当前位置: 首页 > 文章 > 神经网络方法计算参考作物腾发量精度分析 节水灌溉 2012 (9) 37-40
Position: Home > Articles > Precision Analysis of Reference Crop Evapotranspiration Calculation by Neural Network Method Water Saving Irrigation 2012 (9) 37-40

神经网络方法计算参考作物腾发量精度分析

作  者:
王乙江;彭世彰;徐俊增;杨士红
单  位:
河海大学水文水资源与水利工程科学国家重点实验室
关键词:
神经网络;参考作物腾发量;气象因素
摘  要:
参考作物腾发量的准确计算是确定农田灌溉制度的重要依据,比较了神经网络模型和经验公式计算腾发量的精度以及如何使用最少的气象数据实现最优的腾发量计算问题。结果表明,神经网络模型比经验公式的计算精度要高。采用神经网络模型用最少气象数据实现最优腾发量计算是可行的,但输入参数类型需要根据不同地区的气候特点进行选择。在半湿润地区最佳网络输入为日最高气温、最低气温、平均气温和外太空辐射,在气候湿润地区最佳网络输入为日最高气温、最低气温、平均气温、经验相对湿度和外太空辐射。因此,在缺少气象资料时可以使用神经网络模型代替经验公式计算参考作物腾发量。
译  名:
Precision Analysis of Reference Crop Evapotranspiration Calculation by Neural Network Method
作  者:
WANG Yi-jiang,PENG Shi-zhang,XU Jun-zeng,YANG Shi-hong(State Key Laboratory of Hydrology-Water Resources and Hydraulic Engineering,Hohai University,Nanjing 210098,China)
关键词:
artificial neural network;evapotranspiration;meteorological factor
摘  要:
Accurate calculating the reference crop evapotranspiration is important for irrigation system design.This paper mainly compares the difference between artificial neural network methods and empirical equations methods in calculating evapotranspiration,and further modifies the neural network method for evapotranspiraion calculation using the least meteorological data.The results show that the artificial neural network methods have a better result than the empirical equation ways in reference crop evapotranspiration calculation.The evapotranspiration calculation using the least meteorological data by the modified method put forward in this paper is a suitable way.But the selection of model input parameters should be depended on the area climate type.The best input parameters of the modified neural network method are daily maximum temperature,daily minimum temperature,daily mean temperature and extraterrestrial radiation in semi-humid region.And in humid region the best input parameters of the neural network method are daily maximum temperature,daily minimum temperature,daily mean temperature,experiential relative humidity and extraterrestrial radiation.So the neural network method can be used in calculating the reference crop evpotranspiration when there is not enough meteorological data.

相似文章

计量
文章访问数: 13
HTML全文浏览量: 0
PDF下载量: 0

所属期刊

推荐期刊