基于遗传算法的危险滑面搜索

在边坡稳定性评价中，潜在危险滑面（Potential sliding plane）的确定有着重要的作用和意义．在刚体极限平衡法计算边坡稳定性时，PSP的计算方法有很多种，已经比较成熟．有限单元法相比刚体极限平衡法有较多的优点，随着有限单元法的普及，在应力场（或位移场）已知情况下，怎样确定边坡的安全系数及其PSP已经成为一个问题．目前已有许多方法．本文提出的是采用遗传算法搜索PSP．编制程序GA—PSP实现上述方法，并计算了澳大利亚考题1（a）．通过实例计算，表明提出的方法简单易行．此外，关于三维滑体的危险滑面搜索问题也提出了初步算法．

Application of Genetic Algorithm to Searching Potential Sliding Plane

The definition of potential sliding plane（PSP） is very important to the evaluation of slope stability. In the rigid limited equilibrium method for the stability of slopes, there are many methods to define the PSP, which are very perfect. Compared with the rigid limited equilibrium method, the finite element method （FEM） has some advantages in slope stability analysis. With the popularization of computers, FEM is used to analyse the slope stability by more and more people. So the definition of slope safety and how to find the PSP of a slope are more and more concerned. There are many researches on this question. In this paper, a method which is defining the PSP by genetic algorithm（GA）, is put forward. The program of GA-PSP is developed, and the Australia examination 1（a） is calculated by the program. The results show that the method proposed is feasible. Furthermore, the search method of PSP of a slope considering three dimensions is put forward; but this problem needs advanced researches.