Power transmission channel forest fire hidden danger intelligent early warning system and method based on digital twinborn technology

Abstract

The invention provides a power transmission channel forest fire hidden danger intelligent early warning system and method based on a digital twinborn technology, relates to the technical field of power transmission lines, and aims to improve the power transmission line disaster early warning capability. The system comprises an air-ground integrated inspection unit, a remote unified platform and a remote monitoring center, the air-ground integrated inspection unit is composed of an unmanned aerial vehicle device, a ground monitoring device and an unmanned aerial vehicle nest, and is used for collecting real-time multi-dimensional data of a power transmission channel; the remote unified platform is used for fusing data acquired by the air-ground integrated inspection unit, constructing and dynamically updating a four-dimensional digital twinborn model of a power transmission channel, and performing forest fire risk assessment and deduction based on the model; and the remote monitoring center is connected with the remote unified platform and is used for visually displaying a risk assessment result and executing early warning. According to the method, predictive analysis and active early warning of the forest fire risk are realized by constructing the dynamically evolved digital twinborn model.

Description

Technical Field

[0001] This invention relates to the field of power transmission line technology, and in particular to an intelligent early warning system and method for wildfire hazards in power transmission channels based on digital twin technology. Background Technology

[0002] Power transmission lines typically traverse vast mountainous and forested areas, which are densely vegetated and prone to wildfires. Once a wildfire occurs, it can easily cause power transmission lines to trip, equipment to be damaged, and even trigger large-scale power outages, seriously threatening the safe and stable operation of the power grid.

[0003] Currently, wildfire prevention along power transmission lines mainly relies on regular manual inspections and fixed video surveillance. Manual inspections are inefficient, costly, and pose significant safety risks to inspectors, making 24 / 7 coverage difficult. While traditional video surveillance systems provide fixed-point monitoring, their limited field of view, susceptibility to severe weather, and passive monitoring capabilities lack the ability to analyze fire risks, often only detecting fires after they have spread, missing the optimal response time. Furthermore, existing technologies struggle to comprehensively and quantitatively assess dynamic risk factors such as vegetation growth and climate change, failing to achieve true "prevention before the fire starts."

[0004] This shows that existing technologies have problems in addressing the risk of wildfires along power transmission lines, such as limited monitoring range, untimely early warning, and low level of intelligence. Summary of the Invention

[0005] To address the problems of existing technologies, this invention provides an intelligent early warning system and method for potential wildfire hazards in power transmission channels based on digital twin technology. This system can perceive the environment of power transmission channels in a comprehensive, three-dimensional, and dynamic manner, and perform intelligent risk analysis and prediction based on a digital twin model, thereby realizing the transformation from "passive response" to "proactive early warning".

[0006] To achieve the above-mentioned technical features, the present invention aims to provide an intelligent early warning system for potential forest fire hazards in power transmission channels based on digital twin technology, comprising an integrated air-ground inspection unit, a remote unified platform, and a remote monitoring center; wherein the integrated air-ground inspection unit comprises a drone device and a ground monitoring device, and is equipped with a drone nest for collecting physical environmental data of the power transmission channel and ensuring automatic take-off, landing, charging, and discharging of the drone. The remote unified platform is connected to the air-ground integrated inspection unit and the remote monitoring center respectively, and is used to receive the physical environment data, construct and dynamically update the digital twin model of the power transmission channel, and conduct risk assessment based on the model. The remote monitoring center is used to display the risk assessment results of the remote unified platform and issue early warnings according to preset rules.

[0007] Preferably, the drone device is a platform that integrates a lidar, a thermal imaging camera, a data processor, and a wireless information transmission device.

[0008] Preferably, the ground monitoring device includes a binocular camera and a miniature weather sensor, used to collect real-time images and meteorological data such as wind speed, temperature, and humidity of key areas.

[0009] Preferably, the remote unified platform synchronizes the coordinates and time of the data collected by the UAV device and the ground monitoring device, and constructs and dynamically updates a four-dimensional digital twin model.

