Method for classifying urban land by using heavy truck travel data

Abstract

The invention provides a method for classifying urban land by using heavy truck travel data. The method comprises the steps that a city is divided into equal-area grids in space, the truck inflow, outflow and net inflow of each grid within a certain time are calculated, the freight time sequence of each grid is obtained, network community division is conducted on all the grids according to the similarity between the freight time sequences of all the grids, and the freight time sequences of all the grids are obtained. And analyzing the freight time sequence characteristics of each network community, identifying the urban land use type corresponding to each network community, and obtaining the spatial position distribution characteristics of various types of network communities in the city. The method can provide scientific support for urban freight traffic policies and plans, can reveal space-time laws of different areas in freight activities, can provide scientific basis for the urban freight traffic policies and logistics plans, can reflect freight characteristics of urban land, can dynamically update classification results, and can adapt to urbanization requirements of rapid development.

Description

Technical Field

[0001] The present invention relates to the technical field of urban land planning, and in particular to a method for classifying urban land using heavy truck travel data. Background Art

[0002] Classification of urban land refers to the functional division of specific areas within a city to meet the different needs of urban residents and economic activities. With the rapid advancement of global urbanization, the rational planning and efficient use of urban land have become key links in promoting sustainable urban development. Traditional urban land classification methods mainly rely on population mobility patterns, and divide cities into functional areas such as residential areas, commercial areas, and industrial areas by analyzing information such as human mobility trajectories, social media data, and taxi trajectories. These methods can provide effective information about urban functions, but they are significantly insufficient in reflecting the characteristics of cargo flow. Freight is an important support for urban economic and social operations. Its activities involve multiple fields such as production, distribution, circulation, and consumption. In particular, heavy trucks play a vital role in the logistics system. Therefore, using freight data to classify urban land can not only supplement the shortcomings of traditional classification methods, but also provide new perspectives and scientific basis for urban logistics planning and freight policy formulation.

[0003] The relationship between freight activities and urban land use is an important research direction for urban logistics and transportation planning. As an important part of the freight system, the activity trajectory and flow data of heavy trucks can reflect the logistics demand and functional characteristics of different regions. However, most existing studies focus on the assessment of the impact of freight activities on traffic flow and the environment, as well as the site selection and layout of logistics facilities based on freight demand. No research has directly used freight data to classify urban land use.

[0004] At present, the following data sources and technical methods are mainly used in the existing urban land classification research:

[0005] 1. Land use classification using remote sensing data

[0006] Satellite images and remote sensing technology are used to identify land cover types, and different functional areas are determined through manual annotation or specific algorithms. This method has wide applicability, but the classification accuracy depends on the resolution of remote sensing images, the update cycle is long, and it cannot dynamically reflect the characteristics of freight activities.

[0007] 2. Land use classification using passenger transport data

[0008] Population mobility data (such as mobile phone signaling data, taxi trajectory data, and social media geographic information) are used to extract population activity characteristics in the region and classify urban land through clustering algorithms. For example, some scholars used the activity data of Weibo users in Beijing to divide urban land into seven types, including residential areas, commercial areas, and university dormitories; some scholars identified residential, commercial, logistics, and nightlife land types in five Spanish cities based on mobile phone call data. However, population mobility data cannot effectively reflect the laws of cargo flow, which limits the applicability of classification results in freight planning.

[0009] 3. Existing results of freight transport research

[0010] In the field of freight, the research focuses on the optimization of logistics facility layout, freight demand forecasting, and the impact of freight on urban transportation systems. For example, some scholars analyzed the impact of commercial and industrial land use on freight activities in 14 cities in the UK, revealing the regulatory role of logistics facility layout on freight flow. Some scholars used regression analysis to explore the relationship between different land use types and freight trip attraction (FTA) in New York City. These studies analyzed the relationship between freight demand and urban functions from the perspective of freight behavior, but did not conduct in-depth research on urban land use classified by freight data.

