Safety management early warning method and system based on digital twin technology

Abstract

The present invention relates to the field of security management technology, and specifically to a security management early warning method and system based on digital twin technology, comprising the following steps: collecting entity status data of multiple physical entities on campus, and constructing a digital twin model containing the entity status data; defining the current scenario risk factor, and analyzing the interaction between the scenario risk factors by performing simulation interaction impact analysis between the scenario risk factors in the digital twin model; constructing a correlation matrix between the risk factors based on the simulation interaction impact analysis results, establishing a dynamic correlation graph between physical entities, and evaluating the weights of each node and edge through a graph calculation algorithm; calculating a comprehensive risk score based on the dynamic correlation graph, and setting a multi-level early warning level. The present invention realizes real-time dynamic modeling of risk factors, avoids the limitations of traditional single entity monitoring, and provides a data-driven dynamic correlation graph that accurately depicts the risk propagation and impact range between physical entities.

Description

Technical Field

[0001] The present invention relates to the field of security management technology, and in particular to a security management early warning method and system based on digital twin technology. Background Art

[0002] With the continuous advancement of the construction of smart campuses, campus security management technology has gradually introduced information technology, such as video surveillance systems, access control management systems, vehicle monitoring equipment, etc. These traditional technologies have improved the campus security management capabilities to a certain extent, but in the face of complex campus scenarios, diverse security threats and real-time dynamic changes in management needs, there are still obvious limitations.

[0003] Existing technologies suffer from widespread information isolation and insufficient collaboration. Status data for physical entities on campus (such as people, doors, windows, and vehicles) is often scattered across independent systems, lacking unified integration and collaborative analysis capabilities. This leads to insufficient awareness of cross-system risks (such as abnormal human behavior and vehicle path conflicts). Interoperability between different devices is low, making it difficult to share security information in a timely manner and delaying management responses.

[0004] Furthermore, existing technologies are insufficient in multi-entity interaction risk analysis and integrated early warning mechanisms. Current security management systems often analyze single events and lack the ability to model and quantify the complex interactions between multiple entities. Summary of the Invention

[0005] The present invention provides a safety management early warning method and system based on digital twin technology.

[0006] The safety management early warning method based on digital twin technology includes the following steps:

[0007] S1, Multi-entity Digital Twin Model Construction: Collect entity status data of multiple physical entities on campus, including personnel, building doors and windows, and vehicles. The entity status data includes personnel status, door and window opening and closing status, and vehicle status, and construct a digital twin model containing entity status data;

[0008] S2, Scenario Risk Factor Definition and Interaction Analysis: Define the current scenario risk factors and analyze the interaction between scenario risk factors by performing simulation interaction analysis between scenario risk factors in the digital twin model;

[0009] S3, dynamic association graph construction: Based on the simulation interaction impact analysis results, the association matrix between risk factors is constructed, a dynamic association graph between physical entities is established, and the weights of each node and edge are evaluated through graph calculation algorithms;

[0010] S4, calculates the comprehensive risk score based on the dynamic correlation graph and sets a multi-level warning level.

[0011] Optionally, collecting entity status data of multiple physical entities on campus in S1 specifically includes:

[0012] S11, Collection of personnel status data: Access control devices installed on campus collect personnel entry and exit times; smart cameras combined with image recognition technology are used to identify personnel behavior, including gathering, movement, and detention;

[0013] S12, Collection of Door and Window Opening and Closing Status Data: Install intelligent magnetic switches on building doors and windows to collect the opening and closing status of doors and windows in real time, and record the time and frequency of opening or closing;

[0014] S13, vehicle status data collection: The vehicle’s entry and exit time and stay status are recorded through the smart gates at the school gate and parking lot, and the real-time location and driving path of the vehicle are monitored by smart cameras on campus.

[0015] Optionally, constructing a digital twin model containing entity state data specifically includes:

[0016] S14, static characteristic modeling: Based on the design parameters and actual deployment of physical entities on campus, a static basic model is created, including a personnel distribution twin model, a building door and window structure twin model, and a vehicle operation twin model, and corresponding attribute parameters are assigned to different physical entities;

[0017] S15, Dynamic Data Mapping:

[0018] Personnel status mapping: Map the entry and exit times and monitoring data (gathering, movement, and retention) collected by access control devices to the personnel distribution twin model in real time to update the personnel distribution and activity status on campus;

[0019] Door and window status mapping: Mapping the collected door and window opening and closing status, opening time and frequency data to the building door and window structure twin model to reflect the opening or closing status of the doors and windows in real time;

[0020] Vehicle status mapping: Map the acquired vehicle entry and exit time, stop status, and driving path data to the vehicle operation twin model to display the vehicle's operation trajectory and parking distribution.

