DOI QR코드

DOI QR Code

A novel adaptive unscented Kalman Filter with forgetting factor for the identification of the time-variant structural parameters

  • Yanzhe Zhang (State Key Laboratory of Mechanical Behavior and System Safety of Traffic Engineering Structures, Shijiazhuang Tiedao University) ;
  • Yong Ding (School of Civil Engineering, Harbin Institute of Technology) ;
  • Jianqing Bu (State Key Laboratory of Mechanical Behavior and System Safety of Traffic Engineering Structures, Shijiazhuang Tiedao University) ;
  • Lina Guo (College of Water Conservancy and Civil Engineering, Northeast Agricultural University)
  • 투고 : 2022.06.09
  • 심사 : 2023.05.31
  • 발행 : 2023.07.25

초록

The parameters of civil engineering structures have time-variant characteristics during their service. When extremely large external excitations, such as earthquake excitation to buildings or overweight vehicles to bridges, apply to structures, sudden or gradual damage may be caused. It is crucially necessary to detect the occurrence time and severity of the damage. The unscented Kalman filter (UKF), as one efficient estimator, is usually used to conduct the recursive identification of parameters. However, the conventional UKF algorithm has a weak tracking ability for time-variant structural parameters. To improve the identification ability of time-variant parameters, an adaptive UKF with forgetting factor (AUKF-FF) algorithm, in which the state covariance, innovation covariance and cross covariance are updated simultaneously with the help of the forgetting factor, is proposed. To verify the effectiveness of the method, this paper conducted two case studies as follows: the identification of time-variant parameters of a simply supported bridge when the vehicle passing, and the model updating of a six-story concrete frame structure with field test during the Yangbi earthquake excitation in Yunnan Province, China. The comparison results of the numerical studies show that the proposed method is superior to the conventional UKF algorithm for the time-variant parameter identification in convergence speed, accuracy and adaptability to the sampling frequency. The field test studies demonstrate that the proposed method can provide suggestions for solving practical problems.

키워드

과제정보

The authors gratefully acknowledge the financial support by the National Key R&D Program of China [Grant No. 2021YFB2600605, 2021YFB2600600], the Key R&D Program of Hebei Province [Grant No. 19275405D], the Hebei Provincial Transport Bureau Research Program [Grant No. TH-201902] and Scientific Research Fund of Institute of Engineering Mechanics, China Earthquake Administration [Grant No. 2019D22].

