基于PSO-SVM模型的拱坝坝变形预测研究  

Study of Arch Dam Deformation Prediction Based on PSO-SVM Model

作　　者：刘敬洋[1,2,3] 刘何稚[1,2,3] 朱凯[1,2,3] 陈晨[1,2,3] 张凤山[3] 

机构地区：[1]河海大学水文水资源与水利工程科学国家重点实验室,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,南京210098 [3]河海大学水利水电学院,南京210098

出　　处：《三峡大学学报（自然科学版）》2013年第1期30-33,共4页Journal of China Three Gorges University：Natural Sciences

基　　金：国家自然科学基金重点项目(51139001;51279052;51179066;51079046);水文水资源与水利工程科学国家重点实验室专项基金(2010585212)

摘　　要：拱坝已成为大型水利枢纽的主要坝型之一,大坝变形预测是大坝安全监控的重要内容,预测分析的难点之一在于变形监测数据往往具有复杂的非线性特点.支持向量机(SVM)具有良好的泛化能力,可有效地解决小样本、非线性、高维数等问题,因此可将其广泛应用于拱坝变形观测中.由于算法的成功与否很大程度上取决于其参数的选取,本文充分利用粒子群算法快速全局优化的特点,采用粒子群算法来优化支持向量机的模型参数,建立了基于PSO-SVM的大坝变形预测模型.将该模型应用于某拱坝坝基变形预测中,与传统的多元回归模型预测结果进行对比.结果表明,PSO-SVM模型用于拱坝变形预测是可行的.Arch dam has gradually evolved as one of dam types as main large-scale hydraulic project; dam deformation prediction is an important part of dam safety monitoring; and it is difficult to predict because of the complicated nonlinear characteristics of the monitoring data. Support vector machine（SVM）could solve the small sample, nonlinear high dimension problem due to its excellent generalization ability; hence it has been widely used in predicting arch dam deformation. However, the predicting results considerably depend on the choice of SVM model parameters. In the paper, the particle swarm optimization（PSO）, which has the characteristic of fast global optimization, is applied to optimize the parameters in SVM; and then the dam deformation prediction model based on PSO-SVM could be established. The model is applied to a certain arch dam foundation prediction. The accuracy of this employed approach is examined by comparing it with multiple regression method. In a word, the experiment results indicate that the proposed method based on PSO-SVM can be used in arch dam deformation prediction.

关 键 词：拱坝 变形预测 粒子群优化算法 支持向量机 

分 类 号：TV698.11[水利工程—水利水电工程]