[0010] Preferably, the remote unified platform has risk assessment capabilities, including tree obstacle analysis based on acquired point cloud data, abnormal heat source identification based on infrared thermal imaging data, and hazard investigation within the passageway in conjunction with vegetation data.

[0011] Preferably, the risk assessment adopts a multi-index weighted scoring model. The risk score R consists of intrusion depth, combustible load index, abnormal heat source significance, wind speed correction, humidity correction and terrain slope correction, and the threshold is adaptively adjusted according to the season and weather.

[0012] Preferably, the remote unified platform has the ability to predict the spread of fire. Based on parameters such as wind speed and direction, air humidity, terrain slope and orientation, and vegetation type, it simulates the spread direction and rate of potential fire points and outputs a prediction of the impact range. Based on risk assessment and propagation simulation results, the remote unified platform autonomously plans the inspection routes and tasks of the drone devices to achieve key review and inspection and data updates in high-risk areas.

[0013] Preferably, the remote monitoring center includes a 3D visualization module and a multi-level early warning module; the 3D visualization module is used to display the digital twin model, risk level distribution, propagation timeline, and task execution status; the multi-level early warning module is used to send early warning information to operation and maintenance personnel in various ways according to the risk level, and record the handling feedback for threshold optimization.

[0014] Another aspect of the present invention provides an intelligent early warning method for wildfire hazards in power transmission channels based on digital twin technology. The method is implemented based on the intelligent early warning system for wildfire hazards in power transmission channels and includes the following steps: a) Data acquisition: Acquire 3D point cloud, visible light images, thermal imaging, and meteorological data through UAV devices and ground monitoring devices; b) Spatiotemporal registration and twin update: Synchronize the coordinates and time of the collected data, construct and dynamically update the four-dimensional digital twin model, and ensure the consistency of model version and time sequence; c) Risk assessment: Calculate the tree barrier distance and safe distance intrusion depth on the twin model, assess the vegetation combustible load and the significance of abnormal heat sources, combine meteorological and topographic parameters to perform multi-indicator weighted scoring, and output the risk level; d) Spread simulation: Based on wind field, humidity and terrain slope, the spread direction and rate of potential fire points are simulated to generate an estimate of the impact range and arrival time; e) Task scheduling and review: Based on the risks and simulation results, autonomously generate inspection tasks and routes, implement key area review and data collection, and transmit data back to improve the model; f) Early warning and closed-loop handling: Output 3D visualization and hierarchical early warning to the remote monitoring center, record the handling results and use them for continuous optimization of thresholds and models.

[0015] Preferably, the risk score in step c) adopts a weighted model, and the weights can be adaptively adjusted with seasons and weather; the spread projection in step d) is based on a parameterized model of wind speed, wind direction and terrain slope, combined with a vegetation combustibility classification library, to calculate the spread isochron and output the estimated arrival time and priority treatment blocks.

[0016] The present invention has the following beneficial effects: 1. This invention achieves all-round, blind-spot-free coverage monitoring of power transmission channels, taking into account both large-scale inspections and continuous monitoring of key areas, eliminating monitoring blind spots.

[0017] 2. This invention enables early identification and precise location of potential wildfire hazards, and allows for advance prediction of fire development trends, shifting from "passive response" to "proactive early warning," thus avoiding missing the best opportunity for response.

[0018] 3. This invention improves the accuracy and intelligence of risk assessment, enables comprehensive quantitative assessment of dynamic risk factors, and makes risk judgment more scientific.

[0019] 4. This invention significantly reduces the cost and safety risks of manual inspections, improves inspection efficiency, and achieves "less manpower and precise operation and maintenance".

[0020] 5. This invention forms a closed-loop management system of "discovery-analysis-review-early warning-handling-feedback", continuously optimizing the system model and early warning thresholds, and improving long-term operational reliability.

[0021] 6. This invention significantly improves the safety and stability of power transmission lines, effectively preventing line tripping, equipment damage, and large-scale power outages caused by wildfires.

