Abstract:Modal parameter identification via ambient vibration is popular but faces challenges from uncertainties due to unknown inputs and low signal‐to‐noise ratio. Bayesian methods are gaining increasing attention for operational modal identification due to their ability to quantify uncertainties. However, improvements in computational efficiency are needed, particularly when addressing numerous modes and degrees of freedom. To address this challenge, this study proposes an innovative approach, termed the "Bayesian spectral decomposition" method (BSD), employing the decompose‐and‐conquer strategy. This novel method, operating within the frequency domain, identifies each mode individually by exploiting their inherent separated modal characteristics. For each mode, the response spectrum matrix undergoes an eigenvalue decomposition, yielding crucial eigenvalues (incorporating frequency and damping information) and eigenvectors (containing mode shape information). Subsequently, statistical properties of the eigenvalues and eigenvectors are utilized to establish likelihood functions for Bayesian parameter identification. By combining prior information, the posterior probability distribution functions of modal parameters are derived. The optimal solution is then obtained by resolving the maximum posterior probability distribution function problem. To further quantify the uncertainty of modal parameters, Gaussian distributions are employed to approximate the posterior probability distribution functions. The adoption of the decomposition approach circumvents the joint identification of all modal parameters, substantially reducing the parameter dimensions for optimization. Consequently, this strategy leads to decreased computational complexity and significantly improved computational stability. The effectiveness of the BSD is confirmed through simulated data generated from an 8‐story shear building as well as measured data collected from both an experimental shear frame and the Canton Tower. The results demonstrate that the proposed method achieves high accuracy in identifying modal parameters, greatly improves computational efficiency, and effectively quantifies the uncertainties in modal parameters.

A Novel Vibration-Based Two-Stage Bayesian System Identification Method *

Bayesian two-stage structural identification with equivalent formulation and EM algorithm

Two-stage Method for System Identification Based on Asymptotic Theory

Hierarchical Bayesian Operational Modal Analysis: Theory and Computations

Structural modal identification and MCMC-based model updating by a Bayesian approach

A Markov chain Monte Carlo-based Bayesian framework for system identification and uncertainty estimation of full-scale structures

Uncertainty quantification in fast Bayesian modal identification using forced vibration data considering the ambient effect

Bayesian model updating of a 20-story office building utilizing operational modal analysis results

Bayesian Spectral Decomposition for Efficient Modal Identification Using Ambient Vibration

Bayesian FFT Modal Identification for Multi-setup Experimental Modal Analysis

Assessing uncertainty in fast Bayesian modal identification based on seismic structural responses

Optimal Parameter Estimation for Uncertain Structural Systems under Unknown Random Excitations

Efficient model updating and health monitoring methodology using incomplete modal data without mode matching

Engineering Performance of Two Analytical Methodologies for Estimating Modal Parameter Uncertainty for Structures

A Model Identification Method of Vibrating Structures from Incomplete Modal Information

Dynamical System Identification, Model Selection and Model Uncertainty Quantification by Bayesian Inference

Bayesian inference based parametric identification of vortex-excited force using on-site measured vibration data on a long-span bridge

Data-Driven Bayesian Inference for Stochastic Model Identification of Nonlinear Aeroelastic Systems

Model reduction-based Bayesian updating of non-classically damped systems using modal data from multiple setups

Probabilistic model updating via variational Bayesian inference and adaptive Gaussian process modeling

Bayesian System Identification based on Hierarchical Sparse Bayesian Learning and Gibbs Sampling with Application to Structural Damage Assessment