褐飞虱发生的相空间线性回归预测模型

A regression model of phase space for predicting brown planthopper (Nilaparvata lugens Stl ) occurrence

基于混沌理论 ,通过对江苏省太湖地区农科所褐飞虱发生时间序列资料的分析 ,组建了褐飞虱NilaparvatalugensSt l发生时间序列的相空间线性回归预测模型。结果表明 ,7次预报的平均相对误差为 2 2 91% ,转化为发生等级 ,预测准确率为 10 0 % 。

A regression model of phase space for predicting brown planthopper (BPH) occurrence in the rice growing area of Taihu District, Jiangsu Province, was established on the basis of chaos theory. Its accuracy was very high with a relative error of only 22 91%. Therefore, the model provides a new and more efficient method for the prediction of the long term occurrence of this insect pest.