Unmanned aerial vehicle-based distribution line inspection method, device, equipment and medium

Abstract

Embodiments of the present application provide a method, device, equipment and medium for inspecting distribution network lines based on a UAV. The present application relates to the field of distribution network line inspection, and the method comprises: in response to a user terminal's request for inspection of a target inspection area, assigning a UAV to the target inspection area; determining a target inspection path of the UAV based on a twin model and the target inspection area; obtaining inspection data of the target inspection area collected by the UAV based on the inspection path; generating an inspection result of the target inspection area based on the inspection data, and sending the inspection result to the user terminal. The method is used to improve the inspection efficiency of the distribution network lines.

Description

Technical Field

[0001] This application relates to the field of power distribution line inspection, and in particular to a method, device, equipment and medium for power distribution line inspection based on unmanned aerial vehicles (UAVs). Background Technology

[0002] With the rapid development of my country's economy and society, electricity demand continues to rise. As the "last mile" of power supply, the construction quality and efficiency of distribution network projects directly affect the safe and stable operation of the power system and the user's electricity experience. In recent years, investment in distribution network construction has continued to increase, the scale of distribution networks has continued to expand, and the line structure has become increasingly complex. The limitations of traditional manual inspection methods have become increasingly prominent. Against the backdrop of the accelerated construction of new power systems and the deep integration of the energy revolution, the acceptance process of distribution network projects is facing a systemic challenge: traditional models are finding it difficult to adapt to digital transformation.

[0003] Traditional manual inspections rely on personnel's visual range and experience, making it difficult to cover dangerous areas such as high altitudes, remote locations, and complex terrain. This results in numerous blind spots and safety hazards. Construction progress relies on manual statistics, summarization, and reporting, a cumbersome and time-consuming process with significant data lag. Existing drone-mounted equipment is mostly single-function, some only integrating visible light cameras, unable to collect multi-dimensional data such as infrared thermal imaging, electrical parameters, and distance measurements, thus failing to meet the multi-modal data requirements of power distribution network acceptance projects. Therefore, existing technologies suffer from low efficiency in power distribution network line inspections. Summary of the Invention

[0004] This application provides a method, apparatus, equipment, and medium for inspecting distribution network lines based on unmanned aerial vehicles (UAVs), which aims to improve the efficiency of distribution network line inspection.

[0005] In a first aspect, embodiments of this application provide a method for inspecting power distribution lines based on unmanned aerial vehicles (UAVs), comprising:

[0006] In response to the user's inspection request for the target inspection area, a drone is allocated to the target inspection area;

[0007] Based on the twin model and the target inspection area, the target inspection path of the UAV is determined; the twin model is used to simulate the three-dimensional geographical features of the target inspection area.

[0008] Acquire inspection data of the target inspection area collected by the drone based on the inspection path;

[0009] Based on the inspection data, the inspection results for the target inspection area are generated and sent to the user terminal.

[0010] In one possible implementation, the target inspection path of the UAV is determined based on the twin model and the target inspection area, including:

[0011] Obtain weather and lighting information for the target inspection area;

[0012] Based on the three-dimensional geographic features, weather information, and lighting information in the twin model, the initial inspection path of the UAV is generated.

[0013] Based on the heuristic fast exploratory random tree algorithm, the initial inspection path is optimized to obtain the first inspection path;

[0014] The first inspection path is optimized to obtain the target inspection path.

[0015] In one possible implementation, the first inspection path is optimized to obtain the target inspection path, including:

[0016] When the drone performs a pre-set inspection of the target inspection area based on the first inspection path, the dynamic state data and wind speed information of the drone within the pre-set inspection time are obtained.

[0017] Based on dynamic state data and wind speed information, the first inspection path is smoothed and wind interference compensation is applied to obtain the second inspection path.

[0018] Based on a dynamic obstacle avoidance model, the obstacles encountered by the UAV when inspecting the target inspection area based on the second inspection path are simulated. The second inspection path is then optimized based on the obstacles to obtain the target inspection path.

[0019] In one possible implementation, acquiring inspection data of the target inspection area collected by the drone based on the inspection path includes:

[0020] The system acquires visible light images, infrared data, distance parameters, and electrical parameters of the target inspection area by using a drone based on night vision sensors, visible light cameras, miniature laser rangefinders, and electrical parameter sensors.

[0021] In one possible implementation, based on the inspection data, the inspection results for the target inspection area are generated, including:

[0022] Based on the visible light images, infrared data, distance parameters, and electrical parameters in the inspection data, the number of power distribution lines to be constructed in the target inspection area is obtained.

[0023] Based on the number of constructions and the target number of constructions, the construction progress information of the target inspection area is obtained;

[0024] Based on the visible light images in the inspection data, the safety inspection information of the target inspection area is obtained by identifying whether the workers in the visible light images are wearing protective equipment.

[0025] Inspection results are obtained based on construction progress information and safety inspection information.

[0026] In one possible implementation, it also includes:

[0027] Based on the twin model, the target distribution network line is divided into multiple inspection areas according to the terrain complexity, construction progress and functional areas; the target inspection area is one of the multiple inspection areas.

