Automated Modal Identification and Tracking of a Cable-Stayed Bridge with p_LSCF

Abstract

The poly-reference least squares complex frequency-domain estimator(p_LSCF) is one of the most popular modal identification methods, which is employed to track the modal parameter of a cable-stayed bridge in this paper. In order to make p_LSCF more successfully applicable to structural modal automatic tracking, two main contributions are presented: on the one hand, p_LSCF is optimized to obtain clearer stabilization diagram, which is beneficial for physical modes extraction, and on the other hand, automatic analysis on stabilisation diagrams is perform based on density-based spatial clustering to realize continuous identification of modal parameters without the need for manual intervention. Finally, the improved p_LSCF is applied to the modal track of the Yangpu bridge located in Danzhou City, Hainan Province, China. The analysis results suggest that the improved p_LSCF offers notable benefits in mode recognition rate and accuracy compared to the original approach. This improved p_LSCF demonstrates the capability to identify closely spaced modes and delivers exceptionally precise estimations, enabling the detection of minute frequency variations. Moreover, it provides meaningful assessments of modal damping ratios.