A method and system for converting trip flow scale by coordinating multi-dimensional data space-time features

Abstract

The application relates to a travel flow scale conversion method and system cooperating with the spatiotemporal characteristics of multi-element data, and belongs to the technical field of geographic information processing. The method solves the problem that the converted flow and the actual flow are different in the prior art. The method comprises the following steps: acquiring mobile phone signaling data and cooperative data; obtaining a user moving track sequence; determining a dominant land block feature of each base station cell based on the cooperative data; constructing a cooperative feature coefficient of each land block feature in each period based on the dominant land block feature and the user moving track sequence; constructing a spatiotemporal adjustment coefficient of flow distribution from each base station cell in each period to each target scale unit based on the cooperative feature coefficient of each land block feature in each period and the spatial distribution feature of each land block feature in each base station cell; and mapping the travel flow to the target scale unit based on the spatiotemporal adjustment coefficient of flow distribution from each base station cell in each period to each target scale unit. Precise travel flow scale conversion is realized.

Description

Technical Field

[0001] This invention relates to the field of geographic information processing technology, and in particular to a method and system for converting travel flow scales based on the spatiotemporal characteristics of collaborative multi-source data. Background Technology

[0002] The movement of urban residents between different spatial locations within a city is essentially the result of the temporal and spatial superposition of various daily activities (commuting, going to school, business, consumption, leisure, medical treatment, etc.). It not only reflects the diverse needs of individuals but also, as a whole, the rationality of urban spatial design and functional layout. With the acceleration of urbanization and the high degree of coupling within urban space, the spatial boundaries of these activities are becoming increasingly blurred, and their temporal rhythms are becoming more pronounced. Consequently, urban population movement exhibits high dynamism and significant spatiotemporal heterogeneity. Against this backdrop, traditional static travel flow statistics methods based on irregular units such as administrative boundaries, blocks, or base station cells are insufficient for the refined analysis of urban functional structures. These methods not only limit the quantitative expression of urban functional patterns but also restrict applications such as horizontal comparisons between different spatial units and spatiotemporal dynamic visualization and regional statistical analysis. Therefore, there is an urgent need for a collaborative scaling method that can fully consider the spatiotemporal differences between regions and reasonably map travel flows based on irregular base station cells to regular spatial units (such as regular grids). This method can not only construct a unified, comparable, and quantifiable spatial analysis framework, enabling travel flow hotspot analysis, heat map drawing, and quantitative visualization, but also provide key data support for applications such as urban traffic monitoring, functional pattern identification, and smart city crowd behavior modeling, thus possessing significant theoretical and practical value.

[0003] Mobile signaling data, due to its wide coverage, passive collection, and high spatiotemporal resolution, has been widely used in fields such as urban travel analysis and urban spatial structure research. In such research, the core issue lies in how to accurately map the recorded spatiotemporal location information to geographic space (i.e., achieving a conversion from the base station scale to other spatial scales). Currently, some patents employ simple processing methods to establish this mapping relationship, such as directly using the base station's latitude and longitude as the start and end points of travel flows for analysis. Other patents aggregate base station locations into spatial units, filtering stop points and characterizing urban travelers by overlaying them with administrative divisions; or they determine whether to aggregate travel flows into a specific grid unit based on whether the origin or destination falls within that grid.

[0004] While existing technologies have achieved some degree of conversion from base station cells to other spatial scales, significant limitations remain. Due to the dual "source-sink" structure of travel flows, scale conversion must simultaneously consider the "push" effect of the starting area and the "pull" effect of the ending area (i.e., the influence of regional characteristics at both ends of the travel flow on the distribution of pedestrian traffic). Current methods generally fail to comprehensively consider the impact of dynamically changing regional attributes (e.g., the functional differences of land parcels at different times) on travel flow distribution, neglecting the uneven distribution of traffic caused by spatiotemporal heterogeneity. Consequently, the traffic flow after scale conversion differs from the actual traffic flow. Summary of the Invention

[0005] Based on the above analysis, the embodiments of the present invention aim to provide a method and system for travel flow scale conversion based on the spatiotemporal characteristics of collaborative multi-dimensional data, in order to solve the problem that existing scale conversion methods ignore regional spatiotemporal heterogeneity, resulting in differences between the converted flow and the actual flow.

[0006] On one hand, embodiments of the present invention provide a method for travel flow scale transformation based on the spatiotemporal features of collaborative multi-source data, comprising the following steps: Acquire mobile signaling data and collaborative data; obtain user movement trajectory sequences based on mobile signaling data; The dominant land parcel characteristics of each base station cell are determined based on the collaborative data; Based on the dominant land parcel characteristics of each base station cell and the user movement trajectory sequence, construct the collaborative feature coefficients for each land parcel characteristic in each time period; Based on the collaborative feature coefficients of each type of land parcel feature in each time period and the spatial distribution characteristics of each type of land parcel feature in each base station cell, a spatiotemporal adjustment coefficient for allocating traffic from each base station cell to each target scale unit in each time period is constructed. Based on the spatiotemporal adjustment coefficient of each base station cell allocating traffic to each target scale unit in each time period, the travel flow of the base station cell is mapped to the target scale unit.