[0021] Optionally, the S2 specifically includes:

[0022] S21, Simulation Environment Creation: Based on the digital twin model, a simulation environment is created that includes people, building doors and windows, and vehicles. The real-time status data of each physical entity is loaded to simulate its dynamic behavior and interaction in the actual scene.

[0023] S22, combined with the status data of the physical entity, defines the scenario risk factor:

[0024] Excessive gathering of people;

[0025] Doors and windows are opened abnormally;

[0026] Abnormal vehicle paths;

[0027] S23, simulation interaction impact analysis, analyzes the interaction between excessive gathering of people, abnormal opening of doors and windows, and abnormal vehicle paths.

[0028] Optionally, the simulation interaction impact analysis includes:

[0029] Interaction between excessive crowd gathering and abnormally open doors and windows: Through simulation analysis of crowd status (such as lingering or abnormal crowding) and building door and window status (such as doors and windows being open for too long), we can identify the interactive impacts that lead to safety hazards, including excessive crowd gathering at open doors and windows.

[0030] Interaction between abnormal vehicle paths and excessive crowd gathering: Through simulation analysis of the distribution of people based on vehicle status and personnel status, it is possible to identify areas where abnormal vehicle paths cover excessive crowd gathering.

[0031] Optionally, constructing the correlation matrix between risk factors in S3 specifically includes:

[0032] Matrix element definition: Based on the simulation interaction impact analysis results, define the degree of correlation between risk factors and construct the correlation matrix R. The elements R in the matrix ij Represents the strength of association between risk factors i and j:

[0033] Among them, w ij is the correlation strength calculated based on the simulation results;

[0034] Matrix normalization: Normalize the correlation matrix to ensure that all correlation values ​​are in the range [0,1] to facilitate subsequent graph calculations.

[0035] Optionally, the step of establishing a dynamic association graph between physical entities in S3 specifically includes:

[0036] Graph node definition: The physical entities in the campus are regarded as nodes of the graph, and the node set is represented as V;

[0037] Graph edge definition: Based on the non-zero elements of the scenario risk factor association matrix, establish the edge set E between nodes, and the edge weight is the corresponding association strength w ij ;

[0038] Dynamic update: Dynamically adjust the graph structure (nodes and edges) and edge weights based on the real-time status data of the physical entity and the simulation results to ensure that the dynamic association graph is updated over time.

[0039] Optionally, the edge weight (association strength) w ij Calculation based on physical distance: Among them, d ij It represents the physical distance between the occurrence points of scenario risk factors i and j, and σ is the distance attenuation coefficient, which is used to control the attenuation degree of association strength with distance.

[0040] The weight of the node is calculated by graph centrality analysis to obtain the importance of each node: C i =Σ j∈V w ij , where C i represents the centrality weight of node i, w ij is the edge weight between nodes i and j.

[0041] Optionally, the comprehensive risk score in S4 is calculated as:

[0042] Among them, R is the comprehensive risk score, C i is the weight of node i, which represents the influence of a single risk factor, w ij is the weight between edges i and j, indicating the strength of the interaction between risk factors, V is the node set of the dynamic association graph, and E is the edge set of the dynamic association graph.

[0043] Based on the calculated risk scores, risks are divided into multiple warning levels, and graded response measures are taken according to different warning levels.

[0044] The digital twin technology-based safety management and early warning system is used to implement the above-mentioned digital twin technology-based safety management and early warning method, and includes the following modules:

[0045] Multi-entity data acquisition module: collects entity status data of multiple human physical entities, building door and window physical entities, and vehicle physical entities. The entity status data includes the status of people, the opening and closing status of doors and windows, and the status of vehicles, and inputs the collected data into the digital twin modeling module;

[0046] Digital twin modeling module: Based on the collected entity status data, a digital twin model that includes the status of people, building doors and windows, and vehicles is constructed to synchronize the dynamic behavior and status of physical entities in real time;

[0047] Scenario Risk Analysis Module: This module is used to define risk factors in the current scenario, including excessive crowds, abnormal opening of doors and windows, and abnormal vehicle paths. It also simulates and analyzes the interaction between risk factors in the digital twin model to identify the correlation between risk factors.

