基于加速遗传算法的震区边坡稳定性可靠度分析

Reliability Analysis on Slope Stability Based on Accelerating Genetic Algorithm in Seismic Area

针对传统极限平衡理论的传统边坡安全度评价方法——中值安全系数法难以体现边坡工程的不确定性,本文基于随机概率分析的边坡稳定可靠度分析方法,充分考虑岩土体参数的均值及其变异性,采用的加速遗传算法计算边坡可靠度指标,并同时搜索最危险滑动面的算法,避免了遗传算法的嵌套,减少了计算量,从而极大的提高运算效率。结合湖南省对口支援四川理县灾后重建项目——理县至小金公路(理县段)k4+040路堑边坡开展可靠度分析,计算结果表明,该边坡在普通工况条件下边坡安全系数为1.305,满足规范要求,其可靠度指标Pf=10.21%,稳定性较好;但对于该地可能出现的强震及降雨(工况3),边坡安全系数分别为1.060,边坡基本处于极限平衡状态其可靠度指标Pf=5.87%,稳定性较差,在此基础上提出的加固优化方案,为今后类似工程提供参考。

The factor of safety method,which is the traditional evaluating method in slope safety degree based on the limit equilibrium theory,is difficult to reflect the uncertainty slope engineering.However the slope stability reliability analysis method based on probability analysis,can take full account of the mean and variance of rock and soil parameters.The slope reliability index is calculated by accelerating genetic algorithm,and the most dangerous slip surface is searched by it which can avoid the nesting of genetic algorithm,reduce the amount of computation and greatly improve the efficiency in this paper.Combined with the real project of cutting slope in k4＋040 in Lixian to Xiaojin Road（Lixian Section） which is the reconstruction project of partner assistance on Lixian by Hunan Province,reliability analysis shows that under normal working conditions,the slope safety coefficient is 1.305 and meet the specification requirements.Its reliability index Pf is equal to 10.21%,and the slope is in good stability.But under the case 3 which possible exists the earthquake and precipitation,the safety coefficient is 1.060,and the slope basically is in the limit equilibrium state whose reliability index Pf is equal to 5.87% and the slope is in less stable.Therefore,the proposed reinforcement optimization program,provide a reference for future similar projects.