[0011] At present, the existing urban land classification methods include the following:

[0012] 1. Urban land classification method based on taxi trajectories

[0013] For example, some scholars have used taxi trajectory data in Hangzhou to extract freight time series characteristics such as the average daily passenger pick-up and drop-off volume in each area, and classified urban land types through support vector machine (SVM) and K nearest neighbor (KNN) algorithms. The shortcomings of this method include: taxi trajectory data mainly reflects population mobility, and the reflection of freight flow characteristics is not comprehensive enough, and it cannot reveal the spatiotemporal laws of urban land in freight activities, which limits its applicability in urban freight planning.

[0014] 2. Urban land classification method based on multi-source data fusion

[0015] Some studies have attempted to integrate multiple data sources (such as population mobility data and POI data) to classify urban land use. For example, some scholars have used Twitter geo-tagged tweet data combined with regional characteristics and adopted a random forest algorithm to classify Chicago's urban land use types into commercial areas, residential areas, etc. The disadvantages of this method include: these data sources cannot capture freight characteristics, and the classification results are difficult to use for freight-related decisions.

[0016] 3. Urban land classification method based on freight demand forecast

[0017] Some scholars have used linear regression models to predict freight demand across the United States and analyzed the impact of different spatial factors on freight volume. The shortcomings of this method include: although the study revealed the correlation coefficient between freight activities and urban land use, it did not apply freight data (especially heavy truck flow data) to urban land classification, resulting in a lack of accurate basic support for urban freight planning and policy making. Summary of the invention

[0018] The present invention provides a method for classifying urban land by using heavy truck travel data, so as to realize scientific planning and management of urban freight system.

[0019] In order to achieve the above object, the present invention adopts the following technical scheme.

[0020] A method for classifying urban land use using heavy truck travel data includes:

[0021] Divide the city spatially into equal-area grids;

[0022] Calculate the truck inflow, outflow, and net inflow of each grid within a certain period of time to obtain the freight time series of each grid;

[0023] According to the similarity between the freight time series of each grid, all grids are divided into network communities;

[0024] The freight time series characteristics of each network community are analyzed, the urban land use type corresponding to each network community is identified, and the spatial location distribution characteristics of various types of network communities in the city are obtained.

[0025] Preferably, the spatial division of the city into equal-area grids comprises:

[0026] Using the urban geographic boundary data, the urban area is spatially gridded according to the hexagonal grid division algorithm. The city is spatially divided into hexagonal grids of equal area. Each hexagonal grid is used as the basic unit of analysis, and each grid is assigned a unique number.

[0027] Preferably, the calculation of the truck inflow, outflow and net inflow of each grid within a certain period of time to obtain the freight time series of each grid includes:

[0028] Obtain satellite navigation trajectory data of heavy trucks in the city, extract the starting point, end point and time information of each trip of the heavy trucks from the satellite navigation trajectory data, map the trajectory data to the geographical location according to the starting point, end point and time information of each trip, map the starting point and end point of the heavy trucks in the city to the defined hexagonal grid, and calculate the inflow, outflow and net inflow of each grid within a certain period of time, respectively. The inflow is the number of trucks entering the grid per hour, the outflow is the number of trucks leaving the grid per hour, and the net inflow is the inflow minus the outflow. The inflow, outflow and net inflow data of each grid constitute the freight time series characteristics of each grid within a certain period of time.

[0029] Preferably, the network community division of all grids according to the similarity between the freight time series of each grid includes:

[0030] The Pearson correlation coefficient is used to calculate the freight time series correlation coefficient of each pair of grids. The calculation formula of the Pearson correlation coefficient is as follows:

[0031]

[0032] r xy is the correlation coefficient between the two freight time series x and y, x i and i are the i-th values ​​of the two sets of freight time series, and are the average values ​​of the two time series, and n is the length of the time series.

