Method for constructing complex dynamic network of traffic hub based on multi-space data driving

Abstract

The application relates to a traffic hub complex dynamic network construction method based on multi-space data driving. The method comprises the following steps: cleaning multi-space data to obtain personnel space-time information; performing community division according to the personnel space-time information, dividing the personnel into passenger communities and staff communities, adopting a degree priority mechanism and a distance priority mechanism to combine and construct a static network, obtaining passenger static networks and staff static networks; performing passenger mobile behavior modeling according to a hybrid mobile behavior mode, utilizing a Monte Carlo simulation method and a mobile path to model, and obtaining a mobile contact network; and performing network aggregation on the mobile contact network, the passenger static network and the staff static network to obtain a complex dynamic network. The method can improve the traffic demand analysis accuracy.

Description

Technical Field

[0001] This application relates to the field of complex network and complex system modeling technology, and in particular to a method, apparatus, computer equipment and storage medium for constructing complex dynamic networks of transportation hubs based on multi-spatial data-driven approaches. Background Technology

[0002] Complex systems play a vital role in our daily lives, science, and economy. Behind every complex system lies an intricate network that characterizes the interactions between its various components. Society is a complex system, with workplace relationships, friendships, and family relationships forming social networks. Complex networks have become an important modeling method for complex social systems, supporting research on the behavior, phenomena, and problems of complex social systems. Since the discovery of the small-world property in social networks, social network analysis and construction have become research hotspots in network science. Transportation hubs are crucial components of national or regional transportation systems, serving as intersections of different transportation modes and connecting fixed and mobile equipment. They collectively handle through traffic, transfers, hub operations, and related urban external transportation functions within their respective areas. Large transportation hubs typically involve large populations, diverse personnel types, and complex population flow characteristics, forming correspondingly complex dynamic social networks. While my country has established a comprehensive transportation infrastructure system over a long period, each mode of transportation remains largely independent, lacking interconnected and collaborative mechanisms. In order to break the original independent development model of each mode of transportation and promote the integrated development of comprehensive transportation, the trend of development is to shift from a single mode of transportation to multimodal transport.

[0003] However, current specialized transportation demand analysis only focuses on a single transportation network. For example, railway demand analysis only covers the railway network, which has problems such as a small analysis scope and poor analysis accuracy. Summary of the Invention

[0004] Therefore, it is necessary to provide a method, computer equipment, and storage medium for constructing complex dynamic networks of transportation hubs based on multi-spatial data-driven approaches to address the aforementioned technical problems.

[0005] A method for constructing complex dynamic networks of transportation hubs based on multi-spatial data, the method comprising:

[0006] Acquire multi-spatial data; multi-spatial data includes physical spatial data, social spatial data, and information spatial data; physical spatial data includes the geographical location of transportation hubs; social spatial data includes personnel information within transportation hubs; information spatial data includes passenger train schedule finalization information and train schedule implementation status information.

[0007] Multi-spatial data is cleaned to obtain spatiotemporal information of personnel; based on the spatiotemporal information of personnel, communities are divided into passenger communities and employee communities. A combination of degree-first and distance-first mechanisms is used to establish connections between community members. Different community groups evolve independently and simultaneously according to different model parameters to construct static networks, resulting in passenger static networks and employee static networks.

[0008] A hybrid mobility behavior pattern combining random walk and goal orientation is constructed. Passenger mobility behavior is modeled based on the hybrid mobility behavior pattern. The service windows selected by passengers during their movement are combined and arranged according to the necessity and non-necessity, logic and temporality of the service to obtain the passenger's mobility path between various employee community groups. The Monte Carlo simulation method and the mobility path are used to model the mobility contact network.

[0009] By aggregating the mobile contact network, passenger static network, and employee static network, a complex dynamic network is obtained.

