Bayesian updating of model parameters using adaptive Gaussian process regression and particle filter
Bayesian model updating provides a powerful framework for updating and uncertainty quantification of models by making use of observations, following probability rules in the treatment of uncertainty. Particle filter (PF) and Bayesian Updating with Structural Reliability method (BUS) have been develo...
| Main Authors: | , , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2023
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/164713 |
| _version_ | 1826118744548573184 |
|---|---|
| author | Yoshida, Ikumasa Nakamura, Tomoka Au, Siu-Kui |
| author2 | School of Civil and Environmental Engineering |
| author_facet | School of Civil and Environmental Engineering Yoshida, Ikumasa Nakamura, Tomoka Au, Siu-Kui |
| author_sort | Yoshida, Ikumasa |
| collection | NTU |
| description | Bayesian model updating provides a powerful framework for updating and uncertainty quantification of models by making use of observations, following probability rules in the treatment of uncertainty. Particle filter (PF) and Bayesian Updating with Structural Reliability method (BUS) have been developed by researchers as promising computational tools for this purpose. However, reducing computational cost in the updating process, especially for complex models, remains one of the key challenges. Surrogate model approach achieves this by appropriately replacing, possibly adaptively, the evaluation of the original computationally costly models with approximate ones that are much less costly. This study proposes an efficient method to estimate the posterior probability density function (PDF) of model parameters by using a surrogate model constructed using adaptive Gaussian Process Regression and PF. Of critical importance is the development of ‘learning function’, which finds the location of large values of posterior PDF and avoids those that have been visited. The proposed methodology is illustrated using a single-variable example and compared with PF and BUS. Its application is illustrated through an example of structural dynamics and another one on settlement prediction by soil-water coupled FEM with Cam-clay model. |
| first_indexed | 2024-10-01T04:48:28Z |
| format | Journal Article |
| id | ntu-10356/164713 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T04:48:28Z |
| publishDate | 2023 |
| record_format | dspace |
| spelling | ntu-10356/1647132023-02-13T06:53:21Z Bayesian updating of model parameters using adaptive Gaussian process regression and particle filter Yoshida, Ikumasa Nakamura, Tomoka Au, Siu-Kui School of Civil and Environmental Engineering Engineering::Civil engineering::Geotechnical Gaussian Process Regression Active Learning Reliability Surrogate Model Meta Modeling Bayesian model updating provides a powerful framework for updating and uncertainty quantification of models by making use of observations, following probability rules in the treatment of uncertainty. Particle filter (PF) and Bayesian Updating with Structural Reliability method (BUS) have been developed by researchers as promising computational tools for this purpose. However, reducing computational cost in the updating process, especially for complex models, remains one of the key challenges. Surrogate model approach achieves this by appropriately replacing, possibly adaptively, the evaluation of the original computationally costly models with approximate ones that are much less costly. This study proposes an efficient method to estimate the posterior probability density function (PDF) of model parameters by using a surrogate model constructed using adaptive Gaussian Process Regression and PF. Of critical importance is the development of ‘learning function’, which finds the location of large values of posterior PDF and avoids those that have been visited. The proposed methodology is illustrated using a single-variable example and compared with PF and BUS. Its application is illustrated through an example of structural dynamics and another one on settlement prediction by soil-water coupled FEM with Cam-clay model. Submitted/Accepted version 2023-02-13T00:27:21Z 2023-02-13T00:27:21Z 2023 Journal Article Yoshida, I., Nakamura, T. & Au, S. (2023). Bayesian updating of model parameters using adaptive Gaussian process regression and particle filter. Structural Safety, 102, 102328-. https://dx.doi.org/10.1016/j.strusafe.2023.102328 0167-4730 https://hdl.handle.net/10356/164713 10.1016/j.strusafe.2023.102328 102 102328 en Structural Safety © 2023 Elsevier Ltd. All rights reserved. This paper was published in Structural Safety and is made available with permission of Elsevier. application/pdf |
| spellingShingle | Engineering::Civil engineering::Geotechnical Gaussian Process Regression Active Learning Reliability Surrogate Model Meta Modeling Yoshida, Ikumasa Nakamura, Tomoka Au, Siu-Kui Bayesian updating of model parameters using adaptive Gaussian process regression and particle filter |
| title | Bayesian updating of model parameters using adaptive Gaussian process regression and particle filter |
| title_full | Bayesian updating of model parameters using adaptive Gaussian process regression and particle filter |
| title_fullStr | Bayesian updating of model parameters using adaptive Gaussian process regression and particle filter |
| title_full_unstemmed | Bayesian updating of model parameters using adaptive Gaussian process regression and particle filter |
| title_short | Bayesian updating of model parameters using adaptive Gaussian process regression and particle filter |
| title_sort | bayesian updating of model parameters using adaptive gaussian process regression and particle filter |
| topic | Engineering::Civil engineering::Geotechnical Gaussian Process Regression Active Learning Reliability Surrogate Model Meta Modeling |
| url | https://hdl.handle.net/10356/164713 |
| work_keys_str_mv | AT yoshidaikumasa bayesianupdatingofmodelparametersusingadaptivegaussianprocessregressionandparticlefilter AT nakamuratomoka bayesianupdatingofmodelparametersusingadaptivegaussianprocessregressionandparticlefilter AT ausiukui bayesianupdatingofmodelparametersusingadaptivegaussianprocessregressionandparticlefilter |