[0022] Overall, compared with existing technologies, this invention, by constructing a dynamically evolving digital twin model and combining air-ground collaborative inspection technology, can monitor changes in risk factors such as vegetation and weather within the power transmission channel in real time, accurately calculate tree barrier distances, quantitatively assess fire risks, and simulate and predict fire development, thereby achieving early identification and precise location of potential wildfire hazards. Attached Figure Description

[0023] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0024] Figure 1 This is a schematic diagram of the overall architecture of the system of the present invention.

[0025] Figure 2 This is a flowchart of the system of the present invention.

[0026] Figure 3 This is a schematic diagram of the functional modules of the digital twin platform in this invention.

[0027] Figure 4 This is a schematic diagram of the visual early warning interface of the remote monitoring center in this invention.

[0028] Figure 5 This is a structural diagram of the unmanned aerial vehicle (UAV) device of the present invention.

[0029] In the diagram, there are: 1. Integrated air-ground inspection unit; 2. Remote unified platform; 3. Remote monitoring center; 4. Unmanned aerial vehicle (UAV) device; 5. Ground monitoring device; 6. UAV nest. It integrates a lidar 401, a thermal imaging camera 402, a data processor 403, and a wireless information transmission device 404. Detailed Implementation

[0030] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0031] Example 1: See Figure 1-4This embodiment provides an intelligent early warning system for wildfire hazards in power transmission channels based on digital twin technology. The system includes an integrated air-ground inspection unit 1, a remote unified platform 2, and a remote monitoring center 3. The integrated air-ground inspection unit 1 includes a drone device 4 and a ground monitoring device 5, and is equipped with a drone nest 6 for collecting physical environmental data of the power transmission channel and ensuring automatic take-off, landing, charging, and discharging of the drone. The remote unified platform 2 is connected to both the integrated air-ground inspection unit 1 and the remote monitoring center 3, and is used to receive the physical environmental data, construct and dynamically update a digital twin model of the power transmission channel, and perform risk assessment based on the model. The remote monitoring center 3 is used to display the risk assessment results of the remote unified platform 2 and issue early warnings according to preset rules. Through this system, by constructing a dynamically evolving digital twin model and combining air-ground collaborative inspection technology, changes in risk factors such as vegetation and weather within the power transmission channel can be monitored in real time, tree obstruction distances can be accurately calculated, fire risks can be quantitatively assessed, and fire development can be simulated and predicted, thereby achieving early identification and precise location of wildfire hazards.

[0032] Further, see Figure 5 The drone device 4 serves as a platform, integrating a lidar 401, a thermal imaging camera 402, a data processor 403, and a wireless information transmission device 404. Through the aforementioned drone device 4, large-scale, high-precision inspection tasks can be performed to acquire three-dimensional spatial information of power transmission channels, vegetation growth status, and abnormal heat source information.

[0033] Furthermore, the ground monitoring device 5 includes a binocular camera and a miniature weather sensor, used to collect real-time images and meteorological data such as wind speed, temperature, and humidity in key areas. The ground monitoring device 5 facilitates the acquisition of meteorological data for the corresponding areas.

[0034] Furthermore, the remote unified platform 2 synchronizes the coordinates and time of the data collected by the UAV device and the ground monitoring device, constructing and dynamically updating a four-dimensional digital twin model. In specific operation, the remote unified platform receives and processes all front-end data to achieve dynamic updates of the digital twin model, reflecting real-time changes in the power transmission channel environment.

[0035] Furthermore, the remote unified platform 2 possesses risk assessment capabilities, including tree obstacle analysis based on acquired point cloud data, abnormal heat source identification based on infrared thermal imaging data, and hazard investigation within the passageway combined with vegetation data. The platform's built-in intelligent analysis engine can perform tree obstacle analysis, flammable material hazard investigation within the passageway, and abnormal heat source identification based on updated model data, and push out the spread trend once a fire point appears, based on real-time meteorological data.

[0036] Furthermore, the risk assessment adopts a multi-index weighted scoring model. The risk score R consists of intrusion depth, combustible load index, abnormal heat source significance, wind speed correction, humidity correction and terrain slope correction, and the threshold is adaptively adjusted according to the season and weather.

