A confined space safety control method and system based on edge computing

Abstract

The present invention discloses a method and system for safety management and control of confined spaces based on edge computing, and relates to the field of data management technology. It includes deploying edge computing nodes, cameras and gas sensors, and collecting data in real time; building a dynamic perception network, adjusting the sensor network topology in real time, and realizing load balancing and backup mechanisms between edge computing nodes; the edge computing node identifies the location of personnel based on the received data; and comprehensively predicts the movement of personnel positions and environmental factors to perform real-time risk level assessment. The present invention shows significant advantages in many aspects such as reaction speed, warning accuracy, system reliability, data processing capabilities and energy efficiency. These improvements not only improve the safety management level of high-risk environments such as thermal power plants, but also may provide a safer and more reliable working environment for staff.

Description

Technical Field

[0001] The present invention relates to the field of data management technology, and in particular to a method and system for confined space security management and control based on edge computing. Background Art

[0002] As a complex industrial environment, thermal power plants face many challenges in terms of safety management and control. Existing safety monitoring systems have many shortcomings in dealing with the complex environment unique to thermal power plants. First, it is difficult for traditional video monitoring systems to achieve comprehensive coverage. Due to the large number of equipment and complex structures in thermal power plants, there are obstructions between various types of equipment, which makes it difficult for cameras to achieve 360-degree coverage of the power plant without blind spots. The existence of such blind spots makes it impossible for certain work surfaces to be effectively controlled by video, thus creating major safety hazards. Secondly, existing gas detection systems generally use fixed sensors, which lack flexibility and dynamic adaptability.

[0003] In addition, the traditional centralized data processing model often encounters problems such as response delays and network congestion when faced with large-scale real-time data streams. This architecture is difficult to meet the needs of rapid decision-making in the complex environment of thermal power plants, especially in emergency situations, where delays may lead to catastrophic consequences.

[0004] Comparative documents include: CN111262644A A transmission method for improving the safety performance of edge computing systems; CN115359404A A food safety supervision system based on edge computing. Summary of the invention

[0005] In view of the problems existing in the existing confined space safety management and control methods, the present invention is proposed.

[0006] Therefore, the problem to be solved by the present invention is how to achieve safe control of confined spaces.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0008] In the first aspect, an embodiment of the present invention provides a method for security management and control of confined spaces based on edge computing, which includes deploying edge computing nodes, cameras and gas sensors, and collecting data in real time; building a dynamic perception network, adjusting the sensor network topology in real time, and realizing load balancing and backup mechanisms among edge computing nodes; the edge computing nodes identify the location of personnel based on the received data; and conducting real-time risk level assessment based on personnel location movement prediction and environmental factors.

[0009] As a preferred solution of the confined space safety control method based on edge computing described in the present invention, the deployment of the edge computing nodes and gas sensors includes: the formula for the number of camera arrangements is as follows:

[0010] ,

[0011] in, is the number of cameras required, is the total area, is the coverage area of ​​a single camera, is the overlap coefficient; the formula for the number of gas sensors to be arranged is:

[0012] ,

[0013] in, is the coverage area of ​​a single sensor, is the number of high-risk areas; the edge computing node arrangement is: assuming that the coverage range of each sensor is a spherical area with a coverage radius of , divide the entire three-dimensional space into several sub-areas, count the number of sensors in each sub-area, and calculate the distance between sensors , set an edge node at the center of each sub-region, and if the distance between two edge nodes is less than , then merge these two nodes into one edge node.

[0014] As a preferred solution of the confined space security management and control method based on edge computing described in the present invention, the construction of a dynamic perception network and real-time adjustment of the sensor network topology include: using smart devices carried by staff as mobile sensor nodes; edge computing nodes continuously scan the surrounding environment, automatically identify and access mobile sensor nodes within the range; real-time calculation and update of the network topology map, including the relative position relationship of fixed sensors, cameras, mobile sensor nodes and edge computing nodes.

