基于最小二乘支持向量机的大坝力学参数反演  

Back analysis of dam mechanical parameters based on least squares support vector machine

作　　者：李波[1,2] 徐宝松[1,2] 武金坤[1,2] 胡江[1,2] 徐光磊[3] 

机构地区：[1]河海大学水利水电工程学院,江苏南京210098 [2]河海大学水资源高效利用与工程安全国家工程研究中心,江苏南京210098 [3]南京水利科学研究院,江苏南京210029

出　　处：《岩土工程学报》2008年第11期1722-1725,共4页Chinese Journal of Geotechnical Engineering

基　　金：国家自然科学基金项目(50539010);国家自然科学基金项目(50539110);国家自然科学基金项目(50539030-1-3);国家自然科学基金项目(50579010);国家自然科学基金项目(50809025);国家科技支撑计划项目(20006BAC14BO3)

摘　　要：针对大坝力学参数和坝体位移间复杂的非线性关系,将最小二乘支持向量机应用于大坝力学参数的位移反演中。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,建立坝体位移水压分量相对值和力学参数间复杂的非线性关系,同时利用偏最小二乘回归模型分离出实测坝体位移的水压分量相对值,并将其输入到训练好的最小二乘支持向量机模型,即可得到大坝力学参数的反演值。以某混凝土重力拱坝为例,采用最小二乘支持向量机反演了坝体弹性模量、岩体变形模量以及主要断层的弹性模量,经过比较分析发现,该方法是可行的。In view of complex nonlinear relationship between dam mechanical parameters and dam displacements, the least squares support vector machine was applied to the back analysis of displacements of dam mechanical parameters. Firstly, the training samples of the least squares support vector machine were obtained by the finite element model, and the complex nonlinear relationship between relative values of water pressure components of dam displacements and mechanical parameters was established. At the same time, actual relative values of water pressure components were isolated from the measured dam displacements by using the partial least squares regression, which were inputted into the least squares support vector machine model, and the values of back analysis of the dam mechanical parameters could be obtained. A concrete gravity arch dam was taken as an example, and the least squares support vector machine was applied to the back analysis of dam elastic modulus, rock deformation modulus and elastic modulus of main faults. Though comparative analysis, it is found that the method is feasible.

关 键 词：反演分析 大坝力学参数 最小二乘支持向量机 

分 类 号：TV64[水利工程—水利水电工程]