基于最小二乘支持向量机的人民胜利渠灌区灌溉用水量预测  

A Prediction of Irrigation Water in People's Victory Canal Irrigation District Based on the Least Squares Support Vector Machine Model

作　　者：宰松梅[1,2] 郭冬冬[1] 温季[1] 

机构地区：[1]水利部农田灌溉研究所,河南新乡453003 [2]西北农林科技大学,陕西杨凌712100

出　　处：《中国农村水利水电》2009年第12期49-51,共3页China Rural Water and Hydropower

基　　金：国家"863"计划项目(2006AA100213);国家科技支撑计划项目(2007BAD38B04)

摘　　要：基于最小二乘支持向量机(LSSVM)良好的泛化能力和特点,以人民胜利渠灌区需水量为研究对象,选用径向基函数(RBF)作为核函数,建立了最小二乘支持向量机预测模型,对灌区需水量进行了模拟计算,用检验样本与灰色预测和基于RBF的神经网络模型的预测结果进行了比较,LSSVM预测的最大误差8.78%,平均误差4.90%。结果表明最小二乘支持向量机模型有较高的预测精度和较强的泛化能力,可为灌区水资源规划提供科学依据。Based on the good generalization ability of the support vector machine model, the People＇s Victory Canal Irrigation District ＇s water consumption is studied. Radial basis function （RBF） is chosen as the kernel function, a least squares support vector machine （LSSVM） prediction model is established. The irrigation water demands of the People＇s Victory Canal Irrigation District over the 8 years are simulated. The test samples are compared with the results of gray prediction and the RBF-based neural network model. The maximum prediction error of SVM is 8. 78%, and the average error is 4. 90%. The results show that the support vector machine model has higher prediction accuracy and stronger generalization water resources. ability. It can provide a scientific basis for the planning of irrigation

关 键 词：灌区 需水量 最小二乘支持向量机 径向基函数 

分 类 号：S274.4[农业科学—农业水土工程]