基于人工蜂群径向基神经网络预测参考作物需水量  

Reference Crop Water Requirement Forecast Based on Artificial Bee Colony Radial Basis Neural Network

作　　者：孟玮[1] 孙西欢[1,2] 郭向红[1] 马娟娟[1] 马文云 赵文渊 张威贤 MENG Wei;SUN Xi-huan;GUO Xiang-hong;MA Juan-juan;MA Wen-yun;ZHAO Wen-yuan;ZHANG Wei-xian(College of Water Resource Science and Engineering Taiyuan University of Technology,Taiyuan 030024,China;Shanxi Jinzhong College,Jinzhong 030619,Shanxi Province,China)

机构地区：[1]太原理工大学水利科学与工程学院,太原030024 [2]晋中学院,山西晋中030600

出　　处：《节水灌溉》2020年第1期79-83,共5页Water Saving Irrigation

基　　金：国家自然科学基金项目(51579168,U1803112);山西省自然科学基金项目(201601D011053)。

摘　　要：为了能够根据有限的气象数据较为准确的模拟蓄水坑灌苹果园的日参考作物需水量,以山西省农业科学院果树所蓄水坑灌试验基地的逐日气象资料为输入项,以日参考需水量为输出项,在径向基神经网络的基础上构建了基于人工蜂群算法的径向基神经网络模型,以预测蓄水坑灌苹果园的日参考作物需水量,以FAO-56 Penman-Monteith(FAO56-PM)公式的计算结果为标准分析预测模型的适用性。结果表明:经人工蜂群算法优化后的径向基神经网络预测模型的模拟结果与标准方法FAO56-PM公式的计算结果更为接近,更适合于预测山西省农业科学院果树所蓄水坑灌苹果园日参考作物需水量。In order to accurately simulate the daily reference crop water demand of apple orchards under water storage pit irrigation according to limited meteorological data,by taking the daily meteorological data of the water storage pit irrigation orchard at the Institute of Shanxi Academy of Agricultural Sciences as input item and the daily reference water requirement as output items,on the basis of RBF neural network,a RBF neural network model based on artificial bee colony algorithm was established to predict the daily reference crop water demand of apple orchard under water storage pit irrigation.The model was validated and determined by the calculation result of the FAO-56 Penman-Monteith(FAO56-PM)formula.The results showed that the simulation results of the RBF neural network prediction model optimized by the artificial bee colony algorithm were closer to the calculation results of the standard method fao56-pm formula,which was more suitable for predicting the day reference crop water requirement of the water storage pit irrigation orchard in this area.

关 键 词：蓄水坑灌 日参考作物需水量 人工蜂群算法 径向基神经网络 

分 类 号：S157[农业科学—土壤学]