规划导向的CA约束方法与城市用地格局变化预测——以广州市白云区为例

Planning Guidance Constrained CA Model for Simulating Urban Landuse Pattern:A Case Study of Baiyun District,Guangzhou

城市发展是一个复杂系统的演化过程,掌握其发展规律并用于预测城市形态变化在城市理论和城市规划管理中有重要意义。一般约束性元胞自动机(CA)无法模拟存在时空分异的约束条件,文中提出一种时、空可变的规划导向约束条件表达方法,将城市发展战略调整、建设周期等外部条件变化用时间变量和空间变量表达,模拟出动态变化的CA外部环境,逼近城市受动态规划条件影响下的演化过程。以广州市白云区为例,对"十二五"规划中城市发展战略继承和调整并存的现象进行了情景模拟;构造了地铁线路、重点发展区两个规划方案在2010―2015年间建设过程的动态约束条件,模拟出规划导向下2020年时的城市格局。预测结果显示,未来10年城市发展仍然是以沿主要交通网络和以老城区为中心的放射状规律扩张,白云区城市面积将达到301.62km^2。

Urban evolvement is a complex process of urban system. Cellular Automata（CA） has strong spatial modeling capabilities, which can be used to simulate urban landuse pattern and can serve as an useful tool for urban landuse pattern forecast. However, CA model based on global constraint conditions can not derive reasonable results when constraint conditions are variable during simulation. In this paper, a CA model with new constraint conditions is brought up to adapt the situation that urban policies and planning are discrete or discontinuous. With this model, urban development strategic modification or periodic planning can be used as constraint factors to control the simulation process. This method has been applied to the simulation and forecast of urban landuse pattern of fast growing city such as Baiyun district, Guangzhou, from 2010 to 2020, when the Twelfth Five-Year Plan will be executed from 2010 to 2015. It is shown that this model is capable to approach variable constraint conditions and helpful to improve the simulation precision. In next 10 years, traffic network will be main control factor to urban landuse pattern. Urban area of Baiyun district will come to 301.62km^2 by 2020.