基于蒙特卡罗法与多滑动面的拟动力法边坡可靠度分析

Slope Reliability Analysis Using Pseudo-dynamic Method with Monte Carlo Simulation and Multiple Slip Surfaces

摘要:

基于拟动力瑞典圆弧法寻求最危险初始相位下的安全系数以评估给定滑动面的稳定性,将大量可行滑动面引入极限状态函数,最终结合蒙特卡罗法提出了基于拟动力瑞典圆弧法的地震边坡可靠度分析方法。将所提方法应用于某均质土坡稳定性分析中,详细研究了横波波速、水平地震力系数以及滑动面位置对最危险初始相位以及安全系数的影响。研究结果发现:随着横波波速的增加,最危险初始相位t_d对应的安全系数逐渐减小,并最终趋于拟静力方法的结果。随着水平地震力系数的增加,t_d保持不变,但其对应的安全系数逐渐减小;在其他条件均相同的前提下,t_d亦随滑动面位置变化而改变。对于均质边坡,基于临界滑动面S_c的可靠度分析方法(方法一)得到的t_d概率密度函数以及失效概率与基于大量可行滑动面分析方法(方法二)的基本一致,但方法一不能合理考量因滑动面位置发生变化而导致的边坡失效后果,得出的滑动体积会偏小,推荐使用方法二进行地震边坡可靠度分析。

Abstract:

The pseudo-dynamic Swedish circle method for seeking the factor of safety under critical initial phase is proposed to evaluate the stability for a given slip surface. A significant number of feasible slip surfaces are incorporated into the limit state function of slope reliability, and the Monte Carlo simulation method is combined to propose the seismic slope reliability analysis method based on the pseudo-dynamic Swedish circle method. The proposed methodology is applied to the stability analysis of a homogeneous slope, with a detailed investigation of the effects of shear wave velocity, horizontal seismic coefficient, and slip surface location on the critical initial phase (denoted by t_d) and factor of safety. It is found that with an increase in shear wave velocity, the factor of safety corresponding to the critical initial phase t_d gradually decreases and eventually converges to the result from the pseudo-static method. With an increase in the horizontal seismic coefficient, t_d remains constant, but the corresponding factor of safety tends to decrease. Moreover, under the same conditions, t_d also differs with the variation in the slip surface location. For homogeneous slopes, the t_d probability density function and failure probability obtained from the reliability analysis method based on the critical slip surface S_c (method 1) are essentially consistent with those derived from the analysis method based on a large number of feasible slip surfaces (method 2). However, since Method 1 fails to properly quantify the consequences of slope failure resulting from the variation in the slip surface location, it has the potential of underestimating the sliding volume. Method 2 is recommended for seismic slope reliability analysis.