[0037] Furthermore, the remote unified platform 2 has the ability to predict the spread of fire. Based on parameters such as wind speed and direction, air humidity, terrain slope and orientation, and vegetation type, it simulates the spread direction and rate of potential fire points and outputs a prediction of the impact range. According to the risk assessment and spread prediction results, the remote unified platform autonomously plans the inspection route and tasks of the UAV device 4 to realize key review and inspection and data update of high-risk areas.

[0038] Furthermore, the remote monitoring center 3 includes a 3D visualization module and a multi-level early warning module. The 3D visualization module is used to display the digital twin model, risk level distribution, propagation timeline, and task execution status. The multi-level early warning module is used to send early warning information to maintenance personnel in various ways according to the risk level, and record handling feedback for threshold optimization. This provides decision support for them to take corresponding preventive measures.

[0039] Example 2: Another aspect of the present invention provides an intelligent early warning method for wildfire hazards in power transmission channels based on digital twin technology. The method is implemented based on the intelligent early warning system for wildfire hazards in power transmission channels and includes the following steps: a) Data acquisition: Acquire 3D point cloud, visible light images, thermal imaging, and meteorological data through UAV devices and ground monitoring devices; b) Spatiotemporal registration and twin update: Synchronize the coordinates and time of the collected data, construct and dynamically update the four-dimensional digital twin model, and ensure the consistency of model version and time sequence; c) Risk assessment: Calculate the tree barrier distance and safe distance intrusion depth on the twin model, assess the vegetation combustible load and the significance of abnormal heat sources, combine meteorological and topographic parameters to perform multi-indicator weighted scoring, and output the risk level; d) Spread simulation: Based on wind field, humidity and terrain slope, the spread direction and rate of potential fire points are simulated to generate an estimate of the impact range and arrival time; e) Task scheduling and review: Based on the risks and simulation results, autonomously generate inspection tasks and routes, implement key area review and data collection, and transmit data back to improve the model; f) Early warning and closed-loop handling: Output 3D visualization and hierarchical early warning to the remote monitoring center, record the handling results and use them for continuous optimization of thresholds and models.

[0040] Preferably, the risk score in step c) adopts a weighted model, and the weights can be adaptively adjusted with seasons and weather; the spread projection in step d) is based on a parameterized model of wind speed, wind direction and terrain slope, combined with a vegetation combustibility classification library, to calculate the spread isochron and output the estimated arrival time and priority treatment blocks.

[0041] Example 3: This invention provides an intelligent early warning system for potential wildfire hazards in power transmission channels based on digital twin technology. (Refer to...) Figure 1 This system includes an integrated air-ground inspection unit, a remote unified platform, and a remote monitoring center. The integrated air-ground inspection unit consists of unmanned aerial vehicle (UAV) devices and ground monitoring devices, and is equipped with UAV nests to enable automatic take-off, landing, charging, and discharging.

[0042] The integrated air-ground inspection unit forms the data sensing foundation of the system, comprising unmanned aerial vehicle (UAV) devices and ground monitoring equipment. The UAV devices are equipped with visible light cameras, infrared thermal imagers, and lidar, enabling them to perform large-scale, high-precision inspection tasks, acquiring three-dimensional spatial information of the power transmission channel, vegetation growth status, and abnormal heat source information. The ground monitoring equipment, deployed on key towers, includes binocular cameras and miniature weather sensors, used for 24 / 7 uninterrupted monitoring of key areas and collecting real-time meteorological data such as wind speed, temperature, and humidity.