[0028] At least one drone is assigned to each inspection area.

[0029] In one possible implementation, at least one drone is assigned to each inspection area, including:

[0030] Determine if the terrain complexity of the inspection area exceeds a preset threshold. If so, allocate multiple drones to the inspection area. Optimize the target inspection path for each drone based on an anti-interference mode.

[0031] If not, then assign a drone to the inspection area.

[0032] In one possible implementation, the method further includes:

[0033] Based on the inspection results of multiple inspection areas, the overall construction progress information and overall safety inspection information of the target distribution network line are generated.

[0034] Determine whether the overall construction progress information and overall safety inspection information meet the preset conditions. If not, generate an inspection warning and send the warning to the user terminal.

[0035] Secondly, embodiments of this application provide a UAV-based inspection device for power distribution lines, comprising:

[0036] The allocation module is used to allocate drones to the target inspection area in response to the user's inspection request for the target inspection area.

[0037] The determination module is used to determine the target inspection path of the UAV based on the twin model and the target inspection area; the twin model is used to simulate the three-dimensional geographical features of the target inspection area.

[0038] The acquisition module is used to acquire inspection data of the target inspection area collected by the UAV based on the inspection path;

[0039] The generation module is used to generate inspection results for the target inspection area based on the inspection data and send the inspection results to the user terminal.

[0040] In one possible implementation, the determining module is further configured to:

[0041] Obtain weather and lighting information for the target inspection area;

[0042] Based on the three-dimensional geographic features, weather information, and lighting information in the twin model, the initial inspection path of the UAV is generated.

[0043] Based on the heuristic fast exploratory random tree algorithm, the initial inspection path is optimized to obtain the first inspection path;

[0044] The first inspection path is optimized to obtain the target inspection path.

[0045] In one possible implementation, the determining module is further configured to:

[0046] When the drone performs a pre-set inspection of the target inspection area based on the first inspection path, the dynamic state data and wind speed information of the drone within the pre-set inspection time are obtained.

[0047] Based on dynamic state data and wind speed information, the first inspection path is smoothed and wind interference compensation is applied to obtain the second inspection path.

[0048] Based on a dynamic obstacle avoidance model, the obstacles encountered by the UAV when inspecting the target inspection area based on the second inspection path are simulated. The second inspection path is then optimized based on the obstacles to obtain the target inspection path.

[0049] In one possible implementation, the acquisition module is further configured to:

[0050] The system acquires visible light images, infrared data, distance parameters, and electrical parameters of the target inspection area by using a drone based on night vision sensors, visible light cameras, miniature laser rangefinders, and electrical parameter sensors.

[0051] In one possible implementation, the generation module is further configured to:

[0052] Based on the visible light images, infrared data, distance parameters, and electrical parameters in the inspection data, the number of power distribution lines to be constructed in the target inspection area is obtained.

[0053] Based on the number of constructions and the target number of constructions, the construction progress information of the target inspection area is obtained;

[0054] Based on the visible light images in the inspection data, the safety inspection information of the target inspection area is obtained by identifying whether the workers in the visible light images are wearing protective equipment.

[0055] Inspection results are obtained based on construction progress information and safety inspection information.

[0056] In one possible implementation, the allocation module is also used for:

[0057] Based on the twin model, the target distribution network line is divided into multiple inspection areas according to the terrain complexity, construction progress and functional areas; the target inspection area is one of the multiple inspection areas.

[0058] At least one drone is assigned to each inspection area.

[0059] In one possible implementation, the allocation module is also used for:

[0060] Determine if the terrain complexity of the inspection area exceeds a preset threshold. If so, allocate multiple drones to the inspection area. Optimize the target inspection path for each drone based on an anti-interference mode.

[0061] If not, then assign a drone to the inspection area.

[0062] In one possible implementation, the generation module is further configured to:

[0063] Based on the inspection results of multiple inspection areas, the overall construction progress information and overall safety inspection information of the target distribution network line are generated.

[0064] Determine whether the overall construction progress information and overall safety inspection information meet the preset conditions. If not, generate an inspection warning and send the warning to the user terminal.

[0065] Thirdly, embodiments of this application provide an electronic device, including: a memory and a processor;

[0066] The memory stores computer-executed instructions;

[0067] The processor executes computer execution instructions stored in the memory, causing the processor to perform the first aspect and / or various possible implementations of the first aspect as described above.

[0068] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect and / or various possible implementations of the first aspect.

[0069] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.

