摘要
根据神经网络的基本原理,结合福建省尤溪县气象因子及刚竹毒蛾发生面积的实测数据,建立神经网络模型。结果表明:所建立的BP神经网络模型,具有满意的拟合精度和预测精度。2个预留调查点的平均预测精度达96.55%,预测准确率为100%。
Based on the principle of network, combined with climatic data and actural Panatar phyllostachysae occurrence area in Youxi County, Fujian Province, the occurrence area forecast model of P. phyllostachysae was established. The results showed that the established neural network model was good at both simulating and forecasting. For 2 preset chosen, its forecast precision was 96.55% and accuracy was 100%.
出处
《华东昆虫学报》
2006年第1期37-39,共3页
Entomological Journal of East China
关键词
刚竹毒蛾
神经网络
发生面积
预测预报
Panatar phyllostachysae
neural network
occurrence area
forecast