A novel adaptive importance sampling algorithm for Bayesian model updating

•A novel adaptive importance sampling-based Bayesian model updating algorithm is proposed.•A stopping criteria called N-ESS is adopted to terminate the adaptive procedure to reduce computational efforts.•The proposed algorithm successfully constructs the importance sampling density that resembles th...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Structural safety Ročník 97; s. 102230
Hlavní autoři: Xiao, Xiong, Li, Quanwang, Wang, Zeyu
Médium: Journal Article
Jazyk:angličtina
Vydáno: Amsterdam Elsevier Ltd 01.07.2022
Elsevier BV
Témata:
Adaptive algorithms
Adaptive importance sampling
Adaptive sampling
Algorithms
Bayesian analysis
Bayesian updating
Chloride-induced concrete corrosion
Computer applications
Corrosion tests
Cross entropy
Density
Entropy (Information theory)
Gaussian mixture
Importance sampling
Markov chains
Mathematical models
Model updating
Population Monte Carlo
Sampling
System identification
Bayesian updating
Cross entropy
Population Monte Carlo
Chloride-induced concrete corrosion
Gaussian mixture
Adaptive importance sampling
Model updating
ISSN:0167-4730, 1879-3355
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:•A novel adaptive importance sampling-based Bayesian model updating algorithm is proposed.•A stopping criteria called N-ESS is adopted to terminate the adaptive procedure to reduce computational efforts.•The proposed algorithm successfully constructs the importance sampling density that resembles the sophisticated target posterior density.•A chloride and carbonation-induced reinforcement concrete corrosion problem is investigated. Bayesian model updating has computational capability of reducing uncertainties in engineering models and yielding valuable inferences for model predictions from observed data. Importance sampling (IS) and Markov chain Monte Carlo (MCMC) are the two main techniques among the existing simulation-based methods for Bayesian model updating. Motivated by the fact IS outperforms MCMC in terms of computational efficiency once the proposal importance sampling density (ISD) is appropriately chosen, this paper proposes an innovative adaptive importance sampling (AIS) algorithm using Gaussian mixture to overcome the limitations of conventional IS-based methods. Integrating of the concepts of population Monte Carlo (PMC) and cross entropy (CE) contributes to the major novelty of the proposed algorithm. Consequently, the proposed AIS algorithm can successfully construct the ISD that resembles the sophisticated target posterior density. Moreover, a metric quantifying sampling effectiveness called normalized effective sample size (N-ESS) is also adopted to measure the similarity between the ISD and the target density. One benchmark example of system identification and one case study of chloride-induced concrete corrosion are investigated to demonstrate the accuracy and robustness of the proposed algorithm.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0167-4730
1879-3355
DOI:10.1016/j.strusafe.2022.102230