[0048] Dynamic association graph construction module: Based on the results of scenario risk analysis, it constructs an association matrix and generates a dynamic association graph between physical entities. It uses a graph calculation algorithm to evaluate the weights of each node and edge in the association graph to characterize the collaborative risk of multiple entities.

[0049] Risk assessment and early warning module: Calculates comprehensive risk scores based on dynamic correlation graphs and sets multi-level early warning levels based on risk scores.

[0050] Beneficial effects of the present invention:

[0051] This invention uses digital twin technology to collect and map the status data of multiple physical entities on campus, such as people, building doors and windows, and vehicles, in real time, and construct a dynamic digital twin model. This model can reflect the dynamic behavior and status changes of entities in real time, realize multi-entity collaborative risk perception, and dynamically capture complex risk scenarios such as crowd gathering, abnormal opening of doors and windows, and abnormal vehicle paths. It realizes real-time dynamic modeling of risk factors, avoids the limitations of traditional single entity monitoring, and provides a data-driven dynamic association diagram to accurately depict the risk propagation and impact range between physical entities.

[0052] This invention proposes a scenario-based risk factor identification method based on simulation analysis, clarifies the core risk factors and their interactive relationships in specific campus scenarios, identifies the linkage effects and comprehensive risks between risk factors through simulation interaction impact analysis, and provides a quantitative basis for the superposition of multiple factors (such as the interaction between the gathering of people and abnormal vehicle paths). It realizes dynamic and real-time scenario risk identification capabilities, makes the setting of warning levels more accurate and scenario-based, and provides managers with clear and reliable decision-making support. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0054] Figure 1 Schematic diagram of a method flow in an embodiment of the present invention;

[0055] Figure 2 A schematic diagram of building a digital twin model according to an embodiment of the present invention. DETAILED DESCRIPTION

[0056] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments. It is also noted that, to provide a more detailed description, the following embodiments are best and preferred embodiments, and those skilled in the art may employ alternative methods for implementing certain known technologies. Furthermore, the accompanying drawings are intended only to provide a more detailed description of the embodiments and are not intended to limit the present invention.

[0057] It should be noted that references in the specification to "one embodiment," "an embodiment," "an exemplary embodiment," "some embodiments," etc. indicate that the described embodiments may include specific features, structures, or characteristics, but not every embodiment necessarily includes such specific features, structures, or characteristics. In addition, when specific features, structures, or characteristics are described in conjunction with an embodiment, it is within the knowledge of persons skilled in the relevant art to implement such features, structures, or characteristics in conjunction with other embodiments (whether or not explicitly described).

[0058] In general, terms can be understood, at least in part, from their use in context. For example, depending at least in part on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending at least in part on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0059] like Figure 1-Figure 2 As shown in the figure, the safety management early warning method based on digital twin technology includes the following steps:

[0060] S1, Multi-entity Digital Twin Model Construction: Collect entity status data of multiple physical entities on campus, including personnel, building doors and windows, and vehicles. The entity status data includes personnel status, door and window opening and closing status, and vehicle status, and construct a digital twin model containing entity status data;

[0061] S2, Scenario Risk Factor Definition and Interaction Analysis: Define the current scenario risk factors and analyze the interaction between scenario risk factors by performing simulation interaction analysis between scenario risk factors in the digital twin model;

[0062] S3, dynamic association graph construction: Based on the simulation interaction impact analysis results, the association matrix between risk factors is constructed, a dynamic association graph between physical entities is established, and the weights of each node and edge are evaluated through graph calculation algorithms;

[0063] S4, calculates the comprehensive risk score based on the dynamic correlation graph and sets a multi-level warning level.

