Bayesian inference based parametric identification of vortex-excited force using on-site measured vibration data on a long-span bridge. (1st September 2022)

• Record Type: Journal Article
• Authors: Liu, Peng; Chu, Xiaolei; Cui, Wei; Zhao, Lin; Ge, Yaojun
• Is Part Of: Engineering structures. Volume 266(2022)
• Publication Date: 2022-09-01
• Subjects: Vortex-induced vibration -- Parametric identification -- Field monitoring -- Bayesian inference; Structural engineering -- Periodicals; Structural analysis (Engineering) -- Periodicals; Construction, Technique de la -- Périodiques; Génie parasismique -- Périodiques; Pression du vent -- Périodiques; Earthquake engineering; Structural engineering; Wind-pressure; Periodicals; 624.105
• DOI: 10.1016/j.engstruct.2022.114597
• Languages: English
• ISSNs: 0141-0296

Abstract

Abstract: In recent years, vortex-induced vibration (VIV) events have occurred on several long-span suspension bridges around the world. Normally, the VIV of a long-span bridge is investigated in wind tunnel tests or computational fluid dynamics. However, examination of bridge VIV through full-scale field test data has rarely been conducted. Because of the rapid development of high precision sensors and high-frequency data transmission devices, the acquisition of structural modal information utilizing field test data from structural health monitoring systems is emerging as a powerful tool to explore the structural dynamic status and locate potential damage. Therefore, it is possible and necessary to inspect the bridge VEF (vortex-excited force) parameters from full-scale field test data and then to simulate and estimate the structural VIV response based on VEF parameters. Existing VEF parametric identification techniques allow structures (sectional model or full-scale bridges) to be tested under laminar flow in wind tunnel tests with known dynamic properties (inertial frequency and damping ratio), requiring measurement of responsive signals and VEF signals synchronously. However, for the actual field test of the full-scale bridge, the flow field is turbulent, and the structural responsive signal is unavoidably contaminated by measuring noises. Furthermore, it is impractical to synchronously record the aerodynamic force applied on the bridge deck during the field test. In this study, a Bayesian inference approach is introduced for the identification of VEF parameters using field vibration data. Using the fast Fourier transform (FFT) of field vibration data, a frequency domain formulation is proposed focusing on the structural vibration mode excited during VIV events. This method fully considers the influence of random vibration induced by ambient excitation and instrument measurement error on the field vibration data, and only the responsive data are needed without measuring the aerodynamic force information. Highlights: Vortex-excited force parameters were identified using on-site monitoring data. Bayesian FFT inference was employed to derive the posterior PDF when VIV. Relationship of vortex-excited force parameters and wind environment was revealed. Predictive abilities for VIV amplitudes were verified using cross-validation.