【摘要】 压缩指数、回弹指数以及次固结系数是表征土体压缩性的重要变形参数,通常被用来计算工程土体沉降,其值获取目前多采用室内固结蠕变试验,但试验结果受多方面因素影响,耗时长且精确度较低。基于此,提出一种基于支持向量机和非支配排序遗传算法(NSGA-Ⅱ)的多目标反分析方法来获取土体压缩性变形参数,首先通过开展原状土室内标准固结试验获取变形参数的取值范围,并利用支持向量机建立变形参数与水平及竖向位移之间的非线性映射关系,最后使用NSGA-Ⅱ算法求解多目标函数,得到土体变形参数的最优解,比较最优土体变形参数的数值模拟位移与现场监测位移,结果吻合,说明该方法可行。更多还原

【Abstract】 Compression index,rebound index,and secondary consolidation coefficient are important deformation parameters that characterize soil compressibility.They are usually used to calculate the settlement of engineering soils.At present,its value is mostly obtained by indoor consolidation creep test,but the test result is affected by many factors,which takes a long time and has low accuracy.Based on this,the paper proposes a multi-objective back analysis method based on support vector machine and non-dominated sorting genetic algorithm(NSGA-Ⅱ)to obtain soil compressive deformation parameters.Firstly,the range of deformation parameters was obtained by carrying out the standard consolidation tests of undisturbed soil,and the nonlinear mapping relationship between deformation parameters and horizontal and vertical displacement was established by support vector machine.Finally,NSGA-Ⅱ algorithm was used to solve the multi-objective function,and the optimal solution of soil deformation parameters was obtained.Comparing the numerical simulation displacement of the optimal soil deformation parameters with the on-site monitoring displacement,the results are consistent,indicating that the method is feasible.更多还原