Method and device for characterizing spatial connection structure of urban agglomeration

Abstract

The application relates to a kind of urban agglomeration spatial connection structure characterization method and device, method includes: based on mobile communication data extraction individual mobile trajectory, identify the activity, activity site and travel of individual, construct the urban agglomeration spatial connection network based on travel;Urban agglomeration spatial connection network uses the non-overlapping community discovery algorithm based on modularity, identifies the community of close connection;Based on the urban agglomeration spatial connection community result identified, the stability of its core-edge structure is tested;Based on the community division result of urban agglomeration spatial connection network, with community as potential hinterland, by node center tendency index determine the location of central node and its covered geographical range.Compared with prior art, the application has the advantages of improving the accuracy of urban agglomeration spatial structure analysis, providing fine decision support for urban agglomeration comprehensive transportation planning.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of urban planning, and in particular to a method and device for representing the spatial connection structure of a city cluster. BACKGROUND

[0002] In the field of urban planning, the study of urban spatial structure has always been an important topic. A city is a complex system, and its spatial structure is closely related to a large number of factors such as the activities of the population in the city, social and economic development, industrial agglomeration, and urban development strategies.

[0003] There are two main approaches to the study of the spatial structure characteristics of a city cluster. One is to study the geometric morphology of the spatial layout of the city through urban morphology, mainly using satellite remote sensing, field surveys, population and economic census statistics, land use data, and other data to analyze the importance of various centers in the city. The other method is based on the discussion of functional connections, and classifies and discusses the spatial structure of the city through the structure of various flows in the city. The functional connection method believes that the "flow space" formed by population flow, material flow, technology flow, information flow, and capital flow can connect spatially discontinuous areas in the city, and form a close connection structure in the "flow space".

[0004] The analysis of the spatial connection structure of a city cluster using the functional connection method requires data support. However, in the traditional analysis of urban transportation networks, the research object is the daily transportation in the city, and the overall time-varying law is relatively simple. Therefore, the spatial division mainly considers that the data is easy to obtain (generally in the form of streets or traffic zones to facilitate the acquisition of residential population data), and is appropriately modified according to the relationship between the planning road network and the spatial division. From a larger scale, the traditional analysis mode focuses on the transportation between "cities" and "cities", and regards the city as a node and the administrative city as the basic spatial unit. However, this division method based on administrative division as the basic spatial unit cannot meet the needs of the comprehensive transportation planning of the city cluster. SUMMARY

[0005] The purpose of the present application is to overcome the defects of the prior art and provide a method and device for representing the spatial connection structure of a city cluster. The method uses mobile communication data to extract individual activity and travel information, and on this basis, constructs a city cluster spatial connection network based on daily travel. The method uses community structure, core-periphery structure, center node, and hinterland measurement indicators in complex network theory to realize the representation and analysis of the spatial structure of the city cluster, improves the accuracy of the analysis of the spatial structure of the city cluster, and provides fine decision support for the comprehensive transportation planning of the city cluster.

[0006] The purpose of the present application can be achieved by the following technical solutions:

[0007] A method for characterizing the spatial connection structure of an urban agglomeration, comprising the following steps:

[0008] S1: Extract individual movement trajectories based on mobile communication data within the study area, reconstruct the spatio-temporal reference system of the individual movement trajectories, identify the activities, activity locations and trips of individuals, and aggregate to obtain a trip-based urban agglomeration spatial connection network;

[0009] S2: Use a non-overlapping community detection algorithm based on modularity to identify communities with close connections in the urban agglomeration spatial connection network;

[0010] S3: Take the complete urban agglomeration spatial connection network as the initial value, randomly remove a certain proportion of edges in the urban agglomeration spatial connection network, and construct a new network based on the method of step S2; By comparing the community division results of the networks constructed under different edge sampling proportions, the stability of the core-periphery structure of the urban agglomeration spatial connection network is tested;

[0011] S4: Based on the community division results of the urban agglomeration spatial connection network, take the community as a potential hinterland, and determine the location of the central node and the geographical range it covers through the node center tendency index.

[0012] Further, step S1 specifically comprises the following steps:

[0013] S11: Preprocess the mobile communication data using the binning method, convert the spatio-temporal reference system of the mobile communication data from space and time dimensions, and obtain position point data containing grid numbers and times;

[0014] S12: Calculate the corresponding residence time based on the obtained position point data, thereby determining the activities, activity locations and trips of individuals;

[0015] S13: Take the grid divided by the binning method as the smallest spatial statistical unit, and take the trip volume between grids as the connection weight between nodes, thereby constructing a trip-based urban agglomeration spatial connection network.

