复杂环境下地铁深基坑变形可靠度分析

Deformation reliability analysis of subway deep foundation pit under complex environment

为计算复杂环境下地铁深基坑变形可靠度,提出了非侵入式随机有限差分法。通过构建随机有限差分接口,实现确定性分析与随机分析相融合,引入Hermite随机多项式,确定3个输出响应量与12个输入变量的显性函数表达式,计算基坑变形可靠度。结果表明:非侵入式随机有限差分法计算精度与蒙特卡罗法相近,但计算效率远高于蒙特卡罗法,通过计算结果验证了非侵入式随机有限差分法的可行性;由输出响应量与输入变量相关性强度可知,砂土层密度对基坑变形可靠度影响较大;将砂土密度变异系数Cov(ρ3)取为初始值的2倍,验证砂土密度的变异性对输出响应量的影响,计算得到的地表沉降、围护结构水平位移失效概率明显提高;为了提高深基坑的变形可靠度,在确定土层参数变异系数前应采集较多的试样,保证土层密度更接近于真实值。

In order to calculate the deformation reliability of subway deep foundation pit under complex environment,a non-intrusive stochastic finite difference method was proposed.By constructing a stochastic finite difference interface,the deterministic analysis and stochastic analysis were integrated.The Hermite random polynomial was introduced to determine the explicit function expressions of three output responses and 12 input variables,and the deformation reliability of foundation pit was calculated.The results show that the calculation accuracy of the non-intrusive stochastic finite difference method is similar to that of the Monte Carlo method,but the calculation efficiency is much higher than that of the Monte Carlo method.The feasibility of the non-intrusive stochastic finite difference method is verified by the calculation results.According to the correlation strength between the output response and the input variable,the density of sand layer has a great influence on the reliability of foundation pit deformation.The coefficient of variation of sand density Cov(ρ3)is set to 2 times of the initial value,the influence of the variability of sand density on the output response is verified,and the failure probability of surface settlement and horizontal displacement of retaining structure is obviously improved.In order to improve the deformation reliability of deep foundation pit,more samples should be collected before determining the coefficient of variation of soil parameters to ensure that the soil density is closer to the real value.