短信平台下基于神经网络的施肥模型建立  

Model establishment of fertilization based on neural network in SMS platform

作　　者：田茁[1] 

机构地区：[1]吉林农业大学资源与环境学院,长春市130118

出　　处：《中国农机化学报》2016年第7期171-173,共3页Journal of Chinese Agricultural Mechanization

基　　金：吉林省科技发展项目(20140312020ZG);吉林省教育厅十二五项目(2012060)

摘　　要：利用神经网络对40组样本数据进行观测和统计;通过氮肥、磷肥、钾肥的加权,同时对玉米产量和肥料用量的实际比较,建立土壤施肥模型;最后利用GSM短信猫和中国移动公司手机接收信号相匹配的应用软件进行施肥决策,将所需要的肥量用量以短信的形式反馈到用户手机上。结果表明,这种将神经网络应用于建立土壤施肥模型,具有较好的学习精度。In this paper,the neural network is used to observe and statistic the data of the 40 groups.Through the weight of nitrogen,phosphate,potash,the actual yield of corn and amount of fertilizer were compared,and it was establish of soil fertilization model.Finally,the model of soil fertilizer application is set up by using GSM and China Mobile,and the amount of fertilizer is fed to the user＇s mobile phone.The results show that the neural network is applied in the establishment of soil fertilization model,which has better learning precision.

关 键 词：短信平台 权重 神经网络 施肥决策 

分 类 号：S147.2[农业科学—肥料学]