考虑不确定度的POT模型在洪水重现期分析中的应用研究
发布时间：2018-11-27 16:01
【摘要】：以淮河紫罗山子流域出口日平均流量数据为研究对象,基于超阈值(POT)模型,采用最大似然法估计广义Pareto(GP)分布参数并计算出重现期水平和相应的置信区间范围。拟合优度检验结果显示POT模型在扩大洪水样本提高使用效率的同时,对样本经验点据的适线性也较好。通过对5种时段长度的水文实测流量数据重现期计算发现:实测数据长度对重现期计算结果不确定性有重要影响,在工程水文中推荐选取恰当的置信区间上界作为设计值加以解决。
[Abstract]:Based on the data of daily average discharge at the outlet of Ziluoshanzi basin of Huaihe River, the maximum likelihood method is used to estimate the generalized Pareto (GP) distribution parameters based on the super-threshold (POT) model, and the level of recurrence period and the corresponding confidence interval range are calculated. The results of goodness of fit test show that the POT model not only expands the flood sample to improve the efficiency, but also has a good linearity for the sample experience points. It is found that the length of the measured data has an important effect on the uncertainty of the calculated results through the calculation of the recurrence period of the hydrological measured discharge data of five periods of time. In engineering hydrology, it is recommended to select the appropriate upper bound of confidence interval as the design value to solve the problem.
【作者单位】： 环境保护部核与辐射安全中心;
【基金】：科技部国家软科学研究计划(2013GXS4B075) 环保部公益性行业科研专项(201309056)
【分类号】：TV122
本文编号：2361383
[Abstract]:Based on the data of daily average discharge at the outlet of Ziluoshanzi basin of Huaihe River, the maximum likelihood method is used to estimate the generalized Pareto (GP) distribution parameters based on the super-threshold (POT) model, and the level of recurrence period and the corresponding confidence interval range are calculated. The results of goodness of fit test show that the POT model not only expands the flood sample to improve the efficiency, but also has a good linearity for the sample experience points. It is found that the length of the measured data has an important effect on the uncertainty of the calculated results through the calculation of the recurrence period of the hydrological measured discharge data of five periods of time. In engineering hydrology, it is recommended to select the appropriate upper bound of confidence interval as the design value to solve the problem.
【作者单位】： 环境保护部核与辐射安全中心;
【基金】：科技部国家软科学研究计划(2013GXS4B075) 环保部公益性行业科研专项(201309056)
【分类号】：TV122
【相似文献】
相关期刊论文 前1条
1 E·赖尼斯 ,宋萌勃;用置信区间进行极值洪水的统计估算[J];长江水利教育;1991年04期
，本文编号：2361383
本文链接：https://www.wllwen.com/kejilunwen/shuiwenshuili/2361383.html
