大数据城市人口时空流动特征分析与模拟  

Analysis and simulation of spatiotemporal urban population flow based on big data

作　　者：向晓芸 肖东升[1,2] 戴小军[1,2] 于冰[1,2] 马德英[1,2] XIANG Xiaoyun;XIAO Dongsheng;DAI Xiaojun;YU Bing;MA Deying(School of Civil Engineering and Geomatics,Southwest Petroleum University,Chengdu 610500,China;Disaster Emergency Research Centre for Geomatics and Remote Sensing Geographic Information,Southwest Petroleum University,Chengdu 610500,China)

机构地区：[1]西南石油大学土木工程与测绘学院,成都610500 [2]西南石油大学测绘遥感地理信息防灾应急研究中心,成都610500

出　　处：《测绘科学》2022年第3期174-185,201,共13页Science of Surveying and Mapping

基　　金：国家自然科学基金项目(51774250,41804077,41801399,41801297);四川省科技计划项目(2019JDR0112);西南石油大学测绘遥感地信与防灾应急青年科技创新团队项目(2019CXTD07);四川省科技创新苗子工程项目(2020046)。

摘　　要：针对精细化城市人口时空动态分析在智慧城市发展与建设中的发展需求问题,提出了一种基于多元大数据的城市人口精细时空流动特征分析与模拟方法。以腾讯位置大数据为基础,融合交通路网和POI数据,构建了城市内部30min时间分辨率的人口时间流动强度指数和1km空间分辨率的人口空间流动引力指数,用以从时间和空间两个角度量化分析城市人口的流动特征,并基于元胞自动机技术,建立了城市人群流动路径预测模型。基于成都市的实验结果表明:构建的城市内部人口时间流动强度指数和人口空间流动引力指数能够有效地量化分析城市人口在时间和空间上的流动特征,且基于元胞自动机的城市人群流动路径预测模型所预测的流动路径定位人口涵盖比重达75%以上,有效地模拟了成都市内部人群的日流动格局,体现了其在城市人口精细时空流动特征把握方面的应用潜质。In view of the importance and development demand of refined urban population spatiotemporal dynamic analysis in the development and construction of smart cities,this paper proposes a method for analyzing and simulating the characteristics of refined urban population spatiotemporal flow based on multivariate big data.Based on the Tencent’s location big data,combining the traffic network and POI data,this paper constructs the temporal population flow intensity index of 30 minutes time resolution and the spatial population flow gravity index of 1km spatial resolution of urban population,which are used to quantitatively analyze the flow characteristics of urban population from both time and space,and establishes the prediction model of urban populations flow paths based on cellular automata technology.The experimental results based on Chengdu city show that the temporal population flow intensity index and the spatial population flow gravity index within the city can effectively quantitatively analyze the flow characteristics of the urban population in time and space,and the positioning population proportion of the floating paths predicted by the prediction model based on cellular automata is more than 75%,which effectively simulates the daily flow pattern of the urban populations in Chengdu and reflects its application potential in understanding the fine spatiotemporal flow characteristics of the urban population.

关 键 词：人口时空流动 腾讯位置大数据 POI 元胞自动机 时间序列分解 成都市 

分 类 号：P208[天文地球—地图制图学与地理信息工程]