Research on Multi-source Information Fusion and Uncertainty Modeling of Flood Risk Assessment
|School||Huazhong University of Science and Technology|
|Course||System Analysis and Integration|
|Keywords||Flood Risk Assessment Multi-source Information Fusion UncertaintyModeling D-S Evidence Theory Random Set Theory Simulation|
Flood is a vital damage to the life and property of people, which is concerned by more and more experts and the society. Flood risk assessment (FRA) is a measure of flood management, it makes sense in economic and social aspects because FRA has been widely used in the application of flood insurance, flood area management, flood refuge, flood damage early warning, flood damage assessment, flood influence assessment, improvement on public flood risk consciousness etc. in order to provide important scientific basis and support for flood risk management. Because flood has caused serious damage to the life and property of people, social, and economic activities recently, flood has been recognized as one of the most serious damage. China is one of countries which suffer most serious flood. However, because of the complexity and uncertainty of flood, the analysis of uncertain in FRA is a current hot research topic. To solve the problems above, this study analyze the uncertain factors in FRA and the shortcomings of current methods, make the uncertain modeling based on DS-AHP method, Random set theory and other new achievement in order to reduce the uncertainty of FRA and improve the accuracy of FRA and assure the availability in FRA.The main research and creative work are as following:(1) Uncertainty and modeling methods in FRATo solve the uncertain problem in current FRA, we analyze the principle of uncertainty and the common universality, and then we classify the uncertainty into two classes:aleatory and epistemic uncertainty. An uncertainty frame in FRA and uncertain modeling flow chart are proposed to provide the support of FRA.(2) Uncertainty modeling and flood risk assessment research based on DS-AHPTo solve the different uncertainty and different weight value between them, we propose an uncertainty analysis method based on AHP, which makes the factors in FRA in a form of hierarchy and clear structure. We get the judgment matrix according to AHP, then we can get the biggest eigenvalue and eignevector and the weight order of each factor. Finally we make an objective flood risk assessment based on DS-AHP.(3) Multi-source information fusion and urban flood risk assessment based on random set theoryAn information fusion model and an integration operator are proposed in this research based on random set theory in order to consider the multi-source information. And the research takes the urbanization factor, DEM factor and rainfall factor in Wuhan city as an example to make the urban flood risk assessment.(4) Flood risk assessment multi-source information integration methodology We propose a decision fusion method based on D-S theory, and we take the Hanjiang Basin as study area to make the population factors fusion including population age and population education, so we can get the decision fusion. Finally, we give a hydraulic simulation based on HEC-RAS and history flood and the flood risk assessment results of Hanjiang River.