基于优化选点的土层剪切波速随机性对地铁车站结构抗震性能的影响研究

Subway station structure is the important node of urban lifeline engineering. The seismic response of underground structure is mainly controlled by the soil deformation, while the distribution of natural soil layer has strong uncertainty. It is very necessary to analyze the influence of the randomness of soil layer parameters on the seismic performance of subway station structure. However, using the traditional Monte Carlo simulation method to solve the stochastic dynamics problem of nonlinear soil subway station structural system will cause much computational cost. Therefore, this paper establishes the finite element model to simulate the soil-structure interaction, and constructs the random ground motion variables and random soil shear wave velocity variables. An optimal point selection strategy is adopted to decrease the computational cost of nonlinear dynamic time history analysis, and the influence of the randomness of soil shear wave velocity on the seismic performance of subway station structure is discussed. The analysis results show that compared with the traditional sampling method, Sobol sequence sampling method effectively can reduce the calculation cost of nonlinear dynamic time history analysis with enough calculation accuracy. When the randomness of soil shear wave velocity is not considered, the distribution range of inter-storey drift ratio response is small and concentrated. The potential seismic damage risk will be ignored, while the seismic reliability of subway station structure will be overestimated. In the future seismic analysis and design of subway station structure, it is necessary to take the randomness of soil shear wave velocity into account.