摘要
介绍了用于制作三峡区间面雨量预报的各种预报方法 ,分析了天气学方法、数值预报方法及以数值预报产品为基础的人工神经网络方法在 1 999年汛期的试验效果 ,结果表明 :人工神经网络方法及 MAPS数值预报方法预报效果较好。并提出了对几种面雨量预报方法的改进措施。
Four kinds of approaches were introduced.They are of synoptic pattern,MAPS numerical model,artificial neural network and Japanese numerical model for forecasting area rainfall in the Sanxia region and analyses impact of the experiment during the summer in 1999.Results show that the forecasting effects are better for artificial neural network and MAPS numerical model.Finally improving measures are given for the first three kinds of approaches.
出处
《气象》
CSCD
北大核心
2000年第11期19-23,共5页
Meteorological Monthly
关键词
面雨量
预报方法
数值预报
天气学模型
三峡区间
area rainfall forecast approach numerical model forecasting artificial neural network synoptic pattern