[0016] Further, step S11 specifically comprises the following steps:

[0017] Divide the study area into equidistant grids;

[0018] For mobile communication data of each Δt minute time bin, calculate the corresponding average position, and take the grid code of the average position as the position of the individual in the time bin, thereby obtaining position point data containing grid numbers and times;

[0019] In step S12, the process of calculating the residence time is specifically as follows: Given the i-th position point, if the subsequent k position points satisfy grid i+k≠grid i+k-1 =grid i+k-2 =…=grid i ≠grid i-1 If k≥1, then the individual is at position point grid i The dwell time d is expressed as:

[0020] d = t i+k-1 -t i

[0021] The dwell time is less than the threshold t s The location of the stay is taken as the activity, and the corresponding position point is grid. i The activity location; the grid where an individual participates in two consecutive activities. i and grid i+1 Movement between these points constitutes travel.

[0022] Further, in step S13, the urban cluster spatial connection network for travel is represented by a weighted directed network G(V,E,W), where V={v1,v2,…,v…} n} is a set of nodes in the network, representing n spatial aggregate units within the research scope; It is a set of edges in a network, representing the functional relationships between spatial cluster units; e ij = <v i ,v j >∈E represents a residential spatial aggregation unit v i With the work site spatial aggregation unit v j Work-residence relationship; W = {w ij} represents the weight of an edge in the network, w ij The value is defined as:

[0023]

[0024] In the formula, q ij Let i be the number of trips starting from grid i and ending at grid j.

[0025] Furthermore, the non-overlapping community detection algorithm includes the following steps:

[0026] The steps to maximize modularity are: merge the nodes in the spatial connection network of the urban agglomeration, and recalculate the modularity after merging to maximize the modularity;

[0027] Community aggregation steps: After maximizing modularity, the communities merged in the modularity maximization step are regarded as nodes. The connections between communities are processed in a certain aggregation manner and transformed into edges between communities.

[0028] The modularity maximization step and the community set counting step are iterated until the modularity no longer increases.

[0029] Further, in the step S3, the stability index is calculated for describing the stability of the node in the process of testing the stability of the core-edge structure of the spatial connection network of the urban agglomeration, and the calculation expression of the stability index is:

[0030]

[0031] In the formula, s i is the stability of the node i, n i is the number of times that the node i is allocated to the community with the number of nodes greater than tau in the multiple sampling, and n is the total number of tests.

[0032] Further, the step S4 further includes adjusting the community division result of the spatial connection network of the urban agglomeration, and the adjustment process is specifically:

[0033] The grids belonging to the same community are merged into the vector surface of the community;

[0034] The outer boundary of each connected region of the vector surface of the same community is taken to form a new vector surface;

[0035] The connected regions of the community are judged, and if the area of the connected region is less than the preset area threshold a, it is considered that the region cannot form an effective agglomeration region and is removed.

[0036] Further, in the step S4, the grid with the minimum node center tendency index is taken as the center node, and the calculation expression of the center node is:

[0037]

[0038]

[0039]

[0040] In the formula, w i , x i , y i are the weight, longitude and latitude of the point i in the community respectively, n is the number of nodes in the community, w ij is the daily travel volume of the point i to the point j, and m is the number of all points.

[0041] Further, according to the location of the center node and the geographical range covered thereby, the comprehensive transportation planning of the urban agglomeration is carried out.

[0042] The application further provides a device for characterizing a spatial connection structure of an urban agglomeration, comprising a memory and a processor, wherein the memory stores a computer program, and the processor invokes the computer program to execute the steps of the method.

[0043] Compared with the prior art, the application has the following advantages:

[0044] The application extracts individual daily travel information based on mobile communication data, constructs an urban agglomeration spatial connection network based on travel, uses a non-overlapping community discovery algorithm based on modularity to identify spatial connection communities with close connections, tests the stability of the core-edge structure of the urban agglomeration spatial connection network, determines the geographic location of the central node by taking the urban agglomeration spatial connection community as a potential hinterland and using a node central tendency index, and thus determines the central node in the urban agglomeration spatial connection network and the geographic range covered thereby. The mobile communication data generated by mobile communication users in the process of social and economic activities in the urban agglomeration is fully utilized, the analysis precision of the spatial structure of the urban agglomeration is effectively improved, and fine decision support is provided for comprehensive transportation planning of the urban agglomeration. BRIEF DESCRIPTION OF DRAWINGS

[0045] Figure 1 A flowchart of a method for characterizing a spatial connection structure of an urban agglomeration is provided in the embodiments of the application. DETAILED DESCRIPTION

[0046] To make the objectives, technical solutions, and advantages of the embodiments of the application clearer, the technical solutions in the embodiments of the application will be described below in connection with the drawings of the embodiments of the application. Obviously, the described embodiments are only some of the embodiments of the application, rather than all the embodiments of the application. The components of the embodiments of the application described and shown in the drawings can be arranged and designed in various different configurations.

