Automatic bridge cable force tracking strategy based on Bayesian operational modal analysis

Stay cables are a crucial component of cable-stayed bridges, where continuous cable force monitoring is significant in integrity health assessment. Additionally, the design life of cables is commonly shorter than that of bridges, leading cable replacement inevitable. Therefore, the cable force tracking method should be feasible in different operational conditions. Vibration-based methods are widely used in force evaluation. However, frequency-domain methods, such as Bayesian operational modal analysis (BAYOMA), require the natural frequency and frequency band of the interested mode for analysis. Due to the short cable replacement period and the significant force changes, it is intractable to extract the frequency and corresponding band in real-time for human manipulation. To address these problems, this article proposes an automatic cable force tracking method considering different operational conditions based on BAYOMA. An automatic peak-picking algorithm is proposed, allowing extraction of multiple resonant peaks from the power spectral density diagram simultaneously. At the same time, the uncertainty law is used to determine the width of frequency band. Since the modal parameters are identified by conducting BAYOMA, the real-time cable force can be derived from the taut string theory and the moving window manner. The proposed method is applied to the in-situ monitoring data of two cable-stayed bridges with different operational scenarios considered. The variation of cable force can be well tracked, demonstrating its reliability and efficiency for real-time monitoring system.