基于改进粒子群算法和元胞自动机的城市扩张模拟——以南京为例

发布时间：2018-10-08 20:08

【摘要】：为合理利用多智能体算法解决城市扩张动态模拟问题,基于地理学理论和社会学规律对粒子群算法进行有针对性的改进,提出分段式粒子群算法(SPSO),并结合元胞自动机模拟复杂时空过程的能力,构建出适用于城市扩张模拟的地理元胞自动机SPSO-CA。在SPSO-CA中我们利用多时像的土地利用数据、交通路网数据和地形数据,挖掘出1995~2000年南京城市扩张的土地转换规则。再由此规则实现1995~2008年的南京市城市扩张过程的动态模拟。最后对比SPSO-CA、PSOCA及NULL模型结果得:SPSO-CA总精度86.3%,Kappa系数为0.792,Moran’s I为0.078,PSO-CA总精度83.6%,Kappa系数为0.755,Moran’s I为0.054,NULL模型总精度81.9%,Kappa系数为0.741,真实的Moran’s I为0.072。这表明无论是总精度还是空间一致性,SPSO-CA都优于PSO-CA和NULL模型,即用SPSO-CA模拟城市扩张是可行的。
[Abstract]:In order to solve the problem of dynamic simulation of urban expansion using multi-agent algorithm reasonably, the particle swarm optimization algorithm is improved based on geography theory and sociological law. A segmented particle swarm optimization (SPSO),) algorithm is proposed and combined with the ability of cellular automata (CA) to simulate complex space-time processes. A geo-cellular automaton (SPSO-CA.), which is suitable for urban expansion simulation, is constructed. In SPSO-CA, land use data, traffic network data and terrain data are used to excavate the land conversion rules of Nanjing urban expansion from 1995 to 2000. Then the dynamic simulation of Nanjing urban expansion process from 1995 to 2008 is realized by this rule. Compared with the results of SPSO-CA,PSOCA and NULL models, it was found that the total accuracy of SPSO-CA,PSOCA and NULL models was 86.3and that the total accuracy of the model was 0.792m Morannica and 0.0778 PSO-CA. The total accuracy of the model was 0.755mnnnnl. The total accuracy of the model was 0.741, and the true Moran's I was 0.072.The total accuracy of the model was 0.74 1. The total accuracy of the model was 0.74 1, and that of the real Moran's I was 0.072%. It shows that SPSO-CA is superior to PSO-CA and NULL models in terms of total precision and spatial consistency, that is, it is feasible to use SPSO-CA to simulate urban expansion.
【作者单位】：中国科学院南京地理与湖泊研究所流域地理学重点实验室;中国科学院大学;江苏省土地勘测规划院;
【基金】：国土资源部重点地区土地综合承载力调查评价项目(DCPJ131208-01) 江苏省国土科技项目(201204)~~
【分类号】：TU984.113

【相似文献】