基于二次重现期的城市两级排涝标准衔接的设计暴雨

Design storm for mixed level-1 and level-2 urban drainage standards based on the secondary return period

  • 摘要: 使用珠海市1984—2015年R1h-R6hR1h-R12hR1h-R24h3个历时暴雨组合推算排水排涝两级标准衔接的设计暴雨水平。应用阿基米德极值Copula与Kendall分布函数构建不同历时暴雨组合的联合概率分布模式。分析各历时暴雨组合的遭遇概率、"或"重现期、"且"重现期和二次重现期,以出现最大可能概率的方法推算各组合的设计暴雨值。结果表明:二次重现期所对应的累积频率更准确地代表了特定设计频率情况下不同历时暴雨组合的风险率;重现期分别为2年、3年、5年、10年、20年、50年、100年推算的二次重现期设计值介于"或"重现期和"且"重现期设计值之间,小于相应的边缘分布重现期设计值,R1h-R6 h组合推算的设计值相对误差为3.1%~7.1%;R1h-R12h组合推算的设计值相对误差为3.3%~9.3%;R1 h-R24 h组合推算的设计值相对误差为3.9%~12.0%。二次重现期推算的不同历时暴雨组合的设计暴雨分位值为内涝工程的风险管理和管渠尺寸提供了优选标准和设计参考。

     

    Abstract: In this paper, we estimated the design storm for mixed level-1 and level-2 urban drain standards according to data of the combination of 1-hour rainstorm and 6-hour rainstorm (R1h-R6h), the combination of 1-hour rainstorm and 12-hour rainstorm(R1h-R12h), and the combination of 1-hour rainstorm and 24-hour rainstorm(R1h-R24h) in Zhuhai city between 1984 and 2015. First, the models of joint probability distribution about different combinations were built using the Archimedean extreme value Copula and the Kendall distribution function. Then, the occurrence probability, "OR" return period, "AND" return period and the secondary return period for each combination were analyzed, and the design storms for each combination were estimated by the method with the maximum occurrence probability. The results are as follows: the cumulative frequency corresponding the secondary return period more accurately represents the risk probabilities of rainstorms of different combinations with specific design frequency; the estimated design storm quantiles of the secondary return period based on return periods with 2-year, 3-year, 5-year, 10-year, 20-year, 50-year and 100-year are between the design storm quantiles of the "OR" return period and these of the "AND" return period, and less than these of the marginal distribution return period, respectively. Compared with the case of design storm quantiles of the marginal distribution return period, the relative error of the R1h-R6h is between 3.1% and 7.1%; that of the combination of R1h-R12h is between 3.3% and 9.3%; that of R1h-R24h is between 3.95% and 12.0%. These provide optimization criterion and guide for the risk management of waterlogging project and the design of drainage pipe, respectively.

     

/

返回文章
返回