【摘要】 建立了基于年径流随机模拟的月径流分解模拟模型,以平寨水库入库径流随机模拟为例,分别采用双层模型法和典型解集法分解模拟月径流,分析模拟结果的统计特性和不同频率下的年内分配过程,比较两种方法的分解效果。结果表明:1典型解集分解法较双层模型分解法能更好的保持月径流的统计特性,在各项参数中,两种分解方法对样本均值特性的保持最好,而对偏态特性的保持最差;2典型解集分解法得到的月径流年内分配过程与实测同频率下月径流年内分配的符合程度较双层模型分解法高,两种方法分解的月径流在偏丰水年的年内分配与实测同频率下月径流年内分配的符合程度低于偏枯水年,符合较差的月份主要集中在汛期(5-9月)。更多还原

【Abstract】 Based on the stochastic simulation of annual runoff,the decomposition of monthly runoff simulation model is built. This paper takes the Ping-zhai Reservoir’s monthly runoff stochastic simulation for example. By using the double model simulation method and the typical method of solution set decomposition on annual runoff. the statistical features of the simulation results and the allocation process under different frequency years are analyzed,then two methods’ synthesis effect are compared. In keeping the monthly runoff statistical properties,the typical method of solution is better than the double model simulation method. In various parameters,the sample mean features of two decomposition methods keep the best,but the characteristics of skewness is the worst; The typical method of solution to get the monthly runoff years allocation process with the measured frequency distribution of runoff years next month in accordance with a high degree of the double model simulation method,the decomposition of the two methods of monthly runoff distribution in partial wet years with the measured frequency years next month runoff distribution of the operator is less than the partial dry years. The poor months are mainly concentrated in the flood season(May-September).更多还原