[0033] If r xy <0 sets it to 0; otherwise, retains the original value r xy , forming a positive correlation coefficient matrix, the row elements in the positive correlation coefficient matrix represent the grid numbers, the column elements represent the grid numbers, and each element M of the positive correlation coefficient matrix ij represents the positive Pearson correlation coefficient value between the time series of the ith grid and the jth grid;

[0034] The grids in the positive correlation coefficient matrix are used as nodes, and the positive correlation coefficients are used as edge weights to construct a correlation coefficient network, and a network community detection algorithm is applied to divide the grids into network communities based on the correlation coefficient network.

[0035] Preferably, the freight time series characteristics of each network community are analyzed, the urban land type corresponding to each network community is identified, and the spatial location distribution characteristics of various types of network communities in the city are obtained, including:

[0036] Summarize the freight time series of all grids in each network community and calculate the total freight time series curve of the network community. The horizontal axis of the total freight time series curve represents the time period of a day, in hours, from 0 to 23, representing 24 hours in a day, and the vertical axis of the total freight time series curve represents the number of trucks. According to the shape of the total freight time series curve of the network community and the peak time period of the traffic, the urban land use type corresponding to the network community is classified into four types: early morning type, daytime type, evening type and night type;

[0037] By calculating the Pearson correlation coefficient between the total freight time series of the network community and the freight time series of each grid, the consistency of the time characteristics of the grid and the network community to which it belongs is verified to ensure the accuracy of the classification results. According to the location coordinate information of various types of network communities, the spatial location distribution of various types of network communities in the city is obtained.

[0038] It can be seen from the technical solutions provided by the above embodiments of the present invention that the method of the present invention can provide scientific support for urban freight traffic policies and planning, and can reveal the spatiotemporal laws of different regions in freight activities, thereby providing a scientific basis for urban freight traffic policies and logistics planning. This method can not only reflect the freight characteristics of urban land, but also dynamically update the classification results to adapt to the needs of rapid urbanization.

[0039] Additional aspects and advantages of the present invention will be given in part in the following description, which will become obvious from the following description, or may be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0041] Figure 1 A processing flow chart of a resource management method in a multimedia communication system provided by an embodiment of the present invention;

[0042] Figure 2 A schematic diagram of a 24-hour freight time series of Chengdu grid division and partial grid truck flow provided by an embodiment of the present invention;

[0043] Figure 3 A schematic diagram of a process of establishing a correlation coefficient network of Beijing, Chengdu, Wuhan and Guangzhou provided by an embodiment of the present invention;

[0044] Figure 4A schematic diagram of the distribution of the number of grids in four cities, Beijing, Chengdu, Wuhan and Guangzhou, which are divided into the same network community and the number of grids divided into the most frequently occurring network community in 100 network community divisions provided by an embodiment of the present invention;

[0045] Figure 5 A schematic diagram of classifying network communities into early morning, daytime, evening and nighttime types in Beijing, Chengdu, Guangzhou and Wuhan according to curve shapes and peak time periods provided by an embodiment of the present invention;

[0046] Figure 6 A spatial distribution map of four types of urban land in Beijing, Chengdu, Guangzhou and Wuhan provided in an embodiment of the present invention;

[0047] Figure 7 A time series of truck inflow and outflow freight in all areas of four land use types in four cities, Beijing, Chengdu, Guangzhou and Wuhan, provided by an embodiment of the present invention Figure 5 Schematic diagram of the distribution (overall trend) of the Pearson correlation coefficient between the total inflow and total outflow freight time series for each group shown. DETAILED DESCRIPTION

[0048] The embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be interpreted as limiting the present invention.

[0049] It will be understood by those skilled in the art that, unless expressly stated, the singular forms "one", "said", and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present invention refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof. It should be understood that when we refer to an element as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be intermediate elements. In addition, the "connection" or "coupling" used herein may include wireless connection or coupling. The term "and / or" used herein includes any unit and all combinations of one or more associated listed items.

