欢迎访问中国科学院大学学报,今天是

中国科学院大学学报 ›› 2010, Vol. 27 ›› Issue (1): 17-26.DOI: 10.7523/j.issn.2095-6134.2010.1.003

• 论文 • 上一篇    下一篇

遥测降雨异常值的三步抗差统计探测

赵超1,2, 洪华生2, 朱木兰1   

  1. 1. 厦门理工学院水资源环境研究所, 厦门 361005;
    2. 近海海洋环境科学国家重点实验室, 厦门大学环科中心, 厦门 361005
  • 收稿日期:2008-11-06 修回日期:2009-09-03 发布日期:2010-01-15
  • 通讯作者: 赵超
  • 基金资助:

    国家自然科学基金(50909084)、福建省自然科学基金(2009J05107)和校级引进人才项目(YKJ08015R)资助 

A three-stepwise robust statistical method for outlying rainfall observation

ZHAO Chao1,2, HONG Hua-Sheng2, ZHU Mu-Lan1   

  1. 1. Water Resources and Environmental Institute, Xiamen University of Technology, Xiamen 361005, China;
    2. State Key Laboratory of Marine Environmental Science, Environmental Science Research Center, Xiamen University, Xiamen 361005, China
  • Received:2008-11-06 Revised:2009-09-03 Published:2010-01-15

摘要:

由于遥测降雨系统自身的原因,遥测降雨资料中常有异常值的出现.充分利用降雨分布特征以及抗差统计理论,提出一种三步抗差统计方法探测遥测降雨资料中的异常值.本方法采用Tukey fence统计方法抵御异常值的干扰,用三步的形式以适应降雨资料的分布特征.对面平均降雨量进行的分组,进一步提高方法的探测效率.数据证明,新方法的探测效果较好,且符合水文预报要求.

关键词: 遥测系统, 异常值, Tukey fence, 分布特征, 三步抗差探测

Abstract:

A three-stepwise robust statistical method combining the robust statistical theory with distribution features of rainfall for detection of outliers in telemetry system is described. The proposed robust statistical method adopts the Tukey fence insensitive to outliers as identification bounds and presents a three-stepwise pattern to adapt the distribution of rainfall data. Moreover, the modified method based on dividing precipitation data into several groups further improves detection efficiency. The results show that the new method is suitable to the hydrological need.

Key words: telemetry system, outlier, Tukey fence, distribution feature, three-stepwise robust detection

中图分类号: