最小二乘支持向量机在大坝变形预测中的应用

介绍了基于统计学习理论的一种新的机器学习技术———支持向量机(SVM)和其拓展方法———最小二乘支持向量机(LSSVM),并将LSSVM算法应用于混凝土大坝安全监控中的变形预测。根据实测数据,建立了基于LSSVM算法的大坝变形预测模型,同时与经典SVM预测模型进行分析比较。结果表明,LSSVM和经典SVM算法在大坝变形预测中都具有较好的可行性、有效性及较高的预测精度;LSSVM在算法的学习训练效率上比SVM有较大的优势,更适合于解决大规模的数据建模。

Research on Dam Displacement Forecasting Model Based on Least Squares Support Vector Machine

Support vector machine(SVM),a machine learning algorithm based on statistical learning theory and its continuation algorithm-Least Squares Support Vector Machine(LSSVM)were presented firstly in this paper,then the LSSVM algorithm was applied to the displacement forecast of concrete dam.According to the measured field data of dam,the authors built the forecasting model of dam displacement based on LSSVM,and the traditional SVM-based forecasting model was built to analyze and compare with LSSVM.The computational results show that both LSSVM algorithm and traditional SVM algorithm have the good feasibility and efficiency,and possess the higher precision forecasting;at the same time,by the results we can also conclude that LSSVM algorithm has the absolute advantage in training efficiency and it is more suitable to solve the modelling problem of large scale data.