基于子集模拟的边坡可靠度分析方法研究

边坡稳定受到诸多不确定性因素的影响,比如土体性质的空间变异性以及地层情况的不确定性。这些不确定性因素的影响可以通过蒙特卡洛模拟(Monte Carlo Simulation, MCS)进行定量地分析。MCS概念简单并具有广泛的适用性。但是,在小概率失效区域内,MCS计算效率很低,需要庞大的随机样本量来保证一定的计算精度。本文提出了一种实用的边坡可靠度分析方法。通过采用一种高级的MCS方法(Subset Simulation, 子集模拟)来提高小概率区域内的计算效率以及计算精度,并以EXCEL的表单环境为平台,联合使用Visual Basic Application(VBA)编写计算程序。在该程序中,子集模拟、边坡稳定的确定性分析和不确定性分析分别由三个相对独立的计算模块实现。最后,本文以James Bay 土坝为例,简明地说明了所提出方法的有效性,并探索了临界滑动面的不确定性对边坡可靠度分析的影响。

Probabilistic Slope Stability Analysis Using Subset Simulation

Various uncertainties exist in slope engineering, such as inherent spatial variability of soils and uncertainties in underground stratigraphy. These uncertainties can be taken into account in probabilistic slope stability analysis through Monte Carlo Simulation (MCS). MCS has been widely used in probabilistic slope stability analysis because of its robustness and conceptual simplicity. It, however, suffers from a lack of efficiency and resolution at small probability levels. This paper develops a MCS-based probabilistic slope stability analysis method that implements an advanced MCS called “Subset Simulation” in a spreadsheet environment, Microsoft EXCEL, for improving efficiency and resolution at relatively small probability levels. The reliability analysis is deliberately decoupled from the deterministic slope stability analysis. For illustration, the proposed method is applied to study a design scenario of the James Bay Dyke. Finally, the effect of critical slip surface uncertainty is explored.