A nuclear power plant area dynamic electronic fence early warning system and early warning method

Abstract

The present application relates to the technical field of nuclear power safety monitoring, and particularly relates to a nuclear power plant area dynamic electronic fence early warning system and method. The system comprises a 5G positioning base station array, a personnel terminal, an equipment electronic tag, an edge computing node and a central control platform. The edge computing node calculates a basic warning radius R0=α·L+β·S according to the equipment physical size L and the risk level coefficient S, and dynamically generates a forbidden area, a warning area and a focus area; wherein the forbidden area radius R_no=R0, the warning area radius R_warn=k1·R0 (k1=1.5~2.0), and the focus area radius R_attention=k2·R0 (k2=2.5~3.0). When an unauthorized person enters the focus area, the edge node starts trajectory recording and speed vector calculation, and if it is determined that the motion trend continuously points to the forbidden area and the expected arrival time is less than a threshold value, early warning is given. The present application realizes tripartite linkage early warning of personnel authority, equipment state and distance, and solves the problem that a fixed fence cannot adapt to dynamic focus on important equipment areas in a nuclear power plant.

Description

Technical Field

[0001] This invention relates to the field of nuclear power safety monitoring technology, specifically to a dynamic electronic fence early warning system and method for nuclear power plant areas. Background Technology

[0002] As a special industrial sector with high safety requirements, nuclear power plants contain numerous high-risk equipment areas (such as reactor pressure vessels, steam generators, nuclear fuel transfer devices, and high-voltage electrical equipment). Personnel accidentally entering these areas could trigger serious safety accidents. Traditional nuclear power safety management relies primarily on physical isolation (fences, warning lines) and manual patrols, which suffers from poor management flexibility, delayed response, and the inability to track personnel movements in real time.

[0003] With the application of 5G technology in smart nuclear power plant construction, the high bandwidth, low latency, and wide connectivity of 5G, combined with the positioning capabilities of personnel terminals (5G mobile phones, explosion-proof tablets, etc.), provide a technological foundation for accurate personnel trajectory tracking and dynamic early warning. However, existing technologies still have the following shortcomings: 1. Existing electronic fences are mostly fixed. For example, CN202311412917.8 discloses a 5G communication interaction system based on electronic fences, mainly used for monitoring pastoral boundaries. Existing electronic fence solutions primarily target general scenarios such as livestock farming, pet care, and asset management, monitoring livestock or goods, and do not address the core requirement of managing the identity and access rights of certified personnel in high-risk nuclear power environments. 2. The relative distance determination between personnel and equipment is not precise enough. For example, the edge intelligent security alarm method disclosed in CN202410037220.5, while involving the identification of distances between people and vehicles at construction sites, mainly relies on preset fixed warning distances and fails to dynamically adjust safety thresholds based on equipment type, personnel qualifications, and operational status. 3. There is a lack of dynamic binding of personnel identity and access rights. Existing proximity alarm systems are mostly based on pure distance triggering, without fully considering the matching relationship between personnel identity and equipment security levels. 4. Insufficient adaptability to the special environment of nuclear power plants. Nuclear power plant areas contain a large number of steel structures and thick concrete walls, which significantly interfere with traditional indoor positioning technologies such as UWB and Bluetooth, resulting in poor positioning stability. Summary of the Invention

[0004] This invention provides a dynamic electronic fence early warning system and method for nuclear power plant areas, which solves the technical problems of existing electronic fences in high-risk nuclear power plant areas, such as poor control flexibility, insufficient positioning accuracy, and inability to dynamically adapt to personnel permissions.

[0005] The technical solution of the present invention is as follows: This invention proposes a dynamic electronic fence early warning system for nuclear power plant areas. The system includes a positioning base station array, personnel terminals, equipment electronic tags, and an early warning module. The positioning base station array receives positioning signals reported by personnel terminals and transmits them to the early warning module. Personnel terminals transmit personnel positioning signals to the positioning base station array. Equipment electronic tags store information about the monitored equipment. The early warning module reads personnel information and equipment information from the equipment electronic tags, determines whether personnel have entered the restricted area by judging the distance between personnel and equipment, and issues an early warning to personnel. The early warning module includes edge computing nodes. The edge computing nodes calculate the basic warning radius based on the equipment information from the equipment electronic tags and dynamically generate three levels of restricted areas: prohibited area, warning area, and attention area.

