Modelling And Simulation Of A Heterogeneous Feature-Based Integrated Hydraulic Model For The Processes Of Urban Flooding

With the process of urbanization and global climate change,urban rainstorms have caused flooding more frequently in China,which results in serious property damage and endangers the lives of humans.In recent years,much scientific research and engineering have been widely used to support urban flood disaster risk management and flood control.Among these methods,with the advantages of discovering the laws of waterflow and predicting its evolution,hydrological/hydraulic models have been widely used for urban flood management and urban planning.However,due to the spatial heterogeneity of the urban environment,the hydraulic modelling and simulation of urban floods have become very complicated.Traditionally,to reduce the complexity of urban flood modelling and simulation and to improve the accuracy of the model simulation results,many simplification or generalization methods are used in urban hydrological/hydraulic modelling and simulation.These simplification or generalization methods can be sorted into two groups:generalization of input data and generalization of model structure.The simplification or generalization of hydraulic modelling seriously hinders the analytical capabilities of simulation results and limits the evaluation of how different features impact the urban flooding processes and the complex interactions between these features.This thesis takes the spatial heterogeneity of features as its basic research idea and the expression of the heterogeneity characteristics of features.dynamic process modelling and simulation,and geographic analysis as its basic research points.This thesis focuses on the key technologies of input data adaptation and spatial discretization,coupling of overland flow and sewer system flow.Furthermore,based on the achievements of data adaptation and hydraulic modelling,the goal of this thesis is to discover the impacts of different features on urban stormwater processes and perform spatial-temporal analysis of key features.The main contents of this thesis include the following:(1)Coping with the complexity of data adaptation and spatial discretization of urban stormwater modelling and simulation when considering spatial heterogeneity.This thesis researched the construction of a feature-based integrated urban hydraulic model data adaptation and spatial discretization program.First,based on the analysis of the heterogeneity characteristics of mainly hydrological features that impact the urban flooding process,a data adaptation model was designed based on aspects of the conceptual model and procedural logical model.The data adaptation model is used to analyse,store and manage the spatial information and attributes of heterogeneous features and other related input data for modelling and simulation.Second,this thesis researched the data adaptabilities of a single feature,the topological relationship between features,and the data pre-processing for integrated simulation of overland and sewer systems.These methods are used to make the input data better adapt to spatial discretization programs and model running conditions.Third,this thesis introduced three methods used to make input data that can be conveniently and automatically converted into model-adaptable data formats:unstructured triangle mesh generation based on a consideration of heterogeneous features;automatic projection of the heterogeneous attributes on discretized meshes;and automatic sewer system data conversion from standard GIS data into an SWMM adaptable format.Three cases based on different types of urban areas and the experiments in chapters 4 and 5 demonstrated that the proposed methods can improve the capabilities of data pre-processing for urban flood integrated modelling and simulations.(2)Due to the essential interaction of overland flow and underground sewer water flow,this thesis is based on two popular open source models,SWMM and ANUGA,to research the bi-directional coupling method of overland flow and sewer system water flow processes.From the aspect of model calculation,the coupling method is implemented by dynamically interacting the overland flow and sewer water flow at each water exchangeable manhole in each simulation step.The method of dynamic correction of the source terms in the continuity equation is adopted to cope with the interaction of the shallow water model.The sewer system water flow simulation module of SWMM was realized by a dynamic setting and obtaining the water outflow/inflow flux of each node.For the coupling simulation of the model systems,the ANUGA model was used to simulate the surface water flow process as the main loop control object,and the sewer waterflow transportation module of the SWMM is invoked by ANUGA.To cope with the heterogeneity of manholes,the operable object programming method is used to realize the dynamic exchange of the nodes.Additionally,some interfaces have been developed to control the working states of the operators,which are used to cope with the complex water flow exchange scenarios in coupled simulations.The real storm event-based experimental results in an urban area demonstrated that the integrated model can simulate the surface water accumulation areas well.The coupled hydraulic model can more accurately depict the motion state of water flow and flooding areas.In addition,the method of considering the heterogeneity of different types of features can better support the analysis of the impacts of features on the storm flood process.(3)In terms of urban stormwater process quantification analysis,the thesis studied scenario analysis and spatial-temporal analysis methods to evaluate the impacts of different features on urban flood processes and spatial-temporal processes for the operation states of manholes.This thesis is based on the total surface runoff and surface flood areas and depth to evaluate the impacts of features;on the other hand,this thesis is based on the empirical orthogonal function(EOF)analysis method to discover the spatial-temporal characteristics of the inflow and overflow of the sewer system.This research results demonstrated that the proposed methods(which are heterogeneity-considered modelling and simulation and scenario analysis)and the integrated model can support revealing the impacts of different features on the flood process.Currently,high-resolution simulation of urban flooding processes has become a popular research goal,and the results of these experiments can help other researchers to understand the meaning of considered heterogeneity in urban flood modelling.In addition,the EOF analysis method was used to analyse the results of the urban flooding simulation based on the hydraulic model.The results show that EOF analysis is a useful method to reveal the spatial-temporal evolution processes of manholes.