Flight inspection method and system based on artificial intelligence

Abstract

The invention provides a flight inspection method and system based on artificial intelligence, and the method comprises the steps: carrying out the real-time collection of a power line through a data collection device carried by an unmanned plane, so as to obtain corresponding image data; performing data processing on the image data by using a neural network model, and performing flight path planning based on a data processing result to obtain a flight path planning result; acquiring equipment information of all power line equipment in the power distribution network project, and constructing a power line equipment model library based on a parametric modeling algorithm and the equipment information; and performing flight inspection on a to-be-detected area by using a front-end sight distance algorithm, the flight path planning result and the equipment model library so as to realize automatic inspection of electric power facilities in the to-be-detected area. According to the invention, the real-time identification and dynamic adjustment functions based on the front-end sight distance AI technology enable the inspection process to be more efficient.

Description

Technical Field

[0001] This application relates to the technical field of data processing, and in particular, to a flight inspection method and system based on artificial intelligence. Background Art

[0002] At present, for the autonomous inspection operation of UAVs on distribution network lines, the autonomous inspection route planning methods are mainly divided into three-dimensional route planning based on point cloud models and manual teaching route recording. Among them, the full-line point cloud modeling has high costs, a long cycle, and a short applicable time limit. The manual teaching route planning method is affected by the flight control level of the pilot, with a large recording workload, low efficiency, and significant safety hazards, making it difficult to fully support the autonomous inspection work of distribution network lines:

[0003] 1. High-cost and low-efficiency point cloud modeling

[0004] Traditional three-dimensional route planning of UAV autonomous inspection operations based on point cloud models requires a large amount of upfront modeling work, involving complex point cloud data collection and processing, resulting in high modeling costs, a long cycle, and being applicable only within a limited time window, making it difficult to adapt to dynamic inspection requirements.

[0005] 2. Low efficiency of manual teaching route recording

[0006] The manual teaching route recording method relies on the experience and skills of the pilot, with a large workload and low efficiency. The pilot needs to manually operate and record the route in real time, which is easily restricted by environmental factors and flight control accuracy, increasing the flight risks and operation errors during the inspection process and posing significant safety hazards. Summary of the Invention

[0007] Embodiments of this application provide a flight inspection method and system based on artificial intelligence to at least address the deficiencies in the above related technologies.

[0008] In a first aspect, embodiments of this application provide a flight inspection method based on artificial intelligence, including the following steps:

[0009] Step 1: Use a UAV equipped with data collection equipment to perform real-time collection on a power line to obtain corresponding image data;

[0010] Step 2: Use a neural network model to process the image data and perform flight path planning based on the data processing results to obtain a flight path planning result;

[0011] Step 3: Obtain the device information of all power line devices in the distribution network project, and construct a power line device model library based on parametric modeling algorithms and the device information;

[0012] Step 4: Use the front-end visual range algorithm, the flight path planning result, and the device model library to conduct flight inspections on the area to be detected, so as to achieve the automated inspection of the power facilities in the area to be detected.

[0013] Further, the first step includes:

[0014] Use a drone equipped with data acquisition equipment to collect real-time data of the power line, so as to obtain the image data corresponding to the power line;

[0015] Use the front-end visual range algorithm to automatically identify and track the conductors in the image data, so as to obtain the corresponding image data.

[0016] Further, the second step includes:

[0017] Use a neural network model to conduct target detection and feature extraction on the image data, and combine environmental conditions to conduct image preprocessing and optimization, so as to obtain optimized image data;

[0018] Use an artificial intelligence algorithm to judge the conductor path of the optimized image data, and conduct dynamic tracking through a front-end visual range camera to correct the flight trajectory of the drone in real time, so as to ensure that the drone always conducts inspections on the correct path of the conductor path.

