基于径向基神经网络预测日参考作物需水量
Prediction of Daily Reference Crop Water Requirement Based on Radial Basis Neural Network Model
孟玮 1孙西欢 2郭向红 3马娟娟3
作者信息
- 1. 太原理工大学 机械与运载工程学院,山西 太原 030024
- 2. 太原理工大学 水利科学与工程学院,山西 太原 030024;晋中学院,山西 晋中 030600
- 3. 太原理工大学 水利科学与工程学院,山西 太原 030024
- 折叠
摘要
为了利用有限的气象数据准确预测蓄水坑灌果园的日参考作物需水量,利用蓄水坑灌试验基地逐日温度与湿度数据,构建了基于径向基神经网络的ET0 预测模型,并将其模拟结果及Hargreaves、Priestley-Taylor两种常用ET0 计算模型的计算结果同FAO-56 Penman-Monteith(FAO56-PM)公式计算的标准值进行对比.结果表明:径向基神经网络预测模型的模拟结果与标准方法FAO56-PM公式的计算结果最接近,而Hargreaves、Priestley-Taylor两个常用计算模型的计算结果比标准值偏大,在实际应用中应对其进行校正.
Abstract
This paper aimed to realize the accurate prediction of the daily reference crop water requirement of the water storage pit irrigation apple orchard based on the limited meteorological data.According to the data of daily temperature and humidity data of the water storage pit irrigation orchard at the Institute of Shanxi Academy of Agricultural Sciences,an ET0 prediction model based on radial basis neural network was built.The simulation results and the calculation results of the two commonly used ET0 calculation formulas of Hargreaves and Priestley-Taylor were compared with the standard values calculated by the FAO-PM formula.The results show that the simulation results of the radial basis neural network model are closer to the standard values calculated by the FAO-PM formula.The calculation results of the Hargreaves for-mula and the Priestley-Taylor formula are larger than that of the standard values,which should be corrected by coefficients in practical appli-cations.
关键词
蓄水坑灌/日参考作物需水量/径向基神经网络/Hargreaves公式/Priestley-Taylor公式Key words
water storage pit irrigation/daily reference crop water requirement/radial basis neural network/Hargreaves equation/Priestley-Taylor equation引用本文复制引用
出版年
2024
国家科技期刊平台
NSTL
万方数据