An airport security management method and system

Abstract

The present invention relates to the technical field of airport security management, and in particular to an airport security management method and system. A multi-source heterogeneous data acquisition module is used to obtain various data of airport operations in real time, and a dynamic risk assessment model is used for feature fusion processing to generate a three-dimensional risk heat map and output risk levels and type identifiers. A hierarchical response mechanism is triggered, and a resource allocation plan is generated through a resource scheduling optimization algorithm and a scheduling instruction is sent. In the emergency response stage, a multi-terminal collaborative platform is used to monitor the disposal effect. When the risk diffusion coefficient exceeds a preset threshold, a cross-departmental collaborative disposal protocol is activated to dynamically adjust the resource allocation weights. The present invention improves the risk detection accuracy through multi-source data fusion and spatio-temporal correlation analysis, realizes multi-objective optimal allocation by using a dynamic resource scheduling matrix, constructs a closed-loop mechanism for hierarchical response and cross-departmental collaborative disposal, and combines digital twin and blockchain technologies to strengthen the process traceability, effectively improving the airport security management level.

Description

Technical Field

[0001] The present invention relates to the technical field of airport security management, and in particular to an airport security management method and system. Background Art

[0002] With the rapid development of the global aviation industry, airports, as the core hub of air transportation, are becoming more and more important in terms of safety management. Airport operations are currently facing a complex and ever-changing risk environment, and traditional safety management methods are unable to meet actual needs in many aspects, as shown below:

[0003] Limitations of traditional monitoring methods: In the past, airports mainly relied on a single type of data or independent monitoring systems for risk detection, which was unable to capture all potential risks in airport operations. For example, single video surveillance and security equipment detection could not fully cover all types of risk scenarios.

[0004] Lack of multi-source data integration: Each monitoring system is independent of each other, and the data is not effectively integrated, which makes it impossible to give full play to the complementary role of different types of data. For example, there is no correlation analysis between flight dynamic data, security inspection data, and personnel location data, which affects the accurate assessment and response of risks.

[0005] Disadvantages of the fixed plan model: Traditional resource scheduling is based on fixed plans, which are formulated based on past experience and typical scenarios. It is difficult to cope with complex and changeable risk events in actual airport operations, and it is easy to lead to unreasonable resource allocation;

[0006] Insufficient real-time situation response: lack of ability to dynamically adjust resource allocation based on real-time situation. Airport operation risk conditions change rapidly, and traditional scheduling models cannot perceive and respond in time, resulting in a disconnect between resource scheduling and actual needs.

[0007] The problem of information islands is serious: during emergency response, there are information barriers between airport departments, independent information systems are used, data formats and standards are not unified, and information is difficult to share and interact;

[0008] Imperfect coordinated command system: There is a lack of an efficient coordinated command system, and the responsibilities and collaboration processes of various departments in emergency response are unclear. When faced with complex risk events, uncoordinated actions, duplication of work or buck-passing are likely to occur.

[0009] Therefore, in view of the above problems, an airport safety management method and system are proposed. Summary of the invention

[0010] The purpose of the present invention is to provide an airport security management method and system to solve the problems raised in the above background technology.

[0011] To achieve the above object, the present invention provides the following technical solutions:

[0012] An airport security management method, comprising the following steps:

[0013] S1. Real-time obtain airport operation data through a multi-source heterogeneous data acquisition module, including: video surveillance data, millimeter wave radar data, security inspection equipment detection data, flight dynamic data, personnel positioning tag data, and environmental sensor data;

[0014] S2. Establish a dynamic risk assessment model, perform feature fusion processing on the input data, generate a three-dimensional risk heat map, and output the risk level and type identifier; the risk types at least include the risk of carrying contraband, the risk of personnel gathering, the risk of equipment abnormality, the risk of flight conflict, and the risk of sudden environment;

[0015] S3. Trigger a hierarchical response mechanism based on the risk level and type, generate a resource allocation plan through a resource scheduling optimization algorithm, and send scheduling instructions to security inspection channels, ground handling vehicles, emergency response units, and medical rescue units;

[0016] S4. In the emergency response stage, real-time monitor the disposal effect through a multi-terminal collaborative platform. When the risk diffusion coefficient exceeds the preset threshold, start a cross-departmental collaborative disposal protocol and dynamically adjust the resource allocation weight;

[0017] S5. Generate a traceability report of the disposal process, update the parameters of the risk assessment model, and store the optimized scheduling strategy in the strategy knowledge base.

[0018] As a preferred solution, the data acquisition module in step S1 includes:

[0019] A multi-spectral camera array deployed at key nodes of the terminal building for collecting personnel behavior characteristics and luggage status data;

[0020] A distributed meteorological sensor group deployed in the runway area for real-time monitoring of visibility, wind speed, and runway friction coefficient;

[0021] A Beidou positioning terminal integrated in shuttle buses and service vehicles for obtaining real-time position and operation trajectory data.

[0022] As a preferred solution, in step S2, the dynamic risk assessment model adopts:

[0023] The modified spatio-temporal convolutional neural network M-STCNN processes video stream data to extract abnormal behavior feature vectors; let the video stream data be , in M-STCNN, after the convolutional layer operation, the feature map The calculation formula of is:

[0024] , where, Represents at a specific spatial and temporal index position The feature map elements at respectively represent the indices of the feature map in the spatial and temporal dimensions. is the size of the convolutional kernel. is the index of the convolutional kernel in the first direction of the spatial dimension. is the index of the convolutional kernel in the second direction of the spatial dimension. is the index of the convolutional kernel in the temporal dimension. represents the input video stream data. at the value at the position. is the convolutional kernel weight. is the bias term; the spatio-temporal features in the video stream data are extracted through this formula, and then the abnormal behavior feature vector is obtained.

[0025] The modified graph neural network M-GNN constructs the association topology graph of personnel - equipment - environment and identifies potential risk propagation paths; let the graph , where is the set of nodes. is the set of edges; the feature vector of the node is In M-GNN, the updated feature vector of the node is calculated by the formula:

[0026] , where is the set of neighbor nodes of the node , is the edge weight between the nodes and , is the weight matrix. is the non-linear activation function. is the bias term. represents the feature vector of the node ; the node feature vector is updated through this formula to identify potential risk propagation paths.

[0027] The decision engine based on fuzzy logic calculates the global risk coefficient by synthesizing risk indicators in each dimension; let the set of risk indicators in each dimension be , the weight of each risk indicator is , and the formula for calculating the global risk coefficient is:

[0028] , where represents the number of risk indicators, and the global risk coefficient is calculated comprehensively through this formula.

[0029] As a preferred solution, in step S3, the resource scheduling optimization algorithm includes:

[0030] Establish a multi-objective optimization function, considering response timeliness, resource utilization rate, and disposal cost simultaneously; let the response timeliness be , the resource utilization rate be , the disposal cost be , and the multi-objective optimization function is:

[0031] , where , , are weight coefficients used to balance the relationships between different objectives, and ;

[0032] Use the improved genetic algorithm M-GA to solve the optimal scheduling plan, where the fitness function includes the risk attenuation rate, resource coverage, and collaborative operation efficiency index; let the risk attenuation rate be , the resource coverage be , the collaborative operation efficiency index be , and the fitness function Fitness is:

[0033] , where , , are weight coefficients used to adjust the importance of different indicators in the fitness function, and ;

[0034] Output a three-dimensional scheduling matrix including personnel grouping, equipment configuration, and path planning.

[0035] As a preferred solution, in step S4, the cross-departmental collaborative disposal protocol includes:

[0036] Construct an emergency command chain based on blockchain to record the response timeliness and disposal effects of each unit;

[0037] When a runway incursion event is detected, push an augmented reality situation map to the air traffic control tower, ground handling control center, and security department simultaneously;

[0038] In the medical rescue scenario, automatically generate a rescue plan including the optimal path, casualty classification, and material requirements.

[0039] As a preferred solution, in step S5, the disposal process traceability report includes:

[0040] Risk event spatio-temporal evolution atlas;

[0041] Resource scheduling response delay analysis;

[0042] Disposal strategy effectiveness assessment matrix;

[0043] Model parameter adaptive adjustment records.

[0044] As a preferred solution, abnormal behavior feature vector extraction adopts:

[0045] Abnormal posture recognition algorithm based on skeleton key point detection;

[0046] The optical flow method is used to analyze the conflict probability of personnel movement trajectories;

[0047] The detection sensitivity of suspicious objects in hidden areas is enhanced through the attention mechanism.

[0048] As a preferred solution, the process of generating the three-dimensional scheduling matrix includes:

[0049] Divide the airport area into honeycomb grid cells and calculate the resource demand density of each grid;

[0050] Conduct scheduling simulation based on the digital twin platform;

[0051] Introduce an elastic resource pool mechanism to dynamically allocate resources to meet sudden demands.

[0052] As a preferred solution, it also includes:

[0053] Implement a graded inspection strategy in the passenger security check process and dynamically adjust the inspection intensity of the security check channel based on the risk level;

[0054] Establish equipment health prediction models to trigger maintenance warnings and adjust spare parts inventory in advance;

[0055] Use the digital twin system to simulate contingency plans for major risk scenarios.

[0056] An airport security management system, which is equipped with an airport security management method to manage airport security.