[0064] The entity status data collected by S1 for multiple physical entities on campus specifically includes:

[0065] S11, Collection of personnel status data: Access control devices installed on campus collect personnel entry and exit times; smart cameras combined with image recognition technology are used to identify personnel behavior, including gathering, movement, and detention;

[0066] S12, Collection of Door and Window Opening and Closing Status Data: Install intelligent magnetic switches on building doors and windows to collect the opening and closing status of doors and windows in real time, and record the time and frequency of opening or closing;

[0067] S13, vehicle status data collection: The vehicle’s entry and exit time and stay status are recorded through the smart gates at the school gate and parking lot, and the real-time location and driving path of the vehicle are monitored by smart cameras on campus.

[0068] Building a digital twin model containing entity state data specifically includes:

[0069] S14, Static Property Modeling: Based on the design parameters and actual deployment of physical entities on campus, a static basic model is created, including a personnel distribution twin model, a building door and window structure twin model, and a vehicle operation twin model. Corresponding attribute parameters are assigned to different physical entities, such as personnel identity, activity permissions, the physical location, type, and control status of doors and windows, and vehicle license plate numbers, owner information, and parking areas.

[0070] S15, Dynamic Data Mapping:

[0071] Personnel status mapping: Map the entry and exit times and monitoring data (gathering, movement, and retention) collected by access control devices to the personnel distribution twin model in real time to update the personnel distribution and activity status on campus;

[0072] Door and window status mapping: Mapping the collected door and window opening and closing status, opening time and frequency data to the building door and window structure twin model to reflect the opening or closing status of the doors and windows in real time;

[0073] Vehicle status mapping: Map the acquired vehicle entry and exit time, stop status, and driving path data to the vehicle operation twin model to display the vehicle's operation trajectory and parking distribution.

[0074] Personnel status data mapping includes:

[0075] Real-time location calculation of personnel: P i (t) = f(G x (t),G y (t),G z (t)), where P i (t) is the position coordinate of the i-th person at time t, G x (t),G y(t),G z (t) is the three-dimensional position data extracted through monitoring, and f is the spatial coordinate calculation function used to convert the raw data into a standardized position;

[0076] Behavioral state recognition:

[0077]

[0078] Among them, B q (t) is the behavior state of the qth person at time t, d qu (t) is the distance between the qth and uth persons, D th is the distance threshold for aggregation (within 2 meters), N g (t) is the number of people within the distance threshold, N th is the minimum number of people that can gather (5 people), v q (t) is the instantaneous velocity of the qth person, V min is the minimum speed threshold for movement (0.5 m / s), t 滞留 is the duration of the detention, T th is the retention time threshold (300 seconds);

[0079] Door and window status data mapping includes:

[0080] Open and closed state calculation: Among them, S w (t) is the open / close status of a door or window at time t (1 for closed, 0 for open);

[0081] Abnormal opening frequency detection Among them, F w is the frequency of doors and windows opening during the observation period (in seconds), N is the number of times doors and windows are opened during the observation period, T 观察 is the total duration of the observation period.

[0082] The vehicle status data map includes:

[0083] Vehicle dwell time calculation T 停留 =t 离开 -t 进入 , where T 停留 Indicates the time a vehicle stays in a parking lot, t 进入 represents the time when the vehicle enters the parking lot, t 离开 Indicates the time when the vehicle leaves the parking lot;

[0084] Vehicle real-time location: P v (t)=(x(t),y(t)), where, P v(t) is the two-dimensional plane position coordinate of the vehicle at time t, x(t), y(t) are the horizontal and vertical coordinates of the vehicle in the campus map coordinate system collected by the smart camera;

[0085] Path trajectory construction: in, is the complete driving trajectory of the vehicle, expressed as a sequence of positions at a set of consecutive time points, P v (t i ) is the time t i The vehicle position at time t0, t1,…, t n is the sequence of time points during the vehicle's motion.

[0086] S2 specifically includes:

[0087] S21, Simulation Environment Creation: Based on the digital twin model, a simulation environment is created that includes people, building doors and windows, and vehicles. The real-time status data of each physical entity is loaded to simulate its dynamic behavior and interaction in the actual scene.

[0088] S22, combined with the status data of the physical entity, defines the scenario risk factor:

[0089] Excessive gathering of people (people density exceeds the standard);

[0090] Abnormal opening of doors and windows (if left open for extended periods during unauthorized periods, they may provide a path for outsiders to illegally enter campus classrooms, increasing the possibility of property damage or security incidents. In addition, severe weather may enter the room through open windows, causing damage to equipment or items);

[0091] Unusual vehicle route (failure to follow the designated safe route for campus vehicles);

[0092] S23, simulation interaction impact analysis, analyzes the interaction between excessive gathering of people, abnormal opening of doors and windows, and abnormal vehicle paths.

