The Study Of Urban Land Expansion Simulation Based On Variation Particle Swarm Optimization

Urbanization is the most important trend in contemporary China. Urban land expansion is not only the most significant form of urbanization, but also an important standard to measure the urbanization. The deep study of urban land expansion has a very important significance to urban land’s better and faster development. This study attempts to introduce Variation Particle Swarm Optimization to simulate the urban land spatial expansion, taking Jin Li Metro as the research object, to verify the effectiveness of the application.The study first discusses the urban land expansion theory, method and the previously used models, and uses them as a theoretical basis for the practical research. Secondly, this study introduces a new method, named Variation Particle Swarm Optimization, and designs a idea for using this algorithm to simulating urban land spatial expansion. Then on the basis of analyzing the driving force of urban land expansion and confirming the reasonable scale of expansion, this paper uses Variation Particle Swarm Optimization to simulate the urban land spatial expansion. Last, by contrasting with the current situation of land use and the existing urban land spatial expansion models to prove the validity of Variation Particle Swarm Optimization. The main co nclusions of the study are as following:(1) The existing urban land spatial expansion models includes the Empirical model, Cellular Automata model and Multi-agent model, and the three models all have their own limitations in the simulation of urban land spatial expansion. PSO has some advantages in the simulation of urban land spatial expansion because of its strong capability in spatial searching, comprehensive evaluation ability and fast operation speed.(2) When studying the driving force of urban land scale growth, it mainly discusses the economic development and the urban population’s influence. And it finally discovers that the urban population is the largest influence on urban land expanding. The main influence factors in urban land spatial expansion are the natural environment, traffic guidance, urban radiation force and government policies.(3) By using Matlab software, this study realizes the simulation of urban land spatial expansion based on Variation Particle Swarm Optimization. And by using visual examination, shape index examination, metastasis contrast and Kappa index examination to compare simulation results and the current situation of land use, we can find that the simulation results is good.(4) By comparing Variation Particle Swarm Optimization and CLUE-S model, Cellular Automata mo del and Multi-agent model, founds that the accuracy of the simulation results are similar. But Variation Particle Swarm Optimization is better than the existing model on three aspects: parameter setting, the choice of impact factor, and the effect of simulation.