提出了一种基于数值应力场下的边坡临界滑动面和临界坡高的分析方法。该分析方法首先利用有限元得到边坡内部较为精确的应力场,再通过在边坡内部模拟类似于链式反应过程中粒子产生和运动规律,在此基础上完成临界滑动面的搜索。主要结论如下:(1)提出了一种基于数值应力场的边坡临界滑动面的搜索方法,该方法可在不对边坡滑动面形状和位置进行干预的情况下完成潜在破坏路径的搜索。(2)通过对算法中区域与边界、激发源、产物扩散路径、路径权值4方面要素进行合理控制,可使算法在具有较高精度的同时又能保证较好的分析效率。(3)算法能较为方便地考虑真实应力条件下地层强度各向异性对边坡稳定性的影响,并能对临界坡高进行很好的分析。(4)通过具体算例对本文算法的计算结果进行验证,将所得结果与PSO和Monte Carlo两种方法得到结果进行对比,对本文方法的可靠性进行了验证,取得了较好的效果。
In this paper, an analysis algorithm which can be used to determine critical sliding surface and critical height of slope, based on numerical stress field, is presented. Upon the algorithm, firstly, the stress field of slope can be obtained by using the finite element analysis method to get more accuracy. Secondly, through simulation of slope, the laws of generation and movement of particles similar to the process of chain reaction are studied. Finally, the critical sliding surface can be determined. Therefore four conclusions can be drawn: (1) A method for determining the critical sliding surface of slope on account of numerical stress field is proposed, which can find potential failure path while no disturbance of the shape and location of slope sliding surface. (2) By controlling such four factors as range and boundary, excitation source, diffusion path and path metrics, the algorithm can get higher accuracy and simultaneously higher efficiency. (3) The algorithm takes the effect of strength anisotropy on stability of slope into consideration. And it is valid to compute the critical slope height. (4) A detailed calculation example has verified the proposed algorithm. Compared to calculation results obtained by the two methods of PSO and Monte Carlo, the reliability of the algorithm could also be proved.