مطالعه شبیه سازی در مدل CA بر اساس پارامتر بهینه سازی الگوریتم ژنتیک و توسعه شهری

This paper presents a new method to calibrate urban cellular automata (CA) using genetic algorithms (GA).The GA is used to find the optimal parameter values so that CA models can simulate urban expansion in a more realistic way. Traditional multi-criterion evaluation (MCE) and logistic methods have limitations for deriving the transition rules of CA models. The variables should be independent so that the parameter values (coefficients) can be properly estimated by regression analysis. This assumption is not true in most situations the limitations can be overcome by using GA to estimate these parameter values for these correlated variables. This method is applied to the simulation of urban expansion in Dongguan, a fast developing city in the Pearl River Delta in South China. The model is able to simulate urban Development in 2004-2010 by using the training data from remote sensing data. The analysis indicates that the proposed model can produce better simulation results than MCE based CA models and logistic calibrated CA models.

Cellular Automata (CA) is a cell space consisted of discrete cells with finite state, at the same time, is also a dynamics system evolved in discrete time dimension according to certain local rules [1]. In recent years, CA has already increasingly been applied in geosciences simulation and many significant study results have been acquired [2, 3]. For example, White along with the team workers have simulated the changes of the land utilization in Cincinnati by applying the restricted cellular automata[4]; Wu has simulated the urban sprawl of Guangzhou[5]; Li Xia and Ye Jia have simulated the expansion situation of the land utilization and urban sprawl of Dongguan[6]. All these researches indicate that CA can simulate the characteristics very close to the actual urban and the simulated results are in accordance with the reality. This paper applies the genetic algorithm to optimize the parameters of CA model and has acquired the optimized parameters of CA model with the self-adapting method after the evolution strategies of encoding the chromosome, establishing fitness function, selecting genetic operator and ensuring genetic algorithm. The optimized CA model applied the genetic algorithm simulates the urban sprawl of Dongguan in 2004-2010.

The key problem of CA is how to define the transformation rules and search the best parameter for the model to make the simulated results closer to the actual urban situations. This paper adopts the global optimizing search program based on the genetic algorithm to quickly search for the best parameter and simulates the urban development situations of Dongguan in 2004-2010 applying CA model with optimized parameters. The research results indicate that CA model established by applying the genetic algorithm can acquire higher simulation precision than the common CA model established by applying multi-norms judgment and logical regression model, the model can conveniently simulate the urban development situations of the fast urbanization regions.