Seismic Response Of A Three-Tower Suspension Bridge With Dynamic Soil-Structure Interaction

With the rapid development of China’s land transportation in recent years, we need a lot of long-span bridges to cross wide straits, the rivers or gorges. The three-tower suspension bridges become more and more favored by bridge engineers because of its unique advantages of spanning-capacity and economy. We were also alerted by the frequent strong earthquakes in recent years that seismic design of long-span suspension bridges should be paid sufficient attention. In this thesis, the effect of dynamic soil-structure interaction (SSI) on the seismic response of three-tower suspension bridge is studied by using the finite element method. The following aspects are addressed:(1) Some popular models of dynamic soil-structure interaction are introduced and basic steps for applying NIST six-spring model to simulate soil-structure interaction are briefly introduced;(2) The finite element model of a three-tower suspension bridge with soil-structure interaction considered is developed by using the six-spring SSI model. The modal analyses with and without SSI considered are compared with each other. The results shows that dynamic SSI will affect the overall stiffness of the bridge;(3) By using uniform support excitation method to calculate the seismic response of the suspension bridge model with and without SSI considered, the influence of soil-structure interaction on the seismic response of the suspension bridge is studied. The results shows that after taking the SSI into consideration, the internal forces of members of the suspension bridges become larger, compared to the results without SSI;(4) By using multi-support excitation method, seismic responses of the suspension bridge under traveling, earthquake ground motions are studied with SSI considered The results shows that the traveling wave effect on the response of suspension bridges changes with different wave velocity and, in the practice, the actual velocity based on site geological conditions should be input to the model to get appropriate design values.