分析学习智能元胞自动机及优化的城市模拟

Analytical Learning-Based Intelligent Cellular Automata for Realistic and Idealized Urban Simulation

元胞自动机被广泛应用于城市及其他地理现象的模拟,模拟过程中的最大问题是如何确定模型的结构和参数。该文提出一种基于分析学习的智能优化元胞自动机,该模型在逻辑回归模型的基础上,基于分析学习的智能方法,寻找元胞自动机模型的最佳参数。该方法允许用户控制空间变量影响权重,进而模拟出不同的城市发展模式,可为城市规划提供重要参考。

There is a rapid development of CA models for simulation of land use patterns and urban systems recently. This paper presents a new model to simulate the dynamic change of multiple land uses based on the integration of analytical learning approach, cellular automata and GIS. However, the simulation of multiple land use changes using CA models is difficult because many spatial variables and coefficients need to utilize. Conventional CA models have problems to define simulation coefficients, transition rules and model structures. In this paper, an approach about analytical learning is designed to optimize the parameter values. This approach is a way of constantly consummating the final values by accumulating the experiences. It is a kind of intel- ligent machinery arithmetic. The analytical learning model calculates the results from the samples determined by ARC/INFO. The proposed method can allow users to choose different ways to simulate land use patterns. The model has successfully applied to the simulation of land use in the Pearl River Delta.