[0043] See Figure 2 The system workflow includes: S1 Data Acquisition: UAV devices acquire 3D point clouds, texture images, and thermal radiation information of the passageway; ground devices acquire real-time images and meteorological data such as wind speed, temperature, and humidity. S2 Spatiotemporal Registration and Twin Update: A remote unified platform aligns the coordinates and time of multi-source data, constructs and dynamically updates a four-dimensional digital twin model, including transmission lines, towers, passageway boundaries, terrain, and vegetation layers; and manages model versions to ensure time consistency. S3 Risk Assessment: The system calculates tree barrier distance and safe clearance depth on the twin model, assesses vegetation combustible load and the significance of abnormal heat sources; and combines meteorological and terrain parameters to form a multi-indicator weighted score and output the risk level. S4 Spread Simulation: Based on wind field, humidity, and terrain slope, the system simulates the spread direction and rate of potential fire points, generates an estimate of the impact range and arrival time, and marks priority disposal areas. S5 Task Scheduling and Review: The platform autonomously generates inspection tasks based on risk and simulation results, plans flight routes, and schedules UAV devices to conduct key area review and data collection; after the data collection results are returned, the twin model and risk score are updated. S6 Early Warning and Closed-Loop Response: The remote monitoring center displays the risk level distribution and spread projection timeline in 3D visualization, executes multi-level early warnings according to preset rules, and pushes them through APP, SMS, etc.; records response feedback and effectiveness for continuous optimization of thresholds and models.

[0044] See Figure 3The remote unified platform receives and processes all front-end data to dynamically update the digital twin model, reflecting real-time changes in the power transmission channel environment. The platform's built-in intelligent analysis engine can perform tree obstruction analysis, flammable material hazard investigation within the channel, and abnormal heat source identification based on the updated model data. It can also push out the spread trend of a fire once it occurs, based on real-time meteorological data.

[0045] When the digital twin platform assesses that the risk level of a certain area exceeds the threshold, it will send the risk information and warning level to the remote monitoring center. On the other hand, it can autonomously generate task instructions to dispatch drone devices to the high-risk area for data verification, forming a closed loop of "discovery-analysis-verification-warning".

[0046] See Figure 4 The remote monitoring center serves as a window for human-machine interaction. Maintenance personnel can intuitively view the digital twin model of the entire power transmission channel, real-time monitoring images, risk area distribution, and fire location information through a 3D visualization interface. Upon receiving an early warning, the system will automatically notify relevant personnel via APP push notifications, SMS messages, and other methods according to preset rules, providing decision support and enabling them to take appropriate preventative measures.

[0047] Through the above-described embodiments, this invention integrates digital twin data and combines air-ground integrated inspection technology to achieve dynamic, quantitative, and predictive management of wildfire risks, greatly improving the safety and reliability of power transmission lines.

[0048] Example 3: To make the objectives, technical solutions, and beneficial effects of this invention clearer, the system and method of this invention will be described in detail below with reference to the accompanying drawings and specific application scenarios.

[0049] I. System Hardware Configuration and Deployment: (I) Deployment of integrated air-ground inspection units: 1. Unmanned Aerial Vehicle (UAV) Device 4: A multi-rotor industrial-grade UAV is selected as the platform. The UAV integrates a lidar (401, model: Velodyne VLP-16, ranging accuracy ±2cm, point cloud density 100,000 points / second), a thermal imaging camera (402, model: FLIRVueProR, temperature range -20℃~150℃, thermal sensitivity ≤50mK), a data processor (403, using NVIDIA Jetson Xavier NX, supporting real-time data preprocessing), and a wireless information transmission device (404, supporting 5G+WiFi dual-mode transmission, maximum transmission distance 10km), which can realize large-scale, high-precision data acquisition in the air.

[0050] 2. Ground monitoring device 5: One set is deployed every 5km along the transmission line and installed at the crossarm of the tower. Each set includes a binocular camera (resolution 1920×1080, frame rate 30fps, supports low-light shooting) and a miniature weather sensor (measurement parameters: wind speed 0~60m / s, temperature -40℃~85℃, humidity 0~100%RH, measurement accuracy ±0.1m / s, ±0.2℃, ±2%RH respectively), to achieve 24 / 7 uninterrupted monitoring of key areas.

[0051] 3. Drone Nest 6: Deploy one every 30km along the power transmission channel. It supports automatic take-off and landing of drones, battery replacement and data transmission. The nest has a built-in charging module (output voltage 20V, charging power 600W) and a storage module (capacity 1TB) to ensure the drone's continuous operation.

