基于半分布式GUH模型的台风期水库入库径流模拟

    Reservoir inflow simulation during typhoons based on semi-distributed GUH model

    • 摘要: 台风期间强降雨集中、响应迅速,极易引发洪水灾害,准确模拟水库入库径流对实现科学调度与防洪安全具有重要意义。本文基于广义单位线(General Unit Hydrograph, GUH)理论,构建了一种半分布式径流模拟模型,并设计独立参数化与共享参数化两种设置,系统评估其在台风期入库径流模拟中的适用性。以广东鹤地水库流域为研究区域,划分8个子流域并提取逐时降雨数据,选取2006—2013年6场典型台风事件进行率定与验证。结果表明:独立参数化的GUH模型在所有验证场次中模拟精度均优于共享参数化,平均确定性系数(DC)达0.71,具备良好的泛化能力;GUH模型在保证结构简洁性的同时模拟性能稳定、洪峰响应合理。该模型在数据需求较少的条件下即可实现高效模拟,适用于快速响应型水库调度系统,在防洪减灾和雨洪资源利用中具有良好的应用前景。

       

      Abstract: Typhoon-induced rainfall is typically intense, short-lived, and highly responsive, often leading to severe flood disasters. Accurate simulation of reservoir inflow during typhoon events is therefore critical for scientific reservoir operation and flood control. In this study, a semi-distributed inflow simulation model was developed based on the General Unit Hydrograph (GUH) theory. Two parameterization strategies were designed and evaluated: sub-basin-specific parameterization, where each sub-basin is assigned its own parameter set, and basin-wide parameterization, where all sub-basins share a common parameter set. The Hedi Reservoir Basin in Guangdong Province was selected as the case study area and divided into eight sub-basins with hourly rainfall data extracted for each. Six representative typhoon events from 2006 to 2013 were used for model calibration and validation. The results showed that the GUH model with sub-basin-specific parameterization consistently outperformed the basin-wide scheme across all validation events, achieving an average Deterministic Coefficient (DC) of 0.71 and demonstrating good generalization capability. Moreover, the GUH model maintained structural simplicity, delivered stable simulation performance, and captured flood peak responses reasonably well. Given its relatively low data requirements and robust simulation ability, the model exhibits strong potential for application in real-time reservoir operation, flood mitigation, and rainwater resource management.

       

    /

    返回文章
    返回