[0057] It can be seen from the technical solutions provided by the present invention that the airport safety management method and system provided by the present invention have the following beneficial effects:

[0058] Multi-source data fusion: Through the multi-source heterogeneous data acquisition module, multiple types of data such as video surveillance, millimeter-wave radar, security equipment detection, flight dynamics, personnel location tags and environmental sensors are obtained, and feature fusion processing is performed; different types of data reflect the airport operation status from different dimensions and complement each other, changing the limitations of traditional single-dimensional risk monitoring; for example, video surveillance data can intuitively present personnel behavior, and millimeter-wave radar data can accurately monitor the movement trajectory of objects. The fusion of the two can more comprehensively and accurately identify potential risks;

[0059] Spatiotemporal correlation analysis: Use spatiotemporal convolutional neural networks to process video stream data and extract abnormal behavior feature vectors; use graph neural networks to build a correlation topology of personnel, equipment, and environment to identify potential risk transmission paths; this spatiotemporal correlation analysis of data fully considers the evolution of events in time and space dimensions, greatly improves the accuracy and timeliness of risk detection, and can detect potential risks in advance;

[0060] Multi-objective optimization configuration: Adopt resource scheduling optimization algorithm, establish multi-objective optimization function, and consider response timeliness, resource utilization and disposal cost at the same time; change the traditional fixed plan scheduling mode, dynamically generate resource allocation plan according to real-time situation, and realize the optimal allocation of resources; for example, when facing different types and levels of risks, it can quickly allocate the appropriate number and type of security inspection channels, ground handling vehicles, emergency disposal units and medical rescue units and other resources, which can not only ensure rapid response, but also improve resource utilization efficiency and reduce disposal costs;

[0061] Dynamic resource scheduling matrix: Generate a three-dimensional scheduling matrix including personnel grouping, equipment configuration, and path planning to make resource scheduling more scientific and accurate; by dividing the airport area into honeycomb grid units, calculating the resource demand density of each grid, and dynamically allocating resources based on simulation and deduction on the digital twin platform and combined with the elastic resource pool mechanism, it can flexibly adjust resource allocation according to real-time changes in risks and efficiently respond to various complex situations;

[0062] Graded response mechanism: The graded response mechanism is triggered based on the risk level and type, and different response measures of different intensity are taken for different risk levels to ensure that a quick and effective response can be made when risks occur, and resources are allocated reasonably to avoid over-response or under-response;

[0063] Cross-departmental collaborative disposal: Building a closed-loop mechanism of hierarchical response and cross-departmental collaborative disposal solves the problem of information islands in emergency disposal; by building a blockchain-based emergency command chain, ensure that the response time and disposal effect of each unit can be recorded and traced; in different risk scenarios, such as runway intrusion incidents and medical rescue scenarios, achieve real-time information sharing and collaborative operations among relevant departments, and improve the efficiency and success rate of emergency disposal. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 The figure is a flow chart of an airport safety management method according to the present invention. DETAILED DESCRIPTION

[0065] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0066] To better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings of the specification and specific embodiments.

[0067] As Figure 1 shown, an embodiment of the present invention provides an airport security management method, including the following steps:

[0068] S1. Real-time obtain airport operation data through a multi-source heterogeneous data acquisition module, including: video surveillance data, millimeter-wave radar data, security inspection equipment detection data, flight dynamic data, personnel positioning tag data, and environmental sensor data;

[0069] S2. Establish a dynamic risk assessment model, perform feature fusion processing on the input data, generate a three-dimensional risk heat map, and output the risk level and type identifier; the risk types at least include the risk of carrying contraband, the risk of personnel gathering, the risk of equipment abnormality, the risk of flight conflict, and the risk of sudden environment;

[0070] The dynamic risk assessment model adopts:

[0071] The modified spatio-temporal convolutional neural network M-STCNN processes video stream data and extracts abnormal behavior feature vectors; assuming the video stream data is , in M-STCNN, after the convolutional layer operation, the feature map The calculation formula is:

[0072] , where represents the feature map element at a specific spatial and time index position , respectively represent the indexes of the feature map in the spatial and time dimensions, is the size of the convolutional kernel, is the index of the convolutional kernel in the first direction of the spatial dimension, is the index of the convolutional kernel in the second direction of the spatial dimension, is the index of the convolutional kernel in the time dimension, represents the input video stream data at position, is the convolutional kernel weight, is the bias term; the spatio-temporal features in the video stream data are extracted through this formula, and then the abnormal behavior feature vectors are obtained;

[0073] The modified graph neural network M-GNN constructs an association topology graph of personnel-equipment-environment and identifies potential risk propagation paths; assuming the graph , where is the node set, is the edge set; the node has a feature vector of , in the M-GNN, the node 's updated feature vector is calculated by the formula:

[0074] , where is the set of neighbor nodes of the node , is the edge weight between the node and , is the weight matrix, is the non-linear activation function, is the bias term; represents the feature vector of the node ; by updating the node feature vector with this formula, potential risk propagation paths can be identified;

[0075] Based on the fuzzy logic-based decision engine, the global risk coefficient is calculated by integrating risk indicators in each dimension; let the set of risk indicators in each dimension be , and the weight of each risk indicator is , the global risk coefficient is calculated by the formula:

[0076] , where represents the number of risk indicators, and the global risk coefficient is obtained by comprehensive calculation with this formula;

[0077] S3. Trigger a hierarchical response mechanism based on the risk level and type, generate a resource allocation plan through the resource scheduling optimization algorithm, and send scheduling instructions to the security inspection channels, ground service vehicles, emergency response units, and medical rescue units;

[0078] The resource scheduling optimization algorithm includes:

[0079] Establish a multi-objective optimization function, considering response timeliness, resource utilization rate, and disposal cost at the same time; let the response timeliness be , the resource utilization rate be , the disposal cost be , the multi-objective optimization function is:

[0080] , where , , are the weight coefficients, used to balance the relationship between different objectives, and ;

[0081] The improved genetic algorithm M-GA is used to solve the optimal scheduling scheme. Among them, the fitness function includes the risk attenuation rate, the resource coverage range, and the collaborative operation efficiency index. Let the risk attenuation rate be , the resource coverage range be , and the collaborative operation efficiency index be . The fitness function Fitness is:

[0082] , where , , are weight coefficients used to adjust the importance of different indicators in the fitness function, and ;

[0083] Output a three-dimensional scheduling matrix including personnel grouping, equipment configuration, and path planning;

[0084] S4. In the emergency response stage, the disposal effect is monitored in real time through the multi-terminal collaborative platform. When the risk diffusion coefficient exceeds the preset threshold, start the cross-departmental collaborative disposal protocol and dynamically adjust the resource allocation weight;

[0085] S5. Generate a traceability report of the disposal process, update the parameters of the risk assessment model, and store the optimized scheduling strategy in the strategy knowledge base.

[0086] In this embodiment, the detailed steps of step S1 are as follows:

[0087] Step S1-1: Planning and deployment of the data collection module

[0088] Determine the collection requirements:

[0089] Based on the overall goal of airport security management and the specific requirements of each business link, clarify the types of data to be collected; for example, to ensure passenger safety and security inspection efficiency, it is necessary to collect security equipment detection data; to monitor the airport operation status, it is necessary to obtain flight dynamic data, personnel positioning tag data, and environmental sensor data, etc.;

[0090] Analyze the roles of different data in aspects such as risk assessment, resource scheduling, and emergency response to provide a guiding direction for subsequent data collection and processing;

[0091] Plan the collection locations:

[0092] For video surveillance data, at key nodes where people and luggage flow, such as the entrance of the terminal building, security check area, waiting hall, boarding gate, etc., reasonably plan the installation locations of multi-spectral camera arrays; ensure that the cameras can cover key areas without monitoring blind spots and can clearly capture the behavior characteristics of people and the status data of luggage; for example, in the security check area, the cameras should be able to clearly capture the process of passengers' luggage passing through the security check, the operations of security personnel, and the relevant behaviors of passengers.

[0093] For millimeter-wave radar data, reasonably layout millimeter-wave radars in areas such as around the runway and apron to accurately monitor information such as the position, speed, and attitude of the aircraft during takeoff and landing, as well as the movement of surrounding vehicles and personnel.

[0094] For the collection of security equipment detection data, ensure that security equipment (such as X-ray security scanners, metal detectors, etc.) is installed in appropriate positions in the passenger security check passage to ensure the effective detection of luggage and items carried by people.

[0095] Flight dynamic data is mainly obtained from the airport's flight information management system. Establish a connection with this system through a network interface to obtain real-time data such as the takeoff and landing times, flight status, and route information of flights.

[0096] For the collection of personnel positioning tag data, equip positioning tags for airport staff and passengers in specific areas, and deploy signal receiving base stations in the corresponding areas; the layout of the base stations should ensure that the real-time position information of people can be accurately obtained, such as reasonably distributing base stations in important areas such as each floor of the terminal building, runway area, and parking lot.

[0097] In terms of the collection of environmental sensor data, in the runway area, according to meteorological monitoring requirements and the characteristics of the runway structure, reasonably arrange a distributed meteorological sensor group for real-time monitoring of meteorological parameters such as visibility, wind speed, and runway surface friction coefficient; in the terminal building, install environmental sensors such as temperature and humidity sensors and air quality sensors to monitor the indoor environmental conditions.

[0098] Deploy collection equipment:

[0099] Install and commission the multi-spectral camera array according to the planned positions; ensure that parameters such as the installation angle, height, and focal length of the cameras are set reasonably to obtain clear and complete image data; at the same time, configure the network for the cameras so that they can transmit the collected video data to the data processing center in real time.

[0100] Deploy millimeter-wave radar equipment to ensure its firm installation and adaptability to the complex environmental conditions of the airport, such as strong winds and sandstorms; calibrate radar parameters, including range resolution, velocity resolution, etc., to improve the accuracy of radar data; connect the radar equipment to the data transmission network to achieve real-time data upload.

[0101] Install and debug security inspection equipment to ensure that its detection accuracy and stability meet relevant standards; integrate the security inspection equipment with the data acquisition system to achieve automatic acquisition and transmission of detection data;

[0102] Establish a data interface with the flight information management system and ensure the ability to stably and real-time obtain flight dynamic data through a secure network connection; conduct regular tests and maintenance on the data interface to prevent data transmission interruptions or errors;

[0103] Distribute and activate personnel positioning tags, and debug and optimize the signal receiving base stations; ensure that the signal coverage range of the base stations meets the design requirements, can accurately receive the signals sent by the positioning tags, and transmit the personnel location data to the data processing system;

[0104] Install a distributed meteorological sensor group and environmental sensors in the terminal building to ensure that the installation locations of the sensors meet the meteorological monitoring specifications and environmental monitoring requirements; calibrate and initialize the sensors to ensure the accuracy and reliability of the collected data; connect the sensors to the data transmission network by wired or wireless means to achieve real-time acquisition and transmission of environmental data;

[0105] Step S1-2: Data acquisition and preliminary processing

[0106] Video surveillance data acquisition and processing:

[0107] The multi-spectral camera array performs video acquisition at a preset frame rate and resolution; at the same time, according to different scene requirements, the frame rate and resolution can be dynamically adjusted. For example, in crowded areas or during important events, the frame rate is increased to capture more detailed behavior information;

[0108] Perform real-time preprocessing on the collected video data, including image enhancement to improve the clarity and readability of the images; noise reduction processing, using methods based on wavelet transform or Gaussian filtering to remove noise interference in the video and improve the image quality;