[0052] (II) Remote unified platform configuration: The system adopts a cloud-edge collaborative architecture, with edge nodes deployed at maintenance stations close to power transmission channels and the cloud deployed at the provincial power grid dispatch center. Hardware configuration includes: an edge computing server (CPU: Intel Xeon Gold 6330, 64GB RAM, 2TB SSD) for real-time data reception, spatiotemporal registration, and preliminary analysis; and a cloud server cluster (4 high-performance servers supporting load balancing) for building a four-dimensional digital twin model, risk assessment, and fire propagation simulation. The software integrates multi-source data fusion algorithms, a digital twin modeling engine, a multi-indicator weighted scoring model, and a fire propagation simulation module, supporting dynamic model updates and autonomous task planning.

[0053] (III) Remote monitoring center configuration: Deployed in the power grid operation and maintenance command center, the hardware includes a 3D visualization workstation (GPU: NVIDIA RTX A6000, 48GB of video memory), a 4K high-definition display screen, and an alarm terminal; the software includes a 3D visualization module (supporting 1:1 digital twin model display of transmission channels, risk level color marking, and propagation simulation animation playback) and a multi-level early warning module (integrating SMS gateway and APP push server, supporting automatic triggering of different early warning methods according to risk level).

[0054] II. System Workflow Implementation Details: (a) Data collection phase: 1. The drone device performs inspection tasks according to the preset route. The lidar collects three-dimensional point cloud data of the power transmission channel, the thermal imaging camera captures abnormal heat sources in the channel (such as unextinguished cigarette butts and spontaneously combusting dead branches), and the visible light camera records the vegetation growth status and the appearance of the line equipment. After the data is preprocessed by the data processor, it is transmitted back to the remote unified platform in real time through the wireless information transmission device. 2. The ground monitoring device collects image data and meteorological data of key areas every 5 minutes. The binocular camera generates depth images for vegetation distance calculation, and the miniature meteorological sensor uploads environmental parameters in real time to ensure data timeliness.

[0055] (II) Spatiotemporal registration and twin update stage: After receiving multi-source data, the remote unified platform first performs spatiotemporal registration with GPS timestamps and coordinate system 1 (using WGS-84 coordinate system) to eliminate data acquisition time difference and location deviation. Then, based on the registered data, it updates the four-dimensional digital twin model (three-dimensional space + time dimension). The model includes layers such as transmission lines, towers, channel boundaries, terrain, and vegetation. The vegetation layer updates its growth status in real time based on point cloud data, and the meteorological layer synchronizes the latest environmental parameters. The model is updated every 15 minutes to ensure temporal consistency.

[0056] (III) Risk Assessment Phase: 1. Based on the digital twin model, the distance between the tree barrier and the transmission line and the depth of the safe clearance intrusion limit are calculated using point cloud data (safe clearance standard: ≥5m for 110kV lines, ≥6m for 220kV lines, and ≥8m for 500kV lines). 2. Assess the vegetation combustible load index (calculated based on vegetation type, density, and moisture content; the combustible load index for coniferous forests is 1.2–1.8, and for broadleaf forests it is 0.8–1.2) and the significance of abnormal heat sources (scored based on heat source temperature, area, and duration, with a score range of 0–10). 3. A multi-index weighted scoring model is used to calculate the risk score R. The formula is: R = α × intrusion depth + β × combustible load index + γ × abnormal heat source significance + δ × wind speed correction + ε × humidity correction + ζ × terrain slope correction, where α, β, γ, δ, ε, and ζ are weight coefficients (β and δ have higher weights in the hot and dry summer season, and ε has higher weights in the rainy season), which are automatically adjusted according to the season and weather. 4. Risk levels are determined by risk score: R < 3 indicates low risk, 3 ≤ R < 6 indicates medium risk, and R ≥ 6 indicates high risk.

[0057] (iv) Spread and deduction stage: When the risk level is medium to high, the spread simulation module is activated: 1. Input parameters include real-time wind speed and direction (from ground monitoring devices), air humidity, terrain slope and orientation (from the terrain layer of the digital twin model), and vegetation type (from the vegetation layer). 2. Based on a parametric model of wind speed, wind direction, and terrain slope, and combined with a vegetation combustibility classification library (coniferous forest spread rate 1.5~2.5m / min, broadleaf forest 0.5~1.0m / min, weeds 0.8~1.2m / min), calculate the fire spread isochrones. 3. Estimated time of fire point reaching the transmission line (error ≤ 10%) and priority treatment area (the area closest to the line and with the fastest spread rate is marked as the first-level treatment area).

