高阶概率权重矩在洪水频率分布参数估计中的应用  

Application of higher probability weighted moments in the estimation of flood frequency distribution parameters

作　　者：胡素端[1,2] 宋松柏[1] 

机构地区：[1]西北农林科技大学水利与建筑工程学院,陕西杨凌712100 [2]大连理工大学水利工程学院,辽宁大连116024

出　　处：《西北农林科技大学学报（自然科学版）》2013年第6期207-212,220,共7页Journal of Northwest A&F University(Natural Science Edition)

基　　金：国家自然科学基金项目(51179160;50879070;50579065);高等学校博士学科点专项科研基金项目(20110204110017)

摘　　要：【目的】研究高阶概率权重矩在洪水频率分布参数估计中的应用,为推求洪水频率分布参数估计提供合理的计算方法。【方法】选取广义极值分布作为洪水频率分布模型,推求其高阶概率权重矩计算公式,利用MATLAB语言编程实现其参数的数值求解,并以陕西省赵石窑、绥德、安塞和鹦鸽4个水文站的年最大洪峰流量序列为例进行洪水频率分析。【结果】高阶概率权重矩可以较好地拟合序列中大洪水段的经验点据,计算得到了基于高阶概率权重矩的陕西省赵石窑、绥德、安塞和鹦鸽4个水文站的广义极值分布参数的估计结果,获得WLS准则下赵石窑、绥德、安塞和鹦鸽4个水文站推荐采用的高阶概率权重矩的阶数r分别为:r=4,5,6;r=4,5,6;r=3,4,5和r=3,4,5。【结论】高阶概率权重矩会给予大洪水值较大的权重,可以用于洪水频率分布参数的估计。【Objective】 This paper investigated the estimation of flood frequency distribution parameters based on higher probability weighted moments（PWMs）,aiming to provide a reasonable method for calculating flood frequency distribution parameters.【Method】 Choosing the generalized extreme value（GEV） distribution as flood frequency distribution,the calculation formula with higher PWMs was deduced,and a numerical method using MATLAB programs to estimate its parameters was presented.Then,the method was applied to estimate parameters of annual maximum peak flow series at 4 stations in northern Shaanxi：Zhaoshiyao,Suide,Ansai and Yingge.【Result】 The calculated distribution based on higher PWMs were closer to the larger flows in the sample.The estimated parameters of the GEV distribution based on higher PWMs were obtained for Zhaoshiyao,Suide,Ansai and Yingge stations and their recommended orders（r=4,5,6;r=4,5,6;r=3,4,5 and r=3,4,5,respectively） were given under the standard of WLS.【Conclusion】 Higher PWMs can be used for the estimation of flood frequency distribution parameters.

关 键 词：高阶概率权重矩 洪水频率 参数估计 广义极值分布 

分 类 号：P333.2[天文地球—水文科学]