[0019] Further, the third step includes:

[0020] Use a parametric modeling algorithm to model the device information of common devices in the distribution network, and modularly manage the physical parameters, working status, and environmental adaptability of each type of device in the common devices, so as to build an initial device model library;

[0021] Import the geometric shapes, material characteristics, and spatial relationships with other power facilities of each common device into the initial device model library, so as to obtain a power line device model library.

[0022] Further, the fourth step includes:

[0023] Use the front-end visual range algorithm, the flight path planning result, and the device model library to conduct flight identification on the area to be detected, so as to identify the identification results of all power line devices in the area to be detected in real time;

[0024] Autonomously adjust the flight path and shooting parameters of the drone according to the identification results, so as to achieve the automated inspection of the power facilities in the area to be detected.

[0025] The present invention also proposes an artificial intelligence-based flight inspection system, including:

[0026] A data acquisition module, which is used to collect data from power lines in real time by using a data acquisition device carried by a drone to obtain corresponding image data;

[0027] A data processing module, which is used to process the image data by using a neural network model and perform flight path planning based on the data processing results to obtain a flight path planning result;

[0028] A model library construction module, which is used to obtain the device information of all power line equipment in the distribution network project and construct a power line equipment model library based on the parametric modeling algorithm and the device information;

[0029] A flight inspection module, which is used to perform flight inspection on the area to be detected by using the front-end line-of-sight algorithm, the flight path planning result and the device model library to realize the automatic inspection of the power facilities in the area to be detected.

[0030] Furthermore, the data acquisition module includes:

[0031] A data acquisition unit, which is used to collect data from power lines in real time by using a data acquisition device carried by a drone to obtain the image data corresponding to the power lines;

[0032] A wire tracking unit, which is used to automatically identify and track the wires in the image data by using the front-end line-of-sight algorithm to obtain corresponding image data.

[0033] Furthermore, the data processing module includes:

[0034] A feature processing unit, which is used to perform target detection and feature extraction on the image data by using a neural network model, and perform image preprocessing and optimization in combination with environmental conditions to obtain optimized image data;

[0035] A path processing unit, which is used to judge the wire path of the optimized image data by using an artificial intelligence algorithm, and perform dynamic tracking through a front-end line-of-sight camera to correct the flight trajectory of the drone in real time to ensure that the drone always inspects on the correct path of the wire path.

[0036] Furthermore, the model library construction module includes:

[0037] A parameter modeling unit, which is used to model the device information of common devices in the distribution network by using the parametric modeling algorithm, and modularly manage the physical parameters, working states and environmental adaptabilities of each type of device in the common devices to construct an initial device model library;

[0038] A model optimization unit for importing the geometric shapes, material characteristics of each of the common devices, and the spatial relationships with other power facilities into the initial device model library to obtain a power line device model library.

[0039] Further, the flight inspection module includes:

[0040] A flight recognition unit for using the front-end line-of-sight algorithm, the flight path planning result, and the device model library to perform flight recognition on the area to be detected, so as to real-time recognize the recognition results of all power line devices in the area to be detected;

[0041] A flight inspection unit for autonomously adjusting the flight path and shooting parameters of the UAV according to the recognition results to achieve automated inspection of the power facilities in the area to be detected.

[0042] Compared with the related technology, a flight inspection method and system based on artificial intelligence provided by an embodiment of the present application implement autonomous inspection operations of UAVs in the distribution network lines, significantly improving the automation, accuracy, and safety of UAVs in power equipment inspection. The AI recognition model can analyze devices such as poles, cross-arms, and insulators in the live video transmission images in real time, accurately identify the device types, positions, and states using deep learning algorithms, avoiding missed inspections and misjudgments in manual inspections. The UAV autonomously adjusts the flight path and angle according to the recognition results to achieve efficient and accurate device inspection, providing multiple intelligent functions to ensure that the device details are clearly visible, allowing manual auxiliary adjustment according to environmental changes to ensure the best shooting effect, and performing accurate inspections according to the device information such as conductors and poles identified by AI, greatly reducing the costs of manual intervention and flight path recording. The real-time recognition and dynamic adjustment functions based on the front-end line-of-sight AI technology make the inspection process more efficient.