[0010] In one embodiment, passengers are defined as a group that purchases tickets, uses transportation at a transportation hub, has a short travel cycle, strong purpose, a large travel range, and exhibits certain regularity; employees are defined as a group with fixed on-duty hours and fixed work locations, and a stable travel range over a period of time; based on the spatiotemporal information of the personnel, the groups are divided into passenger groups and employee groups, including:

[0011] Based on different behaviors of people in the transportation hub, people are divided into passengers and employees. Utilizing their respective travel or work environments, and combining spatial and temporal constraints, passengers and employees are further divided into communities, resulting in passenger communities and employee communities. Each passenger community and employee community has a unique community ID number, the IDs of the members belonging to the community, and their corresponding personnel types. Spatial constraints include different terminals, floors, and windows; temporal constraints include passengers traveling on different dates and employees working on different days.

[0012] In one embodiment, the degree distribution of the static network follows a power-law distribution; a combination of degree-first and distance-first mechanisms is used to establish connections between community members. Different community groups evolve independently and simultaneously according to different model parameters to construct static networks, including:

[0013] All passenger and employee communities are divided into multiple community groups based on their respective time and space information. Each group generates an initial network model according to a degree-first mechanism. The initial network model is then evolved by considering the community size, the distance between communities, and the degree of nodes to obtain a static network.

[0014] In one embodiment, each group generates an initial network model according to a degree-first mechanism, and evolves the initial network model by considering the community size, the distance between communities, and the degree of nodes to obtain a static network, including:

[0015] Each group generates an initial network model based on a degree-first mechanism. The initial network model has c0 communities, and each community has n0 nodes; the n0 nodes are connected to each other.

[0016] When c0≥2, communities are connected in pairs. First, a community is selected according to the community size and distance priority mechanism, and then a node in the community is selected according to the degree priority mechanism to generate the connection.

[0017] During the evolution of the network, at each step of the evolution, a community is added to the network with probability p, and an add is added to a community in the network with probability 1-p. num When the total number of nodes in the network equals the target total number of nodes, the evolution ends, and a static network is finally generated.

[0018] In one embodiment, goal orientation refers to passengers going to a specific type of service window based on their current needs when making necessary movements, and selecting a specific window based on distance and crowd density; random walk refers to randomly exploring while satisfying one's own needs, and the probability of contact with staff at the service window is also random.

[0019] In one embodiment, a mobile contact network is obtained by modeling using Monte Carlo simulation and movement paths, including:

[0020] Using Monte Carlo simulation, random numbers are generated to simulate the selection of each service window and generate movement paths. Staff members are used as hubs to establish movement contact relationships between passengers and staff. Passengers at different times and in different places establish contact relationships with staff during the process of receiving staff services. Based on these contact relationships, corresponding spatiotemporal network edges between passengers and staff are generated. As time progresses and the transportation hub continues to operate, the edges continue to grow, forming a movement contact network.

[0021] In one embodiment, the mobile contact network, passenger static network, and employee static network are aggregated to obtain a complex dynamic network, which further includes:

[0022] Based on the connections between passengers and employees in the mobile contact network, and by aggregating the passenger static network sequence and employee static network sequence using the IDs and existence times of personnel and communities, a complex dynamic network is obtained.

[0023] In one embodiment, a complex dynamic network is obtained by aggregating passenger static network sequences and employee static network sequences based on the connections between passengers and employees in the mobile contact network and by utilizing the IDs and presence times of individuals and communities. This complex dynamic network includes:

[0024] Based on the connections between passengers and employees in the mobile contact network, can the aggregation relationship be determined using the IDs of personnel and communities to aggregate the passenger static network sequence and the employee static network sequence, thus obtaining the original aggregated network?

[0025] By embedding the temporal attributes of social relationships into the original aggregated network based on the real-time vehicle operation status and transportation hub operation status in the information space data, a complex dynamic network is obtained.

[0026] A computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program performing the following steps:

[0027] Acquire multi-spatial data; multi-spatial data includes physical spatial data, social spatial data, and information spatial data; physical spatial data includes the geographical location of transportation hubs; social spatial data includes personnel information within transportation hubs; information spatial data includes passenger train schedule finalization information and train schedule implementation status information.

