基于最小二乘支持向量机的土石坝渗透系数反演

针对土石坝渗透参数和测压管水位间复杂的非线性关系,应用最小二乘支持向量机于土石坝渗透系数的反演。首先利用有限元模型得到最小二乘支持向量机的训练样本,建立坝体水压分量相对值和渗透系数间复杂的非线性关系,并将其输入到训练好的最小二乘支持向量机模型,即可得到大坝渗透系数的反演值。以某土石坝为例经对比分析,该方法是可行的。

Back Analysis of Dam Seepage Parameters Based on Least Squares Support Vector Machine

In view of complex nonlinear relationship between dam seepage parameters and piezometric tube level,the least squares support vector machine is applied to the back analysis of seepage parameters.Firstly,training samples of the least squares support vector machine are obtained through the finite element model,the complex nonlinear relationship between relative values of water pressure components of dam pressure and seepage parameters is established,which are inputted into the least squares support vector machine model,and the values of back analysis of dam seepage parameters can be obtained.At last an earth rockfill dam is taken as an example,and the least squares support vector machine is applied to the back analysis of dam seepage parameters.Through comparative analysis,it is found that the method is feasible.