[0047] Therefore, the following detailed description of the embodiments of the application provided in the drawings is not intended to limit the scope of the claimed application, but only represents selected embodiments of the application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the application without creative labor are within the scope of protection of the application.

[0048] It should be noted that: similar reference numerals and letters represent similar items in the following drawings, and therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings.

[0049] Embodiment 1

[0050] As shown in the drawings, Figure 1 The embodiments provide a method for characterizing a spatial connection structure of an urban agglomeration, comprising the following steps:

[0051] S1: Extract individual movement trajectories based on mobile communication data within the study area, reconstruct the spatio-temporal reference system of individual movement trajectories, identify individual activities, activity locations and trips, and further aggregate the spatial connections of urban agglomerations based on trips, and construct the spatial connection network of urban agglomerations based on trips;

[0052] S2: Introduce network community structure in complex network theory to describe the spatial structure of urban agglomerations, and use non-overlapping community detection algorithm based on modularity to identify communities with close connections in the spatial connection network of urban agglomerations;

[0053] S3: Take the complete spatial connection network of urban agglomerations as the initial value, randomly remove a certain proportion of edges in the spatial connection network of urban agglomerations, and construct a new network based on the method of step S2; by comparing the community division results of the networks constructed under different edge sampling proportions, the stability of the core-periphery structure of the spatial connection network of urban agglomerations is tested;

[0054] S4: Based on the community division results of the spatial connection network of urban agglomerations, taking the community as a potential hinterland, determine the location of the central node and the geographical range covered by the central node through the node central tendency index.

[0055] Step S1 specifically includes the following steps:

[0056] S11: Preprocess the mobile communication data by using the binning method, convert the spatio-temporal reference system of mobile communication data from space and time dimensions, and obtain location point data containing grid number and time; the processing procedure is as follows:

[0057] (1) Divide the study area into equidistant grids;

[0058] (2) For mobile communication data of each Δt minute time bin, calculate the average position of the signaling data, and take the grid number of the average position as the position of the user in the time bin.

[0059] (3) After binning, the mobile phone signaling data is converted into location point data:

[0060] c i (grid i ,t i ),i=1,2,… (1)

[0061] Where c i represents the i-th location point of the user; grid i represents the grid number of the location point; t i represents the time when the user arrives at the location point.

[0062] S12: According to the obtained position point data, the corresponding residence time is calculated, so as to determine the activity, activity site and travel of the individual;

[0063] Specifically, given the i-th position point, if the subsequent k position points satisfy grid i+k ≠grid i+k-1 =…=grid i ≠grid i-1 ,k≥1, then the residence time of the user at the position point grid i can be expressed as:

[0064] d=t i+k-1 -t i (2)

[0065] The residence time less than the threshold value t s is taken as the activity, and the corresponding position point grid i is the activity site. The movement of the individual between the sites grid i and grid i+1 of the two consecutive activities is the travel.

[0066] S13: The grid divided by the binning method is taken as the minimum spatial statistical unit, and the travel amount between the grids is taken as the connection weight between nodes, so as to construct the spatial connection network of the urban agglomeration based on travel.

[0067] Specifically, the grid with the side length L is taken as the minimum spatial statistical unit, and the OD travel amount between the grids is taken as the connection weight between nodes. The travel amount from the i-th grid to the j-th grid is q ij . In this way, the travel OD can be expressed as an adjacency matrix:

[0068]

[0069] Therefore, the weighted directed network G(V,E,W) can be established to represent the spatial connection in the research area, wherein V={v1,v2,…,v n} is the set of nodes in the network, representing the n spatial aggregate units in the research range; is the set of edges in the network, representing the job-residence connection between the spatial aggregate units; e ij =<v i ,v j >∈E then represents the job-residence connection between the residential space aggregate unit v i and the working space aggregate unit v j ; W={w ij} is the weight of the edge in the network, and the value of w ij is defined as:

[0070]

[0071] In the formula, q ij is the trip volume from the i-th grid to the j-th grid.

[0072] In this embodiment, Δt is 10 minutes, t s is 30 minutes, and L is 1000 meters.

