ANALYSIS OF NOISE ROBUSTNESS IN BAYESIAN MODAL IDENTIFICATION
The aim of this study is to investigate the noise robustness of Bayesian methods in modal identification us-ing time-domain acceleration data.The impact of noise on modal parameter identification of a 6-degree-of-freedom mass-spring model is analyzed.The experimental results demonstrate that Bayesian methods accurately identify modal parameters within an acceptable range for engineering applications when the noise ratio does not exceed 20%.This finding verifies the ability of Bayesian methods to effectively utilize prior information and probability models to extract valuable structural features from noisy data.The demonstrated noise robustness of Bayesian methods in struc-tural dynamic testing has significant implications and promotes the widespread application of Bayesian methods in practical engineering scenarios.
万方数据
维普
