| Continuing health monitoring or regular condition assessment of important bridges is necessary to ensure that the bridges function well. In order to do that, a baseline finite element model of the target bridge is often indispensable not only for complex dynamic response analysis (such as seismic analysis, wind- resistant stability analysis, train-bridge vibration analysis and so on), but also for health monitoring of the bridge. A baseline finite element model must be a full three-dimensional model that can reflect the built-up bridge conditions comprehensively and correctly. In addition, it must be a model properly validated by the bridge field tests. The thesis is aimed at presenting a procedure to establish a baseline finite element model of the Qingzhou cable-stayed bridge over the Ming River, Fuzhou, China that was newly constructed. Its main span length of 605m is the longest in the world among the completed composite-deck cable-stayed bridge. The proposed procedure includes several tasks: initial finite element modeling base on the design drawings, field loading tests and ambient vibration tests, and finite element model calibration with the test results. The calibrated finite element model reflecting the built-up structural conditions of the real undamaged structure will serve as a baseline model to monitor structural health. Some key issues in the modeling of a long span composite cable-stayed bridge are discussed such as cable tension, dead load, large deflection, concrete deck stiffness, concrete deck-steel girder connection, and boundary condition. Some conclusion can drawn from the thesis: 1. The initial equilibrium configuration of the bridge is clarified in the thesis. It is the geometrical position of initial equilibrium due to deal load and initial stay cable tension. It is more reasonable that any succeeding analysis starts from the initial equilibrium configuration.2. The effect of large deflection on the initial equilibrium configuration is studied. It is demonstrated that elastic and small deflection static analysis is adequate to calculate the static initial equilibrium configuration of large span cable-stayed bridges. However, it is always the non-geometrical analysis due to the initial tension of the stay cables. Finite element modal analysis of large span cable-stayed bridges is a "pre-stress modal analysis" based on the deformed equilibrium configuration due to dead load and cable pre-tensions. Starting from the initial equilibrium configuration might have a minimal3. effect on increasing the bridge natural frequencies. However, it is essential to carry out the dynamic response under wind or seismic loadings.4. It is observed that the concrete slab contributes a little to the vertical stiffness, but significantly to the transverse (horizontal) stiffness of the bridge. Therefore, the simplified model of cable-stayed bridge without the bridge deck is inadequate to predict the three-dimensional behavior. Using spring to model the shear connection between concrete slab and steel girders shows that the connection is comparatively strong and can be treated as full restrained.5. The ambient vibration test induced by normal traffic and natural wind is a convenient, fast and cheap way to perform the bridge dynamic test. It is demonstrated that the ambient vibration response measurements only are sufficient enough to identify the most significant modes of large span cable-stayed bridges, in despite of the rather low level of ambient vibration signal captured, the low range (0~1.0 Hz) of natural frequencies of interest, and the relatively dense modes of vibration in that range. 6. The field tests involving static and dynamic test provide a comprehensive investigation on the actual properties of bridges. The analytical analysis through three-dimensional finite element modeling gives a detailed description of the physical and modal characteristics of the bridge. It is right approach to establish the baseline finite element model base on the result from field tests...
|
PDF Full Text Request