基于数值预报产品的地面气温BP-MOS预报方法  

THE FORECASTING OF SURFACE AIR TEMPERATURE USING BP-MOS METHOD BASED ON THE NUMERICAL FORECASTING RESULTS

作　　者：吴君 裴洪芹 石莹 张京英 王庆华 

机构地区：[1]山东省临沂市气象局,临沂276004

出　　处：《气象科学》2007年第4期430-435,共6页Journal of the Meteorological Sciences

基　　金：山东省气象局重点科研课题计划项目(编号:2004SDQXZ02)

摘　　要：在山东省临沂市气象局开发的"中尺度数值预报业务系统"的基础上,利用高分辨率的数值预报产品和地面气温观测资料,建立了地面气温的BP神经网络方法预报模型。检验结果表明BP神经网络模型的气温预报准确率高于逐步回归模型和MM5模式输出的气温预报准确率,可应用于实际预报业务中来制作气温的精细化预报。In this paper, based on the ＂Mesoscale Numerical Forecasting System＂ developed by the Linyi Meteorological Bureau, the BP neural network model was applied to forecast the surface air temperature using the high resolution numerical forecasting data and observation as input. The result shows that accuracy rate of the BP model is higher than the stepwise regression model and MM5 model in surface air temperature forecasting, and the BP model can be applied in realtime surface air temperature forecast

关 键 词：气温预报 数值预报产品 MM5模式 BP神经网络 逐步回归 

分 类 号：P456.9[天文地球—大气科学及气象学]