[0006] In some embodiments, the positioning base station array is a 5G positioning base station array, and the edge computing node is directly connected to the 5G positioning base station array to calculate the three-dimensional position coordinates of the personnel terminal based on the signal arrival time difference or multiple angle of arrival fusion algorithm; the personnel terminal is an explosion-proof mobile phone or explosion-proof tablet, and the personnel terminal logs in through an account, and the personnel uploads location information in real time. The personnel terminal is equipped with an audible and visual alarm module; the equipment electronic tag is set on the nuclear power equipment to be monitored, and the equipment electronic tag stores the equipment number, equipment type, physical size feature value L, risk level coefficient S and current operating status.

[0007] In some embodiments, the early warning module includes an edge computing node, a central control platform, and an early warning output module. The edge computing node is used to calculate the relative distance between the personnel terminal and the device in real time and dynamically adjust the warning zone according to the real-time operating status of the device. The central control platform includes a personnel permission database, a device management module, a map engine, and an early warning strategy engine. The early warning output module includes a local alarm module for the personnel terminal, a field audible and visual alarm module, and the control platform.

[0008] In some embodiments, the personnel permission database of the central control platform is used to store staff information and verify the login information of personnel terminals; the equipment management module is used to store equipment information; the map engine loads the electronic map of the nuclear power plant area, generates and displays the warning zone and personnel location; the early warning strategy engine is used to set the early warning trigger conditions; when the personnel terminal local alarm of the early warning output module receives an early warning information, it transmits the early warning information to the audible and visual alarm module of the personnel terminal; when the on-site audible and visual alarm module receives an early warning information, it controls the on-site audible and visual alarm to flash; when the control platform receives an early warning information, it pushes SMS and pop-up windows to the personnel terminal; the early warning module has a group density alarm, which is triggered when N or more personnel carrying terminals approach the equipment in the same warning zone at the same time, where N is a preset parameter; the edge computing node introduces the group density coefficient ρ to participate in the warning zone calculation, and the actual warning radius R'=R0 / ρ; the group density coefficient ρ is determined according to the relationship between the personnel density d in the warning zone and the preset density benchmark threshold d0, specifically including: when d≤d0, ρ=1, when d>d0, the calculation formula of ρ is as follows: ρ = 1 + λ·ln(d / d0) Where λ is the sensitivity adjustment factor, which is a preset parameter; The early warning module uses 3D bounding box calculations for irregularly shaped nuclear power equipment. For irregularly shaped nuclear power equipment, the edge computing nodes use the minimum bounding box algorithm to calculate the equivalent feature size L of the equipment, as shown in the following formula: L=γ·√S_h+η·H_v Where γ is the weighting coefficient for the horizontal projected area, S_h is the horizontal projected area of ​​the device's minimum bounding box, η is the weighting coefficient for the vertical projected height, and H_v is the vertical projected height.

[0009] This invention proposes a dynamic electronic fence early warning method for nuclear power plant areas, the method comprising: Step 1: When staff enter the nuclear power plant area, they log in to the personnel terminal, verify the personnel's identity on the central control platform of the early warning module, obtain the personnel's authorization level and list of operable equipment, and send the personnel information to the edge computing node of the early warning module; Step 2: The edge computing node obtains the device's attributes through the device's electronic tag or the device management module of the central control platform, and calculates the device's basic warning radius R0; Step 3: Edge computing nodes generate three types of warning zones for personnel based on the basic warning radius R0; Step 4: The edge computing node calculates the Euclidean distance D between the person and the device in real time based on the personnel positioning signal transmitted by the positioning base station array, and retrieves the personnel's permission information; Step 5: Set the early warning trigger conditions in the early warning strategy engine of the central control platform. The edge computing node determines whether to trigger an early warning based on the Euclidean distance D between the personnel and the equipment and the warning zone. When it is determined that the personnel do not have the current equipment operation permission and the corresponding early warning conditions are met, the corresponding level of early warning is triggered. Step Six: The early warning output module outputs early warnings according to the warning level.

[0010] In some embodiments, the device attributes in step two include static attributes and dynamic attributes. Static attributes include device type, physical size characteristic value L, and baseline risk level coefficient S. Dynamic attributes include current operating status and whether it is in a high-risk operating condition. The calculation method for the basic warning radius R0 is shown in the following formula: R0 = α*L + β*S Where L is the physical size characteristic value of the equipment, S is the equipment risk level coefficient, and α and β are weighting coefficients; when the equipment operating status changes, the equipment management module automatically updates R0 and reissues it to the edge computing node.

