降水数值预报和水文预报耦合效果定量评价方法

Quantitative Evaluation Method for Coupling Effect between Rainfall Prediction and Hydrological Forecast

基于信息熵的基本原理,改造了信息熵函数,针对系统耦合中的诸多不确定性因素,提出了一种耦合效果的定量评估方法,利用自信息量评估各节点完成预报工作的能力,选取降水数值预报信息的获取能力、降水数值预报信息的处理能力、预报系统的调用能力和系统耦合预报成果的存储与输出能力描述工作能力,并采用数理统计方法确定各节点内不确定性因素的隶属度。实例应用结果表明,该方法能定量估算由于预见期的延长对水文预报耦合成果不确定性的影响,为耦合成果的精度评定和应用提供了客观的评价。

Based on the basic theory of information entropy, a method is proposed to quantitative evaluate the coupling effect, which is realized through the reform of the information entropy function. The forecasting capabilities of every point are evaluated by the self-information, which include the obtain information ability of rainfall prediction, the informa- tion processing ability of rainfall prediction, the system calling ability and the input and output ability of the coupling sys- tem. The membership of uncertain factors is determined by the mathematical statistics method. An example results show that the proposed method can estimate the uncertain effect of hydrological forecast owing to extending of forecast period, which provides supporting for accuracy evaluation and the application ov the forecast result.