[0070] The method, apparatus, equipment, and medium for inspecting distribution network lines based on unmanned aerial vehicles (UAVs) provided in this application receive inspection task instructions from user terminals for specific line areas and automatically allocate corresponding UAV resources for the target inspection area. Based on a pre-constructed digital twin model capable of accurately simulating real terrain, and combined with the actual conditions of the target area, the method intelligently plans and generates the UAV's flight inspection path. The UAV flies autonomously along this path, comprehensively collecting inspection data of the line and its environment using onboard equipment. After analysis and processing of the collected multi-source data, a structured inspection result report containing key information such as line status and potential hazards is automatically generated and ultimately fed back to the requesting user terminal. This method, through digital twin-driven automatic path planning and intelligent data processing, transforms the traditional manual inspection mode into an unmanned, precise, and intelligent closed-loop operation, significantly improving the operational efficiency, safety, and data value of distribution network line inspections, and providing solid technical support for intelligent operation and maintenance and proactive risk management of the power grid. Attached Figure Description

[0071] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0072] Figure 1 A flowchart illustrating the UAV-based power distribution line inspection method provided in this application. Figure 1 ;

[0073] Figure 2 A flowchart illustrating the UAV-based power distribution line inspection method provided in this application. Figure 2 ;

[0074] Figure 3 A schematic diagram of the structure of the UAV-based power distribution line inspection device provided in this application;

[0075] Figure 4 A hardware schematic diagram of the UAV-based power distribution line inspection device provided in this application.

[0076] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0077] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of methods and approaches consistent with some aspects of this application as detailed in the appended claims.

[0078] Currently, the acceptance of power distribution network projects mainly relies on two methods: traditional manual inspection and single-function drone equipment. Traditional manual inspection primarily involves inspectors climbing poles or walking into the construction area, completing the acceptance through visual inspection, experience-based judgment, and basic tools (such as binoculars and rangefinders). This method has the following main drawbacks: High safety risks: working at heights increases the risk of falls, and complex terrain poses a risk of injury. Low efficiency: manually tracking construction progress is time-consuming, data feedback is delayed, and it's difficult to provide real-time guidance for adjustments. Strong data subjectivity: acceptance results rely on human experience, lack multi-dimensional data support, and quality assessment is prone to bias. The other method relies on single-function drone equipment, but existing drone-mounted devices are mostly simple integrations of visible light cameras or infrared thermal imaging modules, with significant functional limitations. They also cannot simultaneously acquire key data such as electrical parameters and distance measurements, leading to incomplete quality assessments. Therefore, existing technologies suffer from low efficiency in power distribution network acceptance.

[0079] The method, apparatus, equipment, and medium for UAV-based distribution network line inspection provided in this application are based on the construction of a high-precision three-dimensional geographic model using digital twin technology, and the automation of UAV inspection task scheduling and path planning. Specifically, upon receiving an inspection request for a specific area from a user, the system automatically assigns a suitable UAV to perform the task; it simulates and models the terrain, topography, and line layout of the target inspection area based on the twin model, generating an optimal inspection path that balances safety and efficiency; the UAV autonomously flies along the planned path and collects multi-dimensional inspection data, which is then analyzed and processed collaboratively in the cloud, ultimately forming a structured inspection result that is fed back to the user. This method transforms the traditional manual inspection mode into an intelligent operation system driven by digital twins and using UAVs as the carrier, achieving unmanned, precise, and closed-loop management of the inspection process, significantly improving the efficiency and safety of distribution network line operation and maintenance, and providing key technical support for the digital and intelligent transformation of the power grid.

[0080] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0081] Figure 1 A flowchart illustrating the UAV-based power distribution line inspection method provided in this application. Figure 1 ,like Figure 1 As shown, the method includes:

[0082] S101. In response to the user's inspection request for the target inspection area, allocate a drone to the target inspection area.

[0083] In this embodiment, when a user submits an inspection task request for a specific area through the client, the system automatically matches and assigns the optimal drone to perform the task based on parameters such as drone resource status, task priority, and geographical coverage. This mechanism not only replaces the limitations of relying on experience-based judgment in the traditional manual scheduling mode, but also significantly improves task response efficiency and equipment utilization through optimized resource allocation, laying a real-time and reliable foundation for the smooth implementation of subsequent inspection operations.

[0084] S102. Based on the twin model and the target inspection area, determine the target inspection path of the UAV.

[0085] In this embodiment, a twin model is used to simulate the three-dimensional geographical features of the target inspection area. By introducing a high-precision digital twin model, a virtual three-dimensional scene completely synchronized with the real geographical environment is constructed. Based on the terrain undulations, building distribution, route orientation, and other features simulated by the twin model, combined with dynamic parameters such as weather conditions and airspace restrictions, a path planning algorithm is used to generate an optimal inspection route that balances safety, efficiency, and coverage integrity. This step breaks through the traditional UAV operation mode that relies on preset routes or manual remote control, realizing dynamic optimization and adaptive adjustment of inspection routes in complex environments, thereby reducing flight risks while ensuring the systematic nature and accuracy of data collection.

[0086] S103. Obtain inspection data of the target inspection area collected by the UAV based on the inspection path.

[0087] In this embodiment, during the autonomous flight of the UAV along the planned path, multiple sensors, including visible light, infrared, and lidar, are integrated to collect multi-dimensional data on line equipment, tower structures, and the corridor environment within the target area. The key to this step lies in achieving the synchronous acquisition and real-time transmission of multi-source heterogeneous data. Its significance lies not only in replacing manual labor in high-risk and complex on-site inspections, but also in providing a complete information foundation for subsequent analysis through high-resolution, multi-spectral data carriers, encompassing multiple attributes such as apparent condition, temperature distribution, and geometric dimensions. This solves the pain points of traditional inspections, such as limited data, missing blind spots, and delayed results.

