基于径流丰枯分类的月水量平衡模型参数响应研究  

Responses of monthly water balance model parameter based on discharge classification

作　　者：张静文[1] 郭家力[1,2] 章梦杰 戴凌全[1] 林凯荣[3] ZHANG Jingwen;GUO Jiali;ZHANG Mengjie;DAI Lingquan;LIN Kairong(College of Hydraulic and Environmental Engineering, China Three Gorges University, Yichang443002,Hubei, China;Center of Land Hydrological Processes & Adaptive Regulation, China Three Gorges University, Yichang443002, Hubei,China;Department of Water Resources and Environment, Sun Yat-Sen University, Guangzhou510275,Guangdong, China)

机构地区：[1]三峡大学水利与环境学院,湖北宜昌443002 [2]三峡大学陆地水循环过程及适应性调控研究中心,湖北宜昌443002 [3]中山大学水资源与环境系,广东广州510275

出　　处：《水利水电技术》2019年第1期81-89,共9页Water Resources and Hydropower Engineering

基　　金：国家自然科学基金项目(51509141;51779279);中国电建集团重大专项课题(DJ-ZDZX-2016-02-09);梯级水电站运行与控制湖北省重点实验室开放基金项目(2015KJX02);三峡大学学位论文培优基金项目(2018SSPY009)

摘　　要：针对变化环境下水文模型参数的时间变异性如何影响径流模拟的准确性问题,提出依据径流丰枯分类后对水文模型分别率定和验证。通过常规法和集对分析法对清江流域径流丰枯分类,并将不同径流丰枯年份对应降水、蒸发输入到两参数月水量平衡模型中,率定各丰枯类型下的模型参数值,并分析不同频率的径流差异情况。结果表明:集对分析法分类后两参数月水量平衡模型的C参数值规律为枯水年>平水年>丰水年,常规法则为枯水年<平水年<丰水年; SC参数值的规律在两种方法下表现类似。从年际变化看,模型效率系数基于集对分析法分类时较大,且对于不同丰枯年份下的径流模拟更接近实测值。从年内分配看,基于集对分析法的不同频率高、中、低水相对误差综合表现优于常规法。基于此进行的交叉模拟进一步发现,使用比当前状态偏湿润年份的参数值,得到的径流响应均为增大;反之则径流响应均为减小,且所使用参数值条件与基准条件变化的幅度越大,径流响应增大或者减小的幅度就越大。Aiming at the problem of that how the temporal variability of hydrological model parameter under changing environment to impact the accuracy of runoff simulation, it is proposed herein that the parameters of hydrological model are calibrated and verified respectively in accordance with the runoff (wet-dry ) classification. Through traditional method (TRA ) and set pair analysis method (SPA ), the runoff (wet-dry ) classification of Qingjiang River Basin is made, and then the precipitations and evaporations corresponding to the runoffs of different wet-dry years are input into the relevant two-parameter monthly water balance model for calibrating the model parameters under the conditions of all the wet-dry types and analyzing the runoff differences with different frequencies. The result shows that the law of the parameter value C of the two-parameter monthly water balance model after the classification made by set pair analysis method (SPA ) is dry year>normal year>wet year and that from traditional method (TRA ) is the same, while the laws of the parameter value SC from both the methods are similar. From the aspect of the intra-annual variation, the classification of set pair analysis method (SPA )-based model efficiency coefficient is larger and is more close to the measured value. From the aspect of the annual distribution, the relative errors of high, middle and lower waters from set pair analysis method (SPA ) are comprehensively better than those from traditional method (TRA ). It is further found out from the relevant crossing simulations made on the basis of this that all the runoff responses obtained from the adoptions of the parameter values of the year which is a little bit wet are increased, otherwise, they are decreased. Moreover, the larger the changing amplitudes of the conditions of the both the parameter values and the references are, the larger the amplitudes of the increase or decrease of the runoff response are to be.

关 键 词：径流 丰枯类型 集对分析法 变化环境 水文模型参数 

分 类 号：P333[天文地球—水文科学] P339[水利工程—水文学及水资源]