[0109] Millimeter-wave radar data acquisition and processing:

[0110] The millimeter-wave radar scans the target area at a set frequency and real-time collects information such as the distance, speed, and angle of the target object;

[0111] Perform filtering processing on the collected radar echo data, using methods such as Kalman filters or Wiener filters to remove noise signals generated by environmental interference and improve the accuracy of the data;

[0112] Convert the polar coordinate data collected by the radar into Cartesian coordinate data for fusion processing with other types of data; at the same time, classify and identify the targets detected by the radar, distinguish different types of targets such as people, vehicles, and airplanes, and mark their relevant attribute information;

[0113] Collection and processing of security inspection equipment detection data:

[0114] During the passenger security inspection process, the security inspection equipment collects detection data in real time; for example, the X-ray security inspection machine collects image data of the items inside the luggage, and the metal detector detects the information of the metal items carried by the personnel or luggage;

[0115] Conduct preliminary analysis and processing on the collected security inspection data; for the X-ray image data, use image recognition algorithms (such as object detection algorithms based on deep learning) to identify the item outlines and categories in the luggage, determine whether there are prohibited items, and mark the positions and types of the prohibited items; for the metal detector data, record the information such as the positions and intensities of the detected metal items;

[0116] Format and store the processed security inspection data for subsequent data transmission and analysis; at the same time, encrypt the security inspection data to ensure passenger privacy and data security;

[0117] Collection and processing of flight dynamic data:

[0118] Regularly obtain flight dynamic data through the real-time data interface established with the flight information management system; the obtained data includes the planned takeoff and landing times, actual takeoff and landing times, flight status (normal, delayed, cancelled, etc.), flight number, route information, etc. of the flight;

[0119] Clean and verify the obtained flight dynamic data, and remove duplicate, incorrect or invalid data records; for example, check whether the flight time format is correct and whether the flight status conforms to the actual situation, etc.;

[0120] Store the processed flight dynamic data in a certain format (such as JSON or XML format) for subsequent data query and analysis; at the same time, associate the flight dynamic data with other data (such as personnel positioning data, equipment status data, etc.) to provide more comprehensive information support for airport security management;

[0121] Collection and processing of personnel positioning tag data:

[0122] The personnel positioning tag sends position signals to the signal receiving base station at a set time interval (such as once per second or once every 5 seconds); after receiving the signal, the base station calculates the real-time position coordinates of the personnel (such as longitude, latitude, floor, etc.) through positioning algorithms (such as positioning algorithms based on RSSI, TOA or TDOA);

[0123] Smooth the collected personnel location data to remove location jumps caused by signal fluctuations or interference; use methods such as moving average filtering or Kalman filtering to improve the stability and accuracy of the location data;

[0124] Associate the personnel location data with the personnel identity information. By reading the ID information of the positioning tag, query the corresponding personnel identity (such as passengers, staff, security personnel, etc.), affiliated department, job position, etc. in the personnel information database;

[0125] Store the processed personnel positioning tag data, and at the same time transmit it to the data processing center in real time to provide real-time location information support for personnel behavior analysis, resource scheduling, etc.;

[0126] Environmental sensor data collection and processing:

[0127] The distributed meteorological sensor group collects meteorological parameters such as visibility, wind speed, wind direction, and runway surface friction coefficient in the runway area in real time, and the environmental sensors in the terminal building monitor environmental data such as temperature, humidity, and air quality (such as PM2.5, PM10, carbon dioxide concentration, etc.) in real time;

[0128] Calibrate and correct the collected environmental sensor data, and adjust the original data according to the characteristics of the sensor and calibration parameters to improve the accuracy of the data;

[0129] Detect and process outliers in the environmental data, and use statistical analysis methods (such as the 3σ criterion) or machine learning algorithms (such as the IsolationForest algorithm) to identify and remove abnormal data points;

[0130] Store the processed environmental sensor data, and summarize and statistically analyze it at regular time intervals (such as every minute or every 5 minutes) to generate an environmental data report to provide data support for airport environmental monitoring and risk assessment;

[0131] Step S1-3: Data transmission and integration

[0132] Data transmission:

[0133] Video surveillance data is transmitted to the data processing center through a wired network (such as fiber optic Ethernet) or a wireless network (such as 5G, Wi-Fi); for video stream data with high real-time requirements, use an efficient transmission protocol (such as RTSP, RTMP) to ensure the smooth transmission of video data and reduce latency and packet loss;

[0134] Millimeter-wave radar data, security inspection equipment detection data, personnel positioning tag data, and environmental sensor data are transmitted to the data processing center via wired or wireless means. For radar data and positioning tag data with small data volumes but high real-time requirements, low-power, high-reliability wireless transmission protocols (such as LoRa, ZigBee) can be used. For security inspection equipment detection data and environmental sensor data, a wired network or wireless network can be selected for transmission according to the actual situation.

[0135] Flight dynamic data is transmitted from the flight information management system to the data processing center through a dedicated data interface and network channel. To ensure the security and stability of data transmission, an encrypted transmission protocol (such as SSL / TLS) is used to encrypt the data to prevent data leakage and tampering.

[0136] Data integration:

[0137] In the data processing center, a data reception module is set up to be responsible for receiving data from different data sources. The received data is classified and identified, and marked according to the data type (such as video data, radar data, security inspection data, etc.) and source (such as specific cameras, radar devices, security inspection channels, etc.) for subsequent data management and processing.

[0138] The received data is stored in the corresponding database or data storage system. For structured data (such as flight dynamic data, personnel positioning tag data, environmental sensor data, etc.), it is stored in a relational database (such as MySQL, Oracle) for efficient querying and analysis. For unstructured data (such as video surveillance data, X-ray image data, etc.), it is stored in a distributed file system (such as Ceph, GlusterFS) or an object storage system (such as Amazon S3, Alibaba Cloud OSS), and indexes are established to associate with relevant metadata information for convenient data retrieval and invocation.

[0139] A data integration mechanism is established to associate and fuse different types of data. Through key attributes such as the timestamp, geographical location information, and relevant identifiers of the data, data from different data sources is matched and integrated. For example, video surveillance data, personnel positioning tag data, and environmental sensor data within the same time and the same area are associated to provide a basis for subsequent multi-source data fusion analysis.

[0140] In this embodiment, the specific operation steps of step S2 include:

[0141] Step S2-1: Data preprocessing

[0142] Video surveillance data: The video surveillance data obtained from step S1 is first subjected to format standardization processing. Multiple formats of videos collected by different multispectral cameras are converted into a standard format convenient for processing, such as the H.264 encoding format. Then, noise reduction processing is performed on the video. A noise reduction algorithm based on wavelet transform is used to remove noise interference in the video and improve the image quality. Next, cropping and normalization of video frames are carried out. The video frames are cropped according to the areas of concern (such as security check channels, waiting areas, etc.), and at the same time, the image size is normalized to a specific size for subsequent neural network processing.

[0143] Millimeter-wave radar data: Filtering processing is performed on the millimeter-wave radar data to remove clutter signals generated by environmental interference. The Kalman filter is used to process the radar echo data to improve the accuracy of target detection. Then, the radar data is converted from polar coordinates to Cartesian coordinates to facilitate fusion with other data. At the same time, the targets detected by the radar are classified to distinguish whether they are humans, vehicles, or other objects, and information such as their positions, speeds, and movement directions is marked.

[0144] Security inspection equipment detection data: Various types of data detected by security inspection equipment are parsed. For example, for the image data of X-ray security inspection equipment, image recognition algorithms are used to identify the outlines and categories of items in the luggage to determine whether there are prohibited items. For metal detector data, the positions and intensity information of detected metal items are recorded. These data are sorted out, key features are extracted, such as the types and positions of prohibited items, and standardized processing is carried out to make the data of different security inspection equipment on the same scale for subsequent fusion.

[0145] Flight dynamic data: Flight dynamic data is obtained from the flight information system, including the takeoff and landing times of flights, estimated arrival / departure times, and flight status (normal, delayed, cancelled, etc.). These data are cleaned to remove duplicate or incorrect information, such as the situation where the flight status is updated repeatedly and inconsistently, and corrected according to authoritative data sources. At the same time, the flight dynamic data is calibrated with the time system of the airport to ensure the time accuracy of the data.

[0146] Personnel positioning tag data: The real-time position data sent by the personnel positioning tags is resolved. According to the positioning technology of the tags (such as RFID, UWB, etc.), the precise position coordinates of personnel in the airport are determined. Smoothing processing is performed on the position data to remove position jumps caused by signal fluctuations. The moving average filtering algorithm is used to process continuous position data. At the same time, the personnel identity information is associated with the position data, such as matching through the tag ID with the personnel information database to obtain the identity categories of personnel (passengers, staff, etc.).

[0147] Environmental sensor data: Integrate data from distributed meteorological sensor groups in the runway area and environmental sensors inside the terminal building; for meteorological data such as visibility, wind speed, runway surface friction coefficient, etc., perform outlier detection and correction. For example, if the wind speed data shows a value that significantly deviates from the historical mean and does not conform to the current weather conditions, interpolation method is used for correction; for environmental sensor data inside the terminal building, such as temperature, humidity, air quality, etc., perform normalization processing to convert data from different sensors to the same numerical range for convenient comprehensive analysis;

[0148] Step S2-2: Feature extraction

[0149] Process video stream data based on the modified spatio-temporal convolutional neural network (M-STCNN): Input the preprocessed video stream data into the M-STCNN; first, through multiple convolutional layers, perform convolutional operations according to the formula where, is the input video stream data, represents the feature map element at a specific spatial and temporal index position , respectively represent the indices of the feature map in the spatial and temporal dimensions, is the size of the convolutional kernel, is the index of the convolutional kernel in the first direction of the spatial dimension, is the index of the convolutional kernel in the second direction of the spatial dimension, is the index of the convolutional kernel in the temporal dimension, represents the input video stream data at position, is the convolutional kernel weight, is the bias term; extract spatio-temporal features in the video through convolutional operations, such as the actions and behavior patterns of people; then, through the pooling layer, perform dimensionality reduction on the feature map to reduce the data volume while retaining key features; finally, convert the extracted features into an abnormal behavior feature vector through the fully connected layer, and this vector contains the differential features between the behavior of people in the video and the normal behavior pattern for subsequent risk assessment;

