Ambient Excitation Based Modal Parameter Identification Of Anqing Railway Yangtze River Bridge

Anqing Railway Yangtze River Bridge is a triple-truss four-line high speed railway cabled-stayed bridge with its main span 580m,which is the key engineering of the intercity railway combining Nanjing to Anqing. It is quite necessary to acquire the actual dynamics of the bridge for the sake that high speed railway bridge has a strict demand for its dynamic characteristics. Structural modal parameter(including frequency, mode shape, damping ratio) identification is a prerequisite for dynamic analysis and can be applied to finite element model(FEM) updating, structural health monitoring and damage detection, etc. Ambient excitation based modal parameter identification is convenient and time-saving as well as no-break disturbance to the operation of the bridge, however, the response measured is low-amplitude, very random and sensitive to noise. The author identifies the modal parameters of the bridge using peak-picking method(PP), frequency domain decomposition method (FDD), poly-reference least-square complex frequency domain method(PolyMAX), and data-driven stochastic subspace identification(SSI-Data). Then the calculated results are compared to the FEM value, followed by doing research to the applicability and reliability as well as analyzing the dynamics. The research of the paper has theoretical significance as well as engineering practical value.The main content and conclusions are as follows:1. The full and detailed spatial FEM of the whole bridge is modeled according to the design drawings, followed by theoretical modal analysis.The calculable results is prepared for ambient vibration experiment and measured data analysis and identification.2. The theories of the four identification methods are introduced systematically. The modal parameter of the bridge is identified by the calculation code programmed on MATLAB platform.3. Verifying the applicability and reliability of the identification methods by comparing the calculated value with the measured value, and make comment to the identification methods. It indicates that all methods can be applied well into the identification of the modal parameters.4. The identified frequencies are more consistent than damping ratios which are scattered. Due to the dense frequencies of long span cable-stayed bridge, it is tough to choose correct modal frequencies properly by using PP and FDD methods and spurious modal parameters may appear by using PolyMAX and SSI methods. In actual operation, the four methods can be used complementarily to acquire more reliable and correct results.