[0015] As a preferred solution of the confined space security management method based on edge computing described in the present invention, the real-time calculation and updating of the network topology map includes the following steps: calculating the Euclidean distance between any two nodes in the fixed sensor, camera, mobile sensor node and edge computing node in the sub-area; using the adjacency matrix D to represent the topology map, wherein, Represents the Euclidean distance between node i and node j: ,in, is the position of node i, is the location of node j, the initial topology graph ,in is a node set, is a set of edges, passing the distance threshold Determine the existence of an edge: For mobile sensor nodes , update the coordinates according to the real-time collected location information, where nk is the number of mobile sensor nodes; for each time step t, calculate the distance between all nodes and update the adjacency matrix :According to the new adjacency matrix Update topology map ,in, is the edge set at time step t, is the distance between node i and node j at time step t; edge computing node Continuously scan the surrounding environment, automatically identify and access mobile sensor nodes within range , the coverage radius of the edge computing node is ,like ,but Accessed In the managed network, ln is the number of edge computing nodes.

[0016] As a preferred solution of the confined space security management method based on edge computing described in the present invention, the edge computing node identifies the location of personnel based on the received data, including the following steps: integrating heterogeneous data from different types of sensors, synchronizing and aligning the data using timestamps and spatial information; identifying and tracking the location of personnel based on the location of mobile sensor nodes, and predicting the direction and speed of movement.

[0017] As a preferred solution of the confined space safety management method based on edge computing described in the present invention, wherein: the identification and tracking of personnel positions according to the positions of mobile sensor nodes, and the prediction of movement direction and speed include: using the GPS, WiFi or Bluetooth signals of the mobile sensor nodes for preliminary positioning, and constructing the movement trajectory of each staff member based on the time series position data; using the weighted moving average method to smooth the trajectory: smoothed position = (2P1 + P2) / 3, where P1 is the position at the previous moment and P2 is the position at the current moment; the movement direction is calculated as follows: calculate the direction angle of two adjacent position points: ,in, and are the horizontal and vertical coordinates of two adjacent position points; calculate the speed of two adjacent position points:

[0018] ,

[0019] in, are the sampling times of two adjacent position points respectively; considering the moving acceleration to predict the moving direction and speed of the person:

[0020] ,

[0021] ,

[0022] ,

[0023] in, is the prediction time interval, is the time interval between the current moment and the previous moment, is the current speed, is the velocity at the previous moment, is the predicted X-direction position, is the X-direction position at the current moment, is the predicted Y position, is the Y position at the current moment; according to the predefined area boundary where activities are allowed, check whether the predicted position is within the allowed area; if the predicted position exceeds the boundary, adjust the predicted position to the nearest boundary point.

[0024] As a preferred solution of the confined space safety management and control method based on edge computing described in the present invention, the comprehensive personnel position movement prediction and environmental factors are used to conduct real-time risk level assessment, including the following steps: using gas sensor data to monitor toxic gas concentrations and oxygen levels in real time; monitoring temperature and humidity change trends, and warning of possible heat stress or other environmental risks; and conducting risk assessment and warning based on trajectory prediction.

[0025] In the second aspect, an embodiment of the present invention provides a confined space safety management and control system based on edge computing, which includes: a data acquisition module, used to deploy edge computing nodes and gas sensors, and collect data in real time; a network management module, used to build a dynamic perception network, adjust the sensor network topology in real time, and realize load balancing and backup mechanism between edge computing nodes; a location identification module, used for edge computing nodes to identify personnel locations according to received data; a risk assessment module, used to comprehensively predict personnel location movement and environmental factors, and perform real-time risk level assessment.

[0026] In a third aspect, an embodiment of the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program instructions are executed by the processor, the steps of the confined space security management and control method based on edge computing as described in the first aspect of the present invention are implemented.

[0027] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium on which a computer program is stored, wherein: when the computer program instructions are executed by a processor, the steps of the confined space security management and control method based on edge computing as described in the first aspect of the present invention are implemented.

