主办单位：中国气象局沈阳大气环境研究所
国际刊号：ISSN 1673-503X
国内刊号：CN 21-1531/P
气象与环境学报

气象与环境学报

• 论文 • 上一篇    下一篇

投影寻踪回归与BP神经网络方法在前汛期降水预测中的比较研究

杨永生1;何平2   

  1. 1.清远市气象局 广东清远511515;2.本溪市气象局 辽宁本溪117000
  • 收稿日期:2007-02-08 修回日期:2007-10-11 出版日期:2008-02-01 发布日期:2008-02-01

Comparison on early flood season precipitation prediction between projection pursuit regression method and back propagation neural network method

YANG Yong-sheng1 HE Ping2   

  1. 1.Qingyuan Meteorological Bureau;Qingyuan 511515;China;2.Benxi Meteorological Bureau;Benxi 117000;China
  • Received:2007-02-08 Revised:2007-10-11 Online:2008-02-01 Published:2008-02-01

PDF (PC)

719

摘要： 以1962—2006年粤北地区7个站4—6月前汛期降水量资料为基础,将前汛期降水量与74项环流指数资料进行灰色关联度分析,确定了影响粤北地区前汛期降水量的16个关键环流指数因子,分别应用投影寻踪回归、BP神经网络和逐步回归方法,建立前汛期降水趋势预测模型,对粤北地区前汛期降水趋势进行预测。结果表明:投影寻踪回归和BP神经网络方法的预测能力均优于传统的逐步回归模型。其中,PPR模型比BP神经网络方法的预测效果更好。

关键词: 投影寻踪回归, BP神经网络, 灰色关联度, 前汛期降水

Abstract: According to early flood season precipitation data(from April to June) from seven weather stations in northern Guangdong province from 1962 to 2006,the relationships between early flood season precipitation and 74 circumfluence indexes were analyzed by grey interconnect degree method.And 16 key general circulation indexes influencing early flood season precipitation in the study area were extracted.Early flood season precipitation models were established by the projection pursuit regression method,back propagation neural network method and stepwise regression method.And early flood season precipitation trends were predicted by these models.The results show that the prediction accuracy from the projection pursuit regression method and back propagation neural network method is much better than that from the stepwise regression method.Among three methods,the projection pursuit regression method is the best one.