[0007] Based on further improvements to the above method, the spatiotemporal adjustment coefficients for allocating traffic from each base station cell to each target scale unit in each time period are constructed using the following method, based on the collaborative feature coefficients of each type of land parcel feature in each time period and the spatial distribution characteristics of each type of land parcel feature within each base station cell: ; ; ; ; in, This represents the spatiotemporal adjustment coefficient for allocating traffic from the m-th base station cell to the i-th target scale unit during the t-th time period. This represents the spatiotemporal adjustment coefficient for allocating traffic from the m-th base station cell to the i-th target scale unit in the t-th time period without normalization. This represents the number of target scale units superimposed on the m-th base station cell. This indicates that the m-th base station cell in the t-th time period of the unnormalized time period is sending data to the m-th base station cell. Spatiotemporal adjustment coefficient for allocating flow to each target scale unit This represents the synergistic characteristic coefficient of the k-th type of land parcel in the t-th time period. This represents the scale of the k-th type of land parcel feature in the region formed by the overlay of the i-th target scale unit and the m-th base station cell. The scale of the k-th type of land parcel feature in the m-th base station cell proportion, This represents the area of ​​the region formed by the superposition of the i-th target scale cell and the m-th base station cell. Occupying the total area of ​​the m-th base station cell The proportion, K represents the number of land parcel feature types.

[0008] Based on the further improvement of the above method, the travel flow of the base station cell is mapped to the target scale unit in the following way, based on the spatiotemporal adjustment coefficient of the traffic allocated by each base station cell to each target scale unit in each time period: ; in, Indicates from the start time By the end time Travel flow from the i-th target scale unit to the j-th target scale unit, Indicates the start time The spatiotemporal adjustment coefficient for allocating traffic from the m-th base station cell to the i-th target scale unit. Indicates the end time The spatiotemporal adjustment coefficient for allocating traffic from the nth base station cell to the jth target scale unit. This indicates that the m-th base station cell starts from the beginning time. By the end time The travel flow to the nth base station cell, where M represents the number of base station cells intersecting with the i-th target scale unit, and N represents the number of base station cells intersecting with the j-th target scale unit.

[0009] Based on further improvements to the above method, collaborative feature coefficients for each type of land parcel feature in each time period are constructed based on the dominant land parcel features of each base station cell and the user movement trajectory sequence, including: The population size of each base station cell in each time period is determined based on the user's movement trajectory sequence; The collaborative feature coefficients for each type of land parcel feature in each time period are calculated based on the population of each base station cell in each time period and the dominant land parcel feature of each base station cell.

[0010] Based on further improvements to the above method, the dominant land parcel characteristics of each base station cell are determined in the following manner: Calculate the land parcel feature factor for each land parcel feature based on the scale of each land parcel feature in the collaborative data included in each base station cell; For each base station cell, the land parcel feature with the largest land parcel feature factor is selected as the dominant land parcel feature of that base station cell.

[0011] Based on the further improvement of the above method, the land parcel feature factor for each base station cell belonging to each land parcel feature is calculated using the following formula: ; in, Indicates the m-th base station cell. The scale of the plot's characteristics This represents the total scale of all land parcel features within the m-th base station cell. Indicates the first The total size of each type of land parcel. This represents the total size of all land parcel characteristics. This indicates that the m-th base station cell belongs to the m-th base station cell. Plot characteristic factors of various plot characteristics.

[0012] Based on the further improvement of the above method, the collaborative feature coefficients of each type of land parcel feature in each time period are calculated using the following formula, based on the population of each base station cell in each time period and the dominant land parcel feature of each base station cell: ; in, This represents the synergistic characteristic coefficient of the k-th type of land parcel in the t-th time period. Let B represent the cumulative population of all base station cells where the dominant land parcel feature is the k-th type, and let B represent the number of land parcel feature categories. This represents the cumulative population of all base station cells whose dominant land parcel characteristic is type b.

[0013] On the other hand, embodiments of the present invention provide a travel flow scale conversion system based on the spatiotemporal features of collaborative multi-source data, comprising: The data acquisition module is used to acquire mobile signaling data and collaborative data; and to obtain the user's movement trajectory sequence based on the mobile signaling data. A dominant land parcel feature determination module is used to determine the dominant land parcel features of each base station cell based on the collaborative data; The collaborative feature coefficient determination module is used to construct collaborative feature coefficients for each type of land parcel feature in each time period based on the dominant land parcel features of each base station cell and the user movement trajectory sequence. The spatiotemporal adjustment coefficient determination module is used to construct the spatiotemporal adjustment coefficient for allocating traffic from each base station cell to each target scale unit in each time period based on the collaborative feature coefficients of each type of land parcel feature in each time period and the spatial distribution characteristics of each type of land parcel feature in each base station cell. The scale conversion module is used to map the travel flow of the base station cell to the target scale unit based on the spatiotemporal adjustment coefficient of the traffic allocated by each base station cell to each target scale unit in each time period.

