An automated algorithm for identifying the fundamental frequency of bridge cable vibrations

Le résumé fourni par la source

Abstract In practical bridge engineering, cable tension reflects the safety performance of the entire bridge structure. Vibration frequency-based cable tension measurement methods have been proven to be practical and effective. In recent years, research on non-contact measurement of cable vibration frequencies has increased significantly. However, non-contact measurements are susceptible to environmental interference, and the stability of signal acquisition equipment can markedly affect the results, leading to impure spectrum and making it difficult to identify the fundamental frequency (FF) used for cable force calculation. To address this problem, an automatic Fundamental Frequency identification in Spectrum (FF-FAS) algorithm is proposed. First, to obtain potential vibration FF from the spectrum, a feature peak screening algorithm is introduced to automatically identify peak points and assign weights based on the smoothness of the peaks and their equivalent width. Subsequently, to accurately identify the FF from the peaks, a new FF determination mechanism is designed, which automatically extracts the frequency sequence based on string vibration theory and selects the FF by synthesizing the contribution of each peak. The FF-FAS algorithm addresses the limitations of traditional FF identification methods, which rely on peak detection and are highly susceptible to noise. This algorithm has been evaluated through both experiments on a cable-stayed cable in a public square and field tests on operational bridges, demonstrating significant improvements in FF identification compared to existing methods.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.