2维水动力模型中入流边界优化处理方法

在极端暴雨事件频率和强度不断加剧的趋势下，防洪安全面临着极其严峻的考验，研究河道洪水演进、漫溢及淹没过程，有效评估洪水灾害风险就显得尤为重要。为精确模拟河道洪水演进过程，本文采用基于非结构网格有限体积法建立的二维水动力模型，构建了一种基于虚拟单元法的入流边界改进方法。该方法采用基于地形网格数据的纵断面提取法以近似确定深泓线的位置，从而计算出河道平均纵比降；在入流边界各网格上以均匀流的方式入流，计算出不同流量下的对应水位；依据各网格上的水深和入流边长计算权重以合理地分配流量，再结合单宽流量边界条件确定边界外侧虚拟单元上的水力要素，最后耦合至模型通量部分进行计算，实现入流模块的优化。将该方法应用于Toce河物理模型试验的模拟，在入流边界附近计算出的水位、流速和流态都更为合理；河道中各测点水位与试验数据吻合度较高，洪水到达测点时间及各点间洪水运动传播时间与测量值较为接近，纳什效率系数在0.82~0.95之间，验证了该方法的可行性和模型的准确性；并且在单机上用时86s完成计算，效率较高，表明模型能高效准确的模拟洪水演进过程。通过对入流模块的改进优化，模型可以更好地应用于洪水演进过程的模拟预警及灾情评估工作，为洪水灾害管理决策提供技术支撑。

Inflow Boundary Optimized Method in Two-dimensional Hydrodynamic Model

With the increasing frequency and intensity of extreme rainstorms, the flood management situation has become extremely severe. It is particularly important to study the process of river flood propagation, overflow and inundation and evaluate the risk of flood disaster effectively. To simulate the river flood propagation accurately, a two-dimensional hydrodynamic model using the finite volume method based on unstructured grids was applied in this paper, proposing an improved inflow boundary processing method based on Ghost-Cell Method. The method of profile extraction based on terrain grid data was used to determine the position of thalweg approximately, and thus to calculate the average gradient of the river. The corresponding water level under different discharge was calculated by defining the flow on the grid of inflow boundary is the uniform flow. The discharge per unit width was distributed reasonably by calculating the weight according to the water depth and inflow boundary length of each grid, so then determined the hydraulic elements on the ghost cell outside the boundary, and finally coupled to flux part of the model for calculation. The method was applied to the simulation of the physical model test of the Toce River, and the flow state near the inflow boundary was more reasonable. The water level at gauging points in the river was in good agreement with the measured data. The arrival time and travel time of flood at gauging points was close to the measured values. The NSE (Nash -Sutcliffe efficiency coefficient) was between 0.82 and 0.95, which verified the feasibility of the method and the accuracy of the model. And it took 86 s to complete the calculation on the personal computer, which shows that the model can simulate flood propagation process efficiently and accurately. Through the improvement and optimization of the inflow module, this model can be better applied to the simulation, early warning and disaster assessment of flood propagation process, and provides more powerful technical support for flood disaster management and decision.