[0150] Construct an association topology graph based on the modified graph neural network (M-GNN): Use different entities such as people, equipment, and environment as nodes and their mutual relationships as edges to construct a graph structure , where, is the set of nodes, is the set of edges. For each node , assign its corresponding feature vector , for example, the feature vector of a personnel node can include the identity information, location information, etc. of the personnel, and the feature vector of a device node can include the device type, operating status, etc. In the M-GNN, according to the formula update the node feature vector, where represents the node updated feature vector, is the set of neighbor nodes of node , is the node and the edge weight between is the weight matrix, is the non-linear activation function, is the bias term; represents the feature vector of node ; By iteratively updating the node feature vector multiple times, identify potential risk propagation paths;

[0151] Extraction of other data features: For the detection data of security inspection equipment, extract features such as the type, quantity, and appearance frequency of prohibited items; for flight dynamic data, extract features such as flight delay duration and number of flight conflicts; for personnel positioning label data, extract features such as personnel aggregation density and personnel flow speed in specific areas; for environmental sensor data, extract features such as the change trend of meteorological conditions and abnormal fluctuations of environmental parameters;

[0152] Step S2-3: Feature fusion

[0153] Selection of data fusion method: Adopt a feature fusion method based on weighted average, and assign corresponding weights to each feature vector according to the importance of different types of data in risk assessment; for example, for the feature vectors of security inspection equipment detection data and video surveillance data directly related to security, assign higher weights; for flight dynamic data and environmental sensor data, reasonably adjust the weights according to the specific scenario and risk type;

[0154] Fusion process: Fusion the feature vectors extracted from different data sources in a weighted average manner; Let the feature vector from video surveillance data be , the feature vector of millimeter wave radar data be , the feature vector of security inspection equipment detection data be , the feature vector of flight dynamic data be , the feature vector of personnel positioning label data be , the feature vector of environmental sensor data be , and the corresponding weights be , , , , , The fused feature vector The calculation formula is as follows:

[0155] ,

[0156] wherein, , through feature fusion, the features of multi-source heterogeneous data are integrated into a comprehensive feature vector, which can more comprehensively reflect the operation status of the airport;

[0157] Step S2-4: Risk assessment and output

[0158] The decision engine based on fuzzy logic calculates the global risk coefficient: Input the fused feature vector into the decision engine based on fuzzy logic; Let the set of risk indicators for each dimension be , and these risk indicators are parameters related to different risk types further extracted from the fused feature vector. For example can represent the risk indicator of carrying contraband, represents the risk indicator of personnel gathering, etc.; The weight of each risk indicator is , and the weight value is determined according to the importance and relevance of the risk type; Calculate the global risk coefficient according to the formula wherein, , represents the number of risk indicators;

[0159] Determination of risk level and type identifier: According to the calculated global risk coefficient , determine the risk level according to the preset risk level division standard. For example, when is between 0 and 0.3, it is a low risk level, between 0.3 and 0.6 is a medium risk level, and between 0.6 and 1 is a high risk level; At the same time, according to the contribution of the risk indicators in each dimension of the fused feature vector, determine the risk type identifier, such as the risk of carrying contraband, the risk of personnel gathering, the risk of equipment abnormality, the risk of flight conflict, and the risk of sudden environment, etc.;

[0160] Generate a three-dimensional risk heat map: Use the spatial position (two-dimensional plane) and time of the airport area as the three-dimensional coordinates, map the risk level and type information to this three-dimensional space, and generate a three-dimensional risk heat map; In the heat map, different colors and brightness represent different risk levels, and the change of risk in the time dimension is displayed in the form of animation, intuitively presenting the risk distribution and evolution trend of different areas of the airport at different times; Finally, output the risk level, type identifier, and three-dimensional risk heat map, providing a decision-making basis for the subsequent hierarchical response mechanism.

[0161] In this embodiment, the specific operation steps of step S3 are as follows:

[0162] Step S3-1: Risk Level and Type Judgment

[0163] Receive the risk level and type identifier output by Step S2; clarify the risk level, such as low risk, medium risk, and high risk; determine the risk type, for example, the risk of carrying contraband, the risk of personnel gathering, the risk of equipment abnormality, the risk of flight conflict, and the risk of sudden environment, etc.; this step is the basis for triggering the hierarchical response mechanism and generating the resource allocation plan, ensuring that the system can make targeted decisions according to different risk situations;

[0164] Step S3-2: Trigger the Hierarchical Response Mechanism

[0165] Low-risk response: If it is determined to be a low-risk level, initiate basic response measures; for example, for the risk of carrying contraband, appropriately increase the random inspection frequency of the security check channels; for the risk of personnel gathering, arrange a small number of security personnel to go to the gathering area for guidance, and remind personnel to keep an appropriate distance through the public address system; in terms of the risk of equipment abnormality, notify the equipment maintenance personnel to pay attention to the status of relevant equipment and conduct a preliminary inspection; for the risk of flight conflict, the air traffic control department strengthens the monitoring of flight dynamics and adjusts the flight queuing order in a timely manner; for the risk of sudden environment, such as mild meteorological anomalies, issue environmental warning information to remind relevant operators to pay attention to safety;

[0166] Medium-risk response: When the risk level is medium risk, take more proactive response measures; for the risk of carrying contraband, comprehensively strengthen the inspection intensity of the security check channels, increase the number of security personnel, and conduct a detailed inspection of the luggage of passengers on key flights or in key areas; for the risk of personnel gathering, deploy more security forces to the scene, set up isolation areas, and guide personnel to evacuate in an orderly manner; in terms of the risk of equipment abnormality, immediately initiate the equipment emergency repair process, arrange a professional repair team for emergency repair, and call in standby equipment to ensure that the airport operation is not affected greatly; for the risk of flight conflict, the air traffic control department coordinates with the flight crew and airlines of relevant flights to formulate a detailed flight adjustment plan, which may include measures such as flight delays and route changes; for the risk of sudden environment, such as moderate meteorological disasters, close some affected areas and adjust relevant operation processes, such as closing some areas of the runway to deal with bad weather;

[0167] High - risk response: If it is determined to be a high - risk level, a comprehensive emergency response is implemented; for the risk of carrying contraband, immediately block the relevant security inspection areas, conduct a comprehensive inspection of all personnel and luggage, and at the same time activate the emergency plan and jointly dispose of it with multiple departments; in terms of the risk of crowd gathering, activate the large - scale personnel evacuation plan, coordinate all security forces and emergency rescue personnel within the airport, and guide personnel to evacuate quickly according to the pre - determined evacuation route to ensure the safety of personnel lives; for the risk of equipment anomalies, if a critical equipment fails severely, immediately stop the operation activities in the relevant area, evacuate the surrounding personnel, organize an expert team for emergency repair, and at the same time evaluate the impact on the overall airport operation and formulate a comprehensive recovery plan; for the risk of flight conflicts, the air traffic control department issues an emergency order to suspend the take - off and landing of some flights, coordinate with surrounding airports to provide temporary alternate landing services, and make every effort to ensure the safety of airport operations; for the risk of sudden environment, such as severe meteorological disasters or other major environmental events, the airport fully enters the emergency state, stops most non - essential operation activities, activates the overall emergency shelter plan of the airport to ensure the safety of personnel and important facilities;

[0168] Step S3 - 3: Establish a multi - objective optimization function

[0169] Response timeliness objective: Considering the importance of rapid response in different risk scenarios, the response timeliness is incorporated into the optimization function; let the time from risk identification to the completion of resource allocation be , in order to minimize the response time, the part of the objective function regarding response timeliness is ;

[0170] Resource utilization rate objective: In order to rationally utilize the limited resources of the airport, the resource utilization rate is taken as an objective. Let the total amount of resources invested be , and the actually effectively utilized amount of resources be , then the resource utilization rate is , and the part of the objective function regarding resource utilization rate is ;

[0171] Disposal cost objective: Considering the cost factors in the process of resource allocation and disposal, such as labor costs, equipment usage costs, etc. Let the disposal cost be , and the part of the objective function regarding disposal cost is ;

[0172] Comprehensive multi - objective optimization function: Combining the above three objectives, a multi - objective optimization function is constructed. By using the weight coefficients , , to balance the relationships between different objectives, and , the multi - objective optimization function is as follows:

[0173] ;

[0174] Step S3-4: Solve the optimal scheduling plan using an improved genetic algorithm

[0175] Coding: Encode the resource scheduling problem and convert information such as personnel grouping, equipment configuration, and path planning into chromosomes in the genetic algorithm; for example, encode the deployment quantities and combination methods of different types of security inspection personnel, ground handling vehicles, emergency response units, and medical rescue units, where each gene represents a specific resource deployment decision;

[0176] Initializing the population: Randomly generate a set of initial chromosomes as the population, and the population size is determined according to the complexity of the actual problem and computing resources; each chromosome represents a possible resource scheduling plan, and these initial plans form the basis for the genetic algorithm to search for the optimal solution;

[0177] Calculating the fitness function: Based on the multi-objective optimization function in Step S3-3 , combine the risk attenuation rate, resource coverage, and collaborative operation efficiency index to construct the fitness function Fitness; let the risk attenuation rate be , the resource coverage be , the collaborative operation efficiency index be , and adjust the importance of different indicators in the fitness function through the weight coefficients , , , and ; the fitness function Fitness is: ;

[0178] For each chromosome (i.e., each resource scheduling plan), calculate its fitness value. The higher the fitness value, the closer the plan is to the optimal solution;

[0179] Selection operation: According to the fitness value, use methods such as roulette wheel selection or tournament selection to select better chromosomes from the current population to enter the next generation population; the selection process simulates the "survival of the fittest" principle in natural selection, making chromosomes with high fitness more likely to be selected, thus retaining the characteristics of excellent resource scheduling plans;

[0180] Crossover operation: Perform a crossover operation on the selected chromosomes, simulating the gene exchange process in biological inheritance; randomly select two chromosomes and exchange some genes between them to generate new chromosomes; for example, for two chromosomes with different personnel grouping and equipment configuration plans, exchange the parts of genes related to the deployment of security inspection personnel and the allocation of ground handling vehicles to generate a new combination of resource scheduling plans; the crossover operation helps to generate new potential excellent plans and expand the search space;

[0181] Mutation operation: Perform mutation operation on the chromosome with a certain probability, randomly changing some gene values in the chromosome; for example, randomly adjusting the deployment quantity of a certain emergency response unit or changing the path planning of a medical rescue unit to prevent the algorithm from falling into a local optimal solution; the mutation operation introduces new gene features and increases the diversity of the population;

[0182] Iterative optimization: Repeatedly perform selection, crossover, and mutation operations, continuously iterate and update the population until the preset termination conditions are met, such as reaching the maximum number of iterations or the fitness value converges to a certain extent; the optimal resource scheduling plan corresponding to the finally obtained optimal chromosome is the optimal scheduling plan solved by the improved genetic algorithm;

[0183] Step S3-5: Generate a three-dimensional scheduling matrix and output scheduling instructions

[0184] Generate a three-dimensional scheduling matrix: Divide the airport area into honeycomb grid cells, calculate the resource demand density of each grid; according to the optimal scheduling plan, integrate information such as personnel grouping, equipment configuration, and path planning into a three-dimensional space to generate a three-dimensional scheduling matrix containing personnel, equipment, and path information; in this matrix, each element corresponds to a specific resource allocation decision, for example, at a certain time point, the number of security inspection personnel, the type of ground service vehicle, and the driving path that should be allocated in a certain grid cell, etc.

