王小杰, 夏军强, 董柏良, 侯精明, 李启杰. 基于汇水区分级划分的城市洪涝模拟[J]. 水科学进展, 2022, 33(2): 196-207. DOI: 10.14042/j.cnki.32.1309.2022.02.004
引用本文: 王小杰, 夏军强, 董柏良, 侯精明, 李启杰. 基于汇水区分级划分的城市洪涝模拟[J]. 水科学进展, 2022, 33(2): 196-207. DOI: 10.14042/j.cnki.32.1309.2022.02.004
WANG Xiaojie, XIA Junqiang, DONG Boliang, HOU Jingming, LI Qijie. Simulation of urban flood using the SWMM with the hierarchical catchment partition method[J]. Advances in Water Science, 2022, 33(2): 196-207. DOI: 10.14042/j.cnki.32.1309.2022.02.004
Citation: WANG Xiaojie, XIA Junqiang, DONG Boliang, HOU Jingming, LI Qijie. Simulation of urban flood using the SWMM with the hierarchical catchment partition method[J]. Advances in Water Science, 2022, 33(2): 196-207. DOI: 10.14042/j.cnki.32.1309.2022.02.004

基于汇水区分级划分的城市洪涝模拟

Simulation of urban flood using the SWMM with the hierarchical catchment partition method

  • 摘要: 为了有效提高城市暴雨洪涝模拟的精度, 针对城区复杂下垫面和雨水井数据缺失情况, 分别提出雨水井节点数据的确定方法和基于空间信息的汇水区分级划分方法。以武汉市青山区为研究区域, 选取2场典型降水过程, 开展SWMM模型的参数率定和验证工作, 并将基于不同方法划分的汇水区模拟结果与实际渍水数据进行对比。结果表明: ①提出的雨水井节点数据确定方法, 在雨水井实测数据缺失的城市洪涝模拟中具有一定的可靠性和适用性。②基于空间信息分级划分法、水文分析结合泰森多边形法和泰森多边形法所划分的汇水区, 模拟的最大积水深度中分别有100%、63%和75%的典型验证点与实际渍水程度相符, 模拟的溢流点中分别有80.0%、76.4%和77.4%的溢流点位置与5年一遇降雨渍水风险图相符。基于空间信息分级划分法所得的汇水区比较符合真实汇水情况, 且模拟结果比其他2种方法更加准确。③ 5年一遇降雨重现期下, 3种方法划分的汇水区所模拟的积水对研究区域影响程度相对较小, 但遇到高于此重现期的暴雨会出现不同程度的内涝。本研究可为城市暴雨洪涝模拟中雨水井节点数据确定与地表空间离散化提供新方法, 模拟结果可为城市防洪减灾提供参考。

     

    Abstract: In order to effectively improve the accuracy of urban rainstorm flood simulations, the catchment hierarchical partition based on the spatial information method and the determination method of stormwater inlets were proposed in this study, owing to the complicated urban underlying surfaces and the absence of stormwater inlets. Taking the Qingshan district of Wuhan City as a study area, the SWMM model was calibrated and verified for the two typical rainfall hydrographs. Different simulated results were obtained by the validated SWMM model, with the results using different sub-catchment division methods being compared with the actual waterlogging data. The results indicate that: ① The determination method of stormwater inlets had a certain reliability and applicability in urban flood modelling if the measured data of inlet nodes were unavailable. ② The ponding depths and waterlogging nodes were simulated by the SWMM model using the catchment hierarchical partition method, the Thiessen polygon combined with the hydrology method, and the Thiessen polygon method. The simulated maximum ponding depths obtained from these three catchment partition methods accounted for 100%, 63% and 75% of the actual inundation degrees. About 80.0%, 76.4% and 77.4% of the simulated waterlogging nodes were consistent with the waterlogging risk map for a 5-year rainfall return period. The simulated results obtained from the catchment hierarchical partition method were more accurate than the results from other two partition methods. It is found that the simulated results from the catchment hierarchical partition method was relatively consistent with the actual real waterlogging situation. ③ In terms of the 5-year rainfall return period, the simulated flooding obtained from these three methods had a small influence on the urban management. However, different degrees of waterlogging would occur for the rainfall return period greater than 5 years. This study can provide a new method for the discretization of ground surfaces and the determination of inlet nodes in simulating urban rainstorm floods, and provide the reference for urban flood control and disaster reduction.

     

/

返回文章
返回