[0028] The beneficial effects of the present invention are as follows: by monitoring environmental factors such as toxic gases, oxygen levels, temperature and humidity, combined with personnel trajectory prediction, the present invention can timely discover potential environmental risks and issue early warnings, which is of great significance for preventing safety accidents caused by the environment; the confined space safety management and control method based on edge computing of the present invention shows significant advantages in many aspects such as reaction speed, warning accuracy, system reliability, data processing capabilities and energy efficiency. These improvements not only improve the safety management level of high-risk environments such as thermal power plants, but also may provide a safer and more reliable working environment for staff. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0030] Figure 1 The figure is a flow chart of the confined space security control method based on edge computing.

[0031] Figure 2 This is a structural diagram of the confined space security management and control system based on edge computing. DETAILED DESCRIPTION

[0032] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0033] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0034] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0035] Example 1

[0036] Reference Figure 1~Figure 2 , which is the first embodiment of the present invention, and provides a confined space security management method based on edge computing, including:

[0037] S1: Deploy edge computing nodes, cameras, and gas sensors, and collect data in real time.

[0038] Multiple edge computing nodes are set up in key operation areas, equipment maintenance areas and limited spaces. These nodes need to have strong data processing capabilities and network connection functions, and use integrated edge computing devices, combined with local storage, data processing and network connection functions, to reduce latency and improve data processing efficiency.

[0039] Various gas sensors (such as CO, CO2, O2 and combustible gas sensors) are arranged in different working areas to ensure the rapid detection of toxic and flammable gases in a limited space. Temperature and humidity sensors are arranged at corresponding locations to monitor environmental changes in real time and use wide-angle cameras and 360-degree rotating cameras to ensure coverage of key areas. At the same time, multiple small cameras are used for blind spot monitoring.

[0040] Specifically, the edge computing equipment and sensors for thermal power plants are arranged as follows:

[0041] The arrangement of cameras needs to take into account the shooting distance and clarity, the importance of the area, and the obstruction of the equipment.

[0042] As well as the 360-degree coverage requirement, the formula for the number of cameras to be deployed is as follows:

[0043] ,

[0044] in, is the number of cameras required, is the total area, is the coverage area of ​​a single camera, is the overlap coefficient, which is generally 0.2-0.3.

[0045] The gas sensor arrangement includes:

[0046] Multi-gas detectors are installed at the entrance of limited space, around major equipment, in poorly ventilated areas, etc. to detect O2, CO, H2S or combustible gases, etc. The sampling frequency is 10-30 seconds / time;

[0047] Fixed gas detectors are installed in high-risk areas such as boilers, steam turbines, chemical storage areas, etc.

[0048] The formula for the number of gas sensors to be arranged is:

[0049] ,

[0050] in, is the coverage area of ​​a single sensor, The number of high-risk areas.

[0051] The edge computing nodes are arranged as follows:

[0052] Assume that the coverage range of each sensor is a spherical area with a coverage radius of , divide the entire three-dimensional space into several sub-areas, count the number of sensors in each sub-area, and calculate the distance between sensors , set an edge node at the center of each sub-region, and if the distance between two edge nodes is less than , then merge the two nodes into one edge node, and calculate the most suitable number of edge nodes through the above steps.

[0053] After ensuring that the equipment covers all key areas, including confined spaces, equipment maintenance areas, and potentially hazardous areas, start all equipment and begin real-time data collection.

[0054] Edge computing nodes obtain gas concentration, environmental parameters, video data and personnel location information in real time by connecting with sensors and cameras. Each node should have a certain local data caching capability to prevent data loss during network interruptions.

[0055] S2: Build a dynamic perception network, adjust the sensor network topology in real time, and implement load balancing and backup mechanisms among edge computing nodes.

[0056] Specifically, smart devices carried by staff, such as dedicated terminals, are used as mobile sensor nodes. Edge computing nodes continuously scan the surrounding environment, automatically identify and access mobile sensor nodes within the range, and calculate and update the network topology map in real time, including the relative position relationship of fixed sensors, cameras, mobile sensor nodes and edge computing nodes.

