用非线性多因子动态组合方法作降水概率预报  

Using Dynamically Combined Nonlinear Multi-Factor Equation to Forecast Probability of Precititation

作　　者：邵明轩[1] 刘凤辉[2] 程维中[3] 魏建明[2] 窦以文[2] 陈大刚[2] 

机构地区：[1]北京大学物理学院大气科学系 [2]北京市气象局,北京100089 [3]苏州市气象局,苏州215021

出　　处：《气象科技》2003年第4期206-210,共5页Meteorological Science and Technology

基　　金：北京市气象局"北京地区中尺度预报业务系统建设"项目资助

摘　　要：将逐时段定点降水预报这个以往一般作为非连续量处理的预报问题转化为对一个连续的降水可能函数的预报问题。降水可能函数为一由实况值确定的函数。因为现在模式变化较快 ,且许多因子和降水可能函数为非线性关系 ,文章采用一种新的非线性多因子动态组合方法 ,通过对降水可能函数的预报 。As the forecast of precipitation is actually considered as the forecast of a discontinuous variable, the forecast is translated into the forecast of a continuous variable—possibility function of precipitation (PFP). PFP is determined by observational values. Because the model on whose output the objective forecast was made updates frequently now, and relation between some forecast factors and PFP is nonlinear, a new method, dynamically combined equation of nonlinear multi factors (DCENMF), is presented to forecast the probability of precipitation at stations by forecasting PFP.

关 键 词：降水概率 预报 降水可能函数 动态组合 

分 类 号：P457.6[天文地球—大气科学及气象学]