基于BP-NN的温室膜下滴灌茄子蒸腾速率预测模型 |
| |
引用本文: | 葛建坤,李小平,罗金耀.基于BP-NN的温室膜下滴灌茄子蒸腾速率预测模型[J].节水灌溉,2016(8):95-98. |
| |
作者姓名: | 葛建坤 李小平 罗金耀 |
| |
作者单位: | 1. 华北水利水电大学,郑州 450011; 河南省节水农业重点实验室,郑州 450011;2. 武汉大学水资源与水电工程科学国家重点实验室,武汉,430072 |
| |
基金项目: | 河南省科技计划项目(162300410138),郑州市普通科技攻关项目(20130924)。 |
| |
摘 要: | 通过田间试验,对温室膜下滴灌茄子冠层叶片蒸腾速率的变化规律进行了深入研究。通过分析温室内地面温度、相对湿度、植株冠层温度、气压、水面蒸发、太阳辐射等6个环境参数与茄子蒸腾速率的综合影响关系,确定了网络拓扑结构为6-9-1。并应用MATLAB软件,选择Levenberg-Marquardt(L-M)优化算法,建立了基于Back Propagation(BP)神经网络的温室膜下滴灌茄子蒸腾速率预测模型。经模型验证得出,BP神经网络模型预测值与蒸腾速率实测值间拟合效果较好,平均相对误差为0.029 8,达到预测精度要求。该研究成果对温室膜下滴灌作物需水规律及需水量研究具有较好的参考价值。
|
关 键 词: | 温室 膜下滴灌 蒸腾速率 神经网络 |
Prediction Model of Greenhouse Eggplant Transpiration Rate Based on BP Neural Network |
| |
Abstract: | In order to reveal the law of crop transpiration in greenhouse,a field experiment on transpiration rate of greenhouse egg-plant with drip irrigation under mulch was taken in a Venlo type greenhouse in North China University of Water Resources and Elec-tric Power.Through the analysis on the combined influence between eggplant transpiration rate and 6 indoor environmental factors (greenhouse ground temperature,relative humidity,plant canopy temperature,air pressure,evaporation and solar radiation),topol-ogical structure of the model was discussed and determined (6-9-1).And a prediction model of greenhouse eggplant transpiration rate was established based on BP Neural network of L-M optimizing algorithm,by using MATLAB.After the model validation,the re-sults indicated that,the BP neural network prediction model has a high precision,the predicted value fits the measured value well, and average relative error is only 0.0298,which meets the precision requirement.The research result has a certain reference value to the study on crop water requirement in greenhouse with drip irrigation under mulch. |
| |
Keywords: | greenhouse drip irrigation under mulch transpiration rate neural network |
本文献已被 CNKI 万方数据 等数据库收录! |
|