[0088] S104. Based on the inspection data, generate the inspection results for the target inspection area and send the inspection results to the user terminal.

[0089] In this embodiment, by intelligently analyzing and structuring massive amounts of inspection data, key information such as equipment defects, environmental hazards, and abnormal conditions is automatically identified, and standardized inspection reports containing problem location, severity assessment, and handling recommendations are generated. The results are pushed to the user terminal in real time through a visual interface or data interface, achieving a closed loop from data collection to decision support. The significance of this process lies in transforming the traditional inspection conclusion generation model, which is mainly based on "manual interpretation and paper records," into an algorithm-driven, standardized, and traceable intelligent output, greatly improving the accuracy of fault detection and the timeliness of maintenance response, providing core support for preventative maintenance and lean management of distribution network lines.

[0090] The UAV-based distribution network line inspection method provided in this application constructs a twin model of the target inspection area to accurately simulate three-dimensional geographical features and intelligently plans the UAV's target inspection path based on this model. Upon responding to an inspection request from a user, a suitable UAV is automatically assigned to the target inspection area to perform the task. The UAV collects multi-dimensional inspection data along the predetermined inspection path, performs intelligent analysis and processing of the data, generates structured inspection results, and feeds them back to the user. This method upgrades the traditional manual inspection mode to a closed-loop operation system driven by digital twins, using UAVs as the carrier, and centered on intelligent analysis. It significantly improves inspection efficiency and data accuracy, reduces the safety risks and subjective errors of manual inspection, provides key technical support for the intelligent operation and maintenance and refined management of distribution network lines, and effectively promotes the digital transformation of the power system.

[0091] Figure 2 A flowchart illustrating the UAV-based power distribution line inspection method provided in this application. Figure 2 ,like Figure 2 As shown, in this embodiment... Figure 1 Based on the embodiments, a detailed description of the UAV-based distribution network line inspection method is provided. The method includes: assigning a UAV to the target inspection area according to steps S201 to S202; obtaining the target inspection path of the UAV according to steps S203 to S204; acquiring inspection data of the target inspection area collected by the UAV based on the inspection path according to step S205; and generating overall construction progress information and overall safety inspection information of the target distribution network line according to steps S206 to S207. The method comprises:

[0092] S201. Based on the twin model, the target distribution network line is divided into multiple inspection areas according to the terrain complexity, construction progress and functional areas of the target distribution network line.

[0093] In this embodiment, a high-precision 3D geographic model constructed using digital twin technology can realistically reflect the terrain undulations, landform features, and functional area distribution along the route, thus providing a dynamic and adaptive spatial basis for regional division. Specifically, geographic feature data such as terrain elevation, slope, and land cover type are first extracted from the twin model. Combined with construction progress information (such as the completion stage of tower foundations and the progress of line erection) and functional area distribution (such as substation exit sections, crossing sections, and densely populated areas), an irregular triangular mesh algorithm is used to perform 3D spatial partitioning of the route corridor. For example, for mountainous sections with steep terrain and drastic elevation changes, the grid size is automatically reduced, and the inspection sampling density is increased to ensure no blind spots in acceptance coverage under complex terrain; while for plain sections with flat terrain and uniform construction progress, the grid size is appropriately increased to improve inspection efficiency.

[0094] This step breaks through the traditional extensive mode of dividing areas according to fixed lengths or manual experience, and achieves two-way dynamic coupling between grid division and geographical environment and construction status. By adaptively adjusting the grid size and boundaries, it avoids the problem of poor terrain adaptability caused by simple uniform division, and overcomes the subjective arbitrariness of manual division. It provides precise spatial management units for subsequent UAV inspection path planning, resource allocation and refined evaluation, and lays a solid foundation for improving the comprehensiveness and reliability of long-distance power distribution line acceptance.

[0095] S202. In response to the user's inspection request for the target inspection area, determine whether the terrain complexity of the target inspection area is higher than a preset threshold. If so, allocate multiple drones to the target inspection area; otherwise, allocate one drone to the target inspection area.

[0096] In this embodiment, a digital twin model is used to quantitatively assess terrain complexity, which serves as a key basis for drone scheduling decisions, thereby achieving efficient allocation of inspection resources and controllable risks. Specifically, geographical feature parameters such as elevation change rate, slope distribution, and surface obstacle density of the target inspection area are first extracted from the twin model. A pre-set complexity assessment algorithm (such as a weighted model combining terrain undulation coefficient and vegetation cover index) is used to calculate the terrain complexity score of the area. When the score is higher than a preset threshold (e.g., a mountainous section with a slope greater than 30 degrees and an elevation difference exceeding 50 meters), a multi-drone collaborative inspection mode is automatically triggered, allocating two or more drones to the area for parallel operations. One drone is responsible for the main route inspection, while the other focuses on detailed checks of terrain blind spots or crossing points. If the terrain complexity score is lower than the threshold (e.g., a flat plain section), only a single drone is allocated to perform the inspection task in that area to optimize resource allocation.

