一维非恒定水动力学模型的实时洪水预报

本文建立了一维非恒定水动力学模型与最小二乘法耦合的实时洪水预报模型。运用最小二乘的递推算法，引入时变遗忘因子实时跟踪模型中时变参数的变化，并动态修正参数，实时改变一维非恒定水动力学模型结构，对洪水预报进行实时修正。通过长江干流宜昌～螺山段洪水实时预报应用．改进后模型预报精度比原一维非恒定水动力学模型计算结果有很大提高，表明所建立的模型用于洪水预报是行之有效的，在洪水预报中具有推广应用价值。

Real-time forecasting method based on 1D unsteady channel flow

In this paper, a real-time forecasting method coupled with model of ID unsteady flow with recursive least-squares is developed. The ID unsteady flow model is modified by using the time-variant parameter and revised it dynamically through introducing a variable weighted forgetting factor, that the output of the model can he adjusted for real time forecasting of floods. Application of the new real time forecasting model in the reach from Yichang to Luoshan of the Yangtze River is demonstrated. Computational result shows that the forecasting accuracy of the new model is much higher than that of the original hydraulics model of ID unsteady flow. The method developed is effective for flood forecasting, and can he used for practical operation in the flood forecasting.