[0043] Details of one or more embodiments of the present application are set forth in the following drawings and description to make other features, objects, and advantages of the present application more concise and understandable. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The drawings described herein are used to provide a further understanding of the present application and form a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation of the present application. In the drawings:

[0045] Figure 1 is a flowchart of the flight inspection method based on artificial intelligence in the first embodiment of the present invention;

[0046] Figure 2 is Figure 1 a detailed flowchart of step S101 in

[0047] Figure 3 isFigure 1 Detailed flowchart of step S102 in

[0048] Figure 4 is Figure 1 detailed flowchart of step S103 in ;

[0049] Figure 5 is Figure 1 detailed flowchart of step S104 in ;

[0050] Figure 6 is the structural block diagram of a flight inspection system based on artificial intelligence in the second embodiment of the present invention.

[0051] The following specific embodiments will further illustrate the present invention in conjunction with the above-mentioned drawings. Specific Embodiments

[0052] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be described and explained below in conjunction with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments provided in the present application without making creative efforts fall within the scope of protection of the present application.

[0053] Obviously, the drawings in the following description are only some examples or embodiments of the present application. For those of ordinary skill in the art, without making creative efforts, the present application can also be applied to other similar scenarios based on these drawings. In addition, it can also be understood that although the efforts made in this development process may be complex and lengthy, for those of ordinary skill in the art related to the content disclosed in the present application, some design, manufacturing or production changes based on the technical content disclosed in the present application are only conventional technical means and should not be understood that the content disclosed in the present application is insufficient.

[0054] Referring to "embodiments" in the present application means that the specific features, structures or characteristics described in conjunction with the embodiments can be included in at least one embodiment of the present application. The phrase appears in various positions in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those of ordinary skill in the art explicitly and implicitly understand that the embodiments described in the present application can be combined with other embodiments without conflict.

[0055] Unless otherwise defined, the technical terms or scientific terms involved in this application shall have the ordinary meanings understood by those with ordinary skills in the technical field to which this application belongs. The words such as "a", "an", "one kind", "the" and the like involved in this application do not indicate a quantity limitation and may represent a singular or plural number. The terms "comprising", "including", "having" and any variations thereof involved in this application are intended to cover non-exclusive inclusion; for example, a process, method, system, product or device comprising a series of steps or modules (units) is not limited to the listed steps or units, but may further include unlisted steps or units, or may further include other steps or units inherent to these processes, methods, products or devices. The words such as "connected", "coupled" and the like involved in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect.

[0056] Embodiment 1

[0057] Please refer to Figure 1 , which shows a flight inspection method based on artificial intelligence in the first embodiment of the present invention. The method specifically includes steps S101 to S104:

[0058] S101, use a drone to carry a data acquisition device to perform real-time acquisition on a power line to obtain corresponding image data;

[0059] Further, please refer to Figure 2 , the step S101 specifically includes steps S1011 to S1012:

[0060] S1011, use a drone to carry a data acquisition device to perform real-time acquisition on a power line to obtain the image data corresponding to the power line;

[0061] S1012, use the front-end line-of-sight algorithm to automatically identify and track the conductors in the image data to obtain corresponding image data.

[0062] In specific implementation, use a drone to carry a data acquisition device to perform real-time acquisition on a power line to obtain the image data corresponding to the power line. Among them, the data acquisition device includes but is not limited to devices such as high-definition cameras and data sensors. Combine the front-end line-of-sight AI technology to perform automatic identification and tracking of conductors, so as to obtain corresponding image data. Specifically, accurately identify and continuously track the conductors from a complex environment, especially in multi-device, high-density facilities and complex terrains, to ensure the accuracy and stability of conductor identification, and integrate deep learning algorithms and image processing algorithms for real-time data processing.

