不同时期水文模型参数的不确定性研究  

Study on Uncertainty of Hydrological Model Parameters in Different Periods

作　　者：李鑫 闫成山 朱龙腾 孙伟 赵学贤 LI Xin;YAN Cheng-shan;ZHU Long-teng;SUN Wei;ZHAO Xue-xian(Shuifa Planning and Design Company,Jinan 250100,China)

机构地区：[1]水发规划设计有限公司,山东济南250100

出　　处：《水电能源科学》2020年第11期17-21,共5页Water Resources and Power

摘　　要：水文模拟精度在不同气候时期不同量级洪水过程下表现形式不同,以汉江安康以上流域为例,采用综合差分样本分离方法(CDSST)分离出流域不同气候时期,并运用基于贝叶斯框架的DREAM算法探究了参数不确定性对不同气候时期不同量级径流模拟的影响。结果表明,参数不确定性在不同气候时期对径流模拟的影响差异明显,混合期影响最大,湿润期次之,干旱期最小;参数不确定对不同量级径流的影响差异较大,高流量时,参数的不确定性影响小,低流量时影响大,且干旱期和混合期均大于湿润期。Hydrological simulation accuracy has different characteristics in different climate periods under different magnitudes of floods.Taking the upstream of Ankang in the Hanjiang Basin as an example,the comprehensive differential sample separation method(CDSST)was used to separate the different climatic periods,and the DREAM algorithm based on the Bayesian framework was adopted to explore the impact of parameter uncertainty on different magnitude runoff simulations in different climatic periods.The results show that the influence of parameter uncertainty on runoff simulation in different climatic periods is obvious.The mixing period has the largest impact,followed by the wet period and the dry period.Meanwhile,the effect of parameter uncertainty on runoff of different magnitudes is quite different.At high flow,the uncertainty of the parameter has a small effect,and low flow has a large effect,and both the dry and mixed periods are greater than the wet period.

关 键 词：动态参数 DREAM算法 参数不确定性 汉江流域 

分 类 号：TV213.9[水利工程—水文学及水资源]