[0050] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as those generally understood by those skilled in the art in the art to which the present invention belongs. It should also be understood that terms such as those defined in common dictionaries should be understood to have meanings consistent with the meanings in the context of the prior art, and will not be interpreted with idealized or overly formal meanings unless defined as herein.

[0051] To facilitate understanding of the embodiments of the present invention, several specific embodiments will be further explained below with reference to the accompanying drawings, and each embodiment does not constitute a limitation on the embodiments of the present invention.

[0052] The embodiment of the present invention extracts the freight time series characteristics of heavy truck flow and combines the correlation coefficient network with the network community partitioning algorithm to realize the urban land classification method using freight data for the first time. Compared with the existing technology, the present invention can fully reveal the spatiotemporal laws of freight activities and provide scientific support for urban freight traffic policies and logistics facility planning.

[0053] The processing flow of a method for classifying urban land using heavy truck travel data provided by an embodiment of the present invention is as follows: Figure 1 As shown, the following processing steps are included:

[0054] Step S1: Divide the city into equal-area grids in space.

[0055] Step S2: Calculate the truck inflow, outflow, and net inflow of each grid every hour within 24 hours to form a freight time series for each grid, and use the freight time series as the feature of each grid.

[0056] Step S3: divide all grids into network communities according to the similarity between the freight time series of each grid.

[0057] Step S4: Analyze the freight time series characteristics of each network community, identify the urban land type corresponding to each network community, and obtain the spatial location distribution characteristics of various types of network communities in the city.

[0058] Furthermore, in one example, step S1 specifically includes:

[0059] Using the urban geographic boundary data, the urban area is spatially gridded according to the hexagonal grid division algorithm, and the city is spatially divided into equal-area grids. Each hexagonal grid is used as the basic unit of analysis, and each grid is assigned a unique number for subsequent data association. Figure 2 A schematic diagram of a 24-hour freight time series of Chengdu grid division and partial grid truck flow provided by an embodiment of the present invention. Figure 2 (a) is the outline map of Chengdu. Figure 2 (b) Chengdu is divided into equal-area hexagonal grids, each with a side length of 1 km. The reason for using hexagons is that their shape is close to a circle, can be seamlessly spliced, and has good spatial uniformity. Figure 2 (c) is the grid where the daily truck inflow is more than 2.

[0060] Further, in one example, step S2 specifically includes:

[0061] By calling the satellite navigation trajectory data of heavy trucks in the city, the starting point, end point and time information of each trip of heavy trucks are extracted from the satellite navigation trajectory data, and the trajectory data is geographically mapped according to the starting point, end point and time information of each trip, and the starting point and end point of the trip of heavy trucks in the city are mapped to the defined hexagonal grid, and the inflow, outflow and net inflow of each grid within a certain time are calculated respectively. The inflow is the number of trucks entering the grid per hour, the outflow is the number of trucks leaving the grid per hour, and the net inflow is the inflow minus the outflow. The inflow, outflow and net inflow data of each grid constitute the freight time series characteristics of each grid within a certain time. The above-mentioned certain time can be 24 hours. By accumulating the inflow, outflow and net inflow of each grid hourly, a 24-hour freight time series can be generated.

[0062] Exclude grids with low freight traffic (e.g. grids with an average daily truck inflow of less than 2, see Figure 2 (c)), only the grids with frequent freight activities are retained for analysis.

[0063] Furthermore, in one example, the above step S3 specifically includes: constructing a network and dividing the network community

[0064] The Pearson correlation coefficient is used to calculate the correlation coefficient of the freight time series for each pair of grids and generate a correlation coefficient matrix.

[0065] The calculation formula of the above Pearson correlation coefficient is as follows:

[0066]

[0067] r xy is the correlation coefficient between the two freight time series x and y

[0068] x i and i are the i-th values ​​of the two freight time series respectively.

[0069] and are the average values ​​of these two time series.

[0070] n is the length of the time series, in this case the number of time periods of 24 hours.

[0071] In the case of the present invention, on this basis, only the positive correlation coefficients are retained to form a positive correlation coefficient matrix.