[0057] Preferably, real-time calculation and updating of the network topology map includes the following steps:

[0058] Calculate the Euclidean distance between any two nodes among fixed sensors, cameras, mobile sensor nodes, and edge computing nodes in the sub-area;

[0059] The topological graph is represented by the adjacency matrix D, where Represents the Euclidean distance between node i and node j: ,in, is the position of node i, is the location of node j, the initial topology graph ,in is a node set, is a set of edges, passing the distance threshold Determine the existence of an edge: .

[0060] For mobile sensor nodes , update its coordinates according to the real-time collected location information, where nk is the number of mobile sensor nodes;

[0061] For each time step t, calculate the distance between all nodes and update the adjacency matrix :According to the new adjacency matrix Update topology map ,in, is the edge set at time step t, is the distance between node i and node j at time step t.

[0062] Edge computing nodes Continuously scan the surrounding environment, automatically identify and access mobile sensor nodes within range , the coverage radius of the edge computing node is ,like ,but Accessed In the managed network, ln is the number of edge computing nodes.

[0063] Furthermore, according to the distribution of personnel and sensor data load, the allocation of computing resources is dynamically adjusted to calculate the load of each edge computing node:

[0064] ,

[0065] If the load of an edge computing node exceeds the preset threshold , then some mobile sensor nodes are reallocated to the neighboring edge computing nodes.

[0066] Furthermore, in densely populated or high-risk areas, the processing capacity of nearby edge nodes can be temporarily increased, and backup computing units can be started when necessary.

[0067] Furthermore, load balancing and backup mechanisms are implemented between edge computing nodes to ensure that a single point of failure will not cause the entire system to crash. Specifically: the CPU, memory, and network load of each node are monitored in real time, and tasks are assigned using polling, minimum number of connections, or weighted algorithms, automatically assigning new tasks to nodes with lighter loads; maintaining a certain number of standby nodes, pre-configuring backup nodes to ensure rapid takeover of tasks from failed nodes, and regularly updating the configuration and data of backup nodes.

[0068] S3: The edge computing node receives data from cameras, gas sensors, and mobile sensor nodes, and identifies the location of people based on the data.

[0069] S3.1: Integrate heterogeneous data from different types of sensors and synchronize and align the data using timestamps and spatial information.

[0070] Specifically, the received data is preliminarily classified, such as video data, environmental data, location data, etc.; a unified timestamp is added to all received data, and the Network Time Protocol (NTP) is used to ensure the clock synchronization of all devices and nodes. For possible time delays, an interpolation algorithm is used for time alignment.

[0071] S3.2: Identify and track the location of people based on the location of mobile sensor nodes, and predict the movement direction and speed.

[0072] First, the GPS, WiFi or Bluetooth signals of the mobile sensor nodes are used for preliminary positioning. Based on the time series location data, the movement trajectory of each staff member is constructed, and the distance between two location points in the movement trajectory of each staff member is calculated. If the distance exceeds the preset threshold (such as 5 meters / second * sampling interval), it is regarded as an outlier and removed. After processing the data, the movement trajectory is constructed: the storage format is: {timestamp, X coordinate, Y coordinate, Z coordinate}.

[0073] Furthermore, the weighted moving average method is used to smooth the trajectory: smoothed position = (2P1 + P2) / 3, where P1 is the position at the previous moment and P2 is the position at the current moment.

[0074] The moving direction is calculated as follows:

[0075] Calculate the direction angle between two adjacent position points: ,in, and They are the horizontal and vertical coordinates of two adjacent points respectively, and the angle range is adjusted to 0-360 degrees according to the quadrant situation.

[0076] Calculate the speed of two adjacent positions:

[0077] ,

[0078] in, are the sampling times of two adjacent points respectively.

