Bayesian Structural Model Updating with Multimodal Variational Autoencoder

TL;DR

This paper introduces a Bayesian structural model updating method using multimodal variational autoencoders to efficiently handle high-dimensional, correlated observations with limited data, improving computational efficiency and accuracy.

Contribution

It presents a novel framework combining Bayesian updating with multimodal VAE encoders, enabling effective likelihood approximation in small-sample, high-dimensional scenarios.

Findings

Benchmarking shows improved computational efficiency.

Maintains accuracy comparable to original VAE.

Applicable to various dynamic analysis models.

Abstract

A novel framework for Bayesian structural model updating is presented in this study. The proposed method utilizes the surrogate unimodal encoders of a multimodal variational autoencoder (VAE). The method facilitates an approximation of the likelihood when dealing with a small number of observations. It is particularly suitable for high-dimensional correlated simultaneous observations applicable to various dynamic analysis models. The proposed approach was benchmarked using a numerical model of a single-story frame building with acceleration and dynamic strain measurements. Additionally, an example involving a Bayesian update of nonlinear model parameters for a three-degree-of-freedom lumped mass model demonstrates computational efficiency when compared to using the original VAE, while maintaining adequate accuracy for practical applications.