[0028] Multi-spatial data is cleaned to obtain spatiotemporal information of personnel; based on the spatiotemporal information of personnel, communities are divided into passenger communities and employee communities. A combination of degree-first and distance-first mechanisms is used to establish connections between community members. Different community groups evolve independently and simultaneously according to different model parameters to construct static networks, resulting in passenger static networks and employee static networks.

[0029] A hybrid mobility behavior pattern combining random walk and goal orientation is constructed. Passenger mobility behavior is modeled based on the hybrid mobility behavior pattern. The service windows selected by passengers during their movement are combined and arranged according to the necessity and non-necessity, logic and temporality of the service to obtain the passenger's mobility path between various employee community groups. The Monte Carlo simulation method and the mobility path are used to model the mobility contact network.

[0030] By aggregating the mobile contact network, passenger static network, and employee static network, a complex dynamic network is obtained.

[0031] A computer-readable storage medium having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0032] Acquire multi-spatial data; multi-spatial data includes physical spatial data, social spatial data, and information spatial data; physical spatial data includes the geographical location of transportation hubs; social spatial data includes personnel information within transportation hubs; information spatial data includes passenger train schedule finalization information and train schedule implementation status information.

[0033] Multi-spatial data is cleaned to obtain spatiotemporal information of personnel; based on the spatiotemporal information of personnel, communities are divided into passenger communities and employee communities. A combination of degree-first and distance-first mechanisms is used to establish connections between community members. Different community groups evolve independently and simultaneously according to different model parameters to construct static networks, resulting in passenger static networks and employee static networks.

[0034] A hybrid mobility behavior pattern combining random walk and goal orientation is constructed. Passenger mobility behavior is modeled based on the hybrid mobility behavior pattern. The service windows selected by passengers during their movement are combined and arranged according to the necessity and non-necessity, logic and temporality of the service to obtain the passenger's mobility path between various employee community groups. The Monte Carlo simulation method and the mobility path are used to model the mobility contact network.

[0035] By aggregating the mobile contact network, passenger static network, and employee static network, a complex dynamic network is obtained.

[0036] The aforementioned method, computer equipment, and storage medium for constructing complex dynamic networks of transportation hubs based on multi-spatial data enable this application to efficiently complete the modeling of complex dynamic networks of large transportation hubs by comprehensively utilizing multi-spatial data from cyber-physical-social fusion systems. The complex dynamic network contains rich dynamic spatiotemporal information such as geographic space and social relationships, which can provide data support for research on transportation demand analysis, such as the flow of goods, emergency evacuation of personnel, and disease transmission in large transportation hubs, thereby expanding the scope of demand analysis and improving the accuracy of demand analysis. Attached Figure Description

[0037] Figure 1 This is a flowchart illustrating a method for constructing a complex dynamic network of a transportation hub based on multi-spatial data in one embodiment.

[0038] Figure 2 This is a schematic diagram of the logical framework of a method for constructing a complex dynamic network of a transportation hub based on multi-spatial data in one embodiment;

[0039] Figure 3 This is a schematic diagram of the multi-space data collection and cleaning process in one embodiment;

[0040] Figure 4 This is a schematic diagram of the community division process in another embodiment;

[0041] Figure 5 This is a schematic diagram of the network evolution process of static network construction in one embodiment;

[0042] Figure 6 This is a schematic diagram of the mobile contact network construction process in one embodiment;

[0043] Figure 7 This is a schematic diagram of network aggregation in one embodiment;

[0044] Figure 8 This is a schematic diagram of the timing calculation method for network aggregation connection edges in one embodiment;

[0045] Figure 9 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0047] In one embodiment, such as Figure 1 and Figure 2 As shown, a method for constructing complex dynamic networks of transportation hubs based on multi-spatial data is provided, including the following steps:

[0048] Step 102: Obtain multi-spatial data; multi-spatial data includes physical spatial data, social spatial data, and information spatial data; physical spatial data includes the geographical location of transportation hubs; social spatial data includes personnel information within transportation hubs; information spatial data includes passenger train schedule finalization information and train schedule implementation status information.