[0185] Output scheduling instructions: According to the three-dimensional scheduling matrix, send scheduling instructions to security inspection channels, ground service vehicles, emergency response units, and medical rescue units; these instructions clarify specific tasks, action times, and locations of each resource unit, etc.; for example, instruct the security inspection channel to increase the number of security inspection personnel and adjust the inspection process, inform the ground service vehicle to go to a specified location to perform a specific task, and notify the emergency response unit and the medical rescue unit to reach the corresponding area for emergency response according to the planned path and time; through precise scheduling instructions, effectively allocate airport resources to cope with different risk scenarios and ensure the safe operation of the airport.

[0186] Step S4-1: Build a multi-terminal collaboration platform

[0187] Hardware deployment: Deploy terminal devices with high reliability and real-time communication capabilities in key areas of the airport, such as the command center, security inspection station, ground service operation area, medical first aid station, etc.; these terminal devices include but are not limited to industrial tablet computers, intelligent handheld devices, vehicle-mounted terminals, etc., to ensure that relevant departments and personnel can obtain information and receive instructions in real time;

[0188] Software Integration: Develop a unified multi-terminal collaborative software with functions such as real-time data display, interactive operation, and task management; install and adapt the software to each terminal device to achieve seamless collaborative work among different types of terminal devices; at the same time, establish a stable connection between the terminal device and the central data processing system through network communication protocols to ensure fast and accurate data transmission;

[0189] Permission Settings: Set corresponding operation permissions for each terminal device according to the responsibilities of different personnel and departments; for example, personnel in the command center have the highest permissions and can view all risk information and issue global scheduling instructions; security inspection personnel can only view risks and disposal information related to security inspection work and perform corresponding security inspection tasks; medical rescue personnel focus on information and tasks related to medical rescue; through reasonable permission settings, ensure the security of information and the standardization of operations;

[0190] Step S4-2: Real-time monitor the disposal effect

[0191] Data Collection and Transmission: During the emergency disposal process, each terminal device collects data related to the disposal effect in real time; for example, the security inspection channel records the number of contraband items seized and the execution time of the security inspection process through equipment; the ground support vehicle feeds back the arrival time at the designated location and the completion status of the task through in-vehicle sensors; the emergency disposal unit and the medical rescue unit report the on-site situation and the progress of the treatment of the wounded through the handheld devices of on-site personnel; these data are transmitted to the central data processing system in real time through the network;

[0192] Data Processing and Analysis: The central data processing system performs real-time processing and analysis on the large amount of collected data; uses data analysis algorithms to compare the actual disposal data with the preset standards and goals to evaluate the disposal effect; for example, compare the number of contraband items seized with the expected goal to judge whether the security inspection work has achieved the expected effect; analyze the task execution time of the ground support vehicle and the planned time to evaluate its execution efficiency; judge whether the medical rescue work is effective according to the progress of the treatment of the wounded and the standard treatment process;

[0193] Visualization Display: Display the processed and analyzed disposal effect data in an intuitive visual way on each terminal device; through forms such as charts, graphs, and texts, present the disposal progress, effect evaluation results, and existing problems of each link in real time; for example, on the large screen in the command center, display the positions and task execution situations of each emergency disposal unit in the form of a map, and use bar charts to compare the number of contraband items seized in different security inspection channels, etc., so that relevant personnel can quickly understand the overall situation;

[0194] Step S4-3: Calculate the risk diffusion coefficient

[0195] Determine the risk diffusion index: Consider various factors comprehensively to determine the risk diffusion index, which includes but is not limited to the range of personnel flow, the affected area of equipment failure, the scope of flight delay impact, the affected area of environmental factors, etc.; for example, for the risk of personnel gathering, the expansion of the range of personnel flow may mean the spread of risk; for the risk of equipment anomaly, if the equipment failure affects other surrounding equipment or areas, it indicates that the risk has a tendency to spread.

[0196] Data collection and quantification: Through a multi-source data acquisition system, collect data related to the risk diffusion index and perform quantification processing on it; for example, use personnel positioning tag data to obtain the real-time location and movement trajectory of personnel, and calculate the range of personnel flow within a certain period of time; obtain the affected range and degree of equipment failure through the equipment monitoring system; count the impact of flight delays on subsequent flights based on flight dynamic data; determine the affected area of sudden environmental risks with the help of environmental sensor data, etc.

[0197] Calculate the risk diffusion coefficient: Use a specific mathematical model or algorithm to calculate the risk diffusion coefficient based on the quantified risk diffusion index data; this coefficient can comprehensively reflect the degree of risk diffusion in the current disposal process; for example, the weighted average method can be adopted, different weights are assigned to each index according to different risk types, and then the risk diffusion coefficient is calculated. :

[0198] , where is the number of risk diffusion indicators, is the weight of the th indicator, is the th indicator's quantified value; the determination of the weight needs to be scientifically analyzed and adjusted according to the characteristics and actual situation of different risk types.

[0199] Step S4-4: Determine whether to initiate the cross-departmental collaborative disposal protocol

[0200] Set a preset threshold: According to historical data, risk assessment models, and the actual operation situation of the airport, set a reasonable preset threshold for the risk diffusion coefficient; this threshold is a key indicator for determining whether to initiate the cross-departmental collaborative disposal protocol; for example, for different types of risks, different thresholds can be set respectively, such as the risk diffusion coefficient threshold for carrying contraband is , the risk diffusion coefficient threshold for personnel gathering is , etc.; the setting of these thresholds should not only consider initiating the collaborative mechanism in a timely manner to effectively control risks, but also avoid unnecessary resource waste and increased collaborative costs due to excessive sensitivity.

[0201] Comparison and Judgment: Compare the calculated risk diffusion coefficient with the corresponding preset threshold; if the risk diffusion coefficient is less than the preset threshold, it indicates that the current emergency response measures can effectively control the risk, and there is no need to initiate an interdepartmental collaborative response agreement. Continue to monitor and handle according to the existing response plan and process; if the risk diffusion coefficient exceeds the preset threshold, it means that the risk has a tendency to get out of control, and it is necessary to immediately initiate an interdepartmental collaborative response agreement to integrate more resources, strengthen the collaboration between departments, and jointly respond to the risk;

[0202] Step S4-5: Initiate an interdepartmental collaborative response agreement and dynamically adjust the resource allocation weights

[0203] Construct an emergency command chain based on blockchain: When it is determined that an interdepartmental collaborative response agreement needs to be initiated, use blockchain technology to construct an emergency command chain; on the blockchain platform, each participating department and relevant personnel act as nodes, and key information during the emergency response process, such as the response timeliness of each unit, disposal effect data, resource allocation records, etc., is recorded on the blockchain in an encrypted manner; the immutable and traceable characteristics of the blockchain ensure the authenticity and credibility of the information, providing reliable data support for collaborative command;

[0204] Information sharing and situation push: For different types of risk events, initiate corresponding information sharing and situation push mechanisms; for example, when a runway incursion event is detected, synchronously push an augmented reality (AR) situation map to the air traffic control tower, ground service control center, and security department through a multi-terminal collaborative platform; this situation map can intuitively display the real-time situation at the runway site, including information such as the location of the intrusion object, runway status, distribution of surrounding flights and vehicles, etc., enabling each department to make collaborative decisions and actions based on the same accurate information;

[0205] Generate a rescue plan for a specific scenario: In the medical rescue scenario, the system automatically generates a rescue plan including the optimal route, casualty classification, and material requirements based on information such as the situation of the wounded at the scene, distribution of medical resources, and traffic conditions; use intelligent algorithms to plan the optimal driving route for rescue vehicles to send the wounded to a suitable medical treatment point at the fastest speed; classify the wounded according to the severity of their injuries and reasonably allocate medical resources; at the same time, according to the prediction of material requirements, timely allocate corresponding medical supplies;

[0206] Dynamic adjustment of resource allocation weights: Based on the real-time situation of risk diffusion and the handling feedback of each department, use an optimization algorithm to dynamically adjust the resource allocation weights. For example, if it is found that the risk of personnel gathering spreads rapidly in a certain area, it is necessary to increase the investment in security personnel and emergency evacuation equipment, and correspondingly increase the weights of these resources in the overall resource allocation. By continuously adjusting the resource allocation weights in real time, resources can be more accurately invested in the places where they are most needed, improving the efficiency and effectiveness of cross-departmental collaborative handling and effectively controlling the further spread of risks.

[0207] In this embodiment, the specific operation steps of step S5 are as follows:

[0208] Step S5-1: Collect data during the handling process

[0209] Data related to risk events: Collect the starting time and location of the occurrence of risk events from the multi-source data acquisition module and records of each link, and detailedly record the types and evolution processes of risk events. For example, if it is a risk of personnel gathering, it is necessary to record the starting position of the gathering, the change in the number of people, the spread of the gathering area, etc.; for the risk of equipment anomalies, information such as the occurrence time of equipment failures, equipment names, failure types, and the impact of failures on surrounding equipment and business processes should be recorded.

[0210] Resource scheduling data: Obtain the initial resource allocation plan generated by the resource scheduling optimization algorithm, as well as the resource scheduling records adjusted due to various situations during the emergency handling process; including the deployment time, quantity, destination of resources such as security check channels, ground handling vehicles, emergency handling units, and medical rescue units, as well as the actual arrival time and time of being put into use, etc., for analyzing the response delay of resource scheduling.

