基于天气预报和符号回归算法的参考作物腾发量预测研究
|
摘要
以干旱区和湿润区6个典型站点1989-2016年历史气象资料和2013-2016年天气预报数据为依据,以PM公式计算结果为对照,比较分析了率定Hargreaves-Samani(HS)模型和符号回归估算模型(SR)的ET_0预报精度。结果表明:率定后的HS模型在各站点的ET_0预报精度均维持在较高水平,且其在干旱区典型站点的预报精度略高于湿润区站点的值;而与HS公式预报结果相比,采用SR模型在不同气候区的ET_0预报精度均有不同程度的提高,其中在湿润区站点的平均MAE、RMSE值降低了18.98%和20.97%,在干旱区各站点的平均MAE、RMSE值减少了9.79%和7.53%。因此,根据不同模型在不同气候区的预报精度,结合气候特征,建议在湿润区和干旱区分别采用SR模型和HS公式进行ET_0预报,可为实时灌溉预报提供准确依据。
With historical meteorological data during the period of 1989 ~ 2016 and weather forecast messages during the period of 2013-2016 as the basis in six stations which are located in arid zones and humid zones in this paper,with ET_0 calculated by FAO56 Penman-Monteith equation as the contrast,we compare the accuracy of ET_0 prediction of calibrated Hargreaves-Samani formula( HS) with that of symbolic regression model( SR). Results show that the accuracy of ET_0 prediction of calibrated HS formular maintains a higher level in all stations,and that its forecast accuracy is slightly higher in typical stations which are in arid zones than those in humid zones. Compared with HS formula,the ET_0 forecast accruacy of SR model ha been improved to some extent in different climatic zones,the value of mean absolute error( MAE) and root mean square error( RMSE) of SR model reduced by 18.98% and 20.97% in humid zones,respectively,and the value of MAE and RMSE of SR model has been reduced by 9.79% and 7.53% in arid zones. Therefore,according to forecasting accuracy of different models ind different climatic zones,combining with climatic features,SR model and HS formula are recommended for predicting ET_0 in humid zones and arid zones. And it can provide accurate data for real-time irrigation forecasting.
With historical meteorological data during the period of 1989 ~ 2016 and weather forecast messages during the period of 2013-2016 as the basis in six stations which are located in arid zones and humid zones in this paper,with ET_0 calculated by FAO56 Penman-Monteith equation as the contrast,we compare the accuracy of ET_0 prediction of calibrated Hargreaves-Samani formula( HS) with that of symbolic regression model( SR). Results show that the accuracy of ET_0 prediction of calibrated HS formular maintains a higher level in all stations,and that its forecast accuracy is slightly higher in typical stations which are in arid zones than those in humid zones. Compared with HS formula,the ET_0 forecast accruacy of SR model ha been improved to some extent in different climatic zones,the value of mean absolute error( MAE) and root mean square error( RMSE) of SR model reduced by 18.98% and 20.97% in humid zones,respectively,and the value of MAE and RMSE of SR model has been reduced by 9.79% and 7.53% in arid zones. Therefore,according to forecasting accuracy of different models ind different climatic zones,combining with climatic features,SR model and HS formula are recommended for predicting ET_0 in humid zones and arid zones. And it can provide accurate data for real-time irrigation forecasting.
引文
[1]许平.我国农业用水的历史变化及可持续利用问题[J].中国农村水利水电,2008,(10):54-57.
[2]Xu J Z,S Z Peng,S H Yang,et al.Predicting daily reference evapotranspiration in a humid region of China by the locally calibrated hargreaves-samani equation using weather forecast data[J].Journal of Agricultural Science and Technology,2012,14(6):1 331-1 342.
[3]茆智,李远华,李会昌.逐日作物需水量预测数学模型研究[J].武汉水利电力大学学报,1995,(3):253-259.
[4]罗玉峰,崔远来,蔡学良.参考作物腾发量预报的傅立叶级数模型[J].武汉大学学报(工学版),2005,(6):45-47,52.
[5]Marino M A,J C Tracy,S A Taghavi.Forecasting of reference crop evapotranspiration[J].Agricultural Water Management,1993,24(3):163-187.
[6]蔡甲冰,刘钰,雷廷武,等.根据天气预报估算参照腾发量[J].农业工程学报,2005,(11):19-23.
[7]Xu J,S Peng,W Wang,et al.Prediction of daily reference evapotranspiration by a multiple regression method based on weather forecast data[J].Archives of Agronomy and Soil Science,2013,59(11):1 487-1 501.
[8]罗玉峰,李思,彭世彰,等.基于气温预报和HS公式的参考作物腾发量预报[J].排灌机械工程学报,2013,(11):987-992.
[9]刘梦,罗玉峰,汪文超,等.基于天气预报的漳河灌区参考作物腾发量预报方法比较[J].农业工程学报,2017,(19):156-162.
[10]谢梅香,罗玉峰,钱子嘉,等.基于气温预报和神经网络的参考作物腾发量预报[J].节水灌溉,2015,(2):33-36.
[11]Guo X,X Sun,J Ma.Prediction of daily crop reference evapotranspiration(ET0)values through a least-squares support vector machine model[J].Hydrology Research,2011,42(4):268-274.