[0014] Based on further improvements to the above system, the spatiotemporal adjustment coefficient determination module constructs the spatiotemporal adjustment coefficients for allocating traffic from each base station cell to each target scale unit in each time period in the following manner: ; ; ; ; in, This represents the spatiotemporal adjustment coefficient for allocating traffic from the m-th base station cell to the i-th target scale unit during the t-th time period. This represents the spatiotemporal adjustment coefficient for allocating traffic from the m-th base station cell to the i-th target scale unit in the t-th time period without normalization. This represents the number of target scale units superimposed on the m-th base station cell. This indicates that the m-th base station cell in the t-th time period of the unnormalized time period is sending data to the m-th base station cell. Spatiotemporal adjustment coefficient for allocating flow to each target scale unit This represents the synergistic characteristic coefficient of the k-th type of land parcel in the t-th time period. This represents the scale of the k-th type of land parcel feature in the region formed by the overlay of the i-th target scale unit and the m-th base station cell. The scale of the k-th type of land parcel feature in the m-th base station cell proportion, This represents the area of ​​the region formed by the superposition of the i-th target scale cell and the m-th base station cell. Occupying the total area of ​​the m-th base station cell The proportion, K represents the number of land parcel feature types.

[0015] Based on further improvements to the above system, the scale conversion module maps the travel flow of the base station cell to the target scale unit in the following manner: ; in, Indicates from the start time By the end time Travel flow from the i-th target scale unit to the j-th target scale unit, Indicates the start time The spatiotemporal adjustment coefficient for allocating traffic from the m-th base station cell to the i-th target scale unit. Indicates the end time The spatiotemporal adjustment coefficient for allocating traffic from the nth base station cell to the jth target scale unit. This indicates that the m-th base station cell starts from the beginning time. By the end time The travel flow to the nth base station cell, where M represents the number of base station cells intersecting with the i-th target scale unit, and N represents the number of base station cells intersecting with the j-th target scale unit.

[0016] Compared with the prior art, the present invention can achieve at least one of the following beneficial effects: 1. This invention introduces a collaborative data source that reflects the intensity of crowd activity by acquiring mobile phone signaling data and collaborative data; then, it determines the dominant land parcel characteristics of each base station cell based on the collaborative data; it constructs collaborative feature coefficients for each type of land parcel characteristic in each time period based on the dominant land parcel characteristics of each base station cell and user movement trajectory sequences; and it constructs spatiotemporal adjustment coefficients for allocating traffic from each base station cell to each target scale unit in each time period based on the collaborative feature coefficients for each type of land parcel characteristic in each time period and the spatial distribution characteristics of each type of land parcel characteristic in each base station cell, thereby considering the attractiveness of specific land parcel characteristics to the crowd in a specific time period, constructing dynamic spatiotemporal adjustment coefficients that integrate spatial and temporal dimensions, and performing travel flow mapping based on dynamic spatiotemporal adjustment coefficients, thereby overcoming the limitations of traditional methods in characterizing spatiotemporal heterogeneity and significantly improving the consistency and accuracy between the conversion results and actual observation data. 2. Convenient data acquisition: Based on high spatiotemporal accuracy and easily accessible mobile signaling data, and high-precision positioning data, dynamic perception of large-scale user travel behavior is achieved, effectively overcoming the problems of difficult acquisition and low update frequency of traditional travel survey data. 3. Taking into account temporal and spatial heterogeneity: During the scale transformation process, temporal dynamic features and collaborative data features are introduced simultaneously, which significantly improves the spatial distribution accuracy of travel flow data at the target scale; 4. Strong model adaptability: The proposed collaborative transformation framework has good flexibility and scalability. It can replace or supplement collaborative data (such as built environment, population distribution, land use, etc.) according to actual application needs to adapt to scale transformation tasks in different research or application scenarios.

[0017] In this invention, the above-described technical solutions can be combined with each other to achieve more preferred combinations. Other features and advantages of this invention will be set forth in the following description, and some advantages may become apparent from the description or be learned by practicing the invention. The objects and other advantages of this invention can be realized and obtained from what is particularly pointed out in the description and drawings. Attached Figure Description

[0018] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts. Figure 1 This is a flowchart of a collaborative scale conversion method for travel flows based on mobile phone signaling data, according to an embodiment of the present invention. Figure 2 This is a block diagram of a collaborative scale conversion system for travel flows based on mobile phone signaling data, according to an embodiment of the present invention. Detailed Implementation

[0019] Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, which form part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not intended to limit the scope of the present invention.

[0020] A specific embodiment of the present invention discloses a collaborative scale conversion method for travel flows based on mobile phone signaling data, such as... Figure 1 As shown, it includes the following steps: S1. Obtain mobile signaling data and collaborative data; obtain the user's movement trajectory sequence based on the mobile signaling data; S2. Determine the dominant land parcel characteristics of each base station cell based on the collaborative data; S3. Construct collaborative feature coefficients for each type of land parcel feature in each time period based on the dominant land parcel features of each base station cell and the user movement trajectory sequence; S4. Based on the collaborative feature coefficients of each type of land parcel feature in each time period and the spatial distribution characteristics of each type of land parcel feature in each base station cell, construct the spatiotemporal adjustment coefficients for allocating traffic from each base station cell to each target scale unit in each time period; S5. Based on the spatiotemporal adjustment coefficient of the traffic allocation from each base station cell to each target scale unit in each time period, the travel flow of the base station cell is mapped to the target scale unit.