[0011] In some embodiments, in step four, the edge computing node calculates the Euclidean distance D between the person and the device based on a signal time difference of arrival or a multiple angle of arrival fusion algorithm. The calculation method for the Euclidean distance D between the person and the device is shown in the following formula: Where, x p ,y p , z p Let x be the coordinates of the personnel's location. d , y d , z d These are the coordinates of the device's location.

[0012] In some embodiments, step three dynamically generates three types of warning zones based on personnel identity and permissions, device status, and real-time distance. The warning zones include a restricted area, a warning area, and a focus area. The radius R_no of the restricted area is equal to the basic warning radius R0. Once an unauthorized person enters the restricted area, the highest level alarm is triggered. The radius R_warn of the warning area is k1·R0. When an unauthorized person enters the warning area, a primary warning is triggered, where k1 is between 1.5 and 2.0. The radius R_attention of the focus area is k2·R0. When an unauthorized person enters the focus area, the personnel terminal records the trajectory and predicts the trend, where k2 is between 2.5 and 3.0.

[0013] In some embodiments, the specific conditions for the edge computing node to determine and trigger the early warning in step five are as follows: The Euclidean distance D between personnel and equipment is less than or equal to the radius R_no of the restricted area, and personnel do not have operating privileges for the monitored equipment; The radius of the restricted area R_no < the Euclidean distance D between personnel and equipment ≤ the radius of the warning area R_warn, and the time personnel stay in the area exceeds the preset threshold; Based on the predicted movement trajectory trend of the personnel terminal, the personnel's velocity vector points to the restricted area and the expected arrival time is less than the preset threshold T_pre; If a person's terminal enters a manually designated temporary warning zone and meets any of the above conditions, an early warning will be triggered.

[0014] In some embodiments, the warning levels in step six are divided into Level 1, Level 2, and Level 3 warnings. For Level 1 warnings, the warning output module causes the personnel terminal to vibrate and displays a pop-up notification on the personnel terminal screen. For Level 2 warnings, the warning output module issues a voice warning on the personnel terminal and drives the on-site audible and visual alarms to flash. For Level 3 warnings, the warning output module displays an evacuation command on the personnel terminal, and the management platform of the warning output module displays a pop-up notification to the safety officer to link the video surveillance to lock the personnel's trajectory.

[0015] The implementation of this invention has the following beneficial effects: 1. This invention addresses the specific needs of the high-risk environment of nuclear power plants by making personnel access a prerequisite for triggering early warnings, achieving precise protection where "authorized personnel do not interfere, and unauthorized personnel are strictly controlled." Unlike existing electronic fence solutions (such as CN120614567A and CN119964304A), which rely solely on location as the triggering basis, this invention effectively avoids false alarms during normal operations by authorized personnel, significantly improving the system's practicality and reliability.

[0016] 2. This invention proposes a dynamic electronic fence early warning system and method for nuclear power plant areas. This system achieves a three-dimensional linkage between personnel, equipment, and permissions, overcoming the limitations of traditional alarm triggering solely based on distance. It incorporates personnel identity and permissions into the decision-making model, ensuring that authorized personnel (such as maintenance workers) can legally approach equipment without triggering false alarms, while unauthorized personnel immediately trigger an alarm if they cross the boundary, balancing safety and efficiency. Compared to technologies like UWB and Bluetooth, this system's 5G-based positioning scheme offers stronger resistance to metal interference and penetration capabilities, adapting to the complex environment of nuclear power plants.

[0017] 3. This invention achieves quantitative calculation of the basic warning radius through formula calculation, and dynamically adjusts the warning zone radius according to the equipment operating status (normal / maintenance / abnormal), solving the problem that existing fixed fences cannot adapt to the multiple operating conditions of nuclear power equipment, and realizing precise protection of "one policy for one piece of equipment and one policy for one operating condition".

[0018] 4. This invention proposes a dynamic electronic fence early warning system and method for nuclear power plant areas. The system realizes multi-level early warning and trend prediction. The edge computing node issues an early warning before personnel actually enter the restricted area through trajectory prediction, leaving time margin for intervention and significantly improving the active protection capability of high-risk areas of nuclear power plants.

