论文部分内容阅读
对卡曼滤波(KALMAN FILTER)和坦克(Tank)模型的联合应用进行了研究。选用状态矢量代表坦克模型中的参数及其初值,并用试错法对其进行估计。卡曼滤波与坦克模型相联合并应用递推算法求解,滤波允许模型参数随时间而变化,从而减少了流域降雨径流过程的物理不确定性。该方法在海河流域峪河口水文站的洪水预报中应用表明,取得了较为满意的计算精度。
The combined application of KALMAN FILTER and Tank model was studied. The state vector is used to represent the parameters of the tank model and its initial values, and it is estimated by trial and error. The Kaman filter is combined with the tank model and applied to the recursive algorithm. The filtering allows the model parameters to change over time, thus reducing the physical uncertainty of the rainfall runoff process. The application of this method to the flood forecast of the Yuhekou hydrological station in the Haihe River Basin shows that the method has achieved satisfactory calculation accuracy.