[0073] In step S2, the non-overlapping community discovery algorithm includes the following steps:

[0074] Modularity maximization step: merging nodes in the spatial connection network of the urban agglomeration and recalculating modularity after merging to maximize modularity;

[0075] Community set counting step: after modularity maximization, the community merged in the modularity maximization step is regarded as a node, and the connection between communities is subjected to certain set counting processing to convert it into an edge between communities;

[0076] Iteratively performing the modularity maximization step and the community set counting step until the modularity no longer increases.

[0077] Specifically, step S2 includes the following steps:

[0078] Modularity is used as an index for evaluating the community division result. In a complex network, modularity is a measurement value between -1 and 1, which is used to compare the internal connection tightness of the community division result with the connection tightness between communities. The modularity measurement method is to compare the proportion of edges within the community with the proportion of edges in a randomly generated network. The mathematical expression of modularity is:

[0079]

[0080] wherein M is modularity, |E| is the number of edges, w ij is the weight between nodes i and j, k i is the sum of the weights of edges connected to node i, c i is the community to which node i is assigned; when u=v, δ(u,v)=1, and otherwise, δ(u,v)=0.

[0081] The Fast unfolding algorithm based on modularity maximization is used to perform community discovery on the spatial connection network of the urban agglomeration. The main idea of the algorithm is to iteratively merge smaller communities through a threshold to form larger communities. The following two steps are performed until the modularity no longer increases:

[0082] (1) Modularity maximization: in each iteration, nodes are merged and modularity is recalculated in each merging until modularity is maximized, at which time the iteration reaches an optimal result.

[0083] (2) Community set counting: when modularity is maximized, nodes have been merged into respective communities. When the next iteration is performed, the communities merged in the previous step need to be treated as nodes, and therefore, the connections between communities also need to be set counted and converted into edges between communities for the next iteration.

[0084] Step S3 specifically includes the following steps:

[0085] S31: Starting from the complete urban agglomeration spatial connection network, a certain proportion of edges are randomly removed, and a new network is constructed. By comparing the community division results of the sub-networks constructed under different edge sampling proportions, the core and edge structure of the network community is determined. Here, an index is introduced for each node in the network to describe the stability of the node, and the calculation method is as follows:

[0086]

[0087] In the formula, s i is the stability of node i, n i is the number of times that node i is assigned to a community with more than τ nodes in multiple samplings, and n is the total number of tests.

[0088] S32: Set the threshold of the stability index s i to distinguish between core nodes and edge nodes in the network. For example, if s i is set to 1, the nodes that can be assigned to large communities each time are core nodes, and other nodes are edge nodes.

[0089] In this embodiment, s i is 1.

[0090] Step S4 specifically includes the following steps:

[0091] S41: Preferably, considering that there are outliers in the community identification result, the community identification result is adjusted by the following steps:

[0092] (1) Merge the grids belonging to the same community into a vector surface of the community;

[0093] (2) The vector surface of the same community may have multiple connected regions. Take the outer boundary of each connected region of the vector surface of the same community to form a new vector surface to eliminate "enclaves";

[0094] (3) judging each connected region of the community, if the area of the region is less than a preset area threshold a, it is considered that the region cannot form an effective gathering region, and is rejected.

[0095] S42: taking the sum of all trip amounts generated by the point i as its weight, and the calculation method is:

[0096] w i =∑ m w ij (7)

[0097] where w ij is the daily trip amount from point i to point j, and m is the number of all points.

[0098] S43: using the weighted central element to quantify the central tendency of the grid distribution in the community. That is, the sum of the weighted distances of each grid in the community to other grids is calculated, and the calculation method is:

[0099]

[0100]

[0101] where w i , x i , y i are the weight, longitude and latitude of point i in the community respectively, n is the number of nodes in the community, w ij is the daily trip amount from point i to point j, and m is the number of all points. Finally, the grid with the smallest sum of weighted distances is selected as the central node.

[0102] According to the location of the central node and the geographical range it covers, fine decision support is provided for urban agglomeration comprehensive transportation planning.

[0103] The embodiment also provides a device for characterizing the spatial contact structure of an urban agglomeration, which comprises a memory and a processor, the memory stores a computer program, and the processor invokes the computer program to execute the steps of the method described above.

[0104] The preferred embodiments of the application are described in detail above. It should be understood that those skilled in the art can make many modifications and changes without creative labor based on the concept of the application. Therefore, any technical solution obtained by logical analysis, reasoning or limited experiment based on the prior art according to the concept of the application shall be within the protection scope defined by the claims.

