论文部分内容阅读
针对实时洪水预报系统中实时校正模型存在一定的适用范围却又难以在线优选的问题,首先分析了自适应滤波模型、马斯京根矩阵模型和反馈模拟模型3种传统实时校正模型的原理与适用性,并编制算法流程图,为代码编程提供依据。其次,提出了一种实时校正模型的优选方法,适用于大流域多校正断面的实时洪水预报系统,可以为各断面在线优选出最优的实时校正模型。在WDS9002三峡实时洪水预报系统中的应用结果表明:优选方法在校正成功率和各项精度评价指标统计上出现最优的频次高于传统实时校正模型,且很少出现指标统计最差的情况;另外,优选方法有可能减小过校正现象的概率。
In order to solve the problem of real-time correction model in real-time flood forecasting system, it is difficult to optimize on-line. First, the principle and application of three traditional real-time correction models of adaptive filter model, Muskingum matrix model and feedback simulation model are analyzed Sex, and the preparation of algorithm flow chart, provide the basis for code programming. Secondly, a method of real-time correction model is proposed, which is suitable for real-time flood forecasting system with multiple correction sections in large river basins. The optimal real-time correction model can be optimized for each section. The results of the application in WDS9002 Three Gorges real-time flood forecasting system show that the optimum frequency of the optimal method is higher than that of the traditional real-time correction model in terms of the calibration success rate and the accuracy evaluation index, In addition, the preferred method makes it possible to reduce the probability of overcorrection.