| 摘要: |
| 【目的】 考虑洪峰、洪量和历时之间的相依关系，应用Archimedean Copula函数，探讨洪水3变量的联合概率分布和条件概率分布，为水利工程规划设计和风险评估提供依据。【方法】 以陕北地区窟野河流域神木水文站48年的洪水资料为例，选择拟合效果好的Copula函数计算洪峰、洪量和历时的联合概率分布、条件概率分布，并绘制相应的条件重现期图。【结果】 Gumbel-Hougaard Copula函数拟合洪峰、洪量和历时的联合概率分布效果较好；得到不同条件组合下洪水事件的条件概率分布和相应的条件重现期，同时得到确定条件下的重现期均小于超过该条件时的重现期。【结论】 Copula函数全面考虑了洪水特征量之间的相依性，可用于洪峰、洪量和历时联合概率分布的计算。 |
| 关键词: 洪水频率分析 Copula函数 联合概率分布 重现期 |
| DOI： |
| 分类号: |
| 基金项目:国家自然科学基金项目（50879070，50579065） |
|
| Research of trivariate flood frequency based on Copula function |
|
|
| Abstract: |
| 【Objective】 Considering dependent relationship among peak,volume and duration,we analyzed the three variables joint probability distribution and the condition probability distribution by Archimedean Copula function,which can provide the basis for the design of hydraulic engineering program and risk assessment.【Method】 Take 48 years’ flood data in Shenmu in the north of Shaanxi as an example.A good fitting Copula function was chosen to calculate the condition probability distribution of flood peak,volume and duration and drawn figs of corresponding return period.【Result】 Gumbel-Hougaard Copula function can fit joint distribution of flood peak volume and duration better.The condition probability distribution and the corresponding return period were obtained and the return period under fixed condition was less than the one when the conditions were above the value.【Conclusion】 Copula function that considered the dependecy of the flood characteristics comprehensively was used to calculate joint probability distribution among peak,volume and duration. |
| Key words: flood frequency analysis Copula function joint probability distribution return period |
