On-line Bayesian Inference for Structural Health Monitoring under Model Uncertainty using Sequential Ensemble Monte Carlo



Lye, Adolphus ORCID: 0000-0002-1803-8344, Cicirello, Alice and Patelli, Edoardo ORCID: 0000-0002-5007-7247
(2022) On-line Bayesian Inference for Structural Health Monitoring under Model Uncertainty using Sequential Ensemble Monte Carlo. In: 13th International Conference on Structural Safety and Reliability, 2022-9-13 - 2022-9-17, Shanghai, China.

[img] Text
ICOSSAR 2021 Conference Paper [Adolphus Lye].pdf - Author Accepted Manuscript
Access to this file is restricted: awaiting official publication and publisher embargo.

Download (1MB)

Abstract

This paper presents an application of the Sequential Ensemble Monte Carlo (SEMC) sampler to perform on-line Bayesian inference of latent parameters. The SEMC implements the Affine-invariant Ensemble sampler algorithm in place of the traditional Metropolis-Hastings algorithm. The objective of this research is to illustrate the strength of the SEMC, when applied to the analysis of a SDoF Spring-Mass-Damper system to identify the time-varying stiffness and damping coefficient parameters subjected to a random process degradation under model uncertainty. The results not only highlight the ability of the SEMC sampler to identify time-varying parameters at a lower computational cost, but also its robustness in moderating the sample acceptance rates via an adaptive tuning algorithm.

Item Type: Conference or Workshop Item (Unspecified)
Uncontrolled Keywords: Affine-Invariance, Bayesian Inference, Ensemble Sampler, Model Selection, Model Uncertainty, Sequential Monte Carlo, Uncertainty Quantification
Divisions: Faculty of Science and Engineering > School of Engineering
Depositing User: Symplectic Admin
Date Deposited: 12 Apr 2022 12:18
Last Modified: 27 Feb 2023 20:05
URI: https://livrepository.liverpool.ac.uk/id/eprint/3152784