[0058] (V) Task scheduling and review stage: The remote unified platform autonomously generates inspection tasks based on risk assessment results and propagation simulation data. 1. For high-risk areas, plan key review routes for drones (increase the route density by 50% compared to regular inspections), and dispatch the nearest drones to take off from their nests and head to the target area for close-range data collection (increase the lidar sampling frequency to 200,000 points / second and adjust the thermal imaging camera frame rate to 60fps). 2. After the data is reviewed and returned, the digital twin model and risk score are updated. If the score is still ≥6, an early warning process is triggered. If the score drops below 6, continuous monitoring is carried out, and the data is updated every 30 minutes.

[0059] (vi) Early warning and closed-loop response phase: 1. The remote monitoring center's 3D visualization module displays the location, risk level, spread simulation animation, and drone verification footage of high-risk areas, allowing maintenance personnel to intuitively grasp the on-site situation through the large display screen; 2. The multi-level early warning module triggers warnings based on risk level: for medium risk, warning information (including location, risk level, and suggested handling measures) is pushed through the APP; for high risk, SMS notification (sent to the operation and maintenance manager and regional dispatcher) and on-site alarm terminal siren are triggered simultaneously, with a warning information response time of ≤30 seconds; 3. Maintenance personnel organize responses based on early warning information (such as clearing flammable vegetation and extinguishing initial fires). After the responses are completed, the results are reported via the APP. The system records the response time, measures, and results, which are used to optimize risk assessment thresholds and digital twin model parameters (such as adjusting the flammability load index weight of a certain type of vegetation) to form a closed-loop management system.

[0060] Example 4: Implementation effect verification This embodiment was piloted in a 220kV transmission channel of a provincial power grid (280km in total length, including 190km through mountainous and forested areas). The following effects were achieved during the application: 1. Monitoring coverage: Achieved 100% coverage of the power transmission channel without blind spots, and discovered 32 potential tree obstructions and 17 abnormal heat sources that were not detected by traditional manual inspections; 2. Early warning timeliness: The average early warning time for wildfire hazards is 45 minutes earlier than that of traditional technologies, and three initial fire points were successfully dealt with before they spread to the line; 3. Risk Assessment: The accuracy rate of risk level determination reaches 92%, and the threshold adaptive adjustment function ensures that the assessment error is ≤8% under different seasons and weather conditions; 4. Operation and maintenance efficiency: The frequency of manual inspections was reduced by 60%, the inspection cost was reduced by 55%, no line tripping accidents caused by wildfires occurred, and the safety and stability of transmission lines were significantly improved.

[0061] Although the preferred embodiments of the present invention have been described above in conjunction with the accompanying drawings, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many specific modifications under the guidance of the present invention without departing from the spirit of the invention and the scope of protection of the claims, and these modifications all fall within the scope of protection of the present invention.

Claims

1. A smart early warning system for wildfire hazards in power transmission channels based on digital twin technology, characterized in that, It includes an integrated air-ground inspection unit (1), a remote unified platform (2), and a remote monitoring center (3); wherein the integrated air-ground inspection unit (1) includes a drone device (4) and a ground monitoring device (5), and a drone nest (6) is set up to collect physical environment data of the power transmission channel and ensure the automatic take-off, landing and charging / discharging of the drone; The remote unified platform (2) is connected to the air-ground integrated inspection unit (1) and the remote monitoring center (3) respectively, and is used to receive the physical environment data, construct and dynamically update the digital twin model of the power transmission channel, and conduct risk assessment based on the model; The remote monitoring center (3) is used to display the risk assessment results of the remote unified platform (2) and issue warnings according to preset rules.

2. The intelligent early warning system for wildfire hazards in power transmission channels based on digital twin technology according to claim 1, characterized in that, The unmanned aerial vehicle (UAV) device (4) is a platform that integrates a lidar (401), a thermal imaging camera (402), a data processor (403), and a wireless information transmission device (404).

