Aiming at resolving the grid problems caused by the inconsistent resolution requirements when simulating overland flows using the 2D shallow water equations, a novel grid generation method based on multi-resolution data fusion is developed in this work. This method is able to not only reduce the computational burden associated with uniform structured grids but also ensure the simulation accuracy of the hydrodynamic model by reproducing the so-called small-scale effect. The efficiency of the method is assessed using different cases. Theoretical and laboratory cases demonstrate that fused non-uniform structured grids can reproduce hydrographs without appreciable accuracy losses. In addition, a high simulation accuracy (NRMSE ≤ 10.40 %, R² ≥ 0.87) is achieved in the simulation of a real flood event. The performance of this method is very promising in terms of the large-scale flood simulation accuracy, and it significantly reduces the data requirements and computational burden with globally fine uniform grids.

针对二维浅水方程模拟陆上流时分辨率要求不一致导致的网格问题，提出了一种基于多分辨率数据融合的网格生成方法。该方法不仅能够减少与均匀结构化网格相关的计算负担，而且通过再现所谓的小尺度效应来确保水动力模型的模拟精度。使用不同的案例评估该方法的效率。理论和实验室案例表明，融合的非均匀结构化网格可以在没有明显精度损失的情况下重现水位线。此外，高仿真精度（NRMSE  ≤ 10.40 %，R ² ≥ 0.87）是在模拟真实洪水事件中实现的。该方法的性能在大规模洪水模拟精度方面非常有前景，并且通过全局精细均匀网格显着降低了数据需求和计算负担。