[0049] Large transportation hubs rely on big data platforms for operation and management. In the process of fulfilling their functions, they generate massive amounts of spatiotemporal data and information in the physical-social-information space. Constructing a complex dynamic network model of the transportation hub's social network first requires collecting and cleaning this cross-spatial, integrated data. This mainly involves acquiring three types of data: in the physical space, data on the transportation hub's geographical location, functional areas, work windows, and the layout of various facilities; in the social space, data on personnel relationships, personnel flow, and personnel behavior within the transportation hub; and in the information space, data on passenger train or flight bookings and real-time operational status of trains or flights.

[0050] Step 104: Clean the multi-spatial data to obtain the spatiotemporal information of the personnel; divide the personnel into communities based on the spatiotemporal information, and divide the personnel into passenger communities and employee communities. Use a combination of degree-first and distance-first mechanisms to establish connections between community members. Different community groups evolve independently and simultaneously according to different model parameters to construct static networks, resulting in passenger static networks and employee static networks.

[0051] By cleaning the acquired multi-spatial data, removing invalid and missing data, and organizing it into corresponding data lists, the specific location distribution of personnel and service windows within the transportation hub over a certain period of time was extracted. The correspondence between train or flight numbers, personnel, and time was also analyzed, as well as the correspondence between personnel and their spatial environment. Based on their different behaviors within the transportation hub, personnel were divided into two main categories: passengers and employees. Passengers are those who purchase tickets and use transportation within the hub; their movement cycle is short, their purpose is strong, their movement range is large, and the influencing factors are complex, but they exhibit certain regularities. Employees are those engaged in operations, service, and support work within the transportation hub. Employees generally have fixed on-duty hours and fixed work locations; over a period of time, the staff in each department within the transportation hub are relatively stable. Next, based on the respective travel or work environments of passengers and employees, and considering spatial constraints (such as different terminals, floors, and service windows) and temporal constraints (passengers on different travel dates, employees with different on-duty times), passengers and employees were further categorized into groups. Each group has a unique group ID number, along with the IDs of the personnel belonging to that group and their corresponding personnel types.

[0052] Based on the segmented community data list, static networks are constructed by classifying passengers and employees. All passenger and employee communities are divided into a series of community groups according to their respective time and space information, and each group independently undergoes network evolution to generate a static network. Regarding the selection of network evolution methods based on community structure, this invention optimizes existing models that generate networks based on a degree-first mechanism. It considers community size, distance between communities, and node degree as factors influencing network structure, employing a node evolution model combining degree-first and distance-first mechanisms for static network construction. The static network represents relatively stable social relationships within the community groups.

[0053] Step 106: Construct a hybrid mobility behavior pattern that combines random walk and goal orientation. Model passenger mobility behavior based on the hybrid mobility behavior pattern. Utilize the service windows selected by passengers during their movement, and combine and arrange the passenger's movement paths between various employee community groups according to the necessity and non-necessity, logic and temporality of the service received. Model the movement paths using Monte Carlo simulation method to obtain the mobility contact network.

[0054] After independently constructing static networks for each community group, considering the connection mechanisms between static networks, it is necessary to construct a personnel movement contact network. Constructing this network, in addition to considering fixed social behaviors and relationships in the social space, requires combining spatial movement and temporal stay data of people in various transportation hubs and locations in the physical space, and passenger ticketing and train / flight data, employee on-duty information, etc., in the information space to model the social relationships arising from movement contact between people. This invention designs a passenger movement path generation algorithm based on the common, concentrated, and regular behaviors of passengers in transportation hubs. By using these movement paths and treating staff as hubs, it establishes movement contact relationships between passengers and staff. Passengers at different times and in different spaces establish contact relationships with staff during the process of receiving services. Based on these relationships, corresponding spatiotemporal passenger-staff network edges are generated. As time progresses and transportation hubs continue to operate, these edges continuously grow, forming a movement contact network.

[0055] Step 106: Aggregate the mobile contact network, passenger static network, and employee static network to obtain a complex dynamic network.