[0019] 5. This invention introduces a crowd density coefficient to dynamically correct the warning zone in scenarios with multiple people gathering, which solves the problem of false alarms or missed alarms caused by multiple people gathering in traditional electronic fences in scenarios such as nuclear power plant maintenance and other collaborative operations, and fills the technical gap of existing electronic fence solutions in scenarios with multiple people. Attached Figure Description

[0020] Figure 1 This is a system architecture diagram of a dynamic electronic fence early warning system for nuclear power plant areas proposed in an embodiment of the present invention; Figure 2 This is a schematic diagram illustrating the generation of a dynamic warning zone in a dynamic electronic fence early warning system for nuclear power plant areas, as proposed in an embodiment of the present invention. Figure 3 This is a flowchart illustrating the early warning judgment process of a dynamic electronic fence early warning method for nuclear power plant areas, as proposed in an embodiment of the present invention. Figure 4 This is a schematic diagram illustrating the relationship between the minimum bounding box of an equipment and the warning zone in a dynamic electronic fence early warning system for nuclear power plant areas, as proposed in an embodiment of the present invention. Detailed Implementation

[0021] The technical solution of the present invention will be clearly and completely described below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] like Figures 1 to 4 As shown, this invention proposes a dynamic electronic fence early warning system for nuclear power plant areas, including a positioning base station array, personnel terminals, equipment electronic tags, and an early warning module.

[0023] The positioning base station array is a 5G positioning base station array, deployed within the nuclear power plant area to cover personnel activity areas and the vicinity of equipment. It is used to receive positioning signals reported by personnel terminals and transmit them to the early warning module. The edge computing nodes of the early warning module are directly connected to the 5G positioning base station array and calculate the three-dimensional position coordinates of personnel terminals based on signal time difference or multi-angle of arrival fusion algorithms.

[0024] The personnel terminal transmits the staff's location signal to the positioning base station array. The personnel terminal includes 5G explosion-proof mobile phones, 5G explosion-proof tablets, etc. After logging into their personal accounts, they can upload location information in real time and have an audible and visual alarm module.

[0025] The electronic tags for the equipment are set on the nuclear power equipment that needs to be monitored. They store information such as equipment ID, equipment type, baseline safety distance, physical size characteristic value L, risk level coefficient S, and current operating status, and support real-time clock (RTC) synchronization.

[0026] The early warning module includes edge computing nodes, a central control platform, and an early warning output module. The edge computing nodes are deployed at the edge of the factory area and are directly connected to the 5G positioning base station array. They are used to calculate the relative distance between personnel terminals and equipment in real time and generate dynamic warning zones.

[0027] The central control platform includes a personnel permission database, an equipment management module, a map engine, and an early warning strategy engine.

[0028] The personnel access control database of the central management platform is used to store staff information and verify login information of personnel terminals. The database stores staff identity information, job qualifications, and authorization levels. When personnel log in via 5G terminals, identity verification is performed. After successful verification, the personnel ID and its corresponding whitelist of operable devices are synchronized to the edge computing nodes as a prerequisite for triggering edge-side alerts.

[0029] The equipment management module stores static ledger information of nuclear power equipment (including equipment physical dimensions L, baseline risk level S, etc.) and obtains the dynamic operating status (normal, under maintenance, abnormal) of the equipment in real time through communication with the equipment's electronic tags. When the equipment status changes, the equipment management module automatically updates the equipment's basic warning radius R0 and sends it to the edge computing nodes.

[0030] As the core of spatial computing and visualization, the map engine loads the electronic map of the nuclear power plant area and accurately maps the 5G positioning coordinates with the map spatial coordinates; it receives the warning zone parameters issued by the edge nodes and renders the dynamic fence layer in real time; it provides a visual zone delineation tool, allowing managers to select and generate temporary warning zones on the map; and when an alarm is triggered, it automatically locks and highlights the personnel's location and historical trajectory.

[0031] The early warning strategy engine stores an early warning rule base, allowing administrators to configure core parameters such as the radius coefficients (k1, k2) of each warning zone, the dwell time threshold T_warn, and the early warning lead time T_pre. When an early warning event is received from an edge computing node, the engine determines the early warning output method (including terminal pop-ups, voice broadcasts, on-site audible and visual alarms, and SMS notifications from the management end) based on preset strategies.

[0032] The early warning output module includes a local alarm module for personnel terminals, a field audible and visual alarm module, and a management and control platform. When the local alarm module for personnel terminals receives an early warning message, it transmits the message to the audible and visual alarm module on the personnel terminal. When the field audible and visual alarm module receives an early warning message, it controls the on-site audible and visual alarms to flash. When the management and control platform receives an early warning message, it sends SMS messages and pop-up windows to the personnel terminals.