[0079] Considering the movement acceleration, predict the movement direction and speed of the person:

[0080] ,

[0081] ,

[0082] ,

[0083] in, is the prediction time interval, is the time interval between the current moment and the previous moment, is the current speed, is the velocity at the previous moment, is the predicted X-direction position, is the X-direction position at the current moment, is the predicted Y position, It is the Y position at the current moment.

[0084] According to the predefined area boundary where activities are allowed, check whether the predicted position is within the allowed area. If the predicted position exceeds the boundary, adjust the predicted position to the nearest boundary point.

[0085] Optionally, the predicted paths of all persons are calculated to check whether there is a path intersection. If there is an intersection, the time to reach the intersection is calculated. If the difference in arrival time between the two persons is less than a preset value (such as 2 seconds), a collision warning is issued.

[0086] S4: Comprehensive personnel location movement prediction and environmental factors to conduct real-time risk level assessment.

[0087] S4.1: Use gas sensor data to monitor toxic gas concentrations and oxygen levels in real time; monitor temperature and humidity trends, and warn of possible heat stress or other environmental risks.

[0088] Specifically, set a high-frequency sampling rate, such as collecting data every 5-10 seconds, clean the collected data, analyze gas concentration and oxygen level, temperature and humidity data in real time, and analyze the changing trend; comprehensively analyze gas concentration, oxygen level, temperature and humidity data, and assess environmental risks. The specific operations are as follows:

[0089] Extract key features, such as the mean, maximum, minimum, standard deviation of gas concentration, daily change rate of temperature and humidity, etc. Use weighted average method or principal component analysis PCA to fuse multiple sensor data together, and use multivariate regression analysis to calculate environmental risk scores:

[0090] ,

[0091] in, Scoring environmental risks, is a constant term, Score the risk for the ith independent variable

[0092] The degree of influence, is the i-th input feature for predicting risk score, and m is the total number of input features.

[0093] Based on the risk score, the current environment is divided into different risk levels (such as low, medium, and high risk) and marked as dangerous.

[0094] S4.2: Conduct risk assessment and early warning based on trajectory prediction.

[0095] Use the trajectory prediction formula in S3.2 to calculate the predicted trajectory in the future (e.g., 5 minutes), compare the predicted trajectory with the map with danger marking, and determine whether the predicted trajectory passes through or enters the high-risk area. That is, the predicted trajectory of the person is composed of position information at multiple times. The high-risk area is defined as a polygonal area surrounded by boundary coordinates. Whether the person enters the high-risk area is determined by judging whether the trajectory point is within the polygon. The ray method or other geometric algorithms can be used to determine whether the point is inside the polygon.

[0096] If the predicted trajectory passes through a high-risk area, the time spent in the high-risk area is calculated:

[0097]

[0098] in, is the starting time in the high-risk area, The departure time in the high-risk area;

[0099] If the person has been in a high-risk area, the actual stay time is accumulated.

[0100] Estimated time required for safe evacuation:

[0101] ,

[0102] in, is the average moving speed of people in the predicted future period, is the distance from the current position to the nearest safe area.

[0103] Furthermore, the warning lead time is calculated:

[0104] ,

[0105] If the expected stay time in a high-risk area is longer than the evacuation time, the personnel shall evacuate at least Timely warnings are issued to ensure that personnel have enough time to evacuate high-risk areas. If it is less than the preset minimum warning time (such as 2 minutes), the minimum warning time is used.

[0106] Calculate the comprehensive risk index based on the environmental risk score and the predicted risk exposure time:

[0107] ,

[0108] in, is the critical exposure time (e.g. 15 minutes), is the basic risk factor.

[0109] Set the risk level according to the comprehensive risk index: For example: Level 1 risk: ; Secondary risk ; Level 3 risk: .

[0110] And generate warning information based on the risk level and warning time. For example, when the risk is level one, an emergency evacuation alarm is triggered; when the risk is level two, a warning message is sent to the staff's mobile device; when the risk is level three, no warning is required.

[0111] Among them, the warning information includes: risk level, expected exposure time, recommended evacuation route, etc.

[0112] S5: Establish a data access permission system to ensure that sensitive information is visible only to authorized personnel.