참고문헌

  1. Ahmed-Ali, T., Kenne, G. and Lamnabhi-Lagarrigue, F. (2009), "Identification of nonlinear systems with time-varying parameters using a sliding-neural network observer", Neurocomputing, 72, 1611-1620. https://doi.org/10.1016/j.neucom.2008.09.001
  2. Asl, R.M., Hagh, Y.S., Simani, S. and Handroos, H. (2019), "Adaptive square-root unscented Kalman filter: An experimental study of hydraulic actuator state estimation", Mech. Syst. Signal Process., 132, 670-691. https://doi.org/10.1016/j.ymssp.2019.07.021
  3. Astroza, R., Ebrahimian, H. and Conte, J. (2019a), "Performance comparison of Kalman-based filters for nonlinear structural finite element model updating", J. Sound Vib., 438, 520-542. https://doi.org/10.1016/J.JSV.2018.09.023
  4. Astroza, R., Alessandri, A. and Conte, J.P. (2019b), "A dual adaptive filtering approach for nonlinear finite element model updating accounting for modeling uncertainty", Mech. Syst. Signal Process., 115, 782-800. https://doi.org/10.1016/j.ymssp.2018.06.014
  5. Bao, Y., Velni, J.M., Basina, A. and Shahbakhti, M. (2020), "Identification of state-space linear parameter-varying models using artificial neural networks", IFAC-PapersOnLine, 53, 5286-5291. https://doi.org/10.1016/J.IFACOL.2020.12.1209
  6. Bisht, S.S. and Singh, M.P. (2014), "An adaptive unscented Kalman filter for tracking sudden stiffness changes", Mech. Syst. Signal Process., 49, 181-195. https://doi.org/10.1016/j.ymssp.2014.04.009
  7. Calabrese, A., Strano, S. and Terzo, M. (2018), "Adaptive constrained unscented Kalman filtering for real-time nonlinear structural system identification", Struct. Control Health Monitor., 25. https://doi.org/10.1002/STC.2084
  8. Cao, J.X., Xiong, H.B. and Chen, L. (2020), "Procedure for parameter identification and mechanical properties assessment of CLT connections", Eng. Struct., 203. https://doi.org/10.1016/j.engstruct.2019.109867
  9. Chen, Y.Y. and Zhou, Y.W. (2020), "Machine learning based decision making for time varying systems: parameter estimation and performance optimization", Knowl. Based Syst., 190. https://doi.org/10.1016/j.knosys.2020.105479
  10. Chin, R., Maass, A.I., Ulapane, N., Manzie, C., Shames, I., Nesic, D., Rowe, J.E. and Nakada, H. (2020), "Active learning for linear parameter-varying system identification", ArXiv, abs/2005.00711. https://doi.org/10.1016/J.IFACOL.2020.12.1274
  11. Chowdhary, G. and Jategaonkar, R. (2010), "Aerodynamic parameter estimation from flight data applying extended and unscented Kalman filter", Aerosp. Sci. Technol., 14(2), 106-117. https://doi.org/10.1016/j.ast.2009.10.003
  12. Chung, M.J. and Sato, T. (2006), "Structural identification using stochastic filtering techniques based on measurements from wireless data acquisition system", Steel Struct., 6, 353-360. https://doi.org/10.12989/scs.2006.6.4.353
  13. Cui, M., Khodayar, M., Chen, C., Wang, X., Zhang, Y. and Khodayar, M.E. (2019), "Deep learning based time-varying parameter identification for system-wide load modeling", LEEE Transact. Smart Grid, 10, 6102-6114. https://doi.org/10.1109/Tsg.2019.2896493
  14. Ding, Y., Guo, L.N. and Zhao, B. (2017), "Parameter Identification for Nonlinear Structures by a Constrained Kalman Filter with Limited Input Information", Int. J. Struct. Stabil. Dyn., 17, 1750010. https://doi.org/10.1142/S0219455417500109
  15. Doebling, S.W., Farrar, C.R. and Prime, M.B. (1998), "A summary review of vibration-based damage identification methods", Shock Vib. Digest, 30(2), 91-105. https://doi.org/10.1177/058310249803000201
  16. Guo, L.N., Ding, Y., Wang, Z., Xu, G.S. and Wu, B. (2018), "A dynamic load estimation method for nonlinear structures with unscented Kalman filter", Mech. Syst. Signal Process., 101, 254-273. https://doi.org/10.1016/j.ymssp.2017.07.047
  17. Hu, P.D., Zhang, M.Z., Zhang, R., Wu, Q.P. and Yang, A.L. (2021), "Correlation method and Kalman filter-based adaptive angle rate estimation for time-varying periodic signals of the attitude and heading reference system", Mech Syst. Signal Process., 156. https://doi.org/10.1016/j.ymssp.2021.107695
  18. Humar, J., Bagchi, A. and Xu, H.P. (2006), "Performance of vibration-based techniques for the identification of structural damage", Struct. Health Monitor., 5, 215-241. https://doi.org/10.1177/1475921706067738
  19. Jategaonkar, R. and Plaetenschke, E. (1998), "Estimation of aircraft parameters using filter error methods and extended Kalman filter", DFVLR FB, 88-15.
  20. Julier, S.J., Uhlmann, J.K. and Durrant-Whyte, H.F. (1995), "A new approach for filtering nonlinear systems", Proceedings of 1995 American Control Conference-ACC'95, 3, 1628-1632. https://doi.org/10.1109/ACC.1995.529783
  21. Manoach, E., Samborski, S., Mitura, A. and Warminski, J. (2012), "Vibration based damage detection in composite beams under temperature variations using Poincare maps", Int. J. Mech. Sci., 62, 120-132. https://doi.org/10.1016/J.IJMECSCI.2012.06.006
