基于动态权重的水文时间序列相似性分析方法  

Similarity Analysis Method of Hydrological Time Series Based on Dynamic Weight

作　　者：赵圆元 张行南[1] 黄钰凯 舒卫民[2] 夏达忠[1] ZHAO Yuan-yuan;ZHANG Xing-nan;HUANG Yu-kai;SHU Wei-min;XIA Da-zhong(College of Hydrology and Water Resources,Hohai University,Nanjing 210098,China;China Yangtze Power Co.,Ltd.,Yichang 443000,China)

机构地区：[1]河海大学水文水资源学院,江苏南京210098 [2]中国长江电力股份有限公司,湖北宜昌443000

出　　处：《水电能源科学》2021年第2期9-12,17,共5页Water Resources and Power

基　　金：国家重点研发计划(2019YFC0409000)。

摘　　要：水文时间序列相似性分析对于洪水预报、防洪调度等具有参考作用。针对洪水特征的关注重点随研究问题的不同而变化,构建了由过程形态和特征值组成的相似指标体系和加权综合相似性度量指标,提出了基于动态权重的相似性分析方法。引入混合交叉赋权法的原理,在不同研究问题中对相似指标赋予不同权重,然后利用相似指标结合权重查找出相似水文时间序列。以嘉陵江北碚水文站历史洪水过程在不同研究问题时的相似性分析为例,对该方法进行验证,结果表明基于动态权重的水文时间序列相似性分析方法可有效识别相似洪水.在实际问题分析中具有可行性。The similarity analysis of hydrological time series plays an important role in flood forecasting and flood.control operation.The focus on flood characteristics varies with different research issues.The similarity index system composed of process shape and characteristic value and the weighted comprehensive similarity measurement index were constructed.A similarity analysis method was proposed based on dynamic weight.The principle of hybrid cross weighting method was introduced to give different weights to similar indexes for different research problems.The similar indexes combined with weight were used to find out similar hydrological time series.Taking the similarity analysis of historical flood process of Beibei station in Jialingjiang River as an example to verify the method,the results show that the proposed method can effectively identify similar floods,which is feasible in practical problem.

关 键 词：水文时间序列 过程形态相似指标 特征值相似指标 混合交叉赋权法 动态权重 嘉陵江 

分 类 号：TV122[水利工程—水文学及水资源]