[0056] Network aggregation uses mobile contact networks as a bridge to aggregate the static networks of various community groups, with personnel IDs serving as unique identifiers for connections during the aggregation process. Simultaneously, time information is embedded into each connection edge in the network based on the operational information of the train or flight booked by the personnel and their dwell time at transportation hubs, representing the creation and demise of network connections over time, ultimately generating a complex dynamic network.

[0057] In the above-mentioned method for constructing complex dynamic networks of transportation hubs based on multi-spatial data, this application comprehensively utilizes multi-spatial data from cyber-physical-social fusion systems to efficiently complete the complex dynamic network modeling of large transportation hubs. The complex dynamic network contains rich dynamic spatiotemporal information such as geospatial and social relationships, which can provide data support for the study of transportation demand analysis, such as the flow of goods, emergency evacuation of personnel, and disease transmission in large transportation hubs, thereby expanding the scope of demand analysis and improving the accuracy of demand analysis.

[0058] In one embodiment, passengers are defined as a group that purchases tickets, uses transportation at a transportation hub, has a short travel cycle, strong purpose, a large travel range, and exhibits certain regularity; employees are defined as a group with fixed on-duty hours and fixed work locations, and a stable travel range over a period of time; based on the spatiotemporal information of the personnel, the groups are divided into passenger groups and employee groups, including:

[0059] Based on different behaviors of people in the transportation hub, people are divided into passengers and employees. Utilizing their respective travel or work environments, and combining spatial and temporal constraints, passengers and employees are further divided into communities, resulting in passenger communities and employee communities. Each passenger community and employee community has a unique community ID number, the IDs of the members belonging to the community, and their corresponding personnel types. Spatial constraints include different terminals, floors, and windows; temporal constraints include passengers traveling on different dates and employees working on different days.

[0060] In a specific embodiment, such as Figure 3 As shown, the first step is to collect and clean multi-space data. Data collection requires acquiring multi-source data from Nanjing Lukou International Airport, including physical space, social space, and information space. In this embodiment, physical space data mainly includes the terminal building's floor plan and geographical location, the location distribution of all service windows and public facilities within the terminal, and the on-duty status of service window staff. This data can be obtained through on-site surveys, questionnaires, remote access, and business cooperation. Social space data mainly consists of personnel flow, personnel relationships, and personnel behavior data at Nanjing Lukou Airport. It is characterized by strong real-time requirements, large data scale, and numerous errors and omissions, making it difficult to acquire and process. It can be acquired by comprehensively utilizing historical personnel data and empirical research, statistical surveys, and real-time sensing and detection. Information space data mainly includes ticketing data from passengers purchasing tickets online or offline, flight takeoff and landing status data, and the operational status of various windows within the airport. Modern international airports rely on the Internet and big data platforms for intelligent management and operation, with the background recording and processing of relevant data in real time, which greatly facilitates the acquisition of information space data.

[0061] Whether it's field observation and surveys, statistical analysis, or historical experience, the data obtained may contain errors and omissions. It needs to be detected and processed to become usable data that can drive the model.

[0062] like Figure 4As shown, the community segmentation includes passenger community segmentation and employee community segmentation. Passenger community segmentation is based on flight ticketing data and social relationship data. Passengers on different flights at different time intervals are divided into different passenger communities based on their inherent social relationships, and these communities are identified by the most concentrated area of ​​passengers in the terminal—the boarding gate waiting area. Employee community segmentation is based on a combination of employee type, work window, and on-duty time. Employees can be divided into two main categories: flight crew and ground crew. Ground crew has the largest number and most diverse personnel, with specific positions including check-in, security screening, VIP customer service, dispatching, ticketing, easy boarding, and cleaning. Each type of personnel has their own work window or responsible facility, and they are divided into their respective community groups based on their on-duty time, identified by their relatively fixed location.

[0063] In one embodiment, the degree distribution of the static network follows a power-law distribution; a combination of degree-first and distance-first mechanisms is used to establish connections between community members. Different community groups evolve independently and simultaneously according to different model parameters to construct static networks, including:

[0064] All passenger and employee communities are divided into multiple community groups based on their respective time and space information. Each group generates an initial network model according to a degree-first mechanism. The initial network model is then evolved by considering the community size, the distance between communities, and the degree of nodes to obtain a static network.