  22. Mariani, S. and Ghisi, A. (2007), "Unscented Kalman filtering for nonlinear structural dynamics", Nonlinear Dyn., 49, 131-150. https://doi.org/10.1007/s11071-006-9118-9
  23. Masti, D., Bernardini, D. and Bemporada, A. (2021), "A machine-learning approach to synthesize virtual sensors for parameter-varying systems", ArXiv, 61, 40-49. https://doi.org/10.1016/j.ejcon.2021.06.005
  24. Mulay, A., Ben, B.S., Ismail, S. and Kocanda, A. (2019), "Prediction of average surface roughness and formability in single point incremental forming using artificial neural network", Arch. Civil Mech. Eng., 19, 1135-1149. https://doi.org/10.1016/j.acme.2019.06.004
  25. Naranjo-Perez, J., Jimenes Alonso, J.F., Pavic, A. and Saez, A. (2020), "Finite-element-model updating of civil engineering structures using a hybrid UKF-HS algorith", Struct. Infrastr. Eng., 17, 620-637. https://doi.org/10.1080/15732479.2020.1760317
  26. Nguyen, H., Vu, T., Vo, T.P. and Thai, H.T. (2021), "Efficient machine learning models for prediction of concrete strengths", Constr. Build. Mater., 266. https://doi.org/10.1016/j.conbuildmat.2020.120950
  27. Pappalardo, C.M. and Guida, D. (2016), "Control of nonlinear vibrations using the adjoint method", Meccanica, 52, 2503-2526. https://doi.org/10.1007/s11012-016-0601-1
  28. Pappalardo, C.M. and Guida, D. (2017), "Adjoint-Based Optimization Procedure for Active Vibration Control of Nonlinear Mechanical Systems", J. Dyn. Syst. Measure. Control-Transact. ASME, 139, 081010. https://doi.org/10.1115/1.4035609
  29. Rahimi, A., Kumar, K.D. and Alighanbari, H. (2017), "Fault estimation of satellite reaction wheels using covariance based adaptive unscented Kalman filter", Acta Astronautica, 134, 159-169. https://doi.org/10.1016/j.actaastro.2017.02.003
  30. Sadhukhan, C., Mitra, S.K., Naskar, M.K. and Sharifpur, M. (2021), "Fault diagnosis of a nonlinear hybrid system using adaptive unscented Kalman flter bank", Eng. Comput., 38, 2717-2728. https://doi.org/10.1007/s00366-020-01235-0
  31. Schleiter, S. and Altay, O. (2020), "Identification of abrupt stiffness changes of structures with tuned mass dampers under sudden events", Struct. Control Health Monitor., 27. https://doi.org/10.1002/stc.2530
  32. Shu, X.S., Bao, T.F., Li, Y.T., Gong, J. and Zhang, K. (2021), "VAE-TALSTM: a temporal attention and variational autoencoder-based long short-term memory framework for dam displacement prediction", Eng. Comput. https://doi.org/10.1007/s00366-021-01362-2
  33. Song, M., Astroza, R., Ebrahimian, H., Moaveni, B. and Papadimitriou, C. (2020), "Adaptive Kalman filters for nonlinear finite element model updating", Mech. Syst. Signal Process., 143. https://doi.org/10.1016/j.ymssp.2020.106837
  34. Soyoz, S. and Feng, M.Q. (2009), "Long-term monitoring and identification of bridge structural parameters", Comput.-Aided Civil Infrastr. Eng., 24, 82-92. https://doi.org/10.1111/j.1467-8667.2008.00572.x
  35. Taffese, W.Z. and Sistonen, E. (2017), "Machine learning for durability and service-life assessment of reinforced concrete structures: Recent advances and future directions", Automat. Constr., 77, 1-14. https://doi.org/10.1016/j.autcon.2017.01.016
  36. Wang, L.J., Xie, Y.X., Wu, Z.J., Du, Y.X. and He, K.D. (2018), "A new fast convergent iteration regularization method", Eng. Comput., 35, 127-138. https://doi.org/10.1007/s00366-018-0588-4
  37. Wang, N., Li, L.Y. and Wang, Q. (2019), "Adaptive UKF-Based Parameter Estimation for Bouc-Wen Model of Magnetorheological Elastomer Materials", J. Aerosp. Eng., 32, https://doi.org/10.1061/(ASCE)AS.1943-5525.0000961-4
  38. Xiao, X., Xu, X. and Shen, W. (2020), "Simultaneous identification of the frequencies and track irregularities of high-speed railway bridges from vehicle vibration data", Mech. Syst. Signal Process., 152, 107412. https://doi.org/10.1016/j.ymssp.2020.107412
  39. Yan, G., Sun, H. and Buyukozturk, O. (2016), "Impact load identification for composite structures using Bayesian regularization and unscented Kalman filter", Struct. Control Health Monitor., 24(5). https://doi.org/10.1002/stc.1910
  40. Yang, J.N. and Lin, S. (2005), "Identification of parametric variations of structures based on least squares estimation and adaptive tracking technique", J. Eng. Mech., 131, 290-298. https://doi.org/10.1061/(Asc(e)0733-9399(2005)131:3(290)
  41. Yang, J.N., Lin, S.L., Huang, H.W. and Zhou, L. (2006), "An adaptive extended Kalman filter for structural damage identification", Struct. Control Health Monitor., 13, 849-867. https://doi.org/10.1002/stc.84
  42. Zhou, W., Li, X.L., Yi, J. and He, H.B. (2019), "A Novel UKF-RBF Method based on adaptive noise factor for fault diagnosis in pumping unit", IEEE Transact. Indust. Inform., 15. https://doi.org/1415-1424. 10.1109/TII.2018.2839062