[0113] Specifically, the data in the system is classified, sensitive information (such as personal information, location data, operation logs, etc.) and general information are identified, and sensitive information is marked to facilitate identification and processing in subsequent processing.

[0114] Based on job responsibilities and needs, define access rights for different roles, such as administrators, operators, maintenance personnel, etc., to ensure that each user can only access and operate data related to their work, avoiding unnecessary permission allocation; and through permission control, ensure that sensitive information is only visible to authorized personnel.

[0115] Furthermore, the present embodiment also provides a confined space safety management and control system based on edge computing, including a data acquisition module for deploying edge computing nodes and gas sensors and collecting data in real time;

[0116] The network management module is used to build a dynamic sensing network, adjust the sensor network topology in real time, and implement load balancing and backup mechanisms among edge computing nodes;

[0117] A location identification module is used by the edge computing node to identify the location of a person based on the received data;

[0118] The risk assessment module is used to conduct real-time risk level assessment based on personnel location movement prediction and environmental factors.

[0119] This embodiment also provides a computer device, which is suitable for the case of a confined space security management and control method based on edge computing, including a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the confined space security management and control method based on edge computing proposed in the above embodiment.

[0120] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covered on the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0121] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for implementing the confined space security management and control based on edge computing as proposed in the above embodiment is implemented.

[0122] In summary, the present invention can timely discover potential environmental risks and issue early warnings by monitoring environmental factors such as toxic gases, oxygen levels, temperature and humidity, combined with personnel trajectory prediction, which is of great significance for preventing safety accidents caused by the environment; the confined space safety management method based on edge computing of the present invention shows significant advantages in many aspects such as reaction speed, warning accuracy, system reliability, data processing capabilities and energy efficiency. These improvements not only improve the safety management level of high-risk environments such as thermal power plants, but also may provide staff with a safer and more reliable working environment.

[0123] Example 2

[0124] Referring to Table 1, which is the second embodiment of the present invention, this embodiment provides a confined space security management method based on edge computing. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through simulation experiments.

[0125] This experiment selected a large thermal power plant as the test site. The total area of ​​the power plant is about 54,000 square meters, including multiple confined space areas, such as boiler rooms, turbine rooms, chemical storage areas, etc. The experimental team first conducted a detailed survey of the entire site and determined the specific location and scope of key operating areas, equipment maintenance areas and confined spaces.

[0126] In order to determine the number and location of edge computing nodes, the team divided the entire space into multiple sub-areas and finally determined the optimal layout of 25 edge computing nodes by calculating the sensor coverage and density. These nodes are all high-performance integrated edge computing devices with powerful data processing capabilities and network connection functions.

[0127] After the system was deployed, the team conducted a 30-day comprehensive test. During the test, 50 staff members were equipped with dedicated smart terminal devices to participate in the experiment as mobile sensor nodes. These devices can transmit location information and environmental data in real time.

[0128] During the experiment, the system continuously collects, processes and analyzes data. The edge computing nodes receive data from fixed sensors, cameras and mobile sensor nodes in real time and perform the following operations:

[0129] The system updates the network topology every 5 seconds and calculates the distance and connection relationship between nodes in real time. When a new mobile sensor node is detected entering the coverage area, the system automatically incorporates it into the management network.

[0130] The system calculates the load of each edge node every 10 seconds. When the load of a node exceeds 80%, it automatically transfers some tasks to nodes with lighter loads. At the same time, the five backup nodes are always on standby, ready to take over tasks from nodes that may fail at any time.

[0131] All data are attached with a unified timestamp, and NTP is used to ensure the clock synchronization of each device. For data with time delay, the system uses a linear interpolation algorithm for time alignment.

[0132] The system updates the worker's location information once a second, uses a weighted moving average method to smooth the trajectory, and predicts the movement direction and speed within the next 5 minutes based on historical data.

[0133] The system comprehensively analyses environmental data such as gas concentration, oxygen level, temperature and humidity, and combines it with personnel location and predicted trajectory to conduct a comprehensive risk level assessment every 30 seconds.

