中文摘要：

针对复杂边坡特点,提出改进的滑动面搜索变量选取方式,通过左、右出滑点和弧高控制滑动面位置,搜索变量取值范围不再需要经验假定。使用分区间搜索策略,以寻找最危险滑动面和潜在危险滑动面,同时也将搜索区域分割,增大算法搜索到全局最优值的可能性。利用混沌优化算法和复合形法联合搜索算法确定复杂边坡的最小安全系数和滑动面,充分利用混沌搜索和复合形法各自的优点,有效地避免陷入局部最小值,并克服混沌优化耗时较长的不足。通过对复杂边坡实例的计算,结果证明,采用改进搜索变量的基本复合形法寻优成功率比传统搜索变量高出40%～60%。联合搜索法的最小安全系数集中于1.70～1.72,全局搜索能力远远优于基本复合形法。多级边坡工程算例的分析表明,该边坡的4条潜在滑动面,其安全系数与最小安全系数相差不超过7.6%,最小仅相差3.6%,但相应的滑动面位置与最危险滑动面完全不同。

英文摘要：

In view of the characteristic of complex slope,an improved method of selecting search variable is presented.The positions of two exiting points and arc height of the slip surface are taken as search variables,and assumptions aren＇t required.Divided-interval search has been introduced so as to locate the most dangerous slip surfaces and the next critical slip surfaces,of which the obvious advantage is employed to divide the search region,increasing the possibility of finding the global optimum.Chaos optimization method and improved complex method are combined to search the minimum safety factor and critical slip surface.Mixed search algorithm combines the merits of two algorithms,which can effectively avoid local minimum,and overcome time-consuming disadvantage of chaos optimization method.Mixed search algorithm is applied to searching the minimum safety factor of a complex slope;and the result shows that the rate of successful computation of basic complex method with improved search variable is 40%-60% higher than that with the current search variable.The minimum safety factor obtained by mixed search algorithm is ranging from 1.70 to 1.72.So mixed search algorithm is much more likely to find the global optimum of complex slope.And analyses of multi-step slope engineering example show that there are four next critical slip surfaces with small differences in safety factor;the relative difference is 7.6% in maximum,and 3.6% in minimum,but large differences in the location of critical slip surface.