[0065] In one embodiment, each group generates an initial network model according to a degree-first mechanism, and evolves the initial network model by considering the community size, the distance between communities, and the degree of nodes to obtain a static network, including:

[0066] Each group generates an initial network model based on a degree-first mechanism. The initial network model has c0 communities, and each community has n0 nodes; the n0 nodes are connected to each other.

[0067] When c0≥2, communities are connected in pairs. First, a community is selected according to the community size and distance priority mechanism, and then a node in the community is selected according to the degree priority mechanism to generate the connection.

[0068] During the evolution of the network, at each step of the evolution, a community is added to the network with probability p, and an add is added to a community in the network with probability 1-p. num When the total number of nodes in the network equals the target total number of nodes, the evolution ends, and a static network is finally generated.

[0069] In a specific embodiment, such as Figure 5As shown, based on the community structure and the traditional gravity model, a combination of degree-first and distance-first mechanisms is used to establish connections between community members. Different community leaders can evolve independently and simultaneously according to different model parameters. The generated static network degree distribution approximately follows a power-law distribution, consistent with existing research and real-world social network characteristics.

[0070] The network is initially initialized with c0 communities, each containing n0 nodes. These n0 nodes are paired, and when c0 ≥ 2, communities are also paired. Edges between communities are created by randomly selecting nodes. Next, the network is evolved. During this evolution, at each step, a community is added to the network with probability p, and an additional community is added with probability 1-p. num The evolution ends when the total number of nodes in the network equals the target total number of nodes, ultimately generating the target network.

[0071] Network evolution involves two intertwined processes: adding communities and adding nodes. When a community is added to the network, the new community is identical to the initial community, consisting of n0 paired nodes. Simultaneously, m edges are introduced, and a node is randomly selected from this community to connect with other communities through these m edges. When adding a node... num When there are a number of nodes, these nodes first choose a community to join based on community size. Then, for each node, m edges are introduced. These m edges connect the node to other nodes in its own community with probability q, and connect the node to nodes in other communities with probability 1-q. When p=0, no new communities are generated in the network. When a node needs to form an edge with nodes in other communities, it first selects a community based on community size and distance priority, and then selects a node from that community based on degree priority. During network evolution, there will be no duplicate edges between two nodes; that is, there is at most one edge between two nodes.

[0072] In one embodiment, goal orientation refers to passengers going to a specific type of service window based on their current needs when making necessary movements, and selecting a specific window based on distance and crowd density; random walk refers to randomly exploring while satisfying one's own needs, and the probability of contact with staff at the service window is also random.

[0073] In one embodiment, a mobile contact network is obtained by modeling using Monte Carlo simulation and movement paths, including:

[0074] Using Monte Carlo simulation, random numbers are generated to simulate the selection of each service window and generate movement paths. Staff members are used as hubs to establish movement contact relationships between passengers and staff. Passengers at different times and in different places establish contact relationships with staff during the process of receiving staff services. Based on these contact relationships, corresponding spatiotemporal network edges between passengers and staff are generated. As time progresses and the transportation hub continues to operate, the edges continue to grow, forming a movement contact network.

[0075] In a specific embodiment, such as Figure 6 As shown, the movement behavior mainly considers the spatial movement and service dwell time of passengers at various service windows from entering the airport to before the flight takes off; the movement path depicts the employee community groups that are connected as passengers move, and passengers come into contact with the staff in the groups, forming a movement contact network.