[0021] Scale transformation: refers to the process of transforming data from one spatial scale (or resolution) to another. For example, aggregating or allocating travel flows based on irregular base station cells to regular grids or other scales (such as streets or administrative divisions).

[0022] During implementation, mobile signaling data and collaborative data used in the computation are acquired within the study area and time period. Collaborative data may include, for example, POI data, public transportation card swipe records, and data on the scale of industry personnel.

[0023] For example, if the research period is one day, all mobile signaling records of a user within that day would be recorded as follows: ; in, This represents a record point in the user's original mobile signaling data, which is generated only when the user's mobile phone communicates with the base station. This indicates the latitude and longitude of the base station to which the user's mobile phone was connected at the time of the interaction; This indicates the cell number of the base station. This corresponds to a timestamp. A user's mobile signaling data for one day consists of multiple trajectory points generated by the individual, recording the individual's spatiotemporal location information.

[0024] The original mobile signaling data is cleaned to remove duplicate data, process interference from signal drift and ping-pong effect, identify effective user dwell points and construct user movement trajectories, and generate high-precision user trajectory sequences.

[0025] Data cleaning: Delete duplicate records with the same timestamp, as well as invalid data such as missing or incomplete fields, to ensure the integrity and reliability of the signaling data required for subsequent analysis.

[0026] Anomaly Drift Removal: Due to signal instability over a short period, a user's location may be incorrectly associated with a distant base station. To identify such anomalies, the movement speed between two adjacent location records is calculated; if the obtained speed is greater than 120 km / h, the record is determined to be caused by signal drift and is removed from the dataset.

[0027] Ping-Pong Effect Handling: In some areas, user signaling records may frequently switch between multiple nearby base stations, forming a "ping-pong effect." To address this issue, when the distance (calculated based on latitude and longitude) between the base stations corresponding to two consecutive signaling records is less than 500 m, they are considered as oscillation records around the same location, and these records are grouped into a candidate dwell point set.

[0028] Valid stop point identification: For each candidate stop point set, calculate the time difference between its first and last records; if the duration exceeds 30 minutes, the set is considered to correspond to a meaningful stop behavior, and the centroid coordinates of each point in the set are used as the location of the stop.

[0029] After the above processing, each user's original mobile signaling data is converted into a sequence of user movement trajectories consisting of a series of meaningful stop points, as shown below: ; ; in: This represents the i-th stop point of the user; n represents the total number of stop points of the user. : Represents the coordinates of the i-th stop point; This indicates the time it took for the user to arrive at this stop point; This indicates the time the user left this location.

[0030] Based on the collaborative data, the dominant land parcel characteristics of each base station cell are determined. Land parcel characteristics can be obtained by applying corresponding hierarchical rules based on different collaborative data sources. For example, if the collaborative data is POI data, land parcel characteristics may include the following categories: a. Commercial service facilities (shopping malls, supermarkets, convenience stores, etc.), b. Office and industrial facilities (office buildings, factories, industrial parks, etc.), c. Medical and health facilities (hospitals, clinics, pharmacies, etc.), d. Education and training facilities (schools, universities, training institutions, etc.), e. Transportation facilities (subway stations, bus stops, train stations, etc.), f. Entertainment and leisure facilities (parks, cinemas, gyms, etc.), g. Residential facilities (residential communities, apartments, dormitories, etc.), and h. Other unclassified facilities.

[0031] For public transportation card swipe record collaborative data, the land parcel type can be divided according to the station (or line) type as follows: a. No bus / rail station, b. Rail / subway station, c. BRT or rapid transit station, d. Trunk / backbone bus station, e. Feeder / connecting bus station.

[0032] If the collaborative data is data on the number of employees, the characteristics of the land parcels can be divided according to the nature of the unit or position as follows: a. government and public institutions, b. industrial and production parks, c. commercial and office buildings, d. public service and medical and educational institutions.

[0033] This application does not restrict collaborative data sources. For ease of understanding, the POI type will be used as an example in the following explanation.

[0034] The service range of each base station is called the base station cell. During implementation, a Voronoi diagram can be constructed for the base stations within the study area to obtain the service range of each base station.

[0035] Specifically, the dominant land parcel characteristics of each base station cell are determined using the following methods: S21. Calculate the land parcel feature factor of each base station cell belonging to each land parcel feature based on the scale of each land parcel feature in the collaborative data included in each base station cell; S22. For each base station cell, select the plot feature with the largest plot feature factor as the dominant plot feature of that base station cell.

