Identification of Closely Spaced Modes of a Long-Span Suspension Bridge Based on Bayesian Inference

Closely spaced modes commonly observed in long-span suspension bridges can greatly increase the difficulty of identifying and tracking modal parameters. Most existing studies generally focus on identifying the closely spaced modes and quantifying the uncertainties based on numerical and experimental models. Further research focusing on full-scale long-span bridges is still required. A case study on identifying the closely spaced modes of the Qixiashan Yangtze River Bridge, a long-span suspension bridge with a main span of 1 418 m, is conducted in this paper. The effectiveness of the generalized fast Bayesian fast Fourier transform (GFBFFT) method is verified by both the simulated and monitoring data. The results show that a larger coefficient of variation (COV) and higher uncertainty is typically contained in the closely spaced modes than the separated modes. Compared with the FDD and SSI methods, the GFBFFT method guarantees higher identification accuracy of modal parameters and can serve as a reliable tool to identify the closely spaced modes.

Automated summary

In this paper, a case study on the Qixiashan Yangtze River Bridge is conducted to verify the effectiveness of the generalized fast Bayesian fast Fourier transform (GFBFFT) method in identifying closely spaced modes. The results show that closely spaced modes typically have larger coefficients of variation and higher uncertainty compared to separated modes. Compared with the FDD and SSI methods, the GFBFFT method ensures a higher identification accuracy of modal parameters and can be relied upon to identify closely spaced modes.