[0097] During this process, an anti-interference mode is simultaneously activated to optimize the target inspection path for each UAV. Reinforcement learning algorithms dynamically adjust flight parameters to cope with complex electromagnetic environments or weather disturbances, ensuring the stability and integrity of data acquisition. For example, in multi-UAV allocation scenarios, based on a pre-set collaborative strategy using a twin model, complementary inspection sub-areas are divided for each UAV. Real-time communication links enable task synchronization and collision avoidance coordination, preventing flight path conflicts or duplicate data collection. The technical advantage of this allocation mechanism lies in overcoming the limitations of traditional single-UAV inspection modes in terms of terrain adaptability. Through dynamic resource scheduling, it achieves efficient decomposition and parallel execution of inspection tasks in complex environments, ensuring coverage integrity in high-risk areas while avoiding resource waste under simple terrain conditions, significantly improving the operational efficiency and safety of long-distance power distribution line acceptance.

[0098] S203. Obtain weather and illumination information for the target inspection area; generate the initial inspection path for the UAV based on the three-dimensional geographical features, weather and illumination information in the twin model; optimize the initial inspection path based on the heuristic fast exploration random tree algorithm to obtain the first inspection path.

[0099] In this embodiment, the core design principle for generating the first inspection path lies in achieving global optimization and dynamic adaptability of the inspection path through deep fusion of multi-source environmental data and a digital twin model. Specifically, weather information (such as wind speed and precipitation probability) and illumination information (such as light intensity and cloud cover) for the target inspection area are first obtained in real time from a meteorological platform. Simultaneously, three-dimensional geographical features of the route corridor (such as elevation, slope, and obstacle distribution) are extracted from the twin model. Based on this multi-dimensional data, a weighted fusion algorithm generates a basic initial inspection path covering all key acceptance points. This path must meet the basic requirements of avoiding permanent obstacles and adapting to current meteorological conditions. Subsequently, a heuristic fast exploration random tree algorithm is used to optimize this initial path: the algorithm uses the initial path as a guide line, randomly samples in three-dimensional space, and constructs an exploration tree. By introducing heuristic functions such as terrain complexity and meteorological risk coefficient, the expansion value of nodes is evaluated, prioritizing growth towards paths with high safety and good observation angles. Simultaneously, the crossover and mutation operations of a genetic algorithm are used to iteratively optimize the path nodes, ultimately generating a first inspection path that achieves a balance between flight distance, risk avoidance, and data acquisition effectiveness. For example, during mountain route inspections, if there is insufficient sunlight and high wind speed, the algorithm will automatically generate an optimized path that avoids icing areas on the shaded side and reduces the flight altitude on the windward side.

[0100] This step achieves a breakthrough in the traditional path planning model that relies on fixed routes or human experience. Through the synergy of environmental perception and intelligent algorithms, path planning gains the ability to dynamically respond to complex environments. It not only solves the problem of path failure caused by sudden weather changes and terrain obstruction, but also significantly improves the efficiency and quality of path planning through heuristic search. This provides a robust global path benchmark for subsequent real-time obstacle avoidance and precise control, thereby ensuring the efficiency and safety of UAV inspection missions in complex power distribution network environments.

[0101] S204. Obtain the dynamic state data and wind speed information of the UAV during the preset time period when it inspects the target inspection area based on the first inspection path; based on the dynamic state data and wind speed information, perform path smoothing and wind interference compensation on the first inspection path to obtain the second inspection path; based on the dynamic obstacle avoidance model, simulate the obstacles encountered by the UAV when it inspects the target inspection area based on the second inspection path, and optimize the second inspection path according to the obstacles to obtain the target inspection path.

[0102] In this embodiment, the core design principle for generating the target inspection path lies in transforming the theoretical path into a flight trajectory with high anti-interference capability and practical feasibility through a multi-stage optimization strategy. Specifically, firstly, dynamic state data (such as attitude angles and acceleration) and real-time wind speed information collected during a pre-set duration test flight of the UAV based on the first inspection path are acquired. The original path is then smoothed using a model predictive control algorithm, and abrupt changes in the trajectory are eliminated by fitting a fifth-order polynomial curve. Simultaneously, based on an anti-wind interference compensation strategy, waypoint positions and attitude commands are dynamically adjusted according to the wind speed vector and the UAV's aerodynamic parameters to generate a smooth, stable second inspection path that effectively counteracts wind disturbance. Subsequently, a dynamic obstacle avoidance model trained using reinforcement learning is activated. In a digital twin environment, dynamic obstacles (such as temporary construction vehicles and flocks of birds) that the UAV may encounter while flying along the second inspection path are simulated. The obstacle trajectory is predicted in real-time using a convolutional neural network, and combined with a dual reward mechanism of obstacle avoidance success rate and flight path tracking efficiency, local trajectory optimization is performed on the second inspection path. Finally, a target inspection path integrating wind resistance, smoothness, and active obstacle avoidance capabilities is output. For example, when a drone encounters crosswinds while inspecting mountain lines, it will automatically shift its path to the leeward side and increase waypoint density to improve control accuracy. If the simulation detects a crane operation ahead, the dynamic obstacle avoidance model will generate a detour arc trajectory to ensure a safe distance while minimizing path deviation.