[0076] First, passenger movement behavior modeling is required. This example categorizes passenger movement into necessary and unnecessary movements based on the nature of the services received from staff. Necessary movements refer to the inherent procedures passengers must follow to complete their flight tasks, such as entering, checking in, and security checks. Unnecessary movements are behaviors influenced by personal or environmental factors, such as asking questions, shopping, or resting in VIP areas. This application proposes a hybrid movement behavior model combining random walks and goal-oriented approaches. Goal-oriented approaches mean that when making necessary movements, passengers go to specific service windows based on their current needs, choosing windows based on distance and crowd density. Random walks refer to random exploration behavior while fulfilling personal needs, with the probability of contact with staff at service windows also being random. Specifically, during movement, passengers need to access a certain type of service window. If this type of service window was classified during the community segmentation phase... m There are several clubs, and the distance a traveler needs to travel to reach each club is d. i Then the probability that the passenger chooses service window community i is

[0077] Then, based on the service windows selected by passengers during their movement, and considering the necessity and non-necessity of the services, as well as logical and temporal order, the passenger's movement paths among various employee communities are arranged. Each path records the community ID of the service window, the order of access, and the employees with whom the passenger may have contact, from entering the airport to the flight's departure. Considering the randomness of the probability of each selection on the movement path, a Monte Carlo simulation method is used to simulate each service window selection by generating random numbers. The final generated mobile contact network exhibits significant randomness, which is caused by passengers randomly selecting access locations based on probability during their movement. The degree of most staff nodes is much larger than that of passenger nodes; in the mobile contact network, staff members exhibit characteristics similar to hub nodes.

[0078] In one embodiment, the mobile contact network, passenger static network, and employee static network are aggregated to obtain a complex dynamic network, which further includes:

[0079] Based on the connections between passengers and employees in the mobile contact network, and by aggregating the passenger static network sequence and employee static network sequence using the IDs and existence times of personnel and communities, a complex dynamic network is obtained.

[0080] In one embodiment, a complex dynamic network is obtained by aggregating passenger static network sequences and employee static network sequences based on the connections between passengers and employees in the mobile contact network and by utilizing the IDs and presence times of individuals and communities. This complex dynamic network includes:

[0081] Based on the connections between passengers and employees in the mobile contact network, can the aggregation relationship be determined using the IDs of personnel and communities to aggregate the passenger static network sequence and the employee static network sequence, thus obtaining the original aggregated network?

[0082] By embedding the temporal attributes of social relationships into the original aggregated network based on the real-time vehicle operation status and transportation hub operation status in the information space data, a complex dynamic network is obtained.

[0083] In a specific embodiment, such as Figure 7 As shown, network aggregation first aggregates the static network sequences of passenger and employee groups based on the connections between passengers and employees in the mobile contact network, according to the IDs and durations of the individuals and communities. The temporal information of network edges describes the time when network edges begin to grow and disappear, denoted as T. e =(t s ,t e ), t s , t e It's a timestamp, t s It is the moment when network edges begin to be generated, t eIt's the moment when the network connection disappears. T e The calculation is based on information space flight ticketing data and terminal operation data. Service time is calculated according to passenger flight times, the order in which passengers receive services, and the service window. Different calculation methods are used for edges formed by different types of nodes, such as... Figure 8 As shown.

[0084] It should be understood that, although Figure 1 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 1 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0085] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 9 As shown, the computer device includes a processor, memory, network interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for constructing a complex dynamic network of a transportation hub based on multi-spatial data-driven architecture. The display screen can be an LCD screen or an e-ink display screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse.

[0086] Those skilled in the art will understand that Figure 9 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0087] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0088] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0089] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for constructing a complex dynamic network of a transportation hub based on multi-space data driving, characterized in that, The method includes: Acquire multi-spatial data; the multi-spatial data includes physical spatial data, social spatial data, and information spatial data; the physical spatial data includes the geographical location of transportation hubs; the social spatial data includes personnel information within transportation hubs; the information spatial data includes passenger train booking information and real-time train operation status information; The multi-spatial data is cleaned to obtain the spatiotemporal information of the personnel. Based on the spatiotemporal information of the personnel, communities are divided into passenger communities and employee communities. A combination of degree-first and distance-first mechanisms is used to establish connections between community members. The passenger community and the employee community are divided into multiple community groups based on their respective time and space information. Different community groups evolve independently and simultaneously according to different model parameters to construct static networks, resulting in passenger static networks and employee static networks. A hybrid mobility behavior pattern combining random walk and goal orientation is constructed. Passenger mobility behavior is modeled based on the hybrid mobility behavior pattern. The service windows selected by passengers during their movement are combined and arranged according to the necessity and non-necessity, logic and temporality of the service to obtain the passenger's mobility path between various employee community groups. The Monte Carlo simulation method and the mobility path are used to model the mobility contact network. By aggregating the mobile contact network, passenger static network, and employee static network, a complex dynamic network is obtained.

