论文部分内容阅读
针对水文模型参数不确定问题,选取干旱半干旱地区开都河流域为研究对象,提出了基于多项式混沌展开的水文模型参数敏感性分析方法。首先在开都河流域多年气象、水文观测数据的基础上建立SLURP水文模型;然后利用稀疏网格配置法获得配置点对模拟径流量进行多项式混沌展开;最后根据混沌系数计算Sobol指标评估模型参数及其交互效应对径流的影响。结果表明,在参数主效应中选取的参数中Sobol敏感度指标最大的是降雨系数,最小的是深层地下水截留常数;在交互效应中,浅层土壤蓄水容量和降雨系数之间交互效应最强,Sobol敏感度指标最大,浅层土壤截留常数与深层地下水截留常数之间交互效应最弱,敏感度指标最小。
In view of the uncertainty of hydrological model parameters, arid and semi-arid areas of the Kaidu River basin are chosen as research objects, and a hydrological model parameter sensitivity analysis method based on polynomial chaos expansion is proposed. Firstly, the SLURP hydrological model is established based on the meteorological and hydrological observations over the years in the Kaidu River basin. Then the sparse grid configuration method is used to obtain the configuration points for polynomial chaos expansion of the simulated runoff. Finally, the parameters of the Sobol indicator assessment model The effect of interaction on runoff. The results show that the largest Sobol sensitivity index among the parameters selected from the main effects of the parameters is the rainfall coefficient, and the minimum is the groundwater cut-off constant of the deep layer; in the interaction effect, the interaction effect between the shallow soil water storage capacity and the rainfall coefficient is the strongest , Sobol has the highest sensitivity index, the weakest interaction effect between shallow soil cut-off constant and deep groundwater cut-off constant, and the smallest sensitivity index.