Fast Bayesian modal identification based on seismic response considering the ambient effect

Modal identification is the initial step to know the fundamental dynamic characteristics of a target structure, which mainly includes the natural frequency, damping ratio, and mode shape. These modal parameters can be used to assess the discrepancy between the design values and the actual values. Based on these, model updating and damage detection can be carried out in conjunction with the finite element model (FEM). Methods of modal identification will differ depending on the type of input excitation varies. In this work, a novel method for modal identification using seismic data considering ambient effects is proposed. This method follows the Bayesian framework with the merit of quantifying uncertainty in the identified modal parameters. To make the proposed method more practical for use in field structures and improve the computational efficiency, the objective function is reformulated to reduce significantly the number of modal parameters to be optimized during modal identification. The proposed method is verified by a numerical example and then applied in a field structure. The results identified by the proposed method are also compared with the results identified using ambient vibration data exclusively. Furthermore, the dynamic characteristics of the field structure were also investigated under ambient excitation and seismic excitation. © 2024