挡土墙库仑土压力的遗传算法求解分析

Analytical solution of Coulomb earth pressure on retaining wall by genetic algorithm

在对破裂面上滑动土体静力极限平衡分析的基础上,建立了基于优化方法求解无黏性土、黏性土库仑土压力的自变量取值区间和目标函数模型,并采用遗传进化方法进行了实例求解分析。研究结果表明,遗传算法在计算挡土墙库仑主动土压力的过程中,收敛速度快、用时短,并具有较高的计算精度。算例1中5组无黏性土挡土墙的主动土压力的计算结果与经典库仑解析解非常接近,平均误差为1.748%,平均进化代数为15代。算例2中8组黏性土挡土墙的主动土压力计算结果与文献的解答非常吻合,平均误差仅为0.017%,平均进化代数为17.125代。遗传算法具有良好的适应性和强大的搜索性能,非常适合求解岩土工程优化问题。

Based on the static limit equilibrium analyses of sliding soil mass on rupture, the range of variable and the model of objective function are built to calculate the Coulomb earth pressure of cohesionless soil and cohesive soil by optimization method, and the genetic evolutionary method is used to carry out an example analytical solution. The results indicate that the genetic algorithm increases the convergence speed, shortens time and has high calculation precision in the process of calculating Coulomb active earth pressure on retaining wall. The numerical results of cohesionless soil active earth pressure on retaining wall about 5 groups from example 1 are very closer to the classical Coulomb solutions; the average error is 1.748 % and the average number of generation is 15. The numerical results of cohesive soil active earth pressure on retaining wall about 8 groups from example 2 also match well with the documents＇ solutions; the average error is only 0.017 %; and the average number of generation is 17.125. The genetic algorithm is very suitable for solving optimization problem of geotechnical engineering because of its good adaptability and strong search performance.