An automatic determination method for the convergence and required posterior samples is proposed to address the problem that the multi-chain MCMC algorithm mainly relies on setting a large number of iteration steps to generate a sufficient number of posterior samples. The optimal competition strategy was introduced at the beginning of the MCMC algorithm iteration, replacing the random difference by the directional difference, so that the samples could be moved to the target direction quickly to accelerate the computational efficiency. Based on the sampling distribution theorem, a t-distribution determination index was constructed using samples within a period of time to automatically determine whether the multi-chain MCMC converged and automatically terminated the algorithm after the sample size meets the statistical requirements, so as to reduce the computational workload of the smooth period. The numerical examples and the results of the real bridge correction show that the proposed determination method can improve the computational efficiency of the multi-chain MCMC algorithm by 30% with the same calculation accuracy, and the entire iteration process can be accelerated by about 50% considering the preset step size, which provides a method to support the application of Bayesian-based finite element model correction in large civil engineering.
FENGZ Q, WANGW Z, HUAX G, et al. Structural model updating based on modal parameters and modified firefly algorithm[J].Journal of Hunan University (Natural Sciences), 2022, 49(11): 252-259.(in Chinese)
ZHANGY F, PENGZ R, ZHANGX P, et al. Stochastic finite element model updating based on radial basis model and Bhattacharyya distance[J]. Journal of Vibration and Shock, 2021,40(19): 221-229.(in Chinese)
WANH P, RENW X, HUANGT L. Stochastic model updating approach by using Bayesian inference[J]. China Journal of Highway and Transport, 2016, 29(4): 67-76.(in Chinese)
YANGP C, XUES T, XIEL Y. Bayesian finite-element model updating of passively controlled building structures[J]. China Civil Engineering Journal, 2021, 54(Sup.1): 13-19.(in Chinese)
[9]
KATAFYGIOTISL S, BECKJ L. Updating models and their uncertainties.Ⅱ: Model identifiability[J]. Journal of Engineering Mechanics, 1998, 124(4): 463-467.
[10]
JONESG L, QINQ. Markov chain Monte Carlo in practice[J].Annual Review of Statistics and Its Application,2022,9:557-578.
[11]
GONGL, FLEGALJ M .A practical sequential stopping rule for high-dimensional Markov chain Monte Carlo[J].Journal of Computational and Graphical Statistics,2016,25(3):684-700.
[12]
张建新 .基于贝叶斯方法的有限元模型修正研究[D].重庆:重庆大学, 2014.
[13]
ZHANGJ X .Research on the modification of finite element model based on Bayesian method[D].Chongqing:Chongqing University,2014.(in Chinese)
[14]
GEYERC J .Practical Markov chain Monte Carlo[J].Statistical Science,1992,7(4): 473-511.
[15]
YUB, MYKLANDP .Looking at Markov samplers through cusum path plots:a simple diagnostic idea[J].Statistics and Computing,1998,8(3):275-286.
[16]
GEWEKEJ. Evaluating the accuracy of sampling-based approaches to the calculation of posterior moments[J]. Bayesian Statistics, 1992, 4(1): 641-649.
[17]
LAMH F, HUJ, ZHANGF L, et al .Markov chain Monte Carlo-based Bayesian model updating of a sailboat-shaped building using a parallel technique[J]. Engineering Structures, 2019, 193:12-27.
[18]
JONESG L, HARANM, CAFFOB S, et al .Fixed-width output analysis for Markov chain Monte Carlo[J]. Journal of the American Statistical Association, 2006, 101(476): 1537-1547.
[19]
GELMANA, RUBIND B .Inference from iterative simulation using multiple sequences[J].Statistical Science,1992,7:457-472.
ZHOUY, JIAF D, XIS H .Experiment research on multi-model structural identification based on Bayesian theory[J].Journal of Hunan University (Natural Sciences), 2018, 45(5): 36-45.(in Chinese)
FANGS E, CHENS, DONGZ L .Improved approximate Bayesian computation for probabilistic damage identification of structures[J].Journal of Vibration Engineering,2019,32(2):224-233.(in Chinese)
HUZ F, XIAOH X .Applied mathematical statistics and stochastic process[M].Beijing:Publishing House of Electronics Industry, 2021: 102-133.(in Chinese)
[26]
COSMAI A, ASGHARIANM .Principle of detailed balance and convergence assessment of Markov Chain Monte Carlo methods and simulated annealing[J].Statistics,2008: 1-29.
[27]
MEYNS, TWEEDIER L, GLYNNP W .Markov chains and stochastic stability[M].Cambridge:Cambridge University Press,2009:11-15.
[28]
BRAAK C J FTER .A Markov chain Monte Carlo version of the genetic algorithm differential evolution:easy Bayesian computing for real parameter spaces[J].Statistics and Computing, 2006, 16(3): 239-249.
JIANGW, LIUG. Bayesian finite element model updating method based on multi-chain differential evolution[J]. Engineering Mechanics, 2019, 36(6): 101-108.(in Chinese)
[31]
HIZALÇ, AKTAŞE. Probabilistic investigation of error propagation in frequency domain decomposition-based operational modal analysis[J].Structural Control and Health Monitoring,2021,28(8): 1-24.