The Research On Urban Drainage Network Inundation Evalution And Control Based On Modeling Technology

With the global climate change and increasing urbanization the urban drainage network suffers from the inundation a lot. The inundation risk of urban drainage network is different from rivers'. It happens in certain areas when the rainfall of short-term storm is beyond the drainage networks'capability, and our present design method is not able to explain and resolve the problem. Therefore, it is necessary to make a systematic study on the urban inundation evolution process. This research starts from the drainage network modeling theory to explore the urban drainage inundation evaluation and forecast issues by using the model-based techniques, the following results are explained:1) According to systems engineering principles, the basic theory of drainage network model is studied.The dissertation analyzes the basic calculation principle of U.S. EPA's SWMM model, firstly adopts Particle Swarm Optimization (PSO) method to optimize and calibrate SWMM model's parameters, and analyzes SWMM's parameters for their sensitivity by using disturbance method.2) The differences of design criteria between our country and foreign countries, especially America and European Union, are compared from five aspects; they are the design flow calculation method, runoff coefficient, design return period (frequency), reduction factor, and hydrodynamic calculation method. Taking the drainage design method as the basic clue, the dissertation summarizes the questions and disadvantages in our design method. And through the research, the dissertation gets the following result that is American and European countries'design criteria are more detailed and scientific. And the idea of using computer drainage model as main design approach can be referred in our country's design theory.3) Focusing on the shortages in our design criteria, the dissertation proposes a concept of important district which limits the inundation risk evaluation into four types, establishes a drainage network inundation design check and over standard evolution system. For the design check research, it proposes an equivalent principle between model parameter and design condition; it achieves the procedures of modeling design check and over-standard inundation evaluation.4) A drainage network inundation comprehensive evaluation model based on Projection Pursuit method is built considering the hydraulic, economic, environmental and social factors, and the dissertation divides the hazard evaluation into distributing evaluation and disaster-grade evaluation, which provides a technological tool for urban flood control decision-making.5) The concept of drainage network online model is proposed, and the dissertation summarizes the online modeling method, develops the online-SWMM model, which sets up a theoretical basis for real-time warning system.6) Basing on the drainage online model a framework of real-time inundation warning system is proposed, and an integration is achieved among online model, comprehensive evaluation model and decision-making model. And also a demo program of the urban drainage network inundation warning system is developed, which realizes some practical functions like the real-time rainfall-inundation's demonstration, warning time prediction, and automated distribution of flood thematic map and flood grade.7) The concept of model coupling among softwares is introduced into our country, and a framework of model coupling system for urban flood control is proposed. The dissertation makes coupling redevelopment about the computing engines of U.S. SWMM and U.K. LISFLOOD-FD, and realizes urban drainage one-dimension and two-dimension model coupling computation, which provides a technical approach for our country's future multi-model coupling study and large-scale system integration model research.8) A comprehensive model-based urban drainage inundation evaluation and control system is summarized, and a detailed analysis between the system's composition and function is explained.Urban drainage network inundation evaluation and control is a comprehensive issue referring to hydraulics, hydrology, operational research, artificial intelligence, geography information, software engineering and lots of other knowledge fields and frontier science. The innovative research results of the dissertation are:1) An optimal and muti-objective calibration method based on Particle Swarm Optimization for model's parameters is proposed.The method achieves model parameters'optimal calibration for high dimension space(eight-dimensional space in this dissertation), the calibration result considers the flow error demand at the peak time and every time, and the optimal calibration method could get the satisfactory rusults for practical engineering.2) A drainage network inundation comprehensive evaluation model based on Projection Pursuit method is built.The model is able to integrate various influence factors related to inundation disaster such as the social, the economic and the technical, establishes the drainage network design check and over-standard evolution system, achieves the urban inundation objective and comprehensive evaluation, which is more practical than other evaluation methods.3) The drainage network online modeling method is given, and U.S. EPA's SWMM event model is redeveloped and updated for online modeling.The dissertation summarizes the urban drainage inundation evaluation and control system whose core technologies are the comprehensive evaluation model and network online model. The system integrates the evaluation method of the design aspect and the realtime warning technology of the operation aspect into a whole body, realizes the digital inundation control system framework for urban drainage network, which provides a technical approach for the practical disaster prevention.The research results of this dissertation provide a basis and reference for our country's inundation prevention and control, which will play more and more important role in the future.In the end, this dissertation also should acknowledge the assistance of the National Natural Science Foundation of China(50678009) and National Key Technology R&D Program in the 11th Five year Plan of China(2006BAJ16B04).