[0063] S102. Use the neural network model to process the image data, and perform flight path planning based on the data processing results to obtain the flight path planning result;

[0064] Further, please refer to Figure 3 , the step S102 specifically includes steps S1021 to S1022:

[0065] S1021. Use the neural network model to perform target detection and feature extraction on the image data, and perform image preprocessing and optimization in combination with environmental conditions to obtain optimized image data;

[0066] S1022. Use the artificial intelligence algorithm to judge the wire path of the optimized image data, and perform dynamic tracking through the front-end line-of-sight camera to correct the flight trajectory of the UAV in real time to ensure that the UAV always patrols on the correct path of the wire path.

[0067] In specific implementation, use the neural network model to perform target detection and feature extraction of the wire on the captured image, and perform image preprocessing and optimization in combination with environmental conditions (such as weather and light changes) to ensure a high recognition rate in different environments. Through the AI algorithm, the system can automatically judge the path of the wire, and perform dynamic tracking through the front-end line-of-sight camera to correct the flight trajectory in real time to ensure that the UAV always patrols on the correct path of the wire.

[0068] Through systematic data processing, with the help of 3D modeling technology and the automatic flight control system, the UAV can perform precise flight path planning and inspection tasks according to the three-dimensional space information of the wire. At the same time, the AI algorithm can also intelligently identify and analyze the characteristics of the shape, position, and state of the wire, providing data support for subsequent power equipment fault prediction and maintenance decision-making.

[0069] S103. Obtain the device information of all power line devices in the distribution network project, and build a power line device model library based on the parametric modeling algorithm and the device information;

[0070] Further, please refer to Figure 4 , the step S103 specifically includes steps S1031 to S1032:

[0071] S1031. Use the parametric modeling algorithm to model the device information of common devices in the distribution network, and modularize the physical parameters, working status, and environmental adaptability of each type of device in the common devices to build an initial device model library;

[0072] S1032. Import the geometric shapes, material characteristics of each of the common devices, and their spatial relationships with other power facilities into the initial device model library to obtain a power line device model library.

[0073] In specific implementation, based on the size, material, and attribute information of common devices such as conductors, poles, and brackets in the distribution network, a parametric modeling method is used to construct a model library of power line devices. The model library covers the geometric shapes, material properties of common devices, and their spatial relationships with other power facilities, providing a standardized 3D model that can be called by the inspection system. Designers can rely on this model library to conveniently select wire and power facility models that meet the standards, providing support for device identification and status assessment during the inspection process.

[0074] To ensure the high adaptability and accuracy of the device model, a multi-level parametric design is adopted to modularly manage the physical parameters, working status, and environmental adaptability of each type of device, supporting dynamic adjustment under different environments. The model library provides accurate training data for AI recognition algorithms and also provides diverse and flexible device identification bases for actual inspections.

[0075] S104. Use the front-end visual range algorithm, the flight path planning result, and the device model library to conduct flight inspections on the area to be detected, so as to achieve automated inspection of the power facilities in the area to be detected.

[0076] Further, please refer to Figure 5 , the step S104 specifically includes steps S1041 to S1042:

[0077] S1041. Use the front-end visual range algorithm, the flight path planning result, and the device model library to conduct flight identification on the area to be detected, so as to real-time identify the identification results of all power line devices in the area to be detected;

[0078] S1042. Autonomously adjust the flight path and shooting parameters of the drone according to the identification result, so as to achieve automated inspection of the power facilities in the area to be detected.

[0079] In specific implementation, by applying the front-end visual range AI technology, the drone can achieve automated inspection of power facilities. This technology can real-time identify devices such as poles, crossarms, and insulators, and autonomously adjust the flight path and shooting angle according to the identification result. AI intelligent flight not only supports automatic setting of photo-taking points, but also can automatically adjust the focal length and shooting angle according to the device type during the inspection process to ensure that every detail is clearly presented. The drone can real-time adjust the flight direction according to the video transmission data, avoid obstacles and respond to emergencies to ensure flight safety. Technologies such as automatic zoom and centered photo-taking effectively improve the quality of inspection images, providing reliable data for subsequent device inspection and analysis.

