摘要
首先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(Support Vector Machine,SVM),并针对目前支持向量机参数选择时人为选择的盲目性,将具有良好优化性能的混沌优化(Chaos Optimi-zation)技术应用到支持向量机惩罚函数和核函数参数的优化,提出了混沌优化支持向量机(Chaos Optimization Support Vector Machine,COSVM)方法.根据丰满大坝1997-2004年的实际监测数据,建立了混沌优化支持向量机大坝安全监控预测模型,进行了与统计回归模型和BP神经网络模型的分析比较,结果表明,COSVM模型具有更高的预测精度,同时在较长时段的预测中,COSVM模型也表现出更好的泛化推广性能.
Support vector machine （SVM）, a machine learning algorithm based on statistical learning theory is presented firstly. Aiming at the blindness of man made choice of the parameter and kernel function of SVM; a chaos optimization method is applied to select parameters of SVM;and a novel algo- rithm, chaos optimization support vector machine（COSVM） is put forward. According to the measured field data （1997-2004） of the Fengman Dam, the COSVM-based safety monitoring model is applied to forecast the deformation of the dam. Compared with statistical regression model and BP neural network model, a conclusion is drawn that the COSVM-based model possesses not only higher precision of forecasting, but also better generalization ability in long-interval forecasting.
出处
《武汉大学学报（工学版）》
CAS
CSCD
北大核心
2007年第1期53-57,共5页
Engineering Journal of Wuhan University
关键词
支持向量机
混沌优化
大坝
安全监控
预测
统计回归
BP神经网络
support vector machine
chaos optimization
dam
safety monitoring
forecasting
statisticalregression
BP neural network.
