基于联合分布的设计暴雨方法  

Design Storm Method Based on Bivariate Joint Distribution

作　　者：张娜[1] 郭生练[1] 肖义[1] 方彬[1] 闫宝伟[1] 

机构地区：[1]武汉大学水资源与水电工程科学国家重点实验室,湖北武汉430072

出　　处：《水力发电》2008年第1期18-21,共4页Water Power

基　　金：国家自然科学基金资助项目(50679063);科技部国际合作项目(2005DFA20520)

摘　　要：利用Gumbel-Hougaard Copula函数构建边缘分布均为PIII型分布的年最大日雨量与年最大七日雨量之间的联合分布。以隔河岩43年降雨资料为例进行分析,结果表明,年最大日雨量与年最大七日雨量联合观测值的理论分布和经验频率拟合较好。在年最大日雨量和年最大七日雨量的边缘分布和理论分布的基础上,推求两变量的联合分布、联合重现期,并基于联合分布推求设计暴雨值,以更加准确地根据设计条件推求设计暴雨过程。The Gumbel-hougaard Copula function was used to analyze the joint distributions between annual maximum daily storm and seven-day storm amounts which are both with Pearson type Ⅲ marginal distributions. The rainfall data of the Geheyan reservoir during 1960 to 2002 were selected as a case study. The results show that a good agreement can be observed between theoretical joint and empirical distribution of annual maximum daily storm and the seven-day storm amounts. On the basis of the marginal and theoretical distributions of these two variables, the joint probability distribution and the related return periods of two variables can be deduced easily, much more exact design storm and storm processes can be obtained at certain condition.

关 键 词：设计暴雨 COPULA 年最大日雨量 联合分布 

分 类 号：TV125[水利工程—水文学及水资源]