精细刻画城镇自组织扩展过程的CA-ABM耦合模拟及预测模型  

A CA-ABM-coupled simulation and prediction model for finely depicting the local self-organization process of urban expansion

作　　者：张彬 胡守庚 王海军[4] 郭颖 童陆亿 夏天顺 ZHANG Bin;HU Shougeng;WANG Haijun;GUO Ying;TONG Luyi;XIA Tianshun(School of Public Administration,China University of Geosciences,Wuhan 430074,China;Key Laboratory of the Ministry of Natural Resources for Research on Rule of Law,Wuhan 430074,China;Guangdong-Hong Kong-Macao Joint Laboratory for Smart Cities,Shenzhen 518000,China;School of Resource and Environmental Sciences,Wuhan University,Wuhan 430079,China;Wuhan University Library,Wuhan 430072,China)

机构地区：[1]中国地质大学(武汉)公共管理学院,湖北武汉430074 [2]自然资源部法治研究重点实验室,湖北武汉430074 [3]粤港澳智慧城市联合实验室,广东深圳518000 [4]武汉大学资源与环境科学学院,湖北武汉430079 [5]武汉大学图书馆,湖北武汉430072

出　　处：《测绘学报》2024年第7期1429-1443,共15页Acta Geodaetica et Cartographica Sinica

基　　金：湖北省自然科学基金(2023AFB022);教育部人文社会科学研究项目(23YJC630223);广东省科技创新战略专项资金(粤港澳联合实验室)(2020B1212030009);国家自然科学基金(42171272;41901236;42171411);中国地质大学(武汉)“地大学者”人才岗位科研启动经费资助(2022129)。

摘　　要：城镇扩展模拟预测是支撑国土空间规划、实现城镇化高质量发展的重要手段,提高其科学性和实用性有助于精确把握城镇扩展趋势,对国土资源的合理和可持续利用意义重大。现有以元胞自动机(CA)为基础的模型着重于刻画空间因子驱动的城镇自然扩展过程,智能体模型(ABM)则在模拟城镇自组织扩展方面具有理论优势,而且已有的耦合方法多为基于模拟步骤的模型级联,难以实现二者紧密耦合以充分发挥其在模拟城镇自然和自组织扩展方面的优势。本文基于CA的结构开放性和ABM的理论优势,以可达性为媒介表征城镇扩展局部自组织过程的范围差异,并依据博弈论设计利益相关者之间的局部自组织规则,将ABM刻画的人地交互决策融入CA邻域构建,建立一种可以精细刻画城镇自组织扩展过程的CA-ABM耦合模拟及预测模型(CA-ABM-LSO),实现CA和ABM在底层结构的深度集成和城镇自然与自组织扩展过程的耦合模拟。以武汉市为例对其进行应用,结果表明:基于CA和ABM的优势互补特点进行耦合建模,能够充分发挥二者在刻画城镇自然和自组织扩展方面的优势,显著提升城镇扩展模拟精度,并改善模拟城镇斑块的景观形态;基于博弈论的局部自组织规则可有效实现宏观经济政策对微观智能体交互决策的指导,提高城镇扩展模拟的科学性和规划实用性;预计至2035年,武汉市城镇扩展的重点区域集中在高新科技园区和运输集散中心附近,这符合“武鄂黄黄”都市圈的发展预期,可为其空间规划提供决策支持。Urban expansion simulation and prediction are vital for supporting national spatial planning and promoting sustainable urbanization.Improving its scientific and practical applicability is essential to accurately capturing urban expansion trends and sustainable land resource utilization.Present cellular automata(CA)models focus on describing spatially driven urban expansion,while those using the agent-based model(ABM)approach provide theoretical benefits in simulating self-organized urban expansion.Moreover,existing coupling methods predominantly cascade these models based on simulation steps,presenting challenges in deeply integrating them to unleash their potential for simulating both natural and self-organized urban expansion.This study is grounded in the structural openness of CA and the theoretical strengths of ABM.It uses accessibility as an intermediary to define scope variations in local self-organized processes during urban expansion.Additionally,it devises rules for local self-organization to model stakeholder interactions using game theory principles.Then this study integrates the human-land interactions portrayed by ABM,guided by the defined scope and rules,into the CA neighborhood construction.This approach leads to the creation of a CA-ABM-coupled urban expansion simulation and prediction model with a fine depiction of local self-organization processes(CA-ABM-LSO).This model revolves around finely defined localized self-organization and achieves a deep integration of CA and ABM within the foundational structure,which enables a coupled simulation of natural and self-organizational urban expansion processes.Using Wuhan as a case study,the results show that the CA-ABM-LSO effectively leverages its capabilities to depict both natural and self-organized urban expansion.This enhancement significantly improves urban expansion simulation accuracy and refines the landscape patterns of simulated urban patches.Rules based on game theory that govern local self-organization can effectively guide the behaviors of

关 键 词：城镇扩展 元胞自动机 智能体 自组织 

分 类 号：P208[天文地球—地图制图学与地理信息工程]