首页 | 本学科首页   官方微博 | 高级检索  
     

基于Copula函数的柘林水库设计洪水频率分析
引用本文:徐〓琨,陆宝宏,汪集旸,邓〓山,赵〓超,崔冬梅,毛〓豆,吴星鑫.基于Copula函数的柘林水库设计洪水频率分析[J].水电能源科学,2014,32(8):61-64.
作者姓名:徐〓琨  陆宝宏  汪集旸  邓〓山  赵〓超  崔冬梅  毛〓豆  吴星鑫
作者单位:河海大学 水文水资源学院, 江苏 南京 210098;河海大学 水文水资源学院, 江苏 南京 210098;河海大学 水文水资源学院, 江苏 南京 210098; 中国科学院地质与地球物理研究所, 北京 100029;河海大学 水文水资源学院, 江苏 南京 210098;河海大学 水文水资源学院, 江苏 南京 210098;河海大学 水文水资源学院, 江苏 南京 210098;河海大学 水文水资源学院, 江苏 南京 210098;河海大学 水文水资源学院, 江苏 南京 210098
基金项目:国家自然科学基金项目(NSFC 50979023);水利部公益性科研项目(201201026)
摘    要:针对传统设计洪水频率分析采用单变量方法且不考虑变量间相关关系的问题,以江西省柘林水库坝址洪水为研究对象,采用Frank Copula函数构建洪峰与5d洪量的二元联合分布,并引入联合重现期的概念。结果表明,柘林水库坝址洪水主要受洪峰和洪量控制,峰量间存在良好的相关性;两变量方法推求的峰量联合设计值均大于单变量方法,各特征量的重现期均高于设计标准。可见Copula函数拟合度较高,计算所得频率设计值更符合洪水事件的内在规律以及工程实际要求,且同时可得不同情况下的峰量设计值组合,在水文领域具有广阔的应用前景。

关 键 词:设计洪水    Copula函数    频率分析    联合分布    Frank  Copula    极大似然估计

Design Flood Frequency Analysis of Zhelin Reservoir Based on Copula Functions
XU Kun,LU Baohong,WANG Jiyang,DENG Shan,ZHAO Chao,CUI Dongmei,MAO Dou and WU Xingxin.Design Flood Frequency Analysis of Zhelin Reservoir Based on Copula Functions[J].International Journal Hydroelectric Energy,2014,32(8):61-64.
Authors:XU Kun  LU Baohong  WANG Jiyang  DENG Shan  ZHAO Chao  CUI Dongmei  MAO Dou and WU Xingxin
Abstract:Univariate frequency analysis method is commonly used in traditional design flood frequency analysis. This paper takes dam site flood of Zhelin reservoir in Jiangxi province as research object. According to the deficiency of ignoring the relationship between variables, a new method deriving flood frequency design values is proposed. The fundamental theory and key technologies in application of Copulas are introduced. In the method, Frank Copula is adopted to construct the bivariate joint distribution of flood peak and volume and the concept of bivariate return period is defined. The result shows that the dam site flood of Zhelin reservoir is commonly controlled by flood peak and volume and the two values are greatly related; the fitting degree of Frank Copula in the study area is high by inspection; the frequency design values based on the new method are larger than the ones based on traditional method; so are the return periods; taking bivariate return period into consideration, the new method is more feasible and accords with the inherent rules of hydrological events as well as the practical engineering; combination of design values in different cases can be obtained. As an effective tool constructing multivariate joint distribution, it has wide application prospect in hydrology.
Keywords:design flood  Copula functions  frequency analysis  joint distribution  Frank Copula  maximum likelihood estimation
本文献已被 CNKI 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号