改进PSO神经网络的白水苹果树腐烂病预测  

Prediction of Baishui Apple Tree Canker of the Improved PSO Neural Network

作　　者：贾花萍[1] 

机构地区：[1]渭南师范学院,陕西渭南714099

出　　处：《安徽农业科学》2014年第10期2959-2961,共3页Journal of Anhui Agricultural Sciences

基　　金：陕西省自然科学基础研究计划项目(2011JM1010)

摘　　要：准确预测苹果树腐烂病的流行程度,可以科学合理地预防苹果树腐烂病。利用陕西白水县的气象历史数据,将其提取主成分后作为预测因子,采用粒子群(PSO)神经网络建立苹果树腐烂病预测模型。与传统BP神经网络相比,改进PSO神经网络预测结果更接近于实际值,同时避免陷入局部极小的缺点,达到了很好的预测效果。可见,该方法用于预测苹果树腐烂病远优于BP算法,可作为一种新方法预测苹果树腐烂病。The accurate prediction of the prevalence of apple trees canker can be scientific and reasonable to prevent rot. Using weather histo- ry data in Baishui County, Shaanxi Province, after its extraction as the main component predictors, particle swarm （PSO） neural network was adopted to establish forecasting model. Compared with the traditional BP neural network, improved PSO neural network prediction results is closer to the actual value, while avoiding the shortcomings of local minima, to achieve a good prediction. The method is far superior than BP algorithm, which can be used as a new method to predict apple canker.

关 键 词：腐烂病 神经网络 预测 

分 类 号：S5-3[农业科学—作物学]