计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (12): 234-237.

• 工程与应用 • 上一篇    下一篇

基于PCA和改进BP网络的降雨预报模型研究

刘 乐1,王洪国1,2,王宝伟3   

  1. 1.山东师范大学 管理与经济学院,济南 250014
    2.山东省科学技术厅,济南 250011
    3.山东师范大学 信息科学与工程学院,济南 250014
  • 收稿日期:2007-08-29 修回日期:2007-10-22 出版日期:2008-04-21 发布日期:2008-04-21
  • 通讯作者: 刘 乐

Research in rain forecasting model based on PCA and improved BP network

LIU Le1,WANG Hong-guo1,2,WANG Bao-wei3   

  1. 1.School of Management and Economy, Shandong Normal University,Ji’nan 250014,China
    2.Department of Science and Technology of Shandong Province,Ji’nan 250011,China
    3.School of Information Science and Engineering, Shandong Normal University,Ji’nan 250014,China
  • Received:2007-08-29 Revised:2007-10-22 Online:2008-04-21 Published:2008-04-21
  • Contact: LIU Le

摘要: 在主成分分析法和改进BP网络相结合的基础上,进行降雨预报模型的研究。先由主成分分析法降低原始气象数据的维数,然后利用改进BP网络有效地学习气象样本数据中蕴含的内在规律。研究结果显示,该降雨预报模型训练效率高,预报效果好。

关键词: 主成分分析, BP网络, 降雨预报

Abstract: On the base of combining Principal Component Analysis with improved BP network,this paper made a research on the rain forecasting model.First the dimensions of the raw meteorological data were decreased by PCA.Then it was by improved BP network to learn the potential rules which existd in meteorological samples effectively.The result of the research shows that,the rain forecasting model has high training efficiency and good forecasting effect.

Key words: Principal Component Analysis(PCA), Back-Propagation network, rain forecasting