2. The method of claim 1, wherein, The passengers are a group that purchases tickets, takes transportation at transportation hubs, has a short travel cycle, a strong purpose, a large travel range, and a certain regularity; the employees are a group that has fixed on-duty hours and fixed work locations, and whose travel range is stable over a period of time. Based on the aforementioned spatiotemporal information of the personnel, communities are divided into passenger communities and employee communities, including: Based on their different behaviors in transportation hubs, people are divided into passengers and employees; By utilizing the respective travel or work environments of passengers and employees, and combining spatial and temporal constraints, passenger and employee communities are divided into passenger communities and employee communities. Each passenger community and employee community has a unique community ID number, the IDs of the members belonging to the community, and their corresponding personnel types. The spatial constraints include different terminals, floors, and windows. The temporal constraints include passengers traveling on different dates and employees working on different days.

3. The method of claim 1, wherein, The degree distribution of the static network follows a power-law distribution; a combination of degree-first and distance-first mechanisms is used to establish connections between community members. Different community groups evolve independently and simultaneously according to different model parameters to construct static networks, including: All passenger and employee communities are divided into multiple community groups based on their respective time and space information. Each group generates an initial network model according to a degree-first mechanism. The initial network model is then evolved by considering the community size, the distance between communities, and the degree of nodes to obtain a static network.

4. The method of claim 3, wherein, Each group generates an initial network model based on a degree-first mechanism. The initial network model is then evolved by considering the community size, the distance between communities, and the degree of nodes to obtain a static network, including: Each group generates an initial network model according to a degree priority mechanism, the initial network model has a community, each community has nodes; the nodes are connected to each other in pairs; When When the communities are connected two by two, a community is selected according to the community size and distance priority mechanism, and an edge is generated according to the degree priority mechanism. In the process of evolving the network, every step of evolution is with a probability adding a community to the network with a probability adding a node to a community in the network When the total number of nodes in the network is equal to the target total number of nodes, the evolution ends, and a static network is finally generated.

5. The method of claim 1, wherein, The term "goal orientation" refers to passengers heading to specific service windows based on their current needs when making necessary movements, and selecting specific windows based on distance and crowd density; the term "random walk" refers to randomly exploring while satisfying one's own needs, and the probability of contact with staff at the service windows is also random.

6. The method of claim 5, wherein, A mobile contact network is obtained by modeling using the Monte Carlo simulation method and the aforementioned movement path, including: Using Monte Carlo simulation, random numbers are generated to simulate the selection of each service window and generate movement paths. Staff members are used as hubs to establish movement contact relationships between passengers and staff. Passengers at different times and in different places establish contact relationships with staff during the process of receiving staff services. Based on these contact relationships, corresponding spatiotemporal network edges between passengers and staff are generated. As time progresses and the transportation hub continues to operate, the edges continue to grow, forming a movement contact network.

7. The method of claim 2, wherein, By aggregating the mobile contact network, passenger static network, and employee static network, a complex dynamic network is obtained, including: Based on the connections between passengers and employees in the mobile contact network, and by aggregating the passenger static network sequence and the employee static network sequence using the IDs and existence times of personnel and communities, a complex dynamic network is obtained.

8. The method of claim 7, wherein, Based on the connections between passengers and employees in the mobile contact network, and by aggregating passenger static network sequences and employee static network sequences using personnel and community IDs and presence times, a complex dynamic network is obtained, including: Based on the connections between passengers and employees in the mobile contact network, the aggregation relationship is determined using the IDs of personnel and communities to aggregate the passenger static network sequence and the employee static network sequence, thus obtaining the original aggregated network; By embedding the temporal attributes of social relationships into the original aggregated network based on the real-time vehicle operation status and transportation hub operation status in the information space data, a complex dynamic network is obtained.

Images