Study on Probabilistic Identification Method of Structural Physical Parameter Based on Bayesian Estimation

Abstract:

Base on the structural modal parameters, the linear regression models of physical parameters are derived from the dynamic equations. The posterior joint Probability Density Function (PDF) of physical parameters is obtained by using Bayesian statistics theory. Then the Markov Chain Monte Carlo (MCMC) sample method is adopted to get the marginal PDF and optimal estimation of structural physical parameters. A numerical simulation of a 5-stroy structure under the excitation of white noise with different noise level is used to identify the physical parameters by the presented method. It is shown that the method can not only determine the optimal estimate but also get the probability distribution of the structural physical parameters from the known modal parameters of the primary modes. The method is with higher validity, robustness and efficiency, and can be applied to the structural health monitoring, damage evaluation, and so on.

• Periodical: Applied Mechanics and Materials (Volumes 578-579)
• Pages: 1113-1117
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.578-579.1113
• Online since: July 2014
• Authors: Mao Sheng Gong, Xiao Hua Li, Yan Bin Gao, Qi Fang Liu
• Keywords: Bayesian Estimation, Modal Parameter, Parameter Identification, Physical Parameter

References

[1] M. Muto: Application of stochastic simulation methods to system identification. Pasadena, CA: California Institute of Technology, (2006).

[2] C.M. Yang: Statistical system identification and applications to seismic response of structures. Pasadena, CA: California Institute of Technology, (1996).

[3] M.W. Vanik: A Bayesian probabilistic approach to structural health monitoring. Pasadena, CA: California Institute of Technology, (1997).

[4] J.Y. Ching, et al: Computer-Aided Civil and Infrastructure Engineering, 21(2006), 242-257.

[5] J.L. Beck: Proceedings of the Fifth International Conference on Structural Safety and Reliability, New York, 1989, 1395-1402.

[6] J.L. Beck, L,S. Katafygiotis: Journal of Engineering Mechanics, 124 (4) (1998), 455- 461.

[7] F. Jalayer, I. Iervolino, G. Manfredi: Structural Safety, 32(3) (2010), 220-228.

[8] C. Soize: Journal of Sound and Vibration, 332 (10) (2013), 2379-2395.

[9] K.V. Yuen, S.K. Au and J.L. Beck: Journal of Engineering Mechanics, 130 (1) (2004), 16-33.

[10] W.R. Gilks, et al: Markov Chain Monte Carlo in practice. Chapmam & Hall, (1996).