Finite element model updating of building structures under seismic excitation: A parallelized latent space-based Bayesian framework

2026-04-24

AI Summary

This study addresses the bias in seismic response predictions of finite element models caused by parameter uncertainties. To this end, the authors propose a latent-space-based Bayesian updating framework that leverages a multimodal variational autoencoder to compress high-dimensional structural response data and employs sequential Monte Carlo sampling for efficient posterior inference. By integrating GPU acceleration and amortized inference, the method enables robust and rapid estimation of complex, multimodal posterior distributions under sparse observational data—without requiring repeated finite element simulations. Numerical simulations and shake-table experiments demonstrate that the proposed approach significantly enhances the accuracy of structural parameter identification, improves uncertainty quantification, and achieves substantial computational efficiency gains.

Technology Category

• Reasoning under Uncertainty: Relational Probabilistic Models
• Machine Learning: Calibration & Uncertainty Quantification
• Intelligent Robots: State Estimation

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Enhancing seismic fragility and risk assessment of nuclear power plants relies on accurate prediction of reactor building responses to seismic hazards, which can be further improved through dynamic analysis of high-fidelity finite element (FE) models. However, FE models often exhibit non-negligible discrepancies from actual structures due to various sources of uncertainty, necessitating FE model updating with rigorous quantification of associated uncertainties. This paper presents a GPU-accelerated latent space--based Bayesian framework for FE model updating of building structures. In the proposed framework, high-dimensional structural response data (e.g., time histories or frequency response functions) are projected into a low-dimensional latent space using a multimodal variational autoencoder (MVAE), thereby enabling efficient and tractable likelihood evaluation without explicit modeling in the original observation space. Once trained, the surrogate enables amortized inference, allowing posterior sampling to be performed without additional simulator evaluations. We specifically employ a sequential Monte Carlo (SMC) sampler, whose population-based formulation allows parallel evaluation of the approximate likelihood on GPUs, resulting in computational efficiency and robustness against multimodal and complex posterior distributions. The proposed framework is validated through both numerical benchmarking and experimental data from a shaking table test of a reinforced concrete building structure. The results demonstrate that the method accurately estimates structural parameters with well-quantified uncertainties, while achieving fast and efficient inference through GPU-based parallelization, and enabling robust inference even in the presence of sparse observations that induce multimodal and highly complex posterior distributions.

Problem

• finite element model updating
• seismic excitation
• uncertainty quantification
• structural response
• Bayesian inference

Innovation

• latent space
• Bayesian model updating
• variational autoencoder
• GPU acceleration
• sequential Monte Carlo