[0134] Based on the preset permission system, the system strictly controls access to sensitive information by users of different levels to ensure data security.

[0135] Throughout the experiment, the team simulated a variety of possible dangerous situations, such as gas leakage, reduced oxygen content, and sharp temperature increases, to test the system's responsiveness and accuracy.

[0136] Table 1 below shows the key indicators of system performance and effect during the experiment:

[0137] Table 1 Part of the system performance and effect data

[0138] ,

[0139] As can be seen from the table above, the average response time has been greatly reduced from 2.5 seconds in the traditional system to 0.3 seconds, an increase of 88%. This significant improvement is mainly due to the localized data processing capabilities of edge computing nodes, which greatly reduces the delay in data transmission and processing. In a high-risk environment such as a thermal power plant, the shortened response time may mean the protection of life safety, which is of great significance for preventing and responding to emergencies.

[0140] The accuracy of risk warning increased from 82% to 97%, an improvement of 18.29%. This improvement is due to the system's comprehensive use of multi-source data, including real-time information from fixed sensors, cameras, and mobile sensor nodes. Through advanced data fusion and analysis algorithms, the accuracy of risk assessment has been greatly improved. Higher warning accuracy means that staff can trust the system's alarms more and take timely safety measures.

[0141] In addition, the false alarm rate and missed alarm rate were reduced from 15% and 8% to 3% and 1%, respectively, a reduction of 80% and 87.5%. The significant improvement of these two indicators fully demonstrates the excellent performance of this system in data processing and risk assessment. The low false alarm rate reduces unnecessary production interruptions and personnel evacuations, improving work efficiency; while the low missed alarm rate ensures that almost all potential hazards can be discovered and handled in a timely manner, greatly improving the overall safety level.

[0142] It can be seen that the present invention not only improves the safety management level of high-risk environments such as thermal power plants, but also provides a safer and more reliable working environment for workers. In addition, the high efficiency and low energy consumption characteristics of the system also bring potential economic benefits to enterprises.

