论文部分内容阅读
为研究广义极值分布模型(GEV)在黄河源区独特的寒区自然环境与冻土覆盖状态环境中的应用价值,基于黄河源区4个站点1956～2011年逐日径流量资料,分别以两参数对数正态分布曲线(LN2)、皮尔逊Ⅲ型曲线(P3)、对数皮尔逊Ⅲ型曲线(LP3)、GEV 4种枯水径流理论频率分布对最枯1、3、7、30、90d平均枯水流量进行拟合分析,并采用概率点距相关系数法(PPCC)和改进的均方根误差法(RMSE)进行量化裁定。结果表明,GEV已具备适应寒区枯水径流频率分析的能力,并在一定程度上优于LN2、P3、LP3概率分布函数,对高流量、高变差系数、长历时均表现出较好的稳定性能。
In order to study the application value of generalized extreme value distribution model (GEV) in natural environment and permafrost environment of cold source region in the Yellow River, based on daily runoff data from 1956 to 2011 of the Yellow River source region, Lognormal distribution curve (LN2), Pearson Ⅲ curve (P3), logarithm Pearson Ⅲ curve (LP3), GEV four kinds of low water runoff theoretical frequency distribution of the most dry 1,3,7,30 , 90d average dry flow were fitted and analyzed. The PPCC and RMSE were used to quantify the results. The results show that GEV has the ability to adapt to the frequency analysis of dry flow in cold regions and is better than the probability distribution function of LN2, P3 and LP3 to a certain extent, and shows good performance for high flow rate, high variation coefficient and long duration Stable performance.