[0072] If r xy <0 sets it to 0; otherwise, retains the original value r xy

[0073] The above correlation coefficient matrix is ​​constructed based on all grids in the study area. The matrix contains:

[0074] Row element: represents the number of the grid.

[0075] Column element: represents the number of the grid.

[0076] Each element of the matrix M ij Represents the positive Pearson correlation coefficient value between the time series of the ith grid and the jth grid. If the Pearson correlation coefficient between grid 10 and grid 17 is 0.8, then the value of the 10th row and 17th column of this matrix is ​​0.8.

[0077] The negative correlation coefficients in the correlation coefficient matrix are set to zero, and only the positive correlation coefficients are retained to form a positive correlation coefficient matrix.

[0078] The grid is used as the node and the positive correlation coefficient is used as the edge weight to construct the correlation coefficient network.

[0079] Figure 3 The process of establishing correlation coefficient networks in Beijing, Chengdu, Wuhan and Guangzhou is demonstrated. Figure 3 (ad) The four sub-figures show the correlation coefficient matrix of the four cities, with the grid number as the coordinate axis, and the size of the freight time series correlation coefficient between each grid is shown by the change of color. Figure 4 (ae) are schematic diagrams of the correlation coefficient networks of the four cities. These sub-graphs provide a visual representation of the correlation coefficient networks of the corresponding cities. Figure 4 Due to the large size of the network, only the edges with the top 5% weights are shown, while the edges not shown represent other connections with relatively low weights.

[0080] According to the correlation coefficient network, the existing network community detection algorithm (such as Louvain algorithm) is applied to divide the network community of the grid. The process of dividing the network community of the grid based on the correlation coefficient network mentions the use of Louvain algorithm in the present invention. In the present invention, the community detection algorithm is only a technical means of implementation. The following is a detailed description and steps:

[0081] 1. Construction of correlation coefficient network

[0082] Input data:

[0083] The above correlation coefficient matrix is ​​calculated from the freight time series.

[0084] Convert to network:

[0085] Node: Mesh (each mesh as a node).

[0086] Edge weight: the positive correlation coefficient value between any two nodes.

[0087] For example, if the Pearson correlation coefficient between grid 10 and grid 17 is 0.8, then an edge with a weight of 0.8 is constructed between node 10 and node 17.

[0088] 2. Basic ideas of network community division

[0089] After the network is built, there are many existing network community division algorithms, and they are all relatively mature. The Louvain algorithm is used in the case of the present invention. The following is the basic idea of ​​the algorithm

[0090] Core objectives:

[0091] The Louvain algorithm divides the network into several subcommunities by maximizing modularity. Modularity is an indicator of the quality of network partitioning, which represents the difference between the connection density between nodes within a grid and the connection sparsity between grids.

[0092] Definition of modularity:

[0093] For a weighted network, the formula for modularity Q is:

[0094]

[0095] in

[0096] A ij represents the edge weight between nodes i and j

[0097] k i Represents the total weight of node i (the sum of the weights of all connected edges)

[0098] m represents the sum of all edge weights in the network

[0099] c i ,c j Respectively represent the community numbers to which nodes i and j belong

[0100] δ(c i ,c j ): If c i =cj , then it is 1, otherwise it is 0.

[0101] By maximizing Q, the Louvain algorithm is able to identify tightly connected communities.

[0102] 3. Steps of Louvain algorithm

[0103] 1. Initialization:

[0104] Each node individually forms a community.

[0105] Calculate the initial modularity Q.

[0106] 2. Local optimization:

[0107] For each node, we try to move it to the adjacent community and calculate the modularity increment.

[0108] If the modularity increases, the node is added to the adjacent community; otherwise, the original community is retained.

[0109] 3. Community merger:

[0110] Treat each community as a supernode and merge the networks.

[0111] Repeat the local optimization in the new network.

[0112] 4. Termination conditions:

[0113] The iteration stops when the modularity increment is less than a certain threshold (or reaches the maximum modularity).

