流域水文模型在山东省设计洪水计算中的适用性研究

发布时间：2018-11-03 09:05

【摘要】：洪水预报是指根据产汇流机制原理构建水文模型,并对流域内洪水各项特征要素进行模拟预报的一系列过程。流域洪水预报是否可靠主要依据以下两点:模拟结构能否反映流域径流形成物理过程;参数率定、检验及应用是否能否达到所需精度标准。目前,实时洪水预报只能在短期防汛中起到一定的作用,但是针对中小流域,由于汇流时间过快,有效的预报时间相应减少。因此考虑流域内关键的水工建筑物的设计防洪标准,预估灾害规模,减少损失,并对其进行进一步的设计洪水计算是必要的。将水文模型预报的结果应用到流域设计洪水计算,为设计洪水的计算提供了一种新方法,也应成为未来水文预报的一个大的趋势。根据下垫面状况,山东省主要地形地貌可涵盖为一般山丘区、岩溶山丘区、一般平原区三种。与之对应分别选取其典型流域太河流域、田庄流域、刘庄闸流域。本文就是针对以上三种不同地貌形态的流域,分别在其上进行不同模型的产汇流演算,并对历史大洪水进行演算预报与设计洪水的做进一步研究,从而探讨模型推广与设计洪水计算的新思路。在对上述三个流域的地理条件与气候水文资料进行分析后,选取了新安江模型、水箱模型、SCS模型三种模型并确定相应模型结构,而后代入流域相关数据,调整模型参数,进行产汇流模拟与应用比较。在此基础上分析了三种模型在用于设计洪水预报的可行性,研究成果表明:1、在次洪预报精度上,受制于流域洪水资料年限较短与降雨强度不均,三种模型对太河水库流域和田庄水库流域的模拟对次洪的预报只能勉强达到相关预报标准,但对一般平原的刘庄闸流域不能满足洪水预报要求。2、在设计洪水计算中,将分时段设计降雨数据代入率定好参数的三类模型中,并与《山东省水文图集》中设计洪水计算结果与流域原有设计洪水标准进行比较,其中选用的新安江模型与SCS模型设计洪水结果较为可靠,水箱模型模拟结果对洪峰突变的过程模拟效果较差。
[Abstract]:Flood forecasting is a series of processes that construct hydrological model according to the principle of runoff yield and confluence mechanism and simulate and forecast each characteristic element of flood in river basin. The reliability of flood forecasting is mainly based on the following two points: whether the simulated structure can reflect the physical process of runoff formation in the basin, and whether the parameter rate can be determined, tested and applied to meet the required precision standard. At present, the real-time flood forecast can only play a certain role in the short-term flood control, but for the middle and small river basins, the effective forecasting time is reduced because of the rapid confluence time. Therefore, it is necessary to consider the design flood control standards of key hydraulic structures in the basin, to estimate the scale of disasters, to reduce losses, and to carry out further design flood calculation. Applying the results of hydrological model forecast to the design flood calculation of river basin provides a new method for the calculation of design flood, and it should also become a big trend of hydrological forecast in the future. According to the condition of underlying surface, the main landforms of Shandong Province can be divided into three types: general hilly area, karst hilly area and general plain area. The typical watershed of Taihe river, Tianzhuang basin and Liuzhuang sluice basin are selected respectively. In this paper, according to the three kinds of river basins with different geomorphological forms, we carry out different models of runoff generation and confluence calculation, and do further research on the prediction of historical flood and design flood. Thus, a new idea of model generalization and design flood calculation is discussed. After analyzing the geographical conditions and climatic and hydrological data of the three basins mentioned above, three models, Xinanjiang model, water tank model and SCS model, are selected and the corresponding model structure is determined, and then relevant data are added to the basin to adjust the model parameters. The simulation and application of production and confluence were compared. On this basis, the feasibility of the three models in designing flood forecast is analyzed. The results show that: 1, the accuracy of flood forecasting is limited by the short duration of flood data and uneven rainfall intensity. The simulation of three models for Taihe reservoir basin and Tianzhuang reservoir basin can only barely meet the relevant forecast standard, but for the Liuzhuang sluice basin in general plain, it can not meet the flood forecast requirements. 2. In the design flood calculation, In this paper, three kinds of models for determining the input rate of the designed rainfall data in different periods are compared with the design flood calculation results in Shandong Hydrological Atlas and the original design flood standards in the basin. The design flood results of Xin'anjiang model and SCS model are reliable, and the simulation results of water tank model are not good for Hong Feng's sudden change process.
【学位授予单位】：山东农业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TV122;TV124

【相似文献】