Research on Model of Predicting Irrigation Water Requirement Based on Kernel Method
Scientific irrigation is very important for saving water in agriculture,increasing output and benefit in our country.In this paper a method of nonlinear character extraction based on kernel canonical correlation analysis(KCCA) is presented in which information of soil and environment are input vectors of model.Nonlinear character are extracted by KCCA,then main character variables are determined which reflects the complex relationship between original input and output data and the array dimension of input data is simplified.At last the model based on least squares support vector machine (SVM) were completed.By comparing simulation results,precision and rapidity of the prediction model based on KCCA-SVM are higher than those of CCA-SVM and LS-SCM model.The experimental results show that the method is very effective.
Irrigation water requirement Prediction model Nonlinear character extraction Kernel canonical correlation analysis Support vector machine
ZHONG Bingxiang
College of Electrical and Information Engineering Chongqing University of science and technology,Chongqing,China
国际会议
沈阳
英文
1338-1341
2012-09-26（万方平台首次上网日期，不代表论文的发表时间）