[0103] This step achieves a technological breakthrough by overcoming the limitation of traditional path planning where theoretical trajectories are disconnected from actual flight conditions. By combining dynamic data feedback with digital twin pre-simulation, path optimization gains adaptability to complex environments. It not only solves the problem of insufficient trajectory tracking accuracy in strong winds but also significantly reduces the risk of flight interruptions caused by sudden obstacles through forward-looking obstacle avoidance simulation, providing core assurance for the stable and reliable inspection of UAVs in complex power distribution network environments.

[0104] S205. Acquire visible light images, infrared data, distance parameters, and electrical parameters of the target inspection area collected by the UAV based on night vision sensors, visible light cameras, miniature laser rangefinders, and electrical parameter sensors.

[0105] In this embodiment, the UAV achieves full-element data acquisition of the target inspection area through a multi-sensor collaborative acquisition mechanism. Its core design principle lies in dynamically configuring the working modes of heterogeneous sensors according to environmental conditions and acceptance requirements, forming a complementary data acquisition system. In specific implementation, the night vision sensor, visible light camera, miniature laser rangefinder, and electrical parameter sensor mounted on the UAV operate synchronously based on the inspection path preset by the digital twin model: the visible light camera automatically activates high-resolution mode under sufficient lighting conditions to continuously capture high-definition images of key parts such as tower structures and conductor connection points; when the light sensor detects that the ambient illuminance is below the threshold (such as at dusk or in a tunnel), it automatically switches to the night vision sensor to take the lead in acquisition, and captures abnormal equipment temperature distribution through infrared thermal imaging technology; the miniature laser rangefinder emits a laser beam at a fixed frequency and accurately measures spatial parameters such as conductor sag and distance to the ground by calculating the round-trip time of the beam; the electrical parameter sensor collects operating data such as line voltage and current harmonics in real time through non-contact coupling technology. For example, during the acceptance of power lines in mountainous areas, visible light cameras photograph the appearance of insulator strings, laser rangefinders simultaneously measure the distance between them and the towers, electrical parameter sensors detect the leakage current of the insulators, and night vision sensors help identify local overheating phenomena in concealed areas.

[0106] This step overcomes the limitations of traditional single-sensor data acquisition. Through the collaborative operation and adaptive switching of multiple heterogeneous sensors, a multi-dimensional data acquisition network covering visible light images, temperature fields, spatial geometry, and electrical parameters is constructed. It not only solves the problem of data acquisition blind spots under complex lighting conditions but also provides a complete and accurate data foundation for subsequent construction progress statistics, safety behavior identification, and equipment status assessment through the synchronous correlation of spatial and electrical parameters, significantly improving the comprehensiveness and reliability of distribution network line acceptance data.

[0107] S206. Based on the visible light images, infrared data, distance parameters, and electrical parameters in the inspection data, obtain the number of power distribution lines to be constructed in the target inspection area; based on the number of construction lines and the target number of construction lines, obtain the construction progress information of the target inspection area; based on the visible light images in the inspection data, identify whether the workers in the visible light images are wearing protective equipment to obtain the safety inspection information of the target inspection area; based on the construction progress information and the safety inspection information, obtain the inspection results.

[0108] In this embodiment, multimodal data fusion analysis and artificial intelligence algorithms are used to achieve automated and accurate identification of construction progress and safety behaviors. Specifically, target detection is first performed on visible light images collected by drones. Based on a shallowly embedded SE attention mechanism using a hybrid network, construction targets such as towers and foundation pits are accurately identified. Target classification and location are completed using a decoupled detection head. Spatial scale is corrected using distance parameters provided by a miniature laser rangefinder. Finally, the number of construction targets is counted and compared with the total number of targets to calculate the percentage of progress, thus obtaining the construction progress information for the target inspection area. Simultaneously, through an attention mechanism that integrates channel, spatial, and temporal dimensions, continuous detection of protective equipment such as safety helmets and safety belts worn by construction workers in the visible light images is performed to obtain safety inspection information for the target inspection area. For example, during the acceptance of a mountain line, if 12 towers have been erected (the target is 15), and the progress is calculated to be 80%, and a worker in a certain grid is detected not wearing a safety helmet, then the inspection result for the target inspection area is generated.

[0109] This step integrates progress statistics and safety supervision into the inspection process through multi-source data collaboration and intelligent algorithm fusion. It not only solves the problems of low efficiency and easy omissions in manual statistics, but also achieves dynamic linkage between safety risks and progress control by linking violations with construction areas in real time. This provides objective and accurate decision-making basis for power distribution network project acceptance, significantly improving the intelligence level and management efficiency of the acceptance operation.

