基于GA-Sarma算法的边坡最不利滑裂面搜索方法

传统方法在选取潜在滑裂面关键点时，未综合考量剪切滑裂点、受拉滑裂点对搜索过程的影响程度，导致搜索过程易陷入局部最优，从而丢失其他同等级搜索数据。针对这一问题，本研究设计了基于GA-Sarma算法的边坡最不利滑裂面搜索方法。该方法通过计算边坡稳定性系数的实际值提取边坡潜在滑裂面，然后根据剪切滑裂点、受拉滑裂点的特性，选取潜在滑裂面关键点。进而，基于GA-Sarma算法建立目标函数和适应度函数，在考虑变异的前提下搜索边坡的最不利滑裂面。实验测试区域中包含3处最不利滑裂面、5处刚刚形成且危险程度偏低的滑裂面、14处潜在滑裂面。测试结果表明：在20轮测试中，本文方法与两组传统方法搜索到的最不利滑裂面个数平均值分别为3处、1.4处以及0.8处，从而验证了GA-Sarma算法强化了边坡最不利滑裂面搜索效果。该方法可为边坡支护、加固及改造等地质工作提供更加可靠的技术支持。

Search method of the most unfavorable slip surface of slope based on GA-Sarma algorithm

When selecting the key points of potential slip surface, the traditional methods do not take into account the influence degree of shear slip point and pull slip point on the search process, which leads to the search process is easy to fall into local optimum and lose other search data of the same grade. To solve this problem, a search method based on GA-Sarma algorithm for the most adverse slope slip surface is designed. In this method, the potential slip surface of slope is extracted by calculating the actual value of slope stability coefficient, and then the key points of potential slip surface are selected according to the characteristics of shear slip point and tensile slip point. Based on this, the objective function and fitness function were established by GA-Sarma algorithm, and the most unfavorable sliding crack surface of the slope was searched under the premise of considering the variation. The experimental test area includes 3 most unfavorable slip surfaces, 5 newly formed slip surfaces with low risk, and 14 potential slip surfaces. The test results show that, in 20 rounds of tests, the average number of the most unfavorable slip surfaces searched by the proposed method and the two groups of traditional methods is 3, 1.4 and 0.8, respectively, which verifies that the GA-Sarma algorithm strengthens the search effect of the most unfavorable slip surface of the slope. This method provides more reliable technical support for national geological work.