3. The intelligent early warning system for wildfire hazards in power transmission channels based on digital twin technology according to claim 1, characterized in that, The ground monitoring device (5) includes a binocular camera and a miniature meteorological sensor, used to collect real-time images and meteorological data such as wind speed, temperature and humidity in key areas.

4. The intelligent early warning system for wildfire hazards in power transmission channels based on digital twin technology according to claim 1, characterized in that, The remote unified platform (2) synchronizes the coordinates and time of the data collected by the UAV device and the ground monitoring device, and constructs and dynamically updates the four-dimensional digital twin model.

5. The intelligent early warning system for wildfire hazards in power transmission channels based on digital twin technology according to claim 4, characterized in that, The remote unified platform (2) has risk assessment capabilities, including tree obstacle analysis based on acquired point cloud data, abnormal heat source identification based on infrared thermal imaging data, and hazard investigation within the passage combined with vegetation data.

6. The intelligent early warning system for wildfire hazards in power transmission channels based on digital twin technology according to claim 5, characterized in that, The risk assessment adopts a multi-index weighted scoring model. The risk score R consists of intrusion depth, combustible load index, abnormal heat source significance, wind speed correction, humidity correction and terrain slope correction, and the threshold is adaptively adjusted according to the season and weather.

7. The intelligent early warning system for wildfire hazards in power transmission channels based on digital twin technology according to claim 6, characterized in that, The remote unified platform (2) has the ability to predict the spread of fire. Based on wind speed and direction, air humidity, terrain slope and orientation, and vegetation type parameters, it simulates the spread direction and rate of potential fire points and outputs the predicted range of influence. Based on the risk assessment and propagation simulation results, the remote unified platform autonomously plans the inspection routes and tasks of the unmanned aerial vehicle (UAV) device (4) to realize key review inspections and data updates in high-risk areas.

8. The intelligent early warning system for wildfire hazards in power transmission channels based on digital twin technology according to claim 7, characterized in that, The remote monitoring center (3) includes a three-dimensional visualization module and a multi-level early warning module; the three-dimensional visualization module is used to display the digital twin model, risk level distribution, propagation timeline and task execution status; the multi-level early warning module is used to send early warning information to operation and maintenance personnel in various ways according to the risk level, and record the handling feedback for threshold optimization.

9. A method for intelligent early warning of wildfire hazards in power transmission channels based on digital twin technology, characterized in that, The method is based on the intelligent early warning system for wildfire hazards in power transmission channels as described in any one of claims 1-8, and includes the following steps: a) Data acquisition: Acquire 3D point cloud, visible light images, thermal imaging, and meteorological data through UAV devices and ground monitoring devices; b) Spatiotemporal registration and twin update: Synchronize the coordinates and time of the collected data, construct and dynamically update the four-dimensional digital twin model, and ensure the consistency of model version and time sequence; c) Risk assessment: Calculate the tree barrier distance and safe distance intrusion depth on the twin model, assess the vegetation combustible load and the significance of abnormal heat sources, combine meteorological and topographic parameters to perform multi-indicator weighted scoring, and output the risk level; d) Spread simulation: Based on wind field, humidity and terrain slope, the spread direction and rate of potential fire points are simulated to generate an estimate of the impact range and arrival time; e) Task scheduling and review: Based on the risks and simulation results, autonomously generate inspection tasks and routes, implement key area review and data collection, and transmit data back to improve the model; f) Early warning and closed-loop handling: Output 3D visualization and hierarchical early warning to the remote monitoring center, record the handling results and use them for continuous optimization of thresholds and models.

10. The intelligent early warning method for wildfire hazards in power transmission channels based on digital twin technology according to claim 9, characterized in that, The risk scoring in step c) uses a weighted model, and the weights can be adaptively adjusted with seasons and weather. The spread projection in step d) is based on a parameterized model of wind speed, wind direction and terrain slope, combined with a vegetation combustibility classification library, to calculate the spread isochron and output the estimated arrival time and priority treatment blocks.

Images