摘要
根据河南省1990～2007年小麦白粉病病情及相关气象资料,建立了基于主分量分析的BP神经网络模型,得出影响其流行的主要分量,最后利用得到的主要分量作为BP神经网络的输入,对2008～2010年小麦白粉病流行情况进行预测。经与未进行主分量分析而建立的全要素BP网络模型进行比较,结果显示,该模型可以快速准确地预测小麦白粉病的流行程度,有效地减少了小麦的损失。
According to wheat powdery mildew condition and associated weather information in Henan province from 1990 to 2007,the principal component which affected the spread was determined,and the BP neural network model based on principal component analysis was established to forcast the epidemic situation of powdery mildew in wheat from 2008 to 2010.Compared with the full feature BP neural network model without principal component analysis,the experimental results showed that the model could predict the epidemic situation of wheat powdery mildew quickly and accurately,and reduce the production loss of the wheat effectively.
出处
《湖北农业科学》
北大核心
2011年第17期3611-3613,共3页
Hubei Agricultural Sciences
基金
河南省科技攻关重点项目(092102110175)
关键词
小麦白粉病
主分量分析
BP网络模型
预测
wheat powdery mildew
principle component analysis
BP network model
forecast
