论文部分内容阅读
元胞自动机CA(Cellular Automata)和多智能体ABM(Agent-Based Model)模型是土地利用格局和演化模拟的主流方法,两者在模拟自然因素影响和人文驱动机制方面具有突出优势,为LUCC研究提供了重要的工具。当前,ABM无论在模型构建还是应用研究方面,CA和ABM均取得了显著进展。论文从数据基础、模拟尺度、CA转换规则挖掘、ABM行为规则定义、CA和ABM的耦合4个方面梳理土地利用模拟模型和方法的研究进展。并总结这些模型在虚拟城市模拟与理论验证、真实城市模拟与规划预测以及多类用地模拟与辅助决策等方面的应用。最后,总结土地利用模拟模型在精细模拟和全球变化研究方面存在的局限性,认为未来发展将主要集中于解决从2维模型向3维模型发展、大数据与规则精细挖掘以及大尺度模拟与知识迁移等问题。
Cellular Automata and Agent-Based Model (ABM) are the mainstream methods of land use pattern and evolution simulation. Both of them have prominent advantages in simulating the influence of natural factors and humanistic driving mechanism, Research provides an important tool. At present, ABM has made significant progress both in model building and applied research. The paper reviews the research progress of land use simulation models and methods from the aspects of data foundation, simulation scales, CA conversion rule mining, the definition of ABM rules of behavior, and the coupling between CA and ABM. And summarizes the application of these models in virtual city simulation and theoretical verification, real urban simulation and planning and prediction, as well as multi-type land use simulation and decision support. Finally, the paper summarizes the limitation of simulation models of land use in the study of fine simulation and global change, and thinks that the future development will mainly focus on the development from 2D model to 3D model, the fine mining of big data and rules, and the large-scale simulation and knowledge Migration and other issues.