[0114] The algorithm is run multiple times to ensure stability, and for grids with inconsistent results, the network community with the most occurrences is selected as the final result. As an example, Figure 4 (a) is the proportion of Beijing, Chengdu, Guangzhou and Wuhan that are always assigned to the same cluster in 100 Louvain algorithm network community divisions, and the proportion of network community divisions remains unchanged in any one of them. The results show that among the four cities, the proportion of Beijing reaches 98%, and the proportion of Chengdu exceeds 76%. Figure 4 (be) is the cumulative distribution C(n) of the number of times (n) that the grids whose network communities of Beijing, Chengdu, Guangzhou, and Wuhan have changed appear in their most frequent clusters in 100 network community divisions. In the example, although the network communities to which some grids belong have changed, the proportion of grids that are divided into the network communities with the most appearances is generally above 50%. The network community with the most appearances in each grid is taken as the final network community of this grid.

[0115] Furthermore, in one example, step S4 specifically includes: according to the network community division result, by analyzing the freight time series characteristics and spatial distribution of each network community, identifying the urban land type corresponding to each network community.

[0116] 1. Analysis of freight time series characteristics:

[0117] Summarize the freight time series of all grids in each network community and calculate the total freight time series curve of the network community. The horizontal axis (X-axis) of the above total freight time series curve represents the time period of a day, usually in hours, from 0 to 23, representing 24 hours in a day. The vertical axis (Y-axis) of the above total freight time series curve represents the number of trucks (trucks / hour).

[0118] According to the shape of the total freight time series curve of the network community and the peak time period of the flow, the network community is classified into four types: early morning type, daytime type, evening type and night type. This classification is based on the spatiotemporal laws of freight activities, focusing on the role of freight characteristics in the urban logistics system, rather than the traditional land use division based on passenger data (such as commercial areas, residential areas, etc.). Specifically:

[0119] 1. Morning Type

[0120] Time characteristics: Freight traffic reaches its peak in the early morning (5AM to 9AM).

[0121] Corresponding urban land functions:

[0122] This type of freight peak is closely related to the operational needs of early nodes in the logistics chain. For example, factories and storage facilities need to complete the dispatch of raw materials or goods in the early morning to support subsequent production or distribution.

[0123] The functions of this type of land are more reflected in its association with manufacturing, logistics centers and large-scale warehousing facilities.

[0124] 2. Daytime type

[0125] Time characteristics: Freight traffic remains high during the day (8AM to 6PM).

[0126] Corresponding urban land functions:

[0127] Daytime areas usually correspond to industrial sites and wholesale markets, reflecting the concentration of daytime production, processing, and bulk distribution activities.

[0128] Freight peaks on such lands coincide with daytime factory production activities and distribution needs of wholesale markets.

[0129] 3. Evening type

[0130] Time characteristics: Freight traffic reaches a peak in the evening (5PM to 10PM), with a secondary peak at noon.

[0131] Corresponding urban land functions:

[0132] The evening zones are closely related to logistics nodes outside the city. These areas are usually temporary stops for trucks before the daytime traffic restrictions are lifted.

[0133] The functions of this type of land are often related to temporary transit stations, logistics distribution centers or distribution centers close to the city center.

[0134] 4. Night type

[0135] Time characteristics: Freight traffic is concentrated at night (8PM to midnight) and gradually decreases from midnight to early morning.

[0136] Corresponding urban land functions:

[0137] Night-time zones are usually located in urban central areas, reflecting the demand for heavy freight entering the city at night.

[0138] The function of this type of land is mainly to serve commercial retail terminals (such as supermarkets and markets) or other high-demand areas (such as construction sites).

[0139] As an example, Figure 5 A schematic diagram provided in an embodiment of the present invention is provided for classifying network communities into early morning type, daytime type, evening type and nighttime type according to curve shapes and peak time periods in Beijing, Chengdu, Guangzhou and Wuhan.

