二项-广义Pareto复合模型的极端海况要素推算  

Extremesea condition element estimation of binomial-generalized Pareto compound model

作　　者：邱玥 庞亮[1] 董胜[1] QIU Yue;PANG Liang;DONG Sheng(College of Engineering,Ocean University of China,Qingdao 266100,China)

机构地区：[1]中国海洋大学工程学院,山东青岛266100

出　　处：《海洋湖沼通报》2024年第1期47-52,共6页Transactions of Oceanology and Limnology

基　　金：国家自然科学基金项目(52171284)。

摘　　要：在复合极值分布理论的基础上,构造基于短期观测样本的二项-广义Pareto复合极值分布模型,并应用于极端海况要素推算。结果表明:二项-广义Pareto复合极值分布模型具有良好的拟合效果,能够合理反映极端海况的长期概率分布特征,弥补了传统方法需要长期原始海况数据的缺陷,且模拟结果与采用长期数据资料,利用Gumbel模型、对数正态模型得到的结果相差不大,在预测波高方面有很强的适用性。Based on the compound extreme value distribution theory,binomial-generalized Pareto compound extreme value distribution model based on short-term observation samples was constructed and applied to the prediction of extreme sea condition elements.The results showed that the binomial-generalized Pareto compound extreme value distribution model had good fitting effect,could reasonably reflect the long-term probability distribution characteristics of extreme sea conditions and made up for the defect of traditional methods need long-term raw sea condition data.The simulation results were similar with the results using long-time data,the Gumbel model and the lognormal model,and had strong applicability in the prediction of wave height.

关 键 词：复合极值分布 极端海况 短期资料 概率预测 设计波浪 

分 类 号：P731.33[天文地球—海洋科学]