琼州海峡跨海大桥桥址处风雨联合概率分布研究

Research of Wind Velocity and Rainfall Joint Probability Distribution at Qiongzhou Strait Bridge Site

选取琼州海峡跨海大桥桥址处观测的36组台风最大10 min平均风速和最大小时降雨量以及澄迈气象站30年来的年最大10 min平均风速和日最大降雨量为样本,确定了风速、雨量的极值Ⅰ型边缘概率分布。采用Copula函数建立了风速、雨量联合分布模型,通过拟合优度比较,得到了桥址处与气象站处的风雨联合分布模型的最优连接函数分别为Gumbel Copula函数和Clayton Copula函数。根据风雨联合分布模型,对琼州海峡大桥风雨极值重现期进行了分析。结果显示,观测塔处极值风速比澄迈气象站处大72%～95%,表明建立桥址观测系统是十分必要的;边缘分布重现期计算出的风速、雨量值小于联合分布重现期的计算值,随重现期的增大,二者差值有减小的趋势。

Thirty-six groups of measurement typhoon data including the maximum 10-minute average wind speed and maximum hourly rainfall of Qiongzhou Strait Bridge site are selected. At the same time,the annual maximum 10-minute average wind speed and maximum daily rainfall in the past 30 years at Chengmai weather station are sampled. Due to above data,the extreme-I marginal probability distributions of the wind speed and rainfall are determined. The joint distribution model of the wind speed and rainfall are established by Copula function. Then,comparing their goodnesses of fit,their optimal contiguous functions of wind-rain joint distribution model at the bridge site and the weather station are Gumbel Copula function and Clayton Copula function separately. The wind-rain extreme value return period of Qiongzhou Strait Bridge is analyzed based on the wind-rain joint distribution model. It shows that（ 1） the extreme wind speed at observation tower is 72%- 95% greater than that at Chengmai station,which indicates that it is very necessary to eastalish the wind-rain measurement system at bridge site;（ 2） the predicted wind speed and rainfall of the margin distribution return period are less than the calculated values of joint distribution return period,as the return period increasing,it exists a decreasing trend between them.