[0140] Figure 5 The solid cyan line represents the truck inflow, and the dotted gray line represents the truck outflow. Sub-figures (a, b, c, d), (e, f, g, h), (i, j, k, l), and (m, n, o, p) show the truck inflow and outflow time series of the four cities for early morning, daytime, evening, and nighttime land use, respectively.

[0141] According to the location coordinate information of various types of network communities, the spatial location distribution of various types of network communities in the city is obtained. Figure 6 A spatial distribution map of four types of urban land in Beijing, Chengdu, Guangzhou and Wuhan is provided in an embodiment of the present invention. Figure 6Spatial distribution of four types of urban land use in (a) Beijing, (b) Chengdu, (c) Guangzhou, and (d) Wuhan. The height indicates the relative size of the truck flow (the sum of the truck inflow and outflow) of the grid, and the color indicates the type of land use to which the grid belongs. The early morning type is represented by cyan, the daytime type is represented by brown, the evening type is represented by pink, and the night type is represented by midnight blue. Comparing the distribution differences of the four types of land use in the city center and suburbs, it is found that in these four case cities, early morning and daytime types of land use are mostly concentrated in the suburbs of the city. The evening type of land use is distributed around the city center. The night type of land use is mainly located in the city center area.

[0142] By calculating the Pearson correlation coefficient between the total freight time series of the network community and the freight time series of each grid, the consistency of the time characteristics of the grid and the network community to which it belongs is verified, and the accuracy of the classification results is ensured. Figure 7 The distribution of Pearson correlation coefficients is shown for visual analysis. If the correlation coefficients of most grids are close to 1, it means that the time characteristics of the grid are consistent with the total time characteristics of the community to which it belongs; if the correlation coefficients are more dispersed or generally low, it may indicate that there are differences in time characteristics within the community.

[0143] As an example, the truck inflow and outflow freight time series are calculated for all grids in each city. Figure 5 Pearson correlation coefficient of the truck inflow and outflow freight time series (overall trend) for the sum of the various network communities shown in Figure 7 A time series of truck inflow and outflow freight in all areas of four land use types in four cities, Beijing, Chengdu, Guangzhou and Wuhan, provided by an embodiment of the present invention Figure 5 The Pearson correlation coefficient distribution (overall trend) between the total inflow and outflow time series of each group is shown. ad is the morning type of land use in the four cities, eh is the daytime type of land use, il is the evening type of land use, and mp is the night type of land use. Figure 7 As shown in the figure, for the grids of early morning, daytime and evening land use, the correlation coefficients are generally high, indicating that the truck flow freight time series of these grids are consistent with the overall trend, and different grids of the same type of land use have similar time characteristics. For night-time land use, the distribution of the correlation coefficient is relatively scattered, indicating that in night-time land use, the degree of correlation between the truck flow of each grid and the overall trend is different. But overall, the grids with positive correlation coefficients account for the majority.

[0144] In summary, compared with the existing urban land classification technology using population flow data, the embodiments of the present invention make up for the deficiency of traditional methods in revealing the law of cargo flow by utilizing heavy truck flow data. The present invention extracts the characteristics of freight time series, constructs a regional correlation coefficient network, and combines the Louvain algorithm to achieve regional classification, which can comprehensively reveal the spatiotemporal laws of different regions in freight activities, thereby providing a scientific basis for urban freight traffic policies and logistics planning. This method can not only reflect the freight characteristics of urban land, but also dynamically update the classification results to adapt to the needs of rapid urbanization.

[0145] The present invention significantly reduces the high cost and long cycle of traditional classification that relies on remote sensing data or field surveys through data-driven automated analysis methods. Compared with existing technologies, the present invention has significant advantages in classification accuracy, dynamic update capability, and comprehensiveness of freight characteristics analysis, providing a new technical path for urban land classification research and promoting the scientific process of freight system planning and management.

[0146] Those skilled in the art can understand that the accompanying drawings are only schematic diagrams of an embodiment, and the modules or processes in the accompanying drawings are not necessarily required to implement the present invention.