[0033] The early warning module is equipped with a crowd density alarm. When N or more people carrying terminals approach the device simultaneously within the same warning zone, where N is a preset parameter, the alarm will trigger an alarm. The edge computing node incorporates a crowd density coefficient ρ into the warning zone calculation, with the actual warning radius R' = R0 / ρ. The crowd density coefficient ρ is determined based on the relationship between the personnel density d within the warning zone and the preset density threshold d0. Specifically, when d ≤ d0, ρ = 1; when d > d0, the calculation formula for ρ is as follows: ρ = 1 + λ·ln(d / d0) Wherein, λ is the sensitivity adjustment factor, which is a preset parameter; the number of people threshold for the early warning module can be customized by the administrator in the range of 2-10 people through the central control platform.

[0034] The early warning module uses 3D bounding box calculations for irregularly shaped nuclear power equipment. For irregularly shaped nuclear power equipment, the edge computing nodes use the minimum bounding box algorithm to calculate the equivalent feature size L of the equipment, as shown in the following formula: L=γ·√S_h+η·H_v Where γ is the weighting coefficient for the horizontal projected area, S_h is the horizontal projected area of ​​the device's minimum bounding box, η is the weighting coefficient for the vertical projected height, and H_v is the vertical projected height.

[0035] This invention proposes a dynamic electronic fence early warning method for nuclear power plant areas, comprising the following steps: Step 1: When staff enter the nuclear power plant area, they log in to the personnel terminal using their personal account. The central control platform of the early warning module verifies the personnel's identity (such as name, position, safety training validity period, and authorization level), obtains the personnel's authorization level and list of operable equipment, and distributes the personnel permission list to the edge computing nodes.

[0036] Step 2: Equipment Information Collection and Safe Distance Initialization Edge computing nodes acquire the static and dynamic attributes of each device through electronic tags or device management modules. Static attributes include device type, physical dimensions, and baseline risk level. Dynamic attributes include current operating status and whether the device is in a high-risk condition (normal, maintenance, or abnormal). The edge computing node calculates the basic warning radius R0 of the device, using the following formula: R0 = α*L + β*S Where L is the physical dimensional characteristic value of the equipment (such as the diagonal length of the minimum bounding box), S is the equipment risk level coefficient (including levels 1-5), and α and β are weighting coefficients, which are set by the management personnel according to the safety specifications.

[0037] Step 3: Generate three types of personnel alert zones based on the basic alert radius R0. Specifically, this involves dynamically generating three types of alert zones based on personnel identity and permissions, device status, and real-time distance: a restricted area, a warning area, and a watch area. The radius of the restricted area is R_no = R0; unauthorized personnel entering the restricted area immediately trigger the highest-level alarm. The radius of the warning zone is R_warn = k1·R0 (k1>1, default value is 1.5-2.0). Unauthorized personnel entering this zone will trigger a primary warning. The radius of the attention zone, R_attention = k2·R0 (k2>k1, default value is 2.5-3.0), is set. When unauthorized personnel enter this zone, trajectory recording begins, and edge computing nodes simultaneously calculate the personnel's velocity vector to predict their movement trends. The attention zone trend prediction mechanism is as follows: When any personnel terminal (regardless of their permissions) enters the attention zone, the edge computing nodes automatically initiate continuous trajectory recording and motion state analysis for that person. Based on continuous position sampling, the edge computing nodes calculate the personnel's velocity vector (including speed and direction angle) in real time. During this process, the system logically integrates the prediction results with the personnel's permissions. If the person is authorized, only their movement is recorded for backend tracking, without triggering an alert; If the person is unauthorized, they are deemed to have a tendency to enter the restricted area if the following conditions are met simultaneously: (1) The direction angle of the personnel's movement continuously points towards the equipment in multiple consecutive sampling periods (the change in direction angle does not exceed a preset threshold, such as 15°). (2) Extrapolating from the current movement speed, the estimated time T_arrival for personnel to reach the boundary of the restricted area is less than the preset safety threshold T_pre (default 5 seconds, adjustable range 3-10 seconds).

[0038] When the system detects an intrusion trend, it triggers a level-two warning before the person actually enters the restricted area. Furthermore, edge computing nodes trigger warnings even before the person actually enters the restricted area, with the person's terminal issuing a voice prompt, achieving proactive protection from "post-incident alarm" to "pre-incident warning." The sampling period, sliding window length, orientation angle change threshold, and safety threshold T_pre are all customizable by administrators through the central control platform to adapt to the control needs of different devices and scenarios.

[0039] The aforementioned coefficients k1 and k2 can be dynamically adjusted according to manual instructions. For example, during equipment maintenance, the radius of the warning zone can be temporarily expanded to cover the work area, or the radius coefficient of the area of ​​concern can be increased when high-risk equipment is running.