[0110] S207. Based on the inspection results of multiple inspection areas, generate overall construction progress information and overall safety inspection information of the target distribution network line; determine whether the overall construction progress information and overall safety inspection information meet the preset conditions; if not, generate an inspection warning and send the inspection warning to the user terminal.

[0111] In this embodiment, a comprehensive acceptance and evaluation system for distribution network lines is constructed by integrating construction progress information and safety inspection information from multiple inspection areas. Specifically, based on the inspection results of each area, the overall construction progress completion rate of the line is automatically calculated, and the distribution density of safety violations is statistically analyzed. When a progress deviation in a certain section exceeds a threshold (e.g., the progress completion rate is below 70%) or safety violations occur continuously, an inspection early warning mechanism is automatically triggered. The early warning information is pushed to the user terminal through a preset communication link. The early warning information includes the problem location, severity level, and handling suggestions. For example, during the acceptance of a mountain line, if the progress of the section from tower 3 to tower 5 is found to be 40% behind schedule, and there are records of violations such as not wearing safety helmets in this area, a yellow warning is immediately generated and the high-risk area is marked, prompting a key re-inspection.

[0112] This method overcomes the limitations of single-point acceptance by constructing a global control perspective through multi-regional data fusion analysis, forming a two-level acceptance and evaluation system of "region-line". It not only solves the problems of traditional acceptance methods, such as reliance on manual summarization of overall progress and difficulty in comprehensively controlling safety risks, but also achieves proactive problem discovery and precise location through an intelligent early warning mechanism. This provides decision support for the quality control and safety management of distribution network line projects, significantly improving the systematicness and foresight of the acceptance work.

[0113] Figure 3 The structural schematic diagram of the UAV-based power distribution line inspection device provided in this application is as follows: Figure 3 As shown, the UAV-based power distribution line inspection device 30 provided in this embodiment:

[0114] The allocation module 301 is used to allocate drones to the target inspection area in response to the user's inspection request for the target inspection area.

[0115] The determination module 302 is used to determine the target inspection path of the UAV based on the twin model and the target inspection area; the twin model is used to simulate the three-dimensional geographical features of the target inspection area.

[0116] The acquisition module 303 is used to acquire inspection data of the target inspection area collected by the UAV based on the inspection path.

[0117] The generation module 304 is used to generate inspection results for the target inspection area based on the inspection data and send the inspection results to the user terminal.

[0118] In one possible implementation, the determining module 302 is further configured to:

[0119] Obtain weather and lighting information for the target inspection area;

[0120] Based on the three-dimensional geographic features, weather information, and lighting information in the twin model, the initial inspection path of the UAV is generated.

[0121] Based on the heuristic fast exploratory random tree algorithm, the initial inspection path is optimized to obtain the first inspection path;

[0122] The first inspection path is optimized to obtain the target inspection path.

[0123] In one possible implementation, the determining module 302 is further configured to:

[0124] When the drone performs a pre-set inspection of the target inspection area based on the first inspection path, the dynamic state data and wind speed information of the drone within the pre-set inspection time are obtained.

[0125] Based on dynamic state data and wind speed information, the first inspection path is smoothed and wind interference compensation is applied to obtain the second inspection path.

[0126] Based on a dynamic obstacle avoidance model, the obstacles encountered by the UAV when inspecting the target inspection area based on the second inspection path are simulated. The second inspection path is then optimized based on the obstacles to obtain the target inspection path.

[0127] In one possible implementation, the acquisition module is further configured to:

[0128] The system acquires visible light images, infrared data, distance parameters, and electrical parameters of the target inspection area by using a drone based on night vision sensors, visible light cameras, miniature laser rangefinders, and electrical parameter sensors.

[0129] In one possible implementation, the generation module 304 is further configured to:

[0130] Based on the visible light images, infrared data, distance parameters, and electrical parameters in the inspection data, the number of power distribution lines to be constructed in the target inspection area is obtained.

[0131] Based on the number of constructions and the target number of constructions, the construction progress information of the target inspection area is obtained;

[0132] Based on the visible light images in the inspection data, the safety inspection information of the target inspection area is obtained by identifying whether the workers in the visible light images are wearing protective equipment.

[0133] Inspection results are obtained based on construction progress information and safety inspection information.

[0134] In one possible implementation, the allocation module 301 is further configured to:

[0135] Based on the twin model, the target distribution network line is divided into multiple inspection areas according to the terrain complexity, construction progress and functional areas; the target inspection area is one of the multiple inspection areas.

[0136] At least one drone is assigned to each inspection area.

[0137] In one possible implementation, the allocation module 301 is further configured to:

[0138] Determine if the terrain complexity of the inspection area exceeds a preset threshold. If so, allocate multiple drones to the inspection area. Optimize the target inspection path for each drone based on an anti-interference mode.

[0139] If not, then assign a drone to the inspection area.

[0140] In one possible implementation, the generation module 304 is further configured to:

[0141] Based on the inspection results of multiple inspection areas, the overall construction progress information and overall safety inspection information of the target distribution network line are generated.

