基于天气预报的参考作物腾发量LS-SVM预测模型  

Least squares support vector machines model for predicting reference evapotranspiration based on weather forecasts

作　　者：张展羽[1,2] 王声锋[3] 段爱旺[4] 王斌[1,2] 

机构地区：[1]河海大学南方地区高效灌排与农业水土环境教育部重点实验室,江苏南京210098 [2]河海大学水利水电学院,江苏南京210098 [3]华北水利水电学院,河南郑州450011 [4]中国农业科学院农田灌溉研究所,河南新乡453003

出　　处：《水科学进展》2010年第1期63-68,共6页Advances in Water Science

基　　金：公益性行业科研专项经费资助项目(200903001);国家自然科学基金资助项目(50839002)~~

摘　　要：利用最小二乘支持向量机（LS—SVM）方法，建立了基于天气预报的参考作物腾发量（ET0）的预测模型。对广利灌区1997～2006年逐日气象信息中的天气类型和风速等级进行量化后，以不同天气预报信息作为输入量，建立10种验证方案，对2007年的逐日ET0进行预测。经验证，方案1～方案7精度均令人满意，其中方案1精度最高。方案1的输入量为气温、天气类型、风速等级3项的预测值，该方案的模型预测值与计算值的统计参数分别为：均方根偏差ERMS为0．5182mm，相对偏差ER为0．1878，决定系数R^2为0．8648，认同系数IA为0．9669，回归系数RC为0．9867；方案7精度亦较好，且以上指标统计参数依次为0．6576mm、0．2332、0．9866、0．7747及0．9866，该方案输入量只有气温项，实用性很强。A reference evapotranspiration （ET0 ） prediction model is developed based on the least squares support vector machines. Weather forecasts are used for ET0 predictions. The model can be trained with daily weather parameters including quantified weather types and wind grades, etc. Different combinations of daily weather parameters can be tested in the model training processes. In this study, the daily weather parameters are obtained from the Guangli irrigation district during the period 1997-2007. The 1997-2006 data are used for the model training, and a total of 10 daily ET0 forecasting schemes are established. Predictions of daily ET0 using each of the 10 schemes are validated withthe observations from 2007. The results show that the scheme using the air temperature and the quantified weather types and wind grades as model predictors is able to give the best model performance; and the corresponding statistics are the root mean square error （ ERMS ） of 0. 518 2, the relative error （ ER） of 0. 187 8, the coefficient of determination （ R^2 ） of 0. 864 8, the Ia of 0. 966 9, and the regression coefficient （RC ） of 0. 986 8. Acceptable daily ET0 predictionsare also obtained with other 6 schemes, among which the simplest scheme using only the air temperature as the model predictor can also produce fair results as revealed by ERMs = 0. 657 6, ER = 0. 233 2,R^2 = 0. 986 6, IA = 0. 774 7 and RC = 0. 968 0. The latter scheme shows a strong potential in practical applications.

关 键 词：天气预报 参考作物腾发量 最小二乘支持向量机 预测模型 

分 类 号：S161.4[农业科学—农业气象学]