[0040] Step 4: The edge computing node calculates the Euclidean distance D between the person and the device in real time based on the personnel positioning signal transmitted by the positioning base station array. Specifically, this includes: the 5G positioning base station array collecting the real-time position coordinates (x, y, y) of the personnel terminal. p , y p , z p The device's electronic tag or positioning base station synchronously collects the device's location coordinates (x). d , y d , z d Edge computing nodes calculate the Euclidean distance D between personnel and devices based on 5G signal time difference of arrival (TDOA) or multiple angle of arrival (AOA) fusion algorithms. The calculation method is shown in the following formula: Where, x p , y p , z p Let x be the coordinates of the personnel's location. d , y d , z d These are the coordinates of the device's location.

[0041] Step 5: Set the early warning trigger conditions within the early warning strategy engine of the central control platform. The edge computing node determines whether to trigger an early warning based on the Euclidean distance D between the personnel and the device and the warning zone. If it is determined that the personnel do not have operating permissions for the current device and the corresponding early warning conditions are met, the corresponding level of early warning is triggered. The edge computing node determines whether to trigger an early warning based on the following conditions: Condition A: The Euclidean distance D between the personnel and the equipment is less than or equal to the radius R_no of the restricted area, and the personnel do not have permission to operate the equipment; Condition B: The radius of the restricted area R_no < the Euclidean distance D between personnel and equipment ≤ the radius of the warning area R_warn, and the time that personnel stay in this area exceeds the preset threshold T_warn (e.g., 30 seconds); Condition C: Based on the motion trajectory prediction of the personnel terminal, the personnel's velocity vector points to the restricted area and the expected arrival time is less than the preset threshold T_pre (e.g., 5 seconds). Condition D: A temporary restricted area is manually designated (e.g., a certain area is selected through a control platform), and personnel enter the area via their terminals.

[0042] If any condition is met, the corresponding level of warning will be triggered.

[0043] Step Six: The warning output module outputs warnings according to the warning level, specifically including: For a Level 1 warning (prompt), the warning output module causes the personnel terminal to vibrate and displays a pop-up window on the personnel terminal screen to prompt: "You have entered the XX device's attention zone"; For Level 2 warnings, the warning output module issues a voice warning on the personnel terminal, which drives the on-site audible and visual alarm to flash, prompting "Do not approach". For Level 3 (emergency) alerts, the alert output module displays evacuation instructions on the personnel terminal, and the management platform of the alert output module pops up a window and notifies the safety officer to link video surveillance to lock the personnel's trajectory.

[0044] This invention proposes a dynamic electronic fence early warning method for nuclear power plant areas. Taking the protection around the reactor pressure vessel as an example, the specific early warning method is as follows: A positioning base station array is deployed around the reactor pressure vessel, consisting of 6 5G positioning base stations, covering a 50-meter radius around the vessel. An electronic tag is installed on the vessel body, with the equipment type "reactor pressure vessel" entered, the baseline risk level S=5 (highest level), and the physical dimension characteristic value L=8 meters (vessel diameter). The parameters α=0.5 and β=2 are set in formula (1), and the basic warning radius R0=0.5×8+2×5=14 meters is calculated according to the formula.

[0045] When an unauthorized inspector enters the reactor hall, their personnel terminal (i.e., a 5G explosion-proof mobile phone) reports their location in real time. The edge node calculates the distance to the container as D=25 meters, placing them within the attention zone (R_attention=2.5×14=35 meters), and the system initiates trajectory recording. When the personnel continue forward to D=13 meters (i.e., entering the restricted area), a level three warning is triggered. The 5G explosion-proof mobile phone forcibly displays a red warning and provides a voice prompt, "Danger! Please evacuate immediately." Simultaneously, a pop-up window in the control room displays the personnel's location, and the safety officer intervenes via walkie-talkie.

[0046] This invention proposes a dynamic electronic fence early warning method for nuclear power plant areas. Taking temporary maintenance work as an example, the specific early warning method is as follows: A steam generator needs maintenance. The safety officer selects the maintenance area (20 meters in diameter) on the map through the control platform, sets up a temporary warning zone, and sets up a list of personnel allowed to enter (5 people in the maintenance team). The system synchronizes this temporary warning zone to the edge nodes.

[0047] During maintenance, if non-team personnel bring terminals into the area, the edge node will identify that they are not authorized and immediately trigger a level 2 warning; if team personnel enter, the system will identify them as authorized personnel and will not trigger an alarm, but will record the entry and exit time for traceability.