[0211] Data on the execution of handling strategies: Collect the specific operation and effect data executed by each department according to the handling strategies; such as the number of prohibited items seized by security personnel under different inspection intensities, the response measures taken by the emergency handling unit and the effectiveness of risk control, data on the treatment process and treatment effect of the medical rescue unit on the wounded, etc., in order to evaluate the effectiveness of the handling strategies.

[0212] Data related to model parameter adjustment: Record the relevant information on the adjustment of the dynamic risk assessment model parameters due to the emergence of new situations or the update of risk assessment during the handling process; including the names of the adjusted parameters, the values before and after adjustment, the reasons and times of adjustment, etc., to fully present the record of the adaptive adjustment of model parameters.

[0213] Step S5-2: Generate a spatio-temporal evolution map of risk events

[0214] Data sorting and annotation: Sort the data related to risk events collected in chronological order, and annotate key information such as the status of risk events, the scope of influence, the locations of relevant personnel and equipment at each time node; for example, for flight conflict risks, annotate information such as the planned location, actual location, and adjusted flight path of each flight at different time points.

[0215] Atlas drawing: Using geographic information system (GIS) technology and time series analysis tools, based on the airport map, visualize the development process of risk events in the spatial and temporal dimensions; in the atlas, use different colors, lines, icons and other elements to represent different types of risk events, the movement trajectories of relevant personnel and equipment, and the changes in the diffusion scope and impact area of risks over time; for example, use red lines to represent the diffusion path of the personnel gathering area, and use green icons to represent emergency equipment that has reached the designated location and been put into disposal work.

[0216] Step S5-3: Conduct analysis of resource scheduling response delay

[0217] Calculate the response delay time: According to the time when the resource allocation instruction is issued and the time when the resource actually arrives and is put into use in the resource scheduling data, calculate the response delay time of each resource unit; for example, for ground handling vehicles, subtract the time when the scheduling instruction is issued from the time when it actually arrives at the designated operation location and is ready to perform tasks to obtain the response delay time of the vehicle.

[0218] Analyze the reasons for the delay: Combine the actual operation situation of the airport, such as traffic congestion, equipment failure, poor communication among personnel, etc., and conduct a detailed analysis of the reasons for the response delay of each resource unit.

[0219] Summarize the delay rules: Conduct a summary analysis of the response delay times and reasons of all resource units to find out the general rules and main influencing factors of resource scheduling response delay.

[0220] Step S5-4: Carry out evaluation of the effectiveness of disposal strategies

[0221] Determine evaluation indicators: According to different risk types and disposal objectives, determine a series of indicators for evaluating the effectiveness of disposal strategies; for example, for the risk of carrying contraband, the evaluation indicators can include the seizure rate and missed inspection rate of contraband; for the risk of personnel gathering, the evaluation indicators can be the evacuation speed of personnel, the evacuation completion time, the incidence of secondary gathering, etc.; for the risk of equipment anomalies, the evaluation indicators can be the equipment repair time, the failure recurrence rate, the degree of recovery of the business process, etc.

[0222] Calculate the values of evaluation indicators: Based on the collected data on the implementation of disposal strategies, calculate the values of various evaluation indicators; for example, calculate the detection rate of prohibited items by counting the actual number of prohibited items seized during the security check and the total number of items to be inspected; calculate the evacuation speed and evacuation completion time according to the personnel evacuation records.

[0223] Construct an evaluation matrix: Organize the values of various evaluation indicators for different risk types into a matrix form; the rows of the matrix represent different risk types, and the columns represent the corresponding evaluation indicators; through this evaluation matrix, the implementation effects of disposal strategies under different risk types can be intuitively compared, and the effectiveness of disposal strategies can be comprehensively evaluated.

[0224] Step S5-5: Update the parameters of the risk assessment model

[0225] Analyze the data and determine the adjustment direction: Based on the information reflected in the spatio-temporal evolution map of risk events, the analysis of resource scheduling response delay, and the evaluation matrix of the effectiveness of disposal strategies, analyze in which aspects the current risk assessment model deviates from the actual situation; if it is evaluated that the assessment of a certain equipment anomaly risk is inaccurate, the parameters related to the equipment need to be adjusted; determine the parameters to be adjusted and the adjustment direction (increase or decrease the parameter value).

[0226] Parameter adjustment and verification: Adjust the parameters of the risk assessment model according to the determined adjustment direction and amplitude; then, use historical data or simulation data to verify the adjusted model to ensure that the adjusted model can more accurately reflect the actual risk situation; if the verification result is not ideal, further analyze the reasons and fine-tune the parameters until the model meets the accuracy requirements.

[0227] Step S5-6: Store the optimized scheduling strategy

[0228] Strategy sorting and summarization: Combine the analysis of resource scheduling response delay and the evaluation results of the effectiveness of disposal strategies to comprehensively sort and summarize the scheduling strategies formed during this emergency disposal process; extract the effective parts of the strategies, such as the efficient resource allocation methods in specific risk scenarios, the collaborative cooperation modes between various resource units, etc., and reflect on and improve the existing problems.

[0229] Store in the strategy knowledge base: Store the optimized scheduling strategy in the strategy knowledge base in a standardized format; during the storage process, annotate the strategy in detail, including information such as the applicable risk types, application scenarios, advantages and limitations of the strategy, etc., so that it can be quickly retrieved and called when encountering similar risk events in the future, providing more effective decision-making support for airport security management.

[0230] In this embodiment, the extraction of abnormal behavior feature vectors adopts:

[0231] Abnormal Posture Recognition Algorithm Based on Skeleton Key Point Detection:

[0232] Video Frame Preprocessing: Extract video frames from the input video stream data at a fixed frame rate; perform grayscale processing on each extracted frame image to convert the color image into a grayscale image to simplify subsequent calculations; then, use Gaussian filtering to smooth the image, remove noise interference in the image, improve the image quality, and provide a clearer image for subsequent skeleton key point detection;

[0233] Skeleton Key Point Detection: Apply a deep learning-based skeleton key point detection model, such as the OpenPose model; input the preprocessed image into the model, and the model predicts the coordinate positions of each key point on the human skeleton through a series of operations such as convolution and pooling; these key points usually include the head, shoulders, elbows, wrists, hips, knees, and ankles, etc. Different models may detect different numbers and types of key points; each key point represents its position in the image in the form of two-dimensional coordinates (x, y);

[0234] Posture Feature Extraction: Calculate the relative position relationship and angle information between various parts of the human body based on the detected skeleton key point coordinates as posture features; these posture features can reflect the posture state of the human body and are an important basis for judging abnormal postures;

[0235] Abnormal Posture Recognition: Compare the extracted posture features with the pre-set normal posture feature range; the normal posture feature range can be obtained by analyzing and statistically processing a large amount of video data of normal behaviors; if the posture features exceed the normal range, it is determined as an abnormal posture; encode the identified abnormal posture information as a part of the feature vector for subsequent construction of the abnormal behavior feature vector;

[0236] Analyze the Conflict Probability of Personnel Movement Trajectories Using the Optical Flow Method:

[0237] Optical Flow Calculation: For consecutive video frames, use the optical flow method (such as the Lucas-Kanade optical flow method or the Horn-Schunck optical flow method) to calculate the motion vector of each pixel point between adjacent two frames, that is, the optical flow; the optical flow reflects the motion information of objects in the image, and the movement trajectory of personnel in the video frame can be tracked through the optical flow; taking the Lucas-Kanade optical flow method as an example, this method assumes that within a small neighborhood, the motion of the object is consistent, and calculates the optical flow by solving a linear equation system;

[0238] Trajectory Tracking: Track the trajectories of people in the video based on the calculated optical flow; starting from the first frame of the video, determine the positions of people in different frames by matching the optical flow information of the same person in adjacent frames, so as to obtain the movement trajectories of people; to improve the tracking accuracy, other information such as the color features and shape features of people can be combined to optimize the tracking process;

[0239] Conflict Probability Calculation: Analyze the crossing and proximity of different people's movement trajectories and calculate the conflict probability; when two or more trajectories are too close or cross in space, it indicates the possibility of a person conflict; set certain distance thresholds and angle thresholds to determine whether the trajectories conflict; for example, if the minimum distance between two trajectories is less than the set distance threshold and their included angle is within a certain range, it is considered that these two trajectories may generate a conflict, and calculate the conflict probability; the conflict probability can be quantitatively calculated according to factors such as the crossing degree and proximity time of the trajectories;

[0240] Feature Vector Encoding: Encode the calculated conflict probability information as part of the feature vector; the higher the conflict probability, the larger the corresponding value in the feature vector, to highlight the conflict risk existing in the people's movement trajectories and provide an important basis for subsequent abnormal behavior analysis;

[0241] Enhance the detection sensitivity of suspicious items in hidden areas through the attention mechanism:

[0242] Region Division and Feature Extraction: Divide the video frames into multiple sub-regions, including the clearly visible region and the hidden regions where hidden items may exist (such as inside the luggage, parts blocked by people's clothes, etc.); for each sub-region, use a convolutional neural network (CNN) to extract its features; CNN automatically learns the feature representation of the image through operations such as convolutional layers and pooling layers to obtain the feature vector of each sub-region;

[0243] Application of the Attention Mechanism: Introduce the attention mechanism to assign different weights to the feature vectors of different sub-regions; the attention mechanism can make the model pay more attention to the features of the hidden regions; by calculating the similarity between the feature vector of each sub-region and a learnable attention vector, obtain the attention weight of each sub-region; the higher the similarity, the greater the attention weight obtained by the corresponding sub-region, indicating that the model pays more attention to this region;

[0244] Feature Fusion and Enhancement: Fuse the feature vectors of the sub-regions with assigned attention weights to obtain the enhanced feature vector of the entire video frame; during the fusion process, the features of the hidden regions contribute more to the enhanced feature vector because they obtain higher attention weights, thus strengthening the feature representation of the hidden regions;

[0245] Suspicious object detection: Input the enhanced feature vector into the trained suspicious object detection model, which can be a deep learning-based object detection model such as Faster R-CNN, YOLO, etc.; the model determines whether there is a suspicious object in the video frame according to the information in the enhanced feature vector; if a suspicious object is detected, relevant information (such as the category and location of the suspicious object) is encoded as part of the feature vector to further enrich the information of the abnormal behavior feature vector and improve the detection accuracy of abnormal behavior;

[0246] Through the above three specific operation steps, abnormal behavior information in different dimensions is extracted respectively, encoded as feature vectors, and finally combined to form a complete abnormal behavior feature vector, providing comprehensive and accurate feature information for subsequent risk assessment and abnormal behavior analysis.