[0080] The AI front-end recognition application uses an image recognition model to analyze equipment types such as poles, crossarms, and insulators in real time during flight, enabling the drone to accurately identify inspection targets and adjust the flight strategy in real time. Different from the traditional drone inspection method that requires manual intervention, it realizes fully autonomous flight. At the same time, functions such as automatic photo-taking point setting, automatic zoom photography, and photo centering processing are introduced to ensure that each inspection image is clear and complete, and can intelligently adjust the shooting angle and zoom ratio according to environmental changes. The system supports inspection course switching and dynamic adjustment, enabling the drone to flexibly respond to complex terrains and changing inspection requirements. In the face of sudden dangerous situations, the emergency braking function can be quickly activated to ensure safety and reliability during the inspection process.

[0081] The flight inspection method based on artificial intelligence in this embodiment has flexible autonomous flight capabilities, can automatically plan the flight direction and dynamically adjust the course, cope with complex scenarios, and ensure that all line equipment is inspected without omission. The emergency braking function responds quickly when encountering obstacles or sudden dangers to ensure flight safety. Through intelligent flight and recognition, it significantly improves the inspection efficiency, reduces manual intervention, optimizes the quality of data collection and analysis, and enhances operation safety.

[0082] In summary, the flight inspection method based on artificial intelligence in the above embodiments of the present invention implements autonomous drone inspection operations in the distribution network line, significantly improving the automation, accuracy, and safety of drones in power equipment inspection. The AI recognition model can analyze equipment such as poles, crossarms, and insulators in the live video transmission screen in real time, accurately identify the equipment type, location, and status using deep learning algorithms, avoiding missed inspections and misjudgments in manual inspections. The drone autonomously adjusts the flight path and angle according to the recognition results to achieve efficient and accurate equipment inspection, provides multiple intelligent functions to ensure that equipment details are clearly visible, and allows manual auxiliary adjustment according to environmental changes to ensure the best shooting effect. Precise inspection is carried out according to the equipment information such as conductors and poles identified by AI, greatly reducing the costs of manual intervention and flight path recording. The real-time recognition and dynamic adjustment functions based on the front-end line-of-sight AI technology make the inspection process more efficient.

[0083] Embodiment 2

[0084] On the other hand, the present invention also proposes a flight inspection system based on artificial intelligence. Please refer to Figure 6 , which shows a flight inspection system based on artificial intelligence in the second embodiment of the present invention, including:

[0085] A data acquisition module 11, configured to perform real-time acquisition of a power line through a data acquisition device carried by a drone to obtain corresponding image data;

[0086] Further, the data acquisition module 11 includes:

[0087] A data acquisition unit, configured to perform real-time acquisition on a power line through a data acquisition device carried by a drone, so as to obtain image data corresponding to the power line;

[0088] A wire tracking unit, configured to automatically identify and track wires in the image data by using a front-end visual range algorithm, so as to obtain corresponding image data.

[0089] A data processing module 12, configured to perform data processing on the image data by using a neural network model, and perform flight path planning based on the data processing result, so as to obtain a flight path planning result;

[0090] Further, the data processing module 12 includes:

[0091] A feature processing unit, configured to perform target detection and feature extraction on the image data by using a neural network model, and perform image preprocessing and optimization in combination with environmental conditions, so as to obtain optimized image data;

[0092] A path processing unit, configured to judge the wire path on the optimized image data by using an artificial intelligence algorithm, and perform dynamic tracking through a front-end visual range camera, and correct the flight trajectory of the drone in real time, so as to ensure that the drone always performs inspections on the correct path of the wire path.