[0093] Simulation interaction impact analysis includes:

[0094] Interaction between excessive crowd gathering and abnormally open doors and windows: Through simulation analysis of crowd status (such as lingering or abnormal crowding) and building door and window status (such as doors and windows being open for too long), we can identify the interactive impacts that lead to safety hazards, including excessive crowd gathering at open doors and windows.

[0095] Interaction between abnormal vehicle paths and excessive crowd gathering: Through simulation analysis of the distribution of people based on vehicle status and personnel status, it is possible to identify areas where abnormal vehicle paths cover excessive crowd gathering.

[0096] The correlation matrix between risk factors constructed in S3 specifically includes:

[0097] Matrix element definition: Based on the simulation interaction impact analysis results, define the degree of correlation between risk factors and construct the correlation matrix R. The elements R in the matrix ij Represents the strength of association between risk factors i and j:

[0098] Among them, w ij is the correlation strength calculated based on the simulation results;

[0099] Matrix normalization: Normalize the correlation matrix to ensure that all correlation values ​​are in the range [0,1] to facilitate subsequent graph calculations.

[0100] The establishment of a dynamic association graph between physical entities in S3 specifically includes:

[0101] Graph node definition: The physical entities in the campus are regarded as nodes of the graph, and the node set is represented as V;

[0102] Graph edge definition: Based on the non-zero elements of the scenario risk factor association matrix, establish the edge set E between nodes, and the edge weight is the corresponding association strength w ij ;

[0103] Dynamic update: Dynamically adjust the graph structure (nodes and edges) and edge weights based on the real-time status data of the physical entity and the simulation results to ensure that the dynamic association graph is updated over time.

[0104] Edge weight (association strength) w ij Calculation based on physical distance: Among them, d ij It represents the physical distance between the occurrence points of scenario risk factors i and j, and σ is the distance attenuation coefficient, which is used to control the attenuation degree of association strength with distance.

[0105] Application scenarios: Abnormal opening of doors and windows and excessive gathering of people. Calculate the distance between the open doors and windows and the area where people are gathered. The closer the distance, the higher the correlation strength.

[0106] Edge weight calculation can be based not only on physical distance, but also on temporal correlation:

[0107] Among them, S i (t), S j (t) are the status values ​​of risk factors i and j at time t (such as 0 for no occurrence and 1 for occurrence), and T is the total length of the time series.

[0108] Application scenario: Excessive crowd gathering and abnormal vehicle paths: Analyze the correlation between the two occurring in the same time period. The more synchronized the time series, the higher the correlation strength.

[0109] We further adopt a comprehensive scoring method to calculate the edge weight w by combining multiple factors. ij :

[0110] γ and β are weight coefficients, which are adjusted according to the importance of different factors in the actual scenario.

[0111] The weight of the node is calculated through graph centrality analysis to obtain the importance of each node: C i =Σ j∈V w ij , where C i represents the centrality weight of node i, w ij is the edge weight between nodes i and j.

[0112] The composite risk score in S4 is calculated as:

[0113] Among them, R is the comprehensive risk score, C i is the weight of node i, which represents the influence of a single risk factor, w ij is the weight between edges i and j, indicating the strength of the interaction between risk factors, V is the node set (risk factors) of the dynamic association graph, and E is the edge set (interaction influence relationship) of the dynamic association graph.

[0114] Based on the calculated risk scores, risks are divided into multiple warning levels, and graded response measures are taken according to different warning levels:

[0115] Level 1 (Low): Maintain routine monitoring and only record and analyze data.

[0116] Level 2 (Medium): Send a risk reminder notification and advise relevant personnel to pay attention.

[0117] Level 3 (high): Activate local early warning systems, strengthen video surveillance, and prepare emergency resources.

[0118] Level 4 (very high): Initiate emergency response and notify relevant management personnel or emergency teams to arrive for on-site disposal.

