场地土层剪切波速测试不可避免会存在误差，误差达30%左右。目前，在对剪切波速误差对土层地震反应计算影响的研究中，设计波速误差或者设计土层模型的方式偏于理想化。为更接近现实情况，本文基于实测土层模型，用蒙特卡洛方法研究剪切波速误差对土层地震反应计算结果的影响。本文搜集并建立了均匀分布在山东地区的132个场地计算模型，涵盖了山东境内分布广泛的Ⅱ类场地和Ⅲ类场地。根据蒙特卡洛随机模拟原理，对每一场地计算模型，在实测剪切波速值30%的偏差范围内随机生成100组剪切波速值，分别进行土层地震反应计算，分析反应谱值、反应谱平台值和特征周期的变化。研究表明，30%以内的随机波速误差会导致某些周期处的反应谱的变化达30%以上，其对反应谱平台值最大的影响大多在10%～50%之间，对Ⅱ类场地特征周期的最大影响大多在0～50%之间，对Ⅲ类场地特征周期的最大影响大多在20%～90%之间。

The error of soil shear wave velocity is inevitable,and it can be up to 30%. At present,the design of shear wave velocity error or the design of soil layer model is somewhat idealized on the study of the influence of shear wave velocity error on soil layer seismic response calculation results. In order to present the real soil condition more precisely,we built the soil layer model based on the measured shear wave velocity,and used the Monte Carlo method to investigate the influence of the shear wave velocity errors on seismic response spectra. In this paper,soil layer models of 132 sites distributed uniformly in Shandong Province were built,including site type Ⅱ and site type Ⅲ. Based on the Monte Carlo method,100 groups of shear wave velocity values are randomly generated in the deviation range of 30% of the measured shear wave velocity for each site. Then,the seismic response calculation of soil layers models including these shear wave velocity values were carried out,and the changes of response spectra and response spectra platform values and characteristic periods were analyzed. The result shows that,the maximum influence from the random errors less than 30% of soil shear wave velocity on response spectra in some periods is generally more than 30%,and the maximal influence on response spectra platform values is generally 10%~50%,the maximal influence on characteristic periods of site type Ⅱ is generally 0%~50%,and the maximal influence on characteristic periods of site type Ⅲ is generally 20%~90%.

P315

山东省地震局科研基金项目（JJ1701Y）、北京市自然科学基金项目（8182056）、中央级公益性科研院所基本科研业务专项资助项目（ZDJ2017-28）共同资助