改进和声搜索算法及其在土坡稳定分析中的应用

Improved harmonic search algorithm and its application to soil slope stability analysis

土坡非圆临界滑动面的寻求是土坡稳定分析的重要步骤,它是一个多极值的优化问题。利用改进的和声搜索算法和修复策略来寻求复杂土坡的临界滑动面及其对应的安全系数,并对改进和声搜索算法与基本和声搜索算法的结果进行了比较,发现改进和声搜索算法比基本和声搜索算法能搜索到更危险的滑动面,原因在于,改进和声搜索算法在每次迭代步中产生多个新解,而基本和声搜索算法中每次迭代产生一个新解。此外,修复策略比惩罚策略更能有效地搜索解空间。结果表明,改进的和声搜索算法和修复策略可以用来进行土坡稳定分析。

Locating the critical slip surface of soil slopes is an important step for soil slope stability analysis. It is usually an optimization problem with multiple local minima. This study utilized a modified harmonic search algorithm and rectifying procedure to determine the critical slope surface with the minimum factor of safety of soil slopes. Numerical results so obtained were compared with those using the simple harmonic algorithm. It can be concluded that the factor of safety obtained by using the modified harmonic search algorithm is lower than that by the simple harmonic search algorithm. This is because that several solutions can be generated with the modified harmonic search algorithm, instead of just one solution by the simple harmonic search algorithm. In addition, the rectifying procedure is more useful to the slope stability analysis than the penalty procedure.