[0119] The digital twin technology-based safety management and early warning system is used to implement the above-mentioned digital twin technology-based safety management and early warning method, and includes the following modules:

[0120] Multi-entity data acquisition module: collects entity status data of multiple personnel physical entities, building door and window physical entities, and vehicle physical entities. The entity status data includes personnel status, door and window opening and closing status, and vehicle status, and inputs the collected data into the digital twin modeling module;

[0121] Digital twin modeling module: Based on the collected entity status data, a digital twin model that includes the status of people, building doors and windows, and vehicles is constructed to synchronize the dynamic behavior and status of physical entities in real time;

[0122] Scenario Risk Analysis Module: This module is used to define risk factors in the current scenario, including excessive crowds, abnormal opening of doors and windows, and abnormal vehicle paths. It also simulates and analyzes the interaction between risk factors in the digital twin model to identify the correlation between risk factors.

[0123] Dynamic association graph construction module: Based on the results of scenario risk analysis, it constructs an association matrix and generates a dynamic association graph between physical entities. It uses a graph calculation algorithm to evaluate the weights of each node and edge in the association graph to characterize the collaborative risk of multiple entities.

[0124] Risk assessment and early warning module: Calculates comprehensive risk scores based on dynamic correlation graphs and sets multi-level early warning levels based on risk scores.

[0125] The present invention encompasses any alternatives, modifications, equivalents, and solutions that fall within the spirit and scope of the present invention. To provide a thorough understanding of the present invention, specific details are described in detail below in connection with the preferred embodiments of the present invention, but those skilled in the art will be able to fully understand the present invention without these detailed descriptions. Furthermore, to avoid unnecessary confusion regarding the essence of the present invention, well-known methods, processes, procedures, components, and circuits have not been described in detail.

[0126] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A safety management early warning method based on digital twin technology, characterized by: The following steps are involved: S1, Multi-entity Digital Twin Model Construction: Collect entity status data of multiple physical entities on campus, including personnel, building doors and windows, and vehicles. The entity status data includes personnel status, door and window opening and closing status, and vehicle status, and construct a digital twin model containing entity status data; S2, Scenario Risk Factor Definition and Interaction Analysis: Define the current scenario risk factors and analyze the interaction between scenario risk factors by performing simulation interaction analysis between scenario risk factors in the digital twin model; S3, dynamic association graph construction: Based on the simulation interaction impact analysis results, the association matrix between risk factors is constructed, a dynamic association graph between physical entities is established, and the weights of each node and edge are evaluated through graph calculation algorithms; S4, calculates the comprehensive risk score based on the dynamic correlation graph and sets a multi-level warning level; The establishment of a dynamic association graph between physical entities in S3 specifically includes: Graph node definition: The physical entities in the campus are regarded as nodes of the graph, and the node set is represented as v; Graph edge definition: Based on the non-zero elements of the scenario risk factor association matrix, establish the edge set E between nodes, and the edge weight is the corresponding association strength w ij ; Dynamic update: Dynamically adjust the graph structure and edge weights based on real-time status data of physical entities and simulation results to ensure that the dynamic association graph is updated over time; The edge weight w ij Calculation based on physical distance: Among them, d ij represents the physical distance between the occurrence points of scenario risk factors i and j, and σ is the distance attenuation coefficient, which is used to control the attenuation degree of the association strength with distance; The weight of the node is calculated by graph centrality analysis to obtain the importance of each node: C i =∑ j∈v w ij , where C i represents the centrality weight of node i, w ij is the edge weight between nodes i and j.

2. The safety management early warning method based on digital twin technology according to claim 1 is characterized in that: The collection of entity status data of multiple physical entities on campus in S1 specifically includes: S11, Collection of personnel status data: Access control devices installed on campus collect personnel entry and exit times; smart cameras combined with image recognition technology are used to identify personnel behavior, including gathering, movement, and detention; S12, Collection of Door and Window Opening and Closing Status Data: Install intelligent magnetic switches on building doors and windows to collect the opening and closing status of doors and windows in real time, and record the time and frequency of opening or closing; S13, vehicle status data collection: The vehicle’s entry and exit time and stay status are recorded through the smart gates at the school gate and parking lot, and the real-time location and driving path of the vehicle are monitored by smart cameras on campus.