[0143] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A confined space security management and control method based on edge computing, characterized by: include, Deploy edge computing nodes, cameras, and gas sensors, and collect data in real time; Build a dynamic perception network, adjust the sensor network topology in real time, and implement load balancing and backup mechanisms between edge computing nodes; The edge computing node identifies the location of the person based on the received data; Comprehensive personnel movement prediction and environmental factors to conduct real-time risk level assessment; The deployment of the edge computing nodes and gas sensors includes: The formula for the number of cameras to be arranged is as follows: ， in, is the number of cameras required, is the total area, is the coverage area of ​​a single camera, is the overlap coefficient; The formula for the number of gas sensors to be arranged is: ， in, is the coverage area of ​​a single sensor, is the number of high-risk areas; The edge computing nodes are arranged as follows: Assume that the coverage range of each sensor is a spherical area with a coverage radius of , divide the entire three-dimensional space into several sub-areas, count the number of sensors in each sub-area, and calculate the distance between sensors , set an edge node at the center of each sub-region, and if the distance between two edge nodes is less than , then merge these two nodes into one edge node; The real-time calculation and updating of the network topology diagram comprises the following steps: Calculate the Euclidean distance between any two nodes among fixed sensors, cameras, mobile sensor nodes, and edge computing nodes in the sub-area; The topological graph is represented by the adjacency matrix D, where Represents the Euclidean distance between node i and node j: ,in, is the position of node i, is the location of node j, the initial topology graph ,in is a node set, is a set of edges, passing the distance threshold Determine the existence of an edge: ; For mobile sensor nodes , update the coordinates according to the real-time collected location information, where nk is the number of mobile sensor nodes; For each time step t, calculate the distance between all nodes and update the adjacency matrix :According to the new adjacency matrix Update topology map : ,in, is the edge set at time step t, is the distance between node i and node j at time step t; Edge computing nodes Continuously scan the surrounding environment, automatically identify and access mobile sensor nodes within range , the coverage radius of the edge computing node is ,like ,but Accessed In the managed network, ln is the number of edge computing nodes; The identifying and tracking of personnel positions according to the positions of mobile sensor nodes and predicting the moving direction and speed include: Use GPS, WiFi or Bluetooth signals of mobile sensor nodes for preliminary positioning, and construct the movement trajectory of each staff member based on time series location data; Use the weighted moving average method to smooth the trajectory: smoothed position = (2P1 + P2) / 3, where P1 is the position at the previous moment and P2 is the position at the current moment; The moving direction is calculated as follows: Calculate the direction angle between two adjacent position points: ,in, and are the horizontal and vertical coordinates of two adjacent points respectively; Calculate the speed of two adjacent positions: ， in, are the sampling times of two adjacent position points respectively; Considering the movement acceleration, predict the movement direction and speed of the person: ， ， ， in, is the prediction time interval, is the time interval between the current moment and the previous moment, is the current speed, is the velocity at the previous moment, is the predicted X-direction position, is the X-direction position at the current moment, is the predicted Y position, is the Y position at the current moment; According to the predefined area boundary of allowed activities, check whether the predicted position is within the allowed area. If the predicted position exceeds the boundary, adjust the predicted position to the nearest boundary point; The comprehensive personnel position movement prediction and environmental factors to conduct real-time risk level assessment include the following steps: Use gas sensor data to monitor toxic gas concentrations and oxygen levels in real time; monitor temperature and humidity trends to warn of possible heat stress or other environmental risks; Risk assessment and early warning based on trajectory prediction; The risk assessment based on trajectory prediction includes: If the predicted trajectory passes through a high-risk area, the time spent in the high-risk area is calculated: ， in, is the starting time in the high-risk area, The departure time in the high-risk area; If the person has been in a high-risk area, the actual time spent there is accumulated; Estimated time required for safe evacuation: ， in, is the average moving speed of people in the predicted future period, is the distance from the current location to the nearest safe area; Calculate the warning lead time: ， If the expected stay time in a high-risk area is longer than the evacuation time, the personnel shall evacuate at least Timely warnings are issued to ensure that personnel have enough time to evacuate high-risk areas. If it is less than the preset minimum warning time, the minimum warning time will be used; Calculate the comprehensive risk index based on the environmental risk score and the predicted risk exposure time: ， in, is the critical exposure time, is the basic risk factor.

2. The method for safety management and control of confined spaces based on edge computing according to claim 1, characterized in that: The construction of a dynamic sensing network and real-time adjustment of the sensor network topology includes: Using smart devices carried by staff as mobile sensor nodes; The edge computing nodes continuously scan the surrounding environment and automatically identify and access mobile sensor nodes within range; Calculate and update the network topology in real time, including the relative position relationship of fixed sensors, cameras, mobile sensor nodes and edge computing nodes.

3. The method for safety management and control of confined spaces based on edge computing according to claim 2, characterized in that: The edge computing node identifies the location of a person based on the received data, including the following steps: Integrate heterogeneous data from different types of sensors, synchronize and align the data using timestamps and spatial information; Identify and track the location of people based on the location of mobile sensor nodes, and predict the direction and speed of movement.

4. A confined space safety management and control system based on edge computing, based on the confined space safety management and control method based on edge computing according to any one of claims 1 to 3, characterized in that: Also includes, Data acquisition module, used to deploy edge computing nodes and gas sensors and collect data in real time; The network management module is used to build a dynamic sensing network, adjust the sensor network topology in real time, and implement load balancing and backup mechanisms among edge computing nodes; A location identification module is used by the edge computing node to identify the location of a person based on the received data; The risk assessment module is used to conduct real-time risk level assessment based on personnel location movement prediction and environmental factors.

5. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the confined space safety management and control method based on edge computing described in any one of claims 1 to 3 are implemented.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the confined space safety management and control method based on edge computing described in any one of claims 1 to 3 are implemented.

Images