Claims

1. A method for characterizing spatial connection structure of urban agglomeration, characterized in that, The method comprises the following steps: S1: extracting individual movement trajectories based on mobile communication data within the research scope, reconstructing the spatio-temporal reference system of the individual movement trajectories, identifying the activities, activity locations and trips of the individuals, and summarizing to obtain a trip-based urban agglomeration spatial contact network; S2: using a non-overlapping community discovery algorithm based on modularity to identify communities with close contacts in the urban agglomeration spatial contact network; S3: taking the complete urban agglomeration spatial contact network as an initial value, randomly removing a certain proportion of edges in the urban agglomeration spatial contact network, and constructing a new network based on the method of step S2; by comparing the community division results of the networks constructed under different edge sampling proportions, the stability of the core-periphery structure of the urban agglomeration spatial contact network is tested; S4: based on the community division results of the urban agglomeration spatial contact network, taking the community as a potential hinterland, and determining the location of the central node and the geographical range covered by the central node through a node central tendency index; Step S1 specifically comprises the following steps: S11: pre-processing the mobile communication data by using a binning method, converting the spatio-temporal reference system of the mobile communication data from space and time dimensions, and obtaining location point data containing grid numbers and time; S12: calculating the corresponding stay time based on the obtained location point data, so as to determine the activities, activity locations and trips of the individuals; S13: taking the grid divided by the binning method as the smallest spatial statistical unit, taking the trip volume between grids as the contact weight between nodes, and thus constructing a trip-based urban agglomeration spatial contact network; Step S11 specifically comprises the following steps: The research scope is divided into equidistant grids; For each minute time bin of mobile communication data, the corresponding average position is calculated, and the grid in which the average position is located is encoded as the position of the individual in the time bin, thereby obtaining position point data containing grid numbers and time; In step S12, the process of calculating the dwell time specifically involves, given the first... A location point, its subsequent... If a location point satisfies Then the individual is at position point Duration of stay Represented as: a stay with a duration less than a threshold value is considered an activity, the corresponding location point is considered an activity location; movement of an individual between the locations of two consecutive activities is considered a trip and ​ In step S13, the trip-based urban agglomeration spatial contact network is represented as a weighted directed network , where, is the set of nodes in the network, representing the n spatial aggregations in the study area; is the set of edges in the network, representing the job-residence contacts between spatial aggregations; represents the job-residence contact of the residential spatial aggregation with the working spatial aggregation . is the weight of the edge in the network, the value of which is defined as: In the formula, is the number of trips from the origin to the destination, is the number of trips from the origin to the destination, is the number of trips from the origin to the destination, In step S3, the stability index is calculated for describing the stability of the node in the process of testing the stability of the core-periphery structure of the urban agglomeration spatial contact network, and the calculation expression of the stability index is: In the formula, is the stability of the node , is the stability of the node , is the number of times the node is allocated to the community with the number of nodes greater than , is the total number of tests. In step S4, the grid with the smallest node central tendency index is taken as the central node, and the calculation expression of the central node is: wherein, , , are the weight, longitude, latitude of the point i within the community respectively, is the number of nodes within the community, is the daily travel volume from point to point , is the number of all points.

2. The method of claim 1, wherein, The non-overlapping community discovery algorithm comprises the following steps: Modularity maximization step: merging the nodes in the urban agglomeration spatial contact network, and recalculating the modularity after merging to maximize the modularity; Community set counting step: after maximizing the modularity, the community merged in the modularity maximization step is regarded as a node, the contact between the communities is subjected to certain set counting processing, and the contact between the communities is converted into edges between the communities; Iteratively performing the modularity maximization step and the community set counting step until the modularity no longer increases.

3. The method of claim 1, wherein, Step S4 further comprises adjusting the community division results of the urban agglomeration spatial contact network, and the adjustment process specifically comprises: Merging the grids belonging to the same community into a vector surface of the community; Taking the outer boundary of each connected region of the vector surface of the same community to form a new vector surface; The system assesses the area of ​​each connected region within the community; if its area is less than a preset area threshold... If a region is deemed unable to form an effective cluster, it will be excluded.

4. The method of claim 1, wherein, According to the location of the central node and the geographical range covered by the central node, the urban agglomeration comprehensive transportation planning is carried out.

5. An urban agglomeration spatial connection structure characterization device, characterized in that, The system comprises a memory and a processor, the memory stores a computer program, and the processor calls the computer program to execute the steps of the method according to any one of claims 1-4.

Images