[0036] Specifically, the following formula is used to calculate the land parcel feature factor for each base station cell belonging to each type of land parcel feature: ; in, Indicates the m-th base station cell. The scale of the plot's characteristics This represents the total scale of all land parcel features within the m-th base station cell. Indicates the first The total size of each type of land parcel. This represents the total size of all land parcel features. This indicates that the m-th base station cell belongs to the m-th base station cell. Plot characteristic factors of various plot characteristics.

[0037] During implementation, by statistically analyzing the scale of each type of land parcel feature within the range of each base station cell, the land parcel feature factor of which the base station cell belongs to a certain land parcel feature is obtained.

[0038] For example, if the collaborative data is POI data, then the m-th base station cell contains the _____ data. The scale of each land parcel feature is the number of land parcel features of the kth POI type within the m-th base station cell. Based on collaborative data of public transportation card swipe records, the kth POI type within the m-th base station cell... The scale of each land parcel characteristic is the number of card swipes for the k-th site (or line) type within the m-th base station cell. If the collaborative data is data on the scale of employees, then the number of card swipes for the k-th site (or line) type within the m-th base station cell... The scale of a plot of land is the number of personnel of the k-th type of unit or job within the m-th base station cell.

[0039] Total size of all plot features It refers to the total scale of all land parcel characteristics within the study area.

[0040] After calculating the land parcel feature factors for each base station cell belonging to each type of land parcel feature, for each base station cell, the land parcel feature with the largest land parcel feature factor is selected as the dominant land parcel feature of that base station cell.

[0041] Specifically, based on the dominant land parcel characteristics of each base station cell and the user movement trajectory sequence, collaborative feature coefficients for each land parcel characteristic in each time period are constructed, including: The population size of each base station cell in each time period is determined based on the user's movement trajectory sequence; The collaborative feature coefficients for each type of land parcel feature in each time period are calculated based on the population of each base station cell in each time period and the dominant land parcel feature of each base station cell.

[0042] During implementation, the user movement trajectory sequence, i.e., the user's stop point sequence, extracted from S2 is aggregated by base station cell to obtain the population count of each base station cell at different time periods. The time periods can be set according to statistical needs, such as 15 minutes, 30 minutes, or 1 hour.

[0043] Furthermore, based on the population size of each base station cell in each time period and the dominant land parcel characteristics of each base station cell, the collaborative feature coefficients for each type of land parcel characteristic in each time period are calculated. The collaborative feature coefficients are used to adjust the spatial distribution ratio of travel traffic in different time periods during the scale transformation process.

[0044] Specifically, based on the population size of each base station cell in each time period and the dominant land parcel characteristics of each base station cell, the collaborative feature coefficients for each land parcel characteristic in each time period are calculated using the following formula: ; in, This represents the synergistic characteristic coefficient of the k-th type of land parcel in the t-th time period. Let B represent the cumulative population of all base station cells where the dominant land parcel feature is the k-th type, and let B represent the number of land parcel feature categories. This represents the cumulative population of all base station cells whose dominant land parcel characteristic is type b.

[0045] The collaborative characteristic coefficient reflects the attractiveness of a specific plot of land to people at different time periods. By synchronously introducing time dynamic characteristics, the transformed travel flow can reflect the traffic fluctuations and distribution patterns at different times (such as morning peak, evening peak, and off-peak periods), rather than just a static spatial aggregation.

[0046] Based on the collaborative feature coefficient, spatial heterogeneity factors are introduced, and the spatial superposition ratio of base station cells and target scale units, as well as the uneven spatial distribution of land features, are comprehensively considered to construct a standardized adjustment coefficient for comprehensive spatiotemporal changes—the spatiotemporal adjustment coefficient—as the basis for the proportion of travel flow redistribution, so as to more finely control the redistribution of travel flow at the spatial scale.

[0047] The spatial factors considered in this invention mainly include two aspects: Area ratio factor: This refers to the proportion of the area of ​​the target scale unit and the base station cell overlapping in the total area of ​​the base station cell. This is a fundamental method used in most scale conversion studies to estimate the spatial allocation weight of the target scale unit in the traffic flow of the base station cell; Cooperative feature heterogeneity: refers to the differences in activity intensity exhibited by different land parcel features (such as different types of POIs, different levels of bus stops, different types of employment units, etc.) at different times within the service area of ​​the same base station. That is, at the same time, different types of land parcel features may correspond to different levels of population inflow or outflow, thus making the spatial distribution of travel traffic within the base station cell exhibit obvious heterogeneous characteristics.

[0048] Therefore, when redistributing and scaling traffic, it is not sufficient to rely solely on area proportions for division; rather, both land parcel characteristics and temporal factors should be considered. A dynamic spatiotemporal adjustment coefficient integrating both spatial and temporal dimensions should be constructed by comprehensively considering the spatial distribution characteristics of collaborative data (land parcel characteristics) within the base station cell and its corresponding collaborative characteristic coefficients.

