Urban Space Expansion Based On Cellular Automata And Multi-agent Models

At present and in the future, China will be in the stage of rapid development of urbanization, and the process of urbanization will inevitably lead to the great changes of land use and land use structure. A large number of non-urban land into urban land, urban built-up area outward expansion, the internal space re configuration, which will inevitably lead to a great change in the natural environment for human survival. So how to guide the development of the city, city sprawl control, is the urgent problem, and in the face of such a complex dynamic system, establishing the city land use evolution model is reasonable and accurate, the development process of city simulation analysis and prediction, provide theoretical support for the city planning decision, reduce the potential risk of city bring, it has important theoretical and practical significance. In this paper, the cellular automata and multi-agent model are combined, and the Logistic-CA-Agent model is established.In this paper, the land use data and related information in Quanzhou are fully studied, from the point of view of environmental protection, the ecological control area, as a model of the constraints. By using the method of two logistics regression, from the historical data, mining influence the position of the four variables on the characteristics of cellular state transitions, introducing neighborhood factors and random variables, and the government agent and human agent of city land are analyzed, a research model is established and the multi cellular automaton based on Intelligent agent.This paper takes Quanzhou as the background, in the ArcGIS implementation of the Logistic-CA-Agent model under the constraints of ecological control line, the city of Quanzhou city in 2005 and 2010 were simulated by case, accuracy evaluation using confusion matrixm,77.3%and 91.6% respectively with the city simulation precision.