基于BP神经网络的单站总云量预报研究  

Single Station Forecast of Total Cloud Cover Based on BP Neural Net

作　　者：张长卫[1] 

机构地区：[1]解放军理工大学气象学院,南京211101

出　　处：《气象与环境科学》2009年第1期68-71,共4页Meteorological and Environmental Sciences

摘　　要：云是能够对飞行活动产生影响、甚至危及飞行安全的气象要素之一。为了对北京单站总云量进行预报,利用人工神经网络的BP模型,针对不同时次进行关键因子的选取,尝试用多种预报因子的组合,建立了总云量预报模型。试验结果表明:所建立的模型具有较好的拟合、预报精度,并且没有出现"过拟合"现象,对新样本具有较好的泛化能力。Cloud is one of the order to forecast total cloud cover Propagation） neural net. Different meteorological elements which have great influence on the aviation. In for Beijing single station, we use the nonlinear method -BP （Back key factors are chosen and composed for each foretime. The final model has a good performance in fitting and prediction. There is no ＂overfitting＂ phenomenon in the result either, so the model can be used with new samples.

关 键 词：BP神经网络 单站预报 总云量 

分 类 号：P457.1[天文地球—大气科学及气象学]