Copula函数在水库峰量联合分布研究中的应用

Applications of Copula Function in Joint Distribution of Reservoir Flood Peak and Volume

将“7·20”降雨产生的特大洪水特征值纳入单变量洪水频率分析,采用Copula函数构建联合分布,计算两变量联合分布函数和重现期。以常庄水库69年洪水资料为例,绘制洪峰、24小时洪量的两变量联合分布频率图和联合重现期等值线图。结果表明,Clayton Copula函数对峰量联合分布的拟合效果较好;同一设计重现期条件下,两变量联合分布洪水频率分析结果较单变量分析结果偏安全。

The characteristic values of the heavy flood generated by the 7·20 Rainfall were included in the univariate flood frequency analysis.In order to calculate the joint distribution functions and return periods of the two variables,Copula functions were used to build the joint distribution model.Taking 69a flood data of Changzhuang Reservoir as an example,the two-variables joint distribution frequency diagram of the flood peak and 24-hour flood volume and the joint return periods contour map were drawn.The results show that Clayton Copula function fits the flood peak-volume joint distribution well.The results of the two variables flood frequency analysis were safer than those of univariate analysis under the condition of the same design return periods.