Research On Discovering Transition Rules Of Urban Expansion Cellular Automata Based On Biogeography-based Optimization Algorithm

Urban expansion is the objective existence of urban development,construction and improvement.Urban expansion provides the necessary land resources for urban development;while on the other hand,the rapid expansion of cities also brings great pressure on limited land resources,especially arable land resources.Therefore,master the regular pattern of urban land use evolvement and trend and accurately simulating of urban expansion process has important significance to develop scientific and rational urban space development strategy,and ensure sustainable utilization of land resources and urban sustainable development.Cellular gained wide attention in urban expansion simulation application.The transition rules are the core of CA,and the most important factor affecting the degree of geographical simulation.Therefore,how to obtain the transition rules effectively is an important work to apply CA model to urban expansion simulation.This article explores a model based on a new bionic intelligent optimization-biogeography-based optimization-to automatically discovering transition rules,studies the mathematical model of the algorithm,the key operator and implementation method,and analyses the simulation accuracy,validity and advantages of the algorithm model.The research contents and conclusions of this paper are as follows:1.Design the model based on biogeography-based optimization to automatically discovering transition rules of cellular automata.Based on the basic theory of biogeography-based optimization and cellular automata,study and design a cellular automata transition rules mining algorithm(BBO-TR-CA)based on biogeography-based optimization,mainly studies the core idea,mathematical model and key operators of BBO-TR-CA model,completed the design and implementation.2.Empirical study on urban expansion simulation of Wuhan central city based on BBO-TR-CA model.Based on the BBO algorithm to mining the transition rules of cellular automata in Wuhan central city,and realized the dynamic simulation of urban expansion.In addition,in order to verify the ability of BBO-TR-CA model mining transformation rules,a linear logistic regression model was introduced for comparison and demonstration analysis.By comparing the simulation accuracy of the two models and the Moran's I index,the quality of transformation rules constructed by the two models was analyzed.The empirical results show that compared with the linear logistic regression model,the simulation results based on BBO-TR-CA model are closer to the actual situation in the simulation precision and urban spatial pattern characteristics,and illustrate the transition rules based on BBO-TR-CA model mining are of better quality.