[0247] In this embodiment, the generation process of the three-dimensional scheduling matrix includes:

[0248] Divide the airport area into honeycomb grid cells and calculate the resource demand density of each grid:

[0249] Area division:

[0250] According to the actual layout of the airport, including different functional areas such as the terminal building, runway, apron, parking lot, etc., use geographic information system (GIS) technology or coordinate-based division methods to divide the entire airport area into uniformly sized honeycomb grid cells; each grid cell has a specific geographical coordinate range, and the size setting needs to comprehensively consider factors such as airport scale, resource scheduling accuracy requirements, and computational complexity;

[0251] Data collection:

[0252] Collect various types of data related to resource requirements in each grid cell, including but not limited to the number of personnel, equipment distribution, types and frequencies of business activities, etc.; for example, in the waiting area grid of the terminal building, count the number of waiting passengers, service personnel, and the number of various service facilities (such as seats, charging facilities, etc.) in this grid; in the grid near the runway, record the distribution of runway operation equipment (such as snow plows, deicing vehicles, etc.) and the intensity of business activities in this area during flight takeoffs and landings;

[0253] Demand analysis:

[0254] For different types of resources (such as security personnel, ground handling vehicles, emergency response equipment, etc.), analyze the demand characteristics of business activities in each grid cell for them;

[0255] Density calculation:

[0256] According to the collected data and the results of demand analysis, calculate the demand density of each grid cell for different resources; the resource demand density can be expressed by the ratio of the demand for a specific resource to the area of the grid cell; by calculating the resource demand density of each grid, provide basic data support for subsequent resource scheduling;

[0257] Based on the digital twin platform, conduct simulation and deduction of the scheduling plan:

[0258] Digital twin model construction:

[0259] On the digital twin platform, construct a virtual model corresponding to the actual airport; this model should not only accurately reflect the geographical layout and facilities of the airport, but also simulate the operation logic of various business activities within the airport; each element in the model (such as buildings, vehicles, personnel, etc.) has corresponding attributes and behavioral characteristics to the actual situation, and real-time operation data of the actual airport is obtained through data interfaces to keep the virtual model synchronized with the actual situation;

[0260] Scheduling plan input:

[0261] Input the scheduling plan initially formulated based on the resource demand density into the digital twin model; the scheduling plan includes content such as personnel grouping, equipment configuration, and path planning;

[0262] Simulation operation:

[0263] Start the simulation operation on the digital twin platform, and according to the set time step, simulate the operation of the airport under this scheduling plan; during the simulation process, the model dynamically updates the status of each element according to the input scheduling plan and the real-time obtained airport operation data, such as the position movement of personnel, the working status of equipment, and the progress of business processes; at the same time, consider various possible interference factors, such as weather changes, equipment failures, and sudden personnel situations, to more realistically simulate the actual operation scenario;

[0264] Effect evaluation:

[0265] After the simulation operation ends, evaluate the implementation effect of the scheduling plan according to the preset evaluation indicators (such as response timeliness, resource utilization rate, disposal cost, etc.); for example, count the resource arrival time of each grid cell to evaluate the response timeliness; calculate the ratio of the actually used resources to the total resources to measure the resource utilization rate; count the human and material costs generated during the scheduling process to evaluate the disposal cost; through the quantitative analysis of these indicators, comprehensively understand the feasibility and effectiveness of the scheduling plan in actual operation;

[0266] Plan optimization:

[0267] Optimize and adjust the scheduling plan according to the evaluation results; if it is found that the resource response time in a certain area is too long, consider increasing the resource allocation in that area or adjusting the route planning; if the resource utilization rate is low, re-evaluate the rationality of resource allocation and optimize it; input the optimized scheduling plan into the digital twin model again for simulation and deduction, and iterate repeatedly until the optimal scheduling plan that meets the requirements is obtained;

[0268] Introduce an elastic resource pool mechanism to dynamically allocate standby resources to cope with sudden demands:

[0269] Resource pool formation:

[0270] From various existing resources at the airport, set aside a part as the elastic resource pool; these resources can include standby security personnel, ground service vehicles, emergency rescue equipment, etc., and their quantity and type are determined according to the historical operation data and risk assessment results of the airport; the resources in the elastic resource pool are usually on standby but can be put into use at any time;

[0271] Demand monitoring:

[0272] Establish a real-time monitoring system to continuously monitor the changes in resource demands in various areas of the airport; by integrating multi-source data, including real-time personnel positioning data, equipment operation status data, flight dynamic data, etc., promptly detect possible sudden resource demand situations;

[0273] Trigger mechanism:

[0274] Set the conditions for triggering the allocation of the elastic resource pool, generally taking a certain threshold that the resource demand exceeds the normal supply capacity as the trigger condition; once the trigger condition is met, the system automatically starts the resource allocation process;

[0275] Resource allocation:

[0276] According to the type and location of the sudden demand, quickly allocate the corresponding resources from the elastic resource pool; the allocation process follows certain priorities and strategies; at the same time, update the three-dimensional scheduling matrix and incorporate the allocated resource information into it to ensure the real-time adjustment and consistency of the scheduling plan;

[0277] Dynamic adjustment and recovery:

[0278] After resource allocation, continuously monitor the changes in demand, dynamically adjust the quantity and usage time of the allocated resources according to the actual demand; when the sudden demand is alleviated and the resource demand drops to a certain level, recover the excess resources from the site and re-incorporate them into the elastic resource pool for future use; through this dynamic adjustment and recovery mechanism, achieve the efficient utilization and flexible allocation of resources and improve the airport's ability to respond to emergencies;

[0279] Through the above three steps, a three-dimensional scheduling matrix containing information such as personnel grouping, equipment configuration, and path planning is generated to achieve scientific and efficient scheduling of airport resources to cope with various complex operating conditions and emergency demands.

[0280] In this embodiment, the airport security management method further includes:

[0281] Implement a hierarchical inspection strategy in the passenger security inspection link, and dynamically adjust the inspection intensity of the security inspection channels based on the risk level:

[0282] Risk level assessment:

[0283] According to the output result of the dynamic risk assessment model in step S2, obtain the risk level related to the passenger; this risk level comprehensively considers multi-source data such as the passenger's identity information, flight information, security equipment detection data, and personnel positioning tag data;

[0284] The risk level is divided into three levels: low, medium, and high, and each level corresponds to different risk characteristics and probabilities;

[0285] Setting inspection intensity standards:

[0286] For different risk levels, set corresponding security inspection channel inspection intensity standards in advance;

[0287] For passengers with a low risk level, the inspection intensity is relatively low;

[0288] For passengers with a medium risk level, the inspection intensity is moderate; in addition to the regular X-ray and metal detection, manual inspection of some items in the luggage is increased, and the sampling ratio is increased to 30%. The passengers are also questioned more carefully to understand information such as their travel purpose and carried items;

[0289] For passengers with a high risk level, implement high-intensity inspections; conduct a comprehensive and detailed manual inspection of their luggage, including opening all suspicious items for inspection; conduct a comprehensive metal detection of the passenger's body, and may even use more advanced detection technologies such as millimeter-wave body imaging inspection; at the same time, the inspection of passengers and their luggage achieves 100% full coverage, and more in-depth background investigations and inquiries may be carried out;

[0290] Real-time adjustment of inspection intensity:

[0291] At the security inspection site, the security inspection system obtains the risk level information of passengers in real time; according to the risk level, automatically adjusts the inspection process and resource allocation of the security inspection channels to achieve dynamic adjustment of the inspection intensity;

[0292] For example, when the system identifies a certain passenger as a high-risk level, it immediately issues a prompt to the staff at the security checkpoint, assigns more security personnel to this checkpoint, and increases manual inspection equipment and technical means to ensure a comprehensive and detailed inspection of the passenger and their luggage;

[0293] For groups of passengers passing through security continuously, the security system dynamically adjusts the inspection intensity of each checkpoint according to the overall risk level. For example, during a certain period, if the risk levels of passengers from a specific flight are generally high, the security system will automatically assign the passengers of this flight to checkpoints with a higher inspection intensity and correspondingly increase the resource allocation for these checkpoints;

[0294] Establish a prediction model for equipment health, trigger maintenance warnings in advance and adjust spare parts inventory:

[0295] Data collection:

[0296] Collect operation data of various equipment, including basic information of the equipment, real-time operation parameters, historical maintenance records, and equipment failure records;

[0297] These data come from multiple data sources such as sensors built into the equipment, monitoring systems, and maintenance management systems;

[0298] Feature engineering:

[0299] Clean and preprocess the collected data, remove noise data, outliers, and duplicate data to ensure the quality and consistency of the data;

[0300] Extract features related to equipment health from the original data; extract features such as maintenance interval time and maintenance cost from historical maintenance records; determine features such as the frequency and severity of faults based on failure records;

[0301] Normalize the extracted features, convert feature values in different ranges and magnitudes to the same numerical interval for easy model training and comparison;

[0302] Model selection and training:

[0303] According to the characteristics of the equipment and data features, select a suitable prediction model, such as support vector machine (SVM) based on machine learning, Random Forest, or long short-term memory network (LSTM) based on deep learning, etc.;

[0304] Divide the processed data into a training set and a test set, usually in a ratio of 70%-30% or 80%-20%; use the training set to train the selected model, and by continuously adjusting the model parameters, enable the model to learn the relationship between equipment operation data and health;

[0305] During the training process, methods such as cross-validation are used to evaluate the performance of the model, and the model parameters with the best performance are selected;

[0306] Model evaluation and optimization:

[0307] Use the test set to evaluate the trained model, and metrics such as accuracy, recall, mean squared error (MSE), etc. are used to measure the prediction accuracy of the model;

[0308] If the model performance does not meet the requirements, analyze the reasons and optimize the model; possible optimization methods include adjusting the model structure, adding more features, using ensemble learning methods, etc.; for example, for the LSTM model, you can try increasing the number of hidden layers or adjusting the number of hidden units to improve the model's expressive ability;

[0309] Health prediction and maintenance warning:

