonics Research Centre, Deputy Vice Chancellor (Research & Innovation) Office, Universiti Malaya) ;
  • Bee Teng Chew (Department of Mechanical Engineering, Faculty of Engineering, Universiti Malaya)
  • Received : 2022.06.06
  • Accepted : 2023.03.17
  • Published : 2023.05.25
⟨ Previous Next ⟩

Abstract

Machine learning-based structural health monitoring (ML-based SHM) methods are researched extensively in the recent decade due to the availability of advanced information and sensing technology. ML methods are well-known for their pattern recognition capability for complex problems. However, the main obstacle of ML-based SHM is that it often requires pre-collected historical data for model training. In most actual scenarios, damage presence can be detected using the unsupervised learning method through anomaly detection, but to further identify the damage types would require prior knowledge or historical events as references. This creates the cold-start problem, especially for new and unobserved structures. Modal-based methods identify damages based on the changes in the structural global properties but often require dense measurements for accurate results. Therefore, a two-stage hybrid modal-machine learning damage detection scheme is proposed. The first stage detects damage presence using Principal Component Analysis-Frequency Response Function (PCA-FRF) in an unsupervised manner, whereas the second stage further identifies the damage. To solve the cold-start problem, mode shape assessment using the first mode is initiated when no trained model is available yet in the second stage. The damage identified by the modal-based method would be stored for future training. This work highlights the performance of the scheme in alleviating the cold-start issue as it transitions through different phases, starting from zero damage sample available. Results showed that single and multiple damages can be identified at an acceptable accuracy level even when training samples are limited.

Keywords

Acknowledgement

The authors wish to acknowledge the financial support and advice given by Impact-Oriented Interdisciplinary Research Grant (IIRG007B-2019), private funding by SD Advance Engineering Sdn Bhd (PV032-2018), Advanced Shock and Vibration Research (ASVR) Group of University of Malaya and other project collaborators.