[0049] Specifically, based on the collaborative feature coefficients of each type of land parcel feature in each time period and the spatial distribution characteristics of each type of land parcel feature within each base station cell, the spatiotemporal adjustment coefficients for allocating traffic from each base station cell to each target scale unit in each time period are constructed in the following manner: ; ; ; ; in, This represents the spatiotemporal adjustment coefficient for allocating traffic from the m-th base station cell to the i-th target scale unit during the t-th time period. This represents the spatiotemporal adjustment coefficient for allocating traffic from the m-th base station cell to the i-th target scale unit in the t-th time period without normalization. This represents the number of target scale units superimposed on the m-th base station cell. This indicates that the m-th base station cell in the t-th time period of the unnormalized time period is sending data to the m-th base station cell. Spatiotemporal adjustment coefficient for allocating flow to each target scale unit This represents the synergistic characteristic coefficient of the k-th type of land parcel in the t-th time period. This represents the scale of the k-th type of land parcel feature in the region formed by the overlay of the i-th target scale unit and the m-th base station cell. The scale of the k-th type of land parcel feature in the m-th base station cell proportion, This represents the area of ​​the region formed by the superposition of the i-th target scale cell and the m-th base station cell. Occupying the total area of ​​the m-th base station cell The proportion, K represents the number of land parcel feature types.

[0050] During implementation, to ensure that the total flow remains consistent after scaling, it is necessary to... The standardization process (normalization) is performed, and the standardized spatiotemporal adjustment coefficient is denoted as... .

[0051] After obtaining the spatiotemporal adjustment coefficients for the allocation of traffic from each base station cell to each target scale unit in each time period, the travel flow of the base station cell is mapped to the target scale unit based on the spatiotemporal adjustment coefficients for the allocation of traffic from each base station cell to each target scale unit in each time period.

[0052] Specifically, based on the spatiotemporal adjustment coefficient for allocating traffic from each base station cell to each target scale unit in each time period, the travel flow of the base station cell is mapped to the target scale unit in the following manner: ; in, Indicates from the start time By the end time Travel flow from the i-th target scale unit to the j-th target scale unit, Indicates the start time The spatiotemporal adjustment coefficient for allocating traffic from the m-th base station cell to the i-th target scale unit. Indicates the end time The spatiotemporal adjustment coefficient for allocating traffic from the nth base station cell to the jth target scale unit. This indicates that the m-th base station cell starts from the beginning time. By the end time The travel flow to the nth base station cell, where M represents the number of base station cells intersecting with the i-th target scale unit, and N represents the number of base station cells intersecting with the j-th target scale unit.

[0053] Based on the original inter-base station travel flow, and combined with spatiotemporal adjustment coefficients, weighted allocation is performed according to the origin and destination points. This process refines the travel flow originally based on base station cells to the target scale units, achieving a coordinated transformation of travel flow from the base station scale to the target scale. This process ensures the conservation of flow before and after the scale transformation, while preserving spatial and temporal heterogeneity information, thereby constructing a more realistic travel flow pattern.

[0054] Compared with existing technologies, the collaborative scale conversion method for travel flows based on mobile signaling data provided in this embodiment innovatively introduces the concept of collaborative modeling. It fully utilizes the characteristics of auxiliary spatial data, overcoming the limitations of traditional methods that do not consider the spatiotemporal heterogeneity of travel flows during the conversion of mobile signaling data from the base station scale to the target scale unit cell. This significantly improves the consistency between the conversion results and actual observation data. This method provides a refined framework for travel flow scale conversion, offering a comprehensive and accurate data foundation for refined urban traffic management, traffic big data analysis, traffic big data mining, intelligent infrastructure planning, and smart city dynamic decision-making, as well as for training related machine learning, reinforcement learning, and deep learning models.

[0055] A specific embodiment of the present invention discloses a travel flow scale conversion system based on collaborative multi-source data spatiotemporal features, such as... Figure 2 As shown, it includes: The data acquisition module is used to acquire mobile signaling data and collaborative data; and to obtain the user's movement trajectory sequence based on the mobile signaling data. A dominant land parcel feature determination module is used to determine the dominant land parcel features of each base station cell based on the collaborative data; The collaborative feature coefficient determination module is used to construct collaborative feature coefficients for each type of land parcel feature in each time period based on the dominant land parcel features of each base station cell and the user movement trajectory sequence. The spatiotemporal adjustment coefficient determination module is used to construct the spatiotemporal adjustment coefficient for allocating traffic from each base station cell to each target scale unit in each time period based on the collaborative feature coefficients of each type of land parcel feature in each time period and the spatial distribution characteristics of each type of land parcel feature in each base station cell. The scale conversion module is used to map the travel flow of the base station cell to the target scale unit based on the spatiotemporal adjustment coefficient of the traffic allocated by each base station cell to each target scale unit in each time period.