[12]Xu J,J Wang,Q Wei,et al.Symbolic regression equations for calculating daily reference evapotranspiration with the same Input to hargreaves-samani in arid China[J].Water Resources Management,2016,30(6):2 055-2 073.
[13]Hargreaves G H,Z Samani.Reference crop evapotranspiration from temperature[J].Applied Engineering in Agriculture,1985,1(2):96-99.
[14]G Allen,R L Pereira,D Raes,et al.Crop evapotranspirationGuidelines for computing crop water requirements-FAO Irrigation and drainage paper 56[M].1998,56.
[15]Schmidt M D,H Lipson.Distilling free-form natural laws from experimental data[J].Science,2009,324(5 923):81-85.
[16]Dubcakova R.Eureqa:software review[J].Genetic Programming and Evolvable Machines,2011,12(2):173-178.
[17]Cai J,Y Liu,T Lei,et al.Estimating reference evapotranspiration with the FAO Penman-Monteith equation using daily weather forecast messages[J].Agricultural and Forest Meteorology,2007,145(1-2):22-35.
[18]Elmetwally M.Sunshine and global solar radiation estimation at different sites in Egypt[J].Journal of Atmospheric and Solar-Terrestrial Physics,2005,67(14):1 331-1 342.
[19]李倩,胡邦辉,王学忠,等.基于BP人工神经网络的区域温度多模式集成预报试验[J].干旱气象,2011,(2):231-235,250.
[20]李国翠,连志鸾,赵彦厂,等.石家庄温度预报检验及影响因子分析[J].气象与环境学报,2009,(1):15-18.
[21]胡庆芳,杨大文,王银堂,等.Hargreaves公式的全局校正及适用性评价[J].水科学进展,2011,(2):160-167.
[2]Xu J Z,S Z Peng,S H Yang,et al.Predicting daily reference evapotranspiration in a humid region of China by the locally calibrated hargreaves-samani equation using weather forecast data[J].Journal of Agricultural Science and Technology,2012,14(6):1 331-1 342.
[3]茆智,李远华,李会昌.逐日作物需水量预测数学模型研究[J].武汉水利电力大学学报,1995,(3):253-259.
[4]罗玉峰,崔远来,蔡学良.参考作物腾发量预报的傅立叶级数模型[J].武汉大学学报(工学版),2005,(6):45-47,52.
[5]Marino M A,J C Tracy,S A Taghavi.Forecasting of reference crop evapotranspiration[J].Agricultural Water Management,1993,24(3):163-187.
[6]蔡甲冰,刘钰,雷廷武,等.根据天气预报估算参照腾发量[J].农业工程学报,2005,(11):19-23.
[7]Xu J,S Peng,W Wang,et al.Prediction of daily reference evapotranspiration by a multiple regression method based on weather forecast data[J].Archives of Agronomy and Soil Science,2013,59(11):1 487-1 501.
[8]罗玉峰,李思,彭世彰,等.基于气温预报和HS公式的参考作物腾发量预报[J].排灌机械工程学报,2013,(11):987-992.
[9]刘梦,罗玉峰,汪文超,等.基于天气预报的漳河灌区参考作物腾发量预报方法比较[J].农业工程学报,2017,(19):156-162.
[10]谢梅香,罗玉峰,钱子嘉,等.基于气温预报和神经网络的参考作物腾发量预报[J].节水灌溉,2015,(2):33-36.
[11]Guo X,X Sun,J Ma.Prediction of daily crop reference evapotranspiration(ET0)values through a least-squares support vector machine model[J].Hydrology Research,2011,42(4):268-274.
[12]Xu J,J Wang,Q Wei,et al.Symbolic regression equations for calculating daily reference evapotranspiration with the same Input to hargreaves-samani in arid China[J].Water Resources Management,2016,30(6):2 055-2 073.
[13]Hargreaves G H,Z Samani.Reference crop evapotranspiration from temperature[J].Applied Engineering in Agriculture,1985,1(2):96-99.
[14]G Allen,R L Pereira,D Raes,et al.Crop evapotranspirationGuidelines for computing crop water requirements-FAO Irrigation and drainage paper 56[M].1998,56.
[15]Schmidt M D,H Lipson.Distilling free-form natural laws from experimental data[J].Science,2009,324(5 923):81-85.
[16]Dubcakova R.Eureqa:software review[J].Genetic Programming and Evolvable Machines,2011,12(2):173-178.
[17]Cai J,Y Liu,T Lei,et al.Estimating reference evapotranspiration with the FAO Penman-Monteith equation using daily weather forecast messages[J].Agricultural and Forest Meteorology,2007,145(1-2):22-35.
[18]Elmetwally M.Sunshine and global solar radiation estimation at different sites in Egypt[J].Journal of Atmospheric and Solar-Terrestrial Physics,2005,67(14):1 331-1 342.
[19]李倩,胡邦辉,王学忠,等.基于BP人工神经网络的区域温度多模式集成预报试验[J].干旱气象,2011,(2):231-235,250.
[20]李国翠,连志鸾,赵彦厂,等.石家庄温度预报检验及影响因子分析[J].气象与环境学报,2009,(1):15-18.
[21]胡庆芳,杨大文,王银堂,等.Hargreaves公式的全局校正及适用性评价[J].水科学进展,2011,(2):160-167.