A Study On Intelligence System Supported By GIS And Its Application Of Landslide-Forecast

The thesis is one important component of fifteenth project— research anddemonstration on the key problem of supervision, pre-warning and decision-makingassistance supported system of geology disaster in the Three Gorges of Yangtse Rive(2001BA604A02). Directed by the research of project, the thesis uses mainlycomputational intelligence and its' coupling method combined with geographyinformation system (GIS) technique to make intelligent compositive-system on thedisaster-changing forecast of landslide supported by GIS, and analyze the examples---the Wu-jiawan landslide and Shiyanxiaoxue landslide. The research and mainachivement is as follows: (1) The basic method and principle of landslide forecast of computationalintelligence including the artificial neural network (ANN), fuzzy logic (FL) and geneticalgorithms (GA) have been discussed. Especially artificial neural network, fuzzy logicand their coupling method, and also the advantage and feasibility of which are appliedin disaster-changing forecast of landslide have been studied. (2) The method of disaster-changing forecast of landslide included ANN and FIShave been deeply studied. The method of adaptive neuro-fuzzy inference system, whichis coupled by ANN and FIS, applied in landslide disaster-changing is presented firstly.The key technique of making model and steps of system design is also put forward. (3) The criterion of disaster-changing forecast of landslide is studied. Thecompositive-information fuzzy-criterion based on the method of computationalintelligence of landslide disaster-changing forecast is presented for the first time. Alsothe advantage and concrete expression of the compositive-information fuzzy-criterionoflandslide disaster-changing forecast are discussed. 4) The basic method and principle of the geography information system(GIS) isanalyzed. The relative key technique and numerical analysis method, which is used tofound the GIS model of the comprehensive information of landslide forecast, is studiedby ArcView GIS 3.2. The relative technique problem of 3D-GIS model of landslide isalso discussed. 5) The new study route combined GIS technique with computational intelligencetheory, which is applied in study of landslide disaster forecast, is presented for the firsttime. The technique of data exchange and data meet-scoop between the GIS III重庆大学博士学位论文comprehensive information model of landslide disaster-changing and computationintelligence forecasting system is developed. 6) The key technique of performing system program is studied. The computationintelligence integration-system of landslide disaster-changing forecast supported by GISis developed by the GIS sub-exploitation technique and mixed making programmetechnique, which include the Microsoft Visual Studio series VB.NET, Matlabinterpretation language and the object language Avenue of ArcView GIS. (7) For the example of the Wu-jiawan landslide and Shiyanxiaoxue landslide ---the demonstrating site of fifteenth project (2001BA604A02), computation intelligenceforecasting system of landslide disaster-changing supported by GIS in the application oflandslide disaster-changing forecast in the Three Gorges of Yangtse Rive is discussed,and the results from several methods are basically consistent. All the above analysis of both the scientific theory and practicalapplication showsthat coupling method of GIS and computation intelligence theory, which is used to studydisaster-changing forecast of landslide, has immense advantages and widest applicationforeground.