References

  1. Abdulkareem, M., Bakhary, N., Vafaei, M., Noor, N.M. and Mohamed, R.N. (2019), "Application of two-dimensional wavelet transform to detect damage in steel plate structures", Measurement, 146, 912-923. https://doi.org/10.1016/j.measurement.2019.07.027
  2. Bandara, R.P., Chan, T.H.T. and Thambiratnam, D.P. (2014), "Frequency response function based damage identification using principal component analysis and pattern recognition technique", Eng. Struct., 66, 116-128. https://doi.org/10.1016/j.engstruct.2014.01.044
  3. Bao, X.X., Li, C.L. and Xiong, C.B. (2015), "Noise elimination algorithm for modal analysis", Appl. Phys. Lett., 107(4), 5. https://doi.org/10.1063/1.4927642
  4. Bull, L., Worden, K., Manson, G. and Dervilis, N. (2018), "Active learning for semi-supervised structural health monitoring", J. Sound Vib., 437, 373-388. https://doi.org/10.1016/j.jsv.2018.08.040
  5. Bull, L.A., Rogers, T.J., Wickramarachchi, C., Cross, E.J., Worden, K. and Dervilis, N. (2019), "Probabilistic active learning: An online framework for structural health monitoring", Mech. Syst. Signal Process., 134, 20. https://doi.org/10.1016/j.ymssp.2019.106294
  6. Cevasco, D., Tautz-Weinert, J., Richmond, M., Sobey, A. and Kolios, A.J. (2022), "A damage detection and location scheme for offshore wind turbine jacket structures based on global modal properties", ASCE-ASME J. Risk Uncertain. Eng. Syst. Part B-Mech. Eng., 8(2), 12. https://doi.org/10.1115/1.4053659
  7. Chang, C.M., Lin, T.K. and Chang, C.W. (2018), "Applications of neural network models for structural health monitoring based on derived modal properties", Measurement, 129, 457-470. https://doi.org/10.1016/j.measurement.2018.07.051
  8. Chen, S., Ong, Z.C., Lam, W.H., Lim, K.-S. and Lai, K.W. (2020a), "Operational damage identification scheme utilizing de-noised frequency response functions and artificial neural network", J. Nondestr. Eval., 39(3), 66. https://doi.org/10.1007/s10921-020-00709-x
  9. Chen, S.L., Ong, Z.C., Lam, W.H., Lim, K.S. and Lai, K.W. (2020b), "Unsupervised damage identification scheme using pca-reduced frequency response function and waveform chain code analysis", Int. J. Struct. Stab. Dyn., 20(8), 26. https://doi.org/10.1142/s0219455420500911
  10. Chen, Y., Zhao, Z.Y., Wu, H.Z., Chen, X., Xiao, Q.B. and Yu, Y.Q. (2022), "P fault anomaly detection of synchronous machine winding based on isolation forest and impulse frequency response analysis", Measurement, 188, 10. https://doi.org/10.1016/j.measurement.2021.110531
  11. Esfandiari, A., Nabiyan, M.S. and Rofooei, F.R. (2020), "Structural damage detection using principal component analysis of frequency response function data", Struct. Control Health Monit., 27(7), 21. https://doi.org/10.1002/stc.2550
  12. Ghannadi, P. and Kourehli, S.S. (2019), "Data-driven method of damage detection using sparse sensors installation by serepa", J. Civ. Struct. Health Monit., 9(4), 459-475. https://doi.org/10.1007/s13349-019-00345-8
  13. Janeliukstis, R., Rucevskis, S. and Kaewunruen, S. (2019), "Mode shape curvature squares method for crack detection in railway prestressed concrete sleepers", Eng. Fail. Anal., 105, 386-401. https://doi.org/https://doi.org/10.1016/j.engfailanal.2019.07.020
  14. Jayasundara, N., Thambiratnam, D.P., Chan, T.H.T. and Nguyen, A. (2020), "Damage detection and quantification in deck type arch bridges using vibration based methods and artificial neural networks", Eng. Fail. Anal., 109, 19. https://doi.org/10.1016/j.engfailanal.2019.104265
  15. Li, J., Dackermann, U., Xu, Y.-L. and Samali, B. (2011), "Damage identification in civil engineering structures utilizing pca-compressed residual frequency response functions and neural network ensembles", Struct. Control Health Monit., 18(2), 207-226. https://doi.org/10.1002/stc.369
  16. Lim, H.C., Ong, Z.C., Ismail, Z. and Khoo, S.Y. (2019), "A performance study of controlled impact timing on harmonics reduction in operational modal testing", J. Dyn. Syst. Measur. Control-Transact. ASME, 141(3). https://doi.org/10.1115/1.4041609
  17. Liu, C., Nagler, O., Tremmel, F., Unterreitmeier, M., Frick, J.J., Patil, R.P., Gu, X.W. and Senesky, D.G. (2022), "Cluster-based acoustic emission signal processing and loading rate effects study of nanoindentation on thin film stack structures", MSSP, 165, 18. https://doi.org/10.1016/j.ymssp.2021.108301