[0048] This invention proposes a dynamic electronic fence early warning method for nuclear power plant areas. Taking movement trend prediction as an example, the specific early warning method is as follows: When a high-voltage electrical device is in operation, the basic warning radius R0 = 5 meters. Edge computing nodes calculate the velocity vector of a person based on their position over three consecutive sampling cycles (once every 0.5 seconds), predicting that they will enter the restricted area in 2 seconds. The system triggers a level-two warning when the person is actually 6 meters away (still within the warning zone). Upon receiving the warning, the person changes their direction of travel to avoid accidentally entering the restricted area.

[0049] The above embodiments merely illustrate several implementation methods of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this patent should be determined by the appended claims.

Claims

1. A dynamic electronic fence early warning system for nuclear power plant areas, characterized in that, The system includes a positioning base station array, a personnel terminal, an electronic tag for equipment, and an early warning module. The positioning base station array receives positioning signals reported by the personnel terminal and transmits them to the early warning module. The personnel terminal transmits the positioning signals of the personnel to the positioning base station array. The electronic tag for equipment stores information about the monitored equipment. The early warning module reads the personnel information and the equipment information from the electronic tag, determines whether the personnel have entered the warning zone by judging the distance between the personnel and the equipment, and issues an early warning to the personnel. The early warning module includes an edge computing node, which calculates a basic warning radius based on the device information of the device's electronic tag and dynamically generates three levels of warning zones: a restricted area, a warning area, and a focus area.

2. The dynamic electronic fence early warning system for nuclear power plant areas according to claim 1, characterized in that, The positioning base station array is a 5G positioning base station array. The edge computing node is directly connected to the 5G positioning base station array and calculates the three-dimensional position coordinates of the personnel terminal based on the signal time difference or multiple angle of arrival fusion algorithm. The personnel terminal is an explosion-proof mobile phone or explosion-proof tablet. The personnel terminal logs in through an account and uploads location information in real time. The personnel terminal is equipped with an audible and visual alarm module. The equipment electronic tag is set on the nuclear power equipment to be monitored. The equipment electronic tag stores the equipment number, equipment type, physical size characteristic value L, risk level coefficient S, and current operating status.

3. The dynamic electronic fence early warning system for nuclear power plant areas according to claim 2, characterized in that, The early warning module includes an edge computing node, a central control platform, and an early warning output module. The edge computing node is used to calculate the relative distance between personnel terminals and equipment in real time and dynamically adjust the warning zone according to the real-time operating status of the equipment. The central control platform includes a personnel permission database, an equipment management module, a map engine, and an early warning strategy engine. The early warning output module includes a local alarm module for personnel terminals, a field audible and visual alarm module, and a control platform.

4. A dynamic electronic fence early warning system for nuclear power plant areas according to claim 3, characterized in that, The central control platform's personnel access database stores staff information and verifies login information for personnel terminals. The equipment management module stores equipment information. The map engine loads an electronic map of the nuclear power plant area, generates and displays warning zones and personnel locations. The early warning strategy engine sets early warning trigger conditions. When the personnel terminal's local alarm receives an early warning message, the early warning output module transmits the message to the terminal's audible and visual alarm module. When the on-site audible and visual alarm module receives an early warning message, it controls the on-site audible and visual alarm to flash. When the control platform receives an early warning message, it pushes SMS messages and pop-ups to the personnel terminals. The early warning module has a group density alarm; when N or more personnel carrying terminals approach the equipment simultaneously within the same warning zone, N is a preset parameter. The edge computing node incorporates a group density coefficient ρ into the warning zone calculation, with the actual warning radius R' = R0 / ρ. The population density coefficient ρ is determined based on the relationship between the population density d within the warning zone and a preset density threshold d0. Specifically, when d ≤ d0, ρ = 1; when d > d0, the calculation formula for ρ is as follows: ρ = 1 + λ·ln(d / d0) Where λ is the sensitivity adjustment factor, which is a preset parameter; The early warning module uses three-dimensional bounding box calculations for irregularly shaped nuclear power equipment. For irregularly shaped nuclear power equipment, the edge calculation nodes use the minimum bounding box algorithm to calculate the equivalent feature size L of the equipment, as shown in the following formula: L=γ·√S_h+η·H_v Where γ is the weighting coefficient for the horizontal projected area, S_h is the horizontal projected area of ​​the device's minimum bounding box, η is the weighting coefficient for the vertical projected height, H_v is the vertical projected height; √ is the square root of the parameter S_h.