[0056] Based on further improvements to the above system, the spatiotemporal adjustment coefficient determination module constructs the spatiotemporal adjustment coefficients for allocating traffic from each base station cell to each target scale unit in each time period in the following manner: ; ; ; ; in, This represents the spatiotemporal adjustment coefficient for allocating traffic from the m-th base station cell to the i-th target scale unit during the t-th time period. This represents the spatiotemporal adjustment coefficient for allocating traffic from the m-th base station cell to the i-th target scale unit in the t-th time period without normalization. This represents the number of target scale units superimposed on the m-th base station cell. This indicates that the m-th base station cell in the t-th time period of the unnormalized time period is sending data to the m-th base station cell. Spatiotemporal adjustment coefficient for allocating flow to each target scale unit This represents the synergistic characteristic coefficient of the k-th type of land parcel in the t-th time period. This represents the scale of the k-th type of land parcel feature in the region formed by the overlay of the i-th target scale unit and the m-th base station cell. The scale of the k-th type of land parcel feature in the m-th base station cell proportion, This represents the area of ​​the region formed by the superposition of the i-th target scale cell and the m-th base station cell. Occupying the total area of ​​the m-th base station cell The proportion, K represents the number of land parcel feature types.

[0057] Based on further improvements to the above system, the scale conversion module maps the travel flow of the base station cell to the target scale unit in the following manner: ; in, Indicates from the start time By the end time Travel flow from the i-th target scale unit to the j-th target scale unit, Indicates the start time The spatiotemporal adjustment coefficient for allocating traffic from the m-th base station cell to the i-th target scale unit. Indicates the end time The spatiotemporal adjustment coefficient for allocating traffic from the nth base station cell to the jth target scale unit. This indicates that the m-th base station cell starts from the beginning time. By the end time The travel flow to the nth base station cell, where M represents the number of base station cells intersecting with the i-th target scale unit, and N represents the number of base station cells intersecting with the j-th target scale unit.

[0058] The above-described method and system embodiments are based on the same principles, and their related aspects can be referenced from each other to achieve the same technical effects. For specific implementation processes, please refer to the foregoing embodiments, which will not be repeated here.

[0059] Those skilled in the art will understand that all or part of the processes of the methods described in the above embodiments can be implemented by a computer program instructing related hardware, and the program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.

[0060] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for travel flow scale transformation based on synergistic multi-source data spatiotemporal features, characterized in that, Includes the following steps: Acquire mobile signaling data and collaborative data; The user's movement trajectory sequence is obtained based on mobile phone signaling data; The dominant land parcel characteristics of each base station cell are determined based on the collaborative data; Based on the dominant land parcel characteristics of each base station cell and the user movement trajectory sequence, construct the collaborative feature coefficients for each land parcel characteristic in each time period; Based on the collaborative feature coefficients of each type of land parcel feature in each time period and the spatial distribution characteristics of each type of land parcel feature in each base station cell, a spatiotemporal adjustment coefficient for allocating traffic from each base station cell to each target scale unit in each time period is constructed. Based on the spatiotemporal adjustment coefficient of each base station cell allocating traffic to each target scale unit in each time period, the travel flow of the base station cell is mapped to the target scale unit.

2. The method for travel flow scale conversion based on collaborative multi-source data spatiotemporal features according to claim 1, characterized in that, Based on the collaborative feature coefficients of each type of land parcel feature in each time period and the spatial distribution characteristics of each type of land parcel feature within each base station cell, the spatiotemporal adjustment coefficients for allocating traffic from each base station cell to each target scale unit in each time period are constructed in the following manner: ； ； ； ； in, This represents the spatiotemporal adjustment coefficient for allocating traffic from the m-th base station cell to the i-th target scale unit during the t-th time period. This represents the spatiotemporal adjustment coefficient for allocating traffic from the m-th base station cell to the i-th target scale unit in the t-th time period without normalization. This represents the number of target scale units superimposed on the m-th base station cell. This indicates that the m-th base station cell in the t-th time period of the unnormalized time period is sending data to the m-th base station cell. Spatiotemporal adjustment coefficient for allocating flow to each target scale unit This represents the synergistic characteristic coefficient of the k-th type of land parcel in the t-th time period. This represents the scale of the k-th type of land parcel feature in the region formed by the overlay of the i-th target scale unit and the m-th base station cell. The scale of the k-th type of land parcel feature in the m-th base station cell proportion, This represents the area of ​​the region formed by the superposition of the i-th target scale cell and the m-th base station cell. Occupying the total area of ​​the m-th base station cell The proportion, K represents the number of land parcel feature types.

3. The method for travel flow scale conversion based on collaborative multi-source data spatiotemporal features according to claim 1, characterized in that, Based on the spatiotemporal adjustment coefficient of traffic allocation from each base station cell to each target scale unit in each time period, the travel flow of the base station cell is mapped to the target scale unit in the following way: ； in, Indicates from the start time By the end time Travel flow from the i-th target scale unit to the j-th target scale unit, Indicates the start time The spatiotemporal adjustment coefficient for allocating traffic from the m-th base station cell to the i-th target scale unit. Indicates the end time The spatiotemporal adjustment coefficient for allocating traffic from the nth base station cell to the jth target scale unit. This indicates that the m-th base station cell starts from the beginning time. By the end time The travel flow to the nth base station cell, where M represents the number of base station cells intersecting with the i-th target scale unit, and N represents the number of base station cells intersecting with the j-th target scale unit.

