主成分分析与支持向量机相结合的区域降水预测应用  

Regional Rainfall Forecast Based on Principal Component Analysis and Support Vector Machine

作　　者：农吉夫[1,2] 

机构地区：[1]广西民族大学数学与计算机科学学院,广西南宁530006 [2]广西混杂计算与集成电路设计分析重点实验室,广西南宁530006

出　　处：《数学的实践与认识》2011年第22期91-96,共6页Mathematics in Practice and Theory

摘　　要：将主成分分析和支持向量机回归相结合,以广西5、6月区域平均日降水量作为预报对象,进行区域日降水量预测研究.首先,整理分析大量的T213数值预报产品信息数据进行主成分分析,得到主成分数据序列;其次,根据主成分数据序列建立训练集训练支持向量机,并利用遗传算法优化参数;最后,输入支持向量机所需数据,得到主成分预测结果,建立广西日降水预报模型.实例计算结果表明,支持向量机回归模型比逐步回归模型有更好的预测能力.A scheme to forecast Regional Mean Rainfall is presented. This scheme combines the principal component analysis and the support vector regression, taking the regional average precipitation of Guangxi in May and June as the forecast object. First, a large number of information on the numerical forecasting products of T213 are turned into several time series data by the principal component analysis. Then, these principal component data are used to train the support vector machines, and a genetic algorithm is applied to optimize the parameters of the support machines. After the training and optimization, by inputting required data to the support vector machines, the principal components are obtained as the outputs, and daily rainfall forecast in Guangxi is established. The calculation results showed that SVM regression model had better forecast ability than stepwise regression model.

关 键 词：主成分分析 支持向量机 遗传算法 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] O242.1[自动化与计算机技术—控制科学与工程] P457.6[理学—计算数学]