HOU Jingming, ZHANG Zhaoan, MA Liping, ZHANG Dawei, XIANG Liyun, WANG Yu, LI Jicheng. Unstructured numerical model for rainfall-runoff process in watershed based on GPU acceleration technology[J]. Advances in Water Science, 2021, 32(4): 567-576. DOI: 10.14042/j.cnki.32.1309.2021.04.008
Citation: HOU Jingming, ZHANG Zhaoan, MA Liping, ZHANG Dawei, XIANG Liyun, WANG Yu, LI Jicheng. Unstructured numerical model for rainfall-runoff process in watershed based on GPU acceleration technology[J]. Advances in Water Science, 2021, 32(4): 567-576. DOI: 10.14042/j.cnki.32.1309.2021.04.008

Unstructured numerical model for rainfall-runoff process in watershed based on GPU acceleration technology

  • In order to simulate the rainfall-runoff process of the basin with high efficiency and high accuracy, the dynamic wave method is applied to solve the two-dimensional Saint-Venant equation, with the hydrological process coupled. A high-performance watershed rainfall-runoff process numerical model including processes such as rainfall runoff, confluence, infiltration, and flood evolution has been established. The advantage of this model is the use of unstructured grids, which can better deal with irregular boundaries and accurately fit the surface of complex terrain, so that the model can accurately calculate and simulate the rainfall-runoff process of the basin; besides, the GPU technology is introduced to accelerate calculations in this model, making it possible to calculate storm floods in large-scale watersheds. Finally, the model was applied to the V-type classic case and two actual watershed flood cases; the obtained results are in good agreement with the actual measurement, and the calculation time is shorter, showing that the model can quickly and accurately simulate the river basin rainfall-runoff process. The research results are helpful for realizing reasonable and efficient prediction of flood disasters in actual river basins, so as to provide strong support for emergency rescue work.
  • loading

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return