5. A method for early warning of dynamic electronic fences in nuclear power plant areas, wherein the method employs the system described in any one of claims 1-4, characterized in that, The method includes: Step 1: When staff enter the nuclear power plant area, they log in to the personnel terminal, verify the personnel's identity on the central control platform of the early warning module, obtain the personnel's authorization level and list of operable equipment, and send the personnel information to the edge computing node of the early warning module; Step 2: The edge computing node obtains the device's attributes through the device's electronic tag or the device management module of the central control platform, and calculates the device's basic warning radius R0; Step 3: Edge computing nodes generate three types of warning zones for personnel based on the basic warning radius R0; Step 4: The edge computing node calculates the Euclidean distance D between the person and the device in real time based on the personnel positioning signal transmitted by the positioning base station array, and retrieves the personnel's permission information; Step 5: Set the early warning trigger conditions in the early warning strategy engine of the central control platform. The edge computing node determines whether to trigger an early warning based on the Euclidean distance D between the personnel and the equipment and the warning zone. When it is determined that the personnel do not have the current equipment operation permission and the corresponding early warning conditions are met, the corresponding level of early warning is triggered. Step Six: The early warning output module outputs early warnings according to the warning level.

6. The method for early warning of a dynamic electronic fence in a nuclear power plant area according to claim 5, characterized in that, In step two, the equipment attributes include static and dynamic attributes. The static attributes include equipment type, physical dimension characteristic value L, and baseline risk level coefficient S. The dynamic attributes include current operating status and whether it is in a high-risk operating condition. The calculation method for the basic warning radius R0 is shown in the following formula: R0=a L+β S Where L is the physical size characteristic value of the equipment, S is the equipment risk level coefficient, and α and β are weighting coefficients; when the equipment operating status changes, the equipment management module automatically updates R0 and reissues it to the edge computing node.

7. A method for early warning of dynamic electronic fences in nuclear power plant areas according to claim 5, characterized in that, In step four, the edge computing node calculates the Euclidean distance D between the personnel and the equipment based on the signal time difference of arrival or multiple angle of arrival fusion algorithm. The calculation method for the Euclidean distance D between the personnel and the equipment is shown in the following formula: Where, x p ,y p , z p Let x be the coordinates of the personnel's location. d , y d , z d These are the coordinates of the device's location.

8. A method for early warning of dynamic electronic fences in nuclear power plant areas according to claim 6, characterized in that, In step three, three types of alert zones are dynamically generated based on personnel identity and permissions, device status, and real-time distance. These alert zones include a restricted area, a warning area, and a focus area. The radius of the restricted area, R_no, is equal to the basic alert radius, R0. Unauthorized personnel entering the restricted area trigger the highest-level alarm. The radius of the warning area, R_warn = k1·R0, triggers a primary warning when unauthorized personnel enter, where k1 ranges from 1.5 to 2.

0. The radius of the focus area, R_attention = k2·R0, records the trajectory and predicts trends when unauthorized personnel enter the focus area, where k2 ranges from 2.5 to 3.

0.

9. A method for early warning of dynamic electronic fences in nuclear power plant areas according to claim 8, characterized in that, The specific conditions for the edge computing node to determine and trigger the early warning in step five are as follows: The Euclidean distance D between personnel and equipment is less than or equal to the radius R_no of the restricted area, and personnel do not have operating privileges for the monitored equipment; The radius of the restricted area R_no < the Euclidean distance D between personnel and equipment ≤ the radius of the warning area R_warn, and the time personnel stay in the area exceeds the preset threshold; Based on the predicted movement trajectory trend of the personnel terminal, the personnel's velocity vector points to the restricted area and the expected arrival time is less than the preset threshold T_pre; If a person's terminal enters a manually designated temporary warning zone and meets any of the above conditions, an early warning will be triggered.

10. A method for early warning of a dynamic electronic fence in a nuclear power plant area according to claim 9, characterized in that, In step six, the warning levels are divided into Level 1 warning, Level 2 warning and Level 3 warning. For Level 1 warning, the warning output module causes the personnel terminal to vibrate and displays a pop-up window on the personnel terminal screen to provide a notification. For Level 2 warnings, the warning output module issues a voice warning on the personnel terminal, driving the on-site audible and visual alarms to flash; for Level 3 warnings, the warning output module displays an evacuation command on the personnel terminal, and the management platform of the warning output module pops up a window and notifies the safety officer to link the video surveillance to lock the personnel's trajectory.

Images