基于模糊LS-SVM算法的大坝变形预测模型  

A Study on Dam Deformation Prediction of Based on Fuzzy LS-SVM Algorithm

作　　者：彭磊[1] 黄张裕[1] 刘胜男[1] 凌晨阳[1] 

机构地区：[1]河海大学地球科学与工程学院,江苏省南京市210098

出　　处：《水电自动化与大坝监测》2011年第1期51-53,共3页HYDROPOWER AUTOMATION AND DAM MONITORING

摘　　要：在大坝工程变形分析和预测方面,研究了一种基于支持向量度的模糊最小二乘支持向量机(LS-SVM)算法,结合具体实例进行对比分析,结果表明模糊LS-SVM模型的预测精度要高于LS-SVM模型,且支持向量机(SVM)的稀疏性也优于LS-SVM模型,可以很好地应用于大坝变形监测分析。A kind of fuzzy least squares support vector machine（FLS-SVM） algorithm based on the measure of support vector in the analysis and prediction of engineering deformation has been studied by authors.And the results of comparative analysis in a concrete example show that the prediction accuracy of FLS-SVM model is higher than that of LS-SVM model,and the sparseness of FLS-SVM is better than that of LS-SVM model.This shows that FLS-SVM can be applied to dam deformation monitoring and analysis perfectly.

关 键 词：大坝变形分析 模糊最小二乘支持向量机 支持向量度 剪切法 变形预测 

分 类 号：TV698.11[水利工程—水利水电工程]