[0142] Determine whether the overall construction progress information and overall safety inspection information meet the preset conditions. If not, generate an inspection warning and send the warning to the user terminal.

[0143] The UAV-based power distribution line inspection device provided in this embodiment can execute the method provided in the above-described method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0144] Figure 4 This is a hardware schematic diagram of the UAV-based power distribution line inspection device provided in this application. Figure 4 As shown, the electronic device 40 provided in this embodiment includes at least one processor 401 and a memory 402. Optionally, the device 40 further includes a communication component 403. The processor 401, memory 402, and communication component 403 are connected via a bus 404.

[0145] In a specific implementation, at least one processor 401 executes computer execution instructions stored in memory 402, causing at least one processor 401 to perform the above-described method.

[0146] The specific implementation process of processor 401 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0147] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0148] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0149] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0150] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0151] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0152] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0153] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0154] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, methods, or units, and may be electrical, mechanical, or other forms.

[0155] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0156] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0157] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0158] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0159] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A method for inspecting power distribution lines based on unmanned aerial vehicles (UAVs), characterized in that, include: In response to a user's inspection request for a target inspection area, a drone is allocated to the target inspection area; Based on the twin model and the target inspection area, the target inspection path of the UAV is determined; The twin model is used to simulate the three-dimensional geographical features of the target inspection area; Acquire inspection data of the target inspection area collected by the drone based on the inspection path; Based on the inspection data, an inspection result for the target inspection area is generated, and the inspection result is sent to the user terminal.

2. The method according to claim 1, characterized in that, The determination of the target inspection path for the UAV based on the twin model and the target inspection area includes: Obtain weather and light information for the target inspection area; Based on the three-dimensional geographic features in the twin model, the weather information, and the lighting information, the initial inspection path of the UAV is generated; Based on the heuristic fast exploratory random tree algorithm, the initial inspection path is optimized to obtain the first inspection path; The first inspection path is optimized to obtain the target inspection path.

3. The method according to claim 2, characterized in that, The optimization of the first inspection path to obtain the target inspection path includes: When the UAV performs a pre-set inspection of the target inspection area based on the first inspection path, the dynamic state data and wind speed information of the UAV within the pre-set inspection time are obtained. Based on the dynamic state data and the wind speed information, the first inspection path is smoothed and wind interference compensation is applied to the first inspection path to obtain the second inspection path. Based on a dynamic obstacle avoidance model, the obstacles encountered by the UAV when it inspects the target inspection area based on the second inspection path are simulated. The second inspection path is then optimized based on the obstacles to obtain the target inspection path.

4. The method according to claim 1, characterized in that, The step of acquiring the inspection data of the target inspection area collected by the UAV based on the inspection path includes: The UAV acquires visible light images, infrared data, distance parameters, and electrical parameters of the target inspection area based on night vision sensors, visible light cameras, miniature laser rangefinders, and electrical parameter sensors.

5. The method according to claim 1, characterized in that, The step of generating inspection results for the target inspection area based on the inspection data includes: Based on the visible light images, infrared data, distance parameters, and electrical parameters in the inspection data, the number of power distribution lines to be constructed in the target inspection area is obtained. Based on the construction quantity and the construction target quantity, the construction progress information of the target inspection area is obtained; Based on the visible light images in the inspection data, the safety inspection information of the target inspection area is obtained by identifying whether the workers in the visible light images are wearing protective equipment. The inspection results are obtained based on the construction progress information and the safety inspection information.

6. The method according to any one of claims 1-5, characterized in that, Also includes: Based on the twin model, the target distribution network line is divided into multiple inspection areas according to the terrain complexity, construction progress, and functional areas of the target distribution network line; wherein, the target inspection area is one of the multiple inspection areas; At least one drone is assigned to each of the inspection areas.

7. The method according to claim 6, characterized in that, The allocation of at least one drone to each of the inspection areas includes: If the terrain complexity of the inspection area is higher than a preset threshold, multiple drones are assigned to the inspection area; wherein, the target inspection path of each drone is optimized based on an anti-interference mode. If not, then assign a drone to the target inspection area.

8. The method according to claim 6, characterized in that, The method further includes: Based on the inspection results of multiple inspection areas, the overall construction progress information and overall safety inspection information of the target distribution network line are generated. Determine whether the overall construction progress information and the overall safety inspection information meet the preset conditions. If not, generate an inspection warning and send the inspection warning to the user terminal.

9. A UAV-based inspection device for power distribution lines, characterized in that, include: The allocation module is used to allocate drones to the target inspection area in response to the user's inspection request for the target inspection area; The determination module is used to determine the target inspection path of the UAV based on the twin model and the target inspection area; The twin model is used to simulate the three-dimensional geographical features of the target inspection area; The acquisition module is used to acquire the inspection data of the target inspection area collected by the UAV based on the inspection path; The generation module is used to generate inspection results for the target inspection area based on the inspection data, and send the inspection results to the user terminal.

10. An electronic device, characterized in that, include: Memory, processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the method as described in any one of claims 1-8.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-8.

Images