Research On The Spatial Pattern And Interaction Measurement Of Chinese Urban Population Flow Network Based On Multimodal Migration Data

The urban population has formed multimodal traffic flows between cities through different modes of travel.Research on the spatial pattern of inter-city population mobility networks and the measurement of inter-city interactions based on different modes of travel is currently an important content of concern in the areas of urban geography,such as flow space,urban network,and spatial structure.However,the existing research on urban migration networks mainly focuses on inter-provincial migration or intra-urban population migration in specific regions,and lacks the analysis of inter-city population migration laws and interaction laws between cities on a national scale.In recent years,some studies have begun to focus on urban migration networks on the national scale,but mainly focus on single modes of transportation such as roads,railways and aviation,and on the multi-modal urban migration networks under the combined action of multiple transportation modes on the national scale.Research is still in its infancy.The national scale multimodal urban population migration network research mainly faces the following problems:(1)Most of the population mobility data is collected through censuses,dynamic monitoring surveys of floating population,individual questionnaires and interviews,etc.The sample data is insufficient in accuracy and timeliness(2)Existing urban spatial patterns and interaction measurement models are mostly built on small-scale urban agglomerations with limited data volume,and there is insufficient support for the measurement and analysis of large-scale,massive,and real-time inter-city population migration data.The use of massive Internet location migration data to improve the measurement methods of the spatial pattern and spatial interaction of multimodal urban population migration networks can help improve the understanding of the spatial pattern and interaction characteristics of urban population in the country,and obtain the role of different transportation modes.The characteristics of the network relationship between cities across the country and the interaction between cities.Based on Tencent's migration data,this paper analyzes the characteristics of the complex network structure of Chinese cities in multi-modal population flows,including nodes,overall structure and time evolution characteristics;explores the distribution characteristics of urban spatial patterns of multi-modal population flows,including City scale,spatial distribution and spatial agglomeration;based on the combination of multimodal migration data and socio-economic statistics,a multi-level urban scale calculation index system was designed,the traditional gravity model was improved,and the role of multi-modal traffic flow was measured Under the gravitational coefficient and distance attenuation coefficient of urban space interaction,the measurement of urban space interaction is realized.The research content of this article mainly includes the following aspects:(1)Topological characteristics of urban population mobility network with multimodal migration data.Study the mapping relationship between population mobility data and urban networks,and establish a city complex network model based on multimodal population mobility between cities;for the city directed weighted network and topology network formed by this model,analyze the correspondence through complex network analysis methods Urban complex network node characteristics and overall structure characteristics;quantitatively reveal the time evolution of network characteristics under different measures.The results show that the key national firsttier cities have strong centrality,and some cities in remote areas(Ali region)have high intermediary and close to centrality.People choose migration and travel methods more and more in favor of trains;the transfer of different network nodes is less than 2 node.The size of the clustering coefficient,in turn,travel by car,train,and plane,has a greater relationship with the complete transportation facilities and the service distance.(2)Urban spatial pattern of multi-modal migration data.The study calculates the strength of the relationship between cities based on the population flow intensity based on car travel,train travel,airplane travel and total population migration between cities.Through the methods of kernel density estimation,rank-scale law and spatial data analysis,the results are different.The scale,spatial distribution and spatial agglomeration distribution of cities under the measurement have analyzed the spatial pattern characteristics of the Chinese urban system based on different measures of population mobility.The results show that the inter-city population migration in my country is not only a separate high-intensity city,but also a regional distribution that conforms to the band characteristics,showing the spatial distribution of "North,Guangzhou,Shenzhen and Chongqing" as the core city.High-value clusters or lowvalue clusters appear at the same level in urban cold hotspot areas with different population travel intensity.Hotspot areas are prone to appear at the junction of coastal areas and North China and Northeast areas.The only cold spots appear in automobile travel Part of the northeast and northwest regions are distributed in a one-sided manner.(3)Measurement of urban spatial interaction of multimodal migration data.Research and design the entropy method and PCA combined with the hierarchical index weight algorithm to calculate the city size,replace the gravity coefficient with the degree of membership in the economic connection strength,and obtain the global distance attenuation coefficient based on the regression calculation and fitting of the urban population flow strength and the urban area GDP,Then improve the traditional gravity model.Introduce the population mobility of multimodal migration data to replace the urban population size,obtain the urban gravity coefficient matrix,and use the population migration of multimodal migration data and the GDP of the urban area to substitute the gravity model to form the regression equation to obtain the global distance The attenuation coefficient is 1.6181.The improved model was used to measure the interaction of urban space based on population movement.An analysis of the spatial interactions among the five major national agglomerations shows that the positive interactions in the urban agglomeration are all core cities with strong political and economic attributes,and the negative interactions are the economic development in the urban agglomeration.