[0310] Input the real-time collected device operation data into the optimized model to predict the health of the device; the model outputs a value or status representing the health of the device;

[0311] Set a health threshold, and when the predicted device health is lower than a certain threshold, trigger a maintenance warning;

[0312] At the same time, based on the device health prediction results and historical fault data, analyze the possible fault types and probabilities, and prepare the corresponding spare parts in advance for maintenance work;

[0313] Spare parts inventory adjustment:

[0314] Adjust the spare parts inventory according to the maintenance warning and fault analysis results, combined with the current spare parts inventory situation;

[0315] If it is predicted that a certain device may have a certain fault and the spare parts inventory required for this fault is insufficient, the system automatically generates a spare parts purchase order to notify the purchasing department to replenish the spare parts in a timely manner; at the same time, dynamically manage the spare parts inventory and update the inventory quantity and location information in real time;

[0316] Regularly review the usage of spare parts and the inventory turnover rate, and optimize the spare parts inventory strategy according to the actual situation to ensure that the inventory cost is minimized while meeting the device maintenance requirements;

[0317] Conduct contingency plan deduction for major risk scenarios through the digital twin system:

[0318] Definition and modeling of major risk scenarios:

[0319] Identify the major risk scenarios that the airport may face, such as severe meteorological disasters, major equipment failures, etc.;

[0320] In the digital twin system, a detailed virtual model is constructed for each major risk scenario; the model not only includes the physical facilities and layout of the airport, but also needs to simulate various factors during the occurrence of risks, such as the intensity and impact range of meteorological disasters, the specific manifestations of equipment failures, etc.; at the same time, consider the impact of risks on personnel, equipment, flight operations, etc. within the airport; for example, when simulating a hurricane scenario, the model should reflect the wind force, path of the hurricane, and its impact on the terminal building structure, apron equipment, and flight takeoffs and landings;

[0321] Formulation of response plans:

[0322] According to different major risk scenarios, corresponding response plans are formulated; the content of the plans includes the emergency response process, the division of responsibilities of each department, the resource allocation plan, the evacuation routes of personnel, etc.

[0323] Convert the response plan into instructions and parameters recognizable by the digital twin system, including the dispatching instructions for personnel and equipment, the action time nodes, the quantity of resource allocation, etc.;

[0324] Setting of initial conditions:

[0325] In the digital twin system, initial conditions are set for the deduction of each major risk scenario; these initial conditions include specific parameters such as the time, location, and intensity of the risk occurrence, as well as the operating status of the airport before the risk occurs, such as the number of flights, passenger distribution, equipment operating conditions, etc.; for example, when simulating a runway lighting system failure scenario, set the specific time of the failure to be just before a certain flight lands, the flight queue situation on the runway at that time, and the operating status of surrounding equipment, etc.;

[0326] Implementation of plan deduction:

[0327] Start the deduction of the response plan in the digital twin system; the system simulates a series of events and response actions after the risk occurs according to the set initial conditions and response plan;

[0328] During the deduction process, the digital twin system updates the status of the airport virtual model in real time, including the action trajectories of personnel, the changes in the operating status of equipment, the adjustments of flights, etc.; for example, when simulating the evacuation of personnel, the system dynamically displays the evacuation process of personnel within the airport according to the set evacuation route and the movement speed of personnel;

[0329] Record the key data during the deduction process, such as the response time, resource consumption, number of casualties, flight delay duration, etc., and these data will be used to evaluate the effectiveness of the response plan;

[0330] Effect evaluation and plan optimization:

[0331] Evaluate the effectiveness of the disposal plan based on the data recorded in the deduction; use a series of evaluation indicators, such as the casualty rate, the degree of property damage, the time for flights to resume normal operation, etc., to measure the effectiveness of the plan in dealing with major risk scenarios;

[0332] Analyze the evaluation results to identify the problems and deficiencies in the disposal plan; for example, if it is found that the personnel evacuation time is too long, it may be that the evacuation route is unreasonable or the guiding measures are not in place; if the resource consumption is too large, the resource allocation plan may need to be optimized;

[0333] Based on the analysis results, optimize and adjust the disposal plan; re-deduce the optimized plan in the digital twin system and iterate repeatedly until the disposal plan can achieve good results, effectively cope with major risk scenarios, and reduce losses and impacts;

[0334] Plan storage and update:

[0335] Store the optimized disposal plan in the plan library of the digital twin system for quick access when actually facing major risks;

[0336] As the airport develops, the operating conditions change, and new risk factors emerge, regularly re-evaluate and model major risk scenarios, update the corresponding disposal plans, and conduct re-deduction and verification in the digital twin system to ensure the effectiveness and adaptability of the plans;

[0337] An airport safety management system, which uses the airport safety management method to manage airport safety.

[0338] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An airport safety management method, characterized in that: The following steps are involved: S1. Real-time acquisition of airport operation data through multi-source heterogeneous data acquisition modules, including: video surveillance data, millimeter-wave radar data, security equipment detection data, flight dynamic data, personnel location tag data and environmental sensor data; S2. Establish a dynamic risk assessment model, perform feature fusion processing on the input data, generate a three-dimensional risk heat map and output risk level and risk type identification; the risk types at least include the risk of carrying prohibited items, the risk of gathering of people, the risk of equipment abnormality, the risk of flight conflict and the risk of sudden environment. The dynamic risk assessment model adopts: The modified spatiotemporal convolutional neural network M-STCNN processes video stream data and extracts abnormal behavior feature vectors; assuming that the video stream data is ,In M-STCNN, after the convolution layer operation, the feature map The calculation formula is: ,in, Represents a location at a specific spatial and temporal index The feature map elements at Respectively represent the index of the feature map in the spatial and temporal dimensions, is the size of the convolution kernel, is the index of the convolution kernel in the first direction of the spatial dimension, is the index of the convolution kernel in the second direction of the spatial dimension, is the index of the convolution kernel in the time dimension, Represents the input video stream data exist The value at position, is the convolution kernel weight, is the bias term; the temporal and spatial features in the video stream data are extracted through this formula, and then the abnormal behavior feature vector is obtained; The modified graph neural network M-GNN constructs the topological graph of personnel-equipment-environment and identifies the potential risk propagation path; ,in, is a collection of nodes, is a set of edges; nodes The eigenvector of , in M-GNN, the node The updated feature vector The calculation formula is: ,in, Is a node The set of neighbor nodes of Is a node and The edge weights between is the weight matrix, is a nonlinear activation function, is the bias term; Representation Node The node feature vector is updated through this formula to identify the potential risk propagation path; Based on the fuzzy logic decision engine, the global risk coefficient is calculated by integrating the risk indicators of each dimension. Suppose the risk indicator set of each dimension is , the weight of each risk indicator is , global risk factor The calculation formula is: ,in, Represents the number of risk indicators, and the global risk coefficient is calculated through this formula; Abnormal behavior feature vector extraction uses: Abnormal posture recognition algorithm based on skeleton key point detection; The optical flow method is used to analyze the conflict probability of personnel movement trajectories; Enhance the detection sensitivity of suspicious objects in hidden areas through the attention mechanism; S3. Triggering a hierarchical response mechanism based on risk level and type, generating a resource allocation plan through a resource scheduling optimization algorithm, and sending scheduling instructions to security inspection channels, ground handling vehicles, emergency response units, and medical rescue units; S4. During the emergency response phase, the multi-terminal collaborative platform monitors the response effect in real time. When the risk diffusion coefficient exceeds the preset threshold, the cross-departmental collaborative response agreement is initiated and the resource allocation weight is dynamically adjusted; S5. Generate a disposal process traceability report, update the risk assessment model parameters, and store the optimized scheduling strategy in the strategy knowledge base.

2. An airport security management method according to claim 1, characterized in that: The data acquisition module in step S1 includes: Multispectral camera arrays deployed at key points in the terminal to collect data on personnel behavior and baggage status; A distributed meteorological sensor group deployed in the runway area monitors visibility, wind speed and pavement friction coefficient in real time; The Beidou positioning terminal integrated in the shuttle bus and service vehicles obtains real-time location and operation trajectory data.

3. The airport security management method according to claim 1, characterized in that: In step S3, the resource scheduling optimization algorithm includes: Establish a multi-objective optimization function, taking into account response timeliness, resource utilization and disposal cost; suppose response timeliness is , the resource utilization is The disposal cost is , multi-objective optimization function for: ,in, , , is the weight coefficient, which is used to balance the relationship between different objectives, and ; The improved genetic algorithm M-GA is used to solve the optimal scheduling scheme, in which the fitness function includes the risk attenuation rate, resource coverage and collaborative operation efficiency index; the risk attenuation rate is set to , the resource coverage is , the collaborative work efficiency index is , the fitness function Fitness is: ,in, , , is the weight coefficient, which is used to adjust the importance of different indicators in the fitness function, and ; The output includes a three-dimensional scheduling matrix of personnel grouping, equipment configuration, and path planning.

4. The airport security management method according to claim 1, characterized in that: In step S4, the cross-departmental collaborative disposal agreement includes: Build a blockchain-based emergency command chain to record the response time and disposal effect of each unit; When a runway incursion is detected, an augmented reality situation map is simultaneously pushed to the air traffic control tower, ground control center and security department; In medical rescue scenarios, a rescue plan is automatically generated that includes the optimal path, casualty classification, and material requirements.

5. The airport security management method according to claim 1, characterized in that: In step S5, the disposal process traceability report includes: Spatiotemporal evolution of risk events; Resource scheduling response delay analysis; Disposal strategy effectiveness assessment matrix; Model parameter adaptive adjustment records.

6. The airport security management method according to claim 3, characterized in that: The generation process of the three-dimensional scheduling matrix includes: Divide the airport area into honeycomb grid cells and calculate the resource demand density of each grid; Conduct scheduling simulation based on the digital twin platform; Introduce an elastic resource pool mechanism to dynamically allocate resources to meet sudden demands.

7. The airport security management method according to claim 1, characterized in that: Also includes: Implement a graded inspection strategy in the passenger security check process and dynamically adjust the inspection intensity of the security check channel based on the risk level; Establish equipment health prediction models to trigger maintenance warnings and adjust spare parts inventory in advance; Use the digital twin system to simulate contingency plans for major risk scenarios.

8. An airport security management system, characterized by: The airport security management system is equipped with the airport security management method described in any one of claims 1-7 to manage airport security.

Images