4. The method for travel flow scale conversion based on collaborative multi-source data spatiotemporal features according to claim 1, characterized in that, Based on the dominant land parcel characteristics of each base station cell and the user movement trajectory sequence, collaborative feature coefficients for each land parcel characteristic in each time period are constructed, including: The population size of each base station cell in each time period is determined based on the user's movement trajectory sequence; The collaborative feature coefficients for each type of land parcel feature in each time period are calculated based on the population of each base station cell in each time period and the dominant land parcel feature of each base station cell.

5. The method for travel flow scale conversion based on collaborative multi-source data spatiotemporal features according to claim 4, characterized in that, The dominant land parcel characteristics of each base station cell are determined using the following method: Calculate the land parcel feature factor for each land parcel feature based on the scale of each land parcel feature in the collaborative data included in each base station cell; For each base station cell, the land parcel feature with the largest land parcel feature factor is selected as the dominant land parcel feature of that base station cell.

6. The method for travel flow scale conversion based on collaborative multi-source data spatiotemporal features according to claim 5, characterized in that, The following formula is used to calculate the land parcel feature factor for each base station cell belonging to each type of land parcel feature: ； in, Indicates the m-th base station cell. The scale of the plot's characteristics This represents the total scale of all land parcel features within the m-th base station cell. Indicates the first The total size of each type of land parcel. This represents the total size of all land parcel characteristics. This indicates that the m-th base station cell belongs to the m-th base station cell. Plot characteristic factors of various plot characteristics.

7. The method for travel flow scale conversion based on collaborative multi-source data spatiotemporal features according to claim 4, characterized in that, Based on the population size of each base station cell in each time period and the dominant land parcel characteristics of each base station cell, the collaborative characteristic coefficients for each land parcel characteristic in each time period are calculated using the following formula: ； in, This represents the synergistic characteristic coefficient of the k-th type of land parcel in the t-th time period. Let B represent the cumulative population of all base station cells where the dominant land parcel feature is the k-th type, and let B represent the number of land parcel feature categories. This represents the cumulative population of all base station cells whose dominant land parcel characteristic is type b.

8. A travel flow scale conversion system based on collaborative multi-source data spatiotemporal features, characterized in that, include: The data acquisition module is used to acquire mobile signaling data and collaborative data; The user's movement trajectory sequence is obtained based on mobile phone signaling data; A dominant land parcel feature determination module is used to determine the dominant land parcel features of each base station cell based on the collaborative data; The collaborative feature coefficient determination module is used to construct collaborative feature coefficients for each type of land parcel feature in each time period based on the dominant land parcel features of each base station cell and the user movement trajectory sequence. The spatiotemporal adjustment coefficient determination module is used to construct the spatiotemporal adjustment coefficient for allocating traffic from each base station cell to each target scale unit in each time period based on the collaborative feature coefficients of each type of land parcel feature in each time period and the spatial distribution characteristics of each type of land parcel feature in each base station cell. The scale conversion module is used to map the travel flow of the base station cell to the target scale unit based on the spatiotemporal adjustment coefficient of the traffic allocated by each base station cell to each target scale unit in each time period.

9. The travel flow scale conversion system based on collaborative multi-source data spatiotemporal features according to claim 8, characterized in that, The spatiotemporal adjustment coefficient determination module constructs the spatiotemporal adjustment coefficients for allocating traffic from each base station cell to each target scale unit in each time period using the following method: ； ； ； ； in, This represents the spatiotemporal adjustment coefficient for allocating traffic from the m-th base station cell to the i-th target scale unit during the t-th time period. This represents the spatiotemporal adjustment coefficient for allocating traffic from the m-th base station cell to the i-th target scale unit in the t-th time period without normalization. This represents the number of target scale units superimposed on the m-th base station cell. This indicates that the m-th base station cell in the t-th time period of the unnormalized time period is sending data to the m-th base station cell. Spatiotemporal adjustment coefficient for allocating flow to each target scale unit This represents the synergistic characteristic coefficient of the k-th type of land parcel in the t-th time period. This represents the scale of the k-th type of land parcel feature in the region formed by the overlay of the i-th target scale unit and the m-th base station cell. The scale of the k-th type of land parcel feature in the m-th base station cell proportion, This represents the area of ​​the region formed by the superposition of the i-th target scale cell and the m-th base station cell. Occupying the total area of ​​the m-th base station cell The proportion, K represents the number of land parcel feature types.

10. The travel flow scale conversion system based on collaborative multi-source data spatiotemporal features according to claim 8, characterized in that, The scale conversion module maps the travel flow of the base station cell to the target scale unit in the following way: ； in, Indicates from the start time By the end time Travel flow from the i-th target scale unit to the j-th target scale unit, Indicates the start time The spatiotemporal adjustment coefficient for allocating traffic from the m-th base station cell to the i-th target scale unit. Indicates the end time The spatiotemporal adjustment coefficient for allocating traffic from the nth base station cell to the jth target scale unit. This indicates that the m-th base station cell starts from the beginning time. By the end time The travel flow to the nth base station cell, where M represents the number of base station cells intersecting with the i-th target scale unit, and N represents the number of base station cells intersecting with the j-th target scale unit.

Images