Monitoring, intelligent perception and early warning of vortex-induced vibration of suspension bridge

Abstract

Vortex-induced vibration(VIV) is a serious problem of suspension bridges and other long-span bridges during the service period. It can cause the excessive amplitude of the structure under low wind speed, which not only affects the driving comfortableness and safety but also makes the structure face the risk of fatigue failure. The previous research on the identification and evaluation of bridge VIV events during the service period is based on the offline batch processing and analysis of monitoring data, which can not realize real-time perception, calculation, and early warning online. In this paper, according to the vibration characteristics of single-mode sinusoidal-like vibration of engineering structure during VIV, an intelligent monitoring and early warning method for VIV of suspension bridge based on recursive Hilbert transform is proposed. Firstly, the real-time acceleration integral algorithm is used to realize the real-time calculation from the acceleration monitoring data to the dynamic displacement of the stiffening beam, and then the recursive Hilbert transform is used to obtain the real-time analytical signal of the structural displacement during VIV; based on its single-mode near-circular trajectory characteristic, the VIV index and the real-time analysis method are proposed to characterize the development trend of VIV events. This online extraction algorithm can realize the first time warning and the whole process tracking and perception of VIV events. Furthermore, this article also provides a real-time online identification method of key motion parameters such as the instantaneous frequency, phase and amplitude of the structure during VIV, which lays a foundation for real-time monitoring of the whole process of VIV and further evaluation and management decision-making. The accuracy, reliability and engineering feasibility of the proposed method are verified by numerical simulation and VIV monitoring data of a real bridge.