[0147] It can be known from the description of the above implementation methods that those skilled in the art can clearly understand that the present invention can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which can be stored in a storage medium such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in the various embodiments of the present invention or certain parts of the embodiments.

[0148] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device or system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. The device and system embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0149] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for classifying urban land using heavy truck travel data, characterized in that: include: Divide the city spatially into equal-area grids; Calculate the truck inflow, outflow, and net inflow of each grid within a certain period of time to obtain the freight time series of each grid; According to the similarity between the freight time series of each grid, all grids are divided into network communities; The freight time series characteristics of each network community are analyzed, the urban land use type corresponding to each network community is identified, and the spatial location distribution characteristics of various types of network communities in the city are obtained.

2. The method according to claim 1, characterized in that The above-mentioned spatial division of the city into equal-area grids includes: Using the urban geographic boundary data, the urban area is spatially gridded according to the hexagonal grid division algorithm. The city is spatially divided into hexagonal grids of equal area. Each hexagonal grid is used as the basic unit of analysis, and each grid is assigned a unique number.

3. The method according to claim 2, characterized in that The above calculation of the truck inflow, outflow and net inflow of each grid within a certain period of time obtains the freight time series of each grid, including: Obtain satellite navigation trajectory data of heavy trucks in the city, extract the starting point, end point and time information of each trip of the heavy trucks from the satellite navigation trajectory data, map the trajectory data to the geographical location according to the starting point, end point and time information of each trip, map the starting point and end point of the heavy trucks in the city to the defined hexagonal grid, and calculate the inflow, outflow and net inflow of each grid within a certain period of time, respectively. The inflow is the number of trucks entering the grid per hour, the outflow is the number of trucks leaving the grid per hour, and the net inflow is the inflow minus the outflow. The inflow, outflow and net inflow data of each grid constitute the freight time series characteristics of each grid within a certain period of time.

4. The method according to claim 3, characterized in that The above-mentioned network community division of all grids according to the similarity between the freight time series of each grid includes: The Pearson correlation coefficient is used to calculate the freight time series correlation coefficient of each pair of grids. The calculation formula of the Pearson correlation coefficient is as follows: r xy is the correlation coefficient between the two freight time series x and y, x i and i are the i-th values ​​of the two sets of freight time series, and are the average values ​​of the two time series, and n is the length of the time series. If r xy <0 sets it to 0; otherwise, retains the original value r xy , forming a positive correlation coefficient matrix, the row elements in the positive correlation coefficient matrix represent the grid numbers, the column elements represent the grid numbers, and each element M of the positive correlation coefficient matrix ij represents the positive Pearson correlation coefficient value between the time series of the ith grid and the jth grid; The grids in the positive correlation coefficient matrix are used as nodes, and the positive correlation coefficients are used as edge weights to construct a correlation coefficient network, and a network community detection algorithm is applied to divide the grids into network communities based on the correlation coefficient network.

5. The method according to claim 4, characterized in that The freight time series characteristics of each network community are analyzed, the urban land type corresponding to each network community is identified, and the spatial location distribution characteristics of various types of network communities in the city are obtained, including: Summarize the freight time series of all grids in each network community and calculate the total freight time series curve of the network community. The horizontal axis of the total freight time series curve represents the time period of a day, in hours, from 0 to 23, representing 24 hours in a day, and the vertical axis of the total freight time series curve represents the number of trucks. According to the shape of the total freight time series curve of the network community and the peak time period of the traffic, the urban land use type corresponding to the network community is classified into four types: early morning type, daytime type, evening type and night type; By calculating the Pearson correlation coefficient between the total freight time series of the network community and the freight time series of each grid, the consistency of the time characteristics of the grid and the network community to which it belongs is verified to ensure the accuracy of the classification results. According to the location coordinate information of various types of network communities, the spatial location distribution of various types of network communities in the city is obtained.

Images