3. The safety management early warning method based on digital twin technology according to claim 2 is characterized in that: The construction of a digital twin model containing entity state data specifically includes: S14, static characteristic modeling: Based on the design parameters and actual deployment of physical entities on campus, a static basic model is created, including a personnel distribution twin model, a building door and window structure twin model, and a vehicle operation twin model, and corresponding attribute parameters are assigned to different physical entities; S15, Dynamic Data Mapping: Personnel status mapping: Map the entry and exit times and monitoring data collected by access control devices to the personnel distribution twin model in real time to update the personnel distribution and activity status on campus; Door and window status mapping: Mapping the collected door and window opening and closing status, opening time and frequency data to the building door and window structure twin model to reflect the opening or closing status of the doors and windows in real time; Vehicle status mapping: Map the acquired vehicle entry and exit time, stop status, and driving path data to the vehicle operation twin model to display the vehicle's operation trajectory and parking distribution.

4. The safety management early warning method based on digital twin technology according to claim 1 is characterized in that: The S2 specifically includes: S21, Simulation Environment Creation: Based on the digital twin model, a simulation environment is created that includes people, building doors and windows, and vehicles. The real-time status data of each physical entity is loaded to simulate its dynamic behavior and interaction in the actual scene. S22, combining the status data of physical entities, defines scenario risk factors, including excessive gathering of people, abnormal opening of doors and windows, and abnormal vehicle paths; S23, simulation interaction impact analysis, analyzes the interaction between excessive gathering of people, abnormal opening of doors and windows, and abnormal vehicle paths.

5. The safety management early warning method based on digital twin technology according to claim 4 is characterized in that: The simulation interaction impact analysis includes: Interaction between excessive crowd gathering and abnormally open doors and windows: Through simulation analysis of the status of people and building doors and windows, the interactive impact that leads to safety hazards is identified, including excessive crowd gathering at open doors and windows. Interaction between abnormal vehicle paths and excessive crowd gathering: Through simulation analysis of the distribution of people based on vehicle status and personnel status, it is possible to identify areas where abnormal vehicle paths cover excessive crowd gathering.

6. The safety management early warning method based on digital twin technology according to claim 5 is characterized in that: The construction of the correlation matrix between risk factors in S3 specifically includes: Matrix element definition: Based on the simulation interaction impact analysis results, define the degree of correlation between risk factors and construct the correlation matrix R. The elements R in the matrix ij Represents the strength of association between risk factors i and j: Among them, w ij is the correlation strength calculated based on the simulation results; Matrix normalization: Normalize the correlation matrix to ensure that all correlation values ​​are in the range [0,1].

7. The safety management early warning method based on digital twin technology according to claim 1 is characterized in that: The comprehensive risk score in S4 is calculated as: R=∑ i∈v C i +∑ (i，j)∈E w ij , where R is the comprehensive risk score, C i is the weight of node i, which represents the influence of a single risk factor, w ij is the weight between edge i and j, indicating the strength of the interaction between risk factors, V is the node set of the dynamic association graph, and E is the edge set of the dynamic association graph; Based on the calculated risk scores, risks are divided into multiple warning levels, and graded response measures are taken according to different warning levels.

8. A safety management and early warning system based on digital twin technology, used to implement the safety management and early warning method based on digital twin technology as described in any one of claims 1 to 7, characterized in that: Includes the following modules: Multi-entity data acquisition module: collects entity status data of multiple human physical entities, building door and window physical entities, and vehicle physical entities. The entity status data includes the status of people, the opening and closing status of doors and windows, and the status of vehicles, and inputs the collected data into the digital twin modeling module; Digital twin modeling module: Based on the collected entity status data, a digital twin model that includes the status of people, building doors and windows, and vehicles is constructed to synchronize the dynamic behavior and status of physical entities in real time; Scenario risk analysis module: used to define risk factors in the current scenario, including excessive crowds, abnormal opening of doors and windows, and abnormal vehicle paths, and simulate the interaction impact analysis between risk factors in the digital twin model to identify the correlation between risk factors; Dynamic association graph construction module: Based on the results of scenario risk analysis, it constructs an association matrix and generates a dynamic association graph between physical entities. It uses a graph calculation algorithm to evaluate the weights of each node and edge in the association graph to characterize the collaborative risk of multiple entities. Risk assessment and early warning module: Calculates comprehensive risk scores based on dynamic correlation graphs and sets multi-level early warning levels based on risk scores.

Images