  18. Mekjavic, I. and Damjanovic, D. (2017), "Damage assessment in bridges based on measured natural frequencies", Int. J. Struct. Stab. Dyn., 17(2). https://doi.org/10.1142/s0219455417500225
  19. Mousavi, A.A., Zhang, C.W., Masri, S.F. and Gholipour, G. (2021), "Damage detection and localization of a steel truss bridge model subjected to impact and white noise excitations using empirical wavelet transform neural network approach", Measurement, 185, 19. https://doi.org/10.1016/j.measurement.2021.110060
  20. Nguyen, D.H., Tran-Ngoc, H., Bui-Tien, T., De Roeck, G. and Wahab, M.A. (2020), "Damage detection in truss bridges using transmissibility and machine learning algorithm: Application to nam o bridge", Smart Struct. Syst., Int. J., 26(1), 35-47. https://doi.org/10.12989/sss.2020.26.1.035
  21. Nick, H., Aziminejad, A., Hosseini, M.H. and Laknejadi, K. (2021), "Damage identification in steel girder bridges using modal strain energy-based damage index method and artificial neural network", Eng. Fail. Anal., 119, 20. https://doi.org/10.1016/j.engfailanal.2020.105010
  22. Ong, Z.C., Lim, H.C., Brandt, A., Ismail, Z. and Khoo, S.Y. (2019), "An inconsistent phase selection assessment for harmonic peaks elimination in operational modal testing", Arch. Appl. Mech., 89(12), 2415-2430. https://doi.org/10.1007/s00419-019-01584-3
  23. Padil, K.H., Bakhary, N., Abdulkareem, M., Li, J. and Hao, H. (2020), "Non-probabilistic method to consider uncertainties in frequency response function for vibration-based damage detection using artificial neural network", J. Sound Vib., 467, 115069. https://doi.org/https://doi.org/10.1016/j.jsv.2019.115069
  24. Porcu, M.C., Patteri, D.M., Melis, S. and Aymerich, F. (2019), "Effectiveness of the frf curvature technique for structural health monitoring", Constr. Build. Mater., 226, 173-187. https://doi.org/https://doi.org/10.1016/j.conbuildmat.2019.07.123
  25. Sarmadi, H., Entezami, A., Salar, M. and De Michele, C. (2021), "Bridge health monitoring in environmental variability by new clustering and threshold estimation methods", J. Civil Struct. Health Monitor., 11(3), 629-644. https://doi.org/10.1007/s13349-021-00472-1
  26. Sha, G., Radzienski, M., Cao, M. and Ostachowicz, W. (2019), "A novel method for single and multiple damage detection in beams using relative natural frequency changes", MSSP, 132, 335-352. https://doi.org/https://doi.org/10.1016/j.ymssp.2019.06.027
  27. Siow, P.Y., Ong, Z.C., Khoo, S.Y. and Lim, K.S. (2021), "Damage sensitive pca-frf feature in unsupervised machine learning for damage detection of plate-like structures", Int. J. Struct. Stab. Dyn., 21(2), 29. https://doi.org/10.1142/s0219455421500280
  28. Solimine, J., Niezrecki, C. and Inalpolat, M. (2020), "An experimental investigation into passive acoustic damage detection for structural health monitoring of wind turbine blades", Struct. Health Monitor., 19(6), 1711-1725. https://doi.org/10.1177/1475921719895588
  29. Tran, C.J., Mora, O.E., Fayne, J.V. and Lenzano, M.G. (2019), "Unsupervised classification for landslide detection from airborne laser scanning", Geosciences, 9(5), p. 221. https://doi.org/10.3390/geosciences9050221
  30. Vafaei, M. and Alih, S.C. (2018), "Adequacy of first mode shape differences for damage identification of cantilever structures using neural networks", Neural Comput. Applicat., 30(8), 2509-2518. https://doi.org/10.1007/s00521-017-2846-6
  31. Wang, S. and Xu, M. (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
  32. Wickramasinghe, W.R., Thambiratnam, D.P. and Chan, T.H.T. (2020), "Damage detection in a suspension bridge using modal flexibility method", Eng. Fail. Anal., 107, p. 104194. https://doi.org/10.1016/j.engfailanal.2019.104194
  33. Xu, Y.L., Huang, Q., Zhan, S., Su, Z.Q. and Liu, H.J. (2014), "Frf-based structural damage detection of controlled buildings with podium structures: Experimental investigation", J. Sound Vib., 333(13), 2762-2775. https://doi.org/10.1016/j.jsv.2014.02.010
  34. Xu, W., Zhu, W.D., Xu, Y.F. and Cao, M.S. (2020), "A comparative study on structural damage detection using derivatives of laser-measured flexural and longitudinal vibration shapes", J. Nondestr. Eval., 39(3), 17. https://doi.org/10.1007/s10921-020-00702-4
  35. Zhu, X., Wang, Y., Li, Y., Tan, Y., Wang, G. and Song, Q. (2019), "A new unsupervised feature selection algorithm using similarity-based feature clustering", Computat. Intell., 35(1), 2-22. https://doi.org/10.1111/coin.12192