[0093] A model library construction module 13, configured to obtain device information of all power line devices in a distribution network project, and construct a power line device model library based on a parametric modeling algorithm and the device information;

[0094] Further, the model library construction module 13 includes:

[0095] A parameter modeling unit, configured to model the device information of common devices in a distribution network by using a parametric modeling algorithm, and perform modular management on the physical parameters, working states, and environmental adaptabilities of each type of device in the common devices, so as to construct an initial device model library;

[0096] A model optimization unit, configured to import the geometric shapes, material characteristics, and spatial relationships with other power facilities of the common devices into the initial device model library, so as to obtain a power line device model library.

[0097] A flight inspection module 14, configured to perform flight inspection on a to-be-detected area by using a front-end visual range algorithm, the flight path planning result, and the device model library, so as to realize automatic inspection of power facilities in the to-be-detected area.

[0098] Further, the flight inspection module 14 includes:

[0099] A flight recognition unit, configured to use a front-end line-of-sight algorithm, the flight path planning result, and the device model library to perform flight recognition on a detection area, so as to identify in real time the recognition results of all power line devices in the detection area;

[0100] A flight inspection unit, configured to autonomously adjust the flight path and shooting parameters of the unmanned aerial vehicle according to the recognition results, so as to implement automatic inspection of power facilities in the detection area.

[0101] The functions or operation steps implemented when the above modules are executed are substantially the same as those in the above method embodiment, and will not be described in detail here.

[0102] A flight inspection system based on artificial intelligence provided by an embodiment of the present invention has the same implementation principle and technical effects as those in the foregoing method embodiment. For a brief description, for the parts not mentioned in the system embodiment, reference may be made to the corresponding content in the foregoing method embodiment.

[0103] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered that the scope recorded in this specification.

[0104] The above embodiments only represent several implementation manners of the present application, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the invention patent. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the appended claims.

Claims

1. A flight inspection method based on artificial intelligence, characterized in that: The following steps are involved: Step 1: Use the drone to carry out real-time data collection of the power lines to obtain the corresponding image data; Step 2: Processing the image data using a neural network model, and performing flight path planning based on the data processing result to obtain a flight path planning result; Step 3: Obtain equipment information of all power line equipment in the distribution network project, and build a power line equipment model library based on the parameterized modeling algorithm and the equipment information; Step 4: Use the front-end line-of-sight algorithm, the flight path planning result and the equipment model library to perform flight inspection on the area to be inspected, so as to realize the automated inspection of the power facilities in the area to be inspected.

2. The artificial intelligence-based flight inspection method according to claim 1, characterized in that: The step one comprises: The data acquisition equipment carried by the drone is used to collect data of the power line in real time, so as to obtain the image data corresponding to the power line; The front-end sight distance algorithm is used to automatically identify and track the wires in the image data to obtain corresponding image data.

3. The artificial intelligence-based flight inspection method according to claim 1, characterized in that: The second step comprises: Using a neural network model to perform target detection and feature extraction on the image data, and performing image preprocessing and optimization in combination with environmental conditions to obtain optimized image data; The optimized image data is used to determine the wire path through an artificial intelligence algorithm, and dynamic tracking is performed through a front-end line-of-sight camera to correct the flight trajectory of the drone in real time to ensure that the drone is always patrolling on the correct path of the wire path.

4. The artificial intelligence-based flight inspection method according to claim 1, characterized in that: The step three comprises: A parametric modeling algorithm is used to model the equipment information of common equipment in the distribution network, and the physical parameters, working status and environmental adaptability of each type of equipment in the common equipment are modularly managed to build an initial equipment model library; The geometric shape, material characteristics and spatial relationship of each of the common equipment with other power facilities are imported into the initial equipment model library to obtain a power line equipment model library.

5. The artificial intelligence-based flight inspection method according to claim 1, characterized in that: The fourth step comprises: Using the front-end sight-range algorithm, the flight path planning result and the equipment model library to perform flight identification on the area to be detected, so as to identify the identification results of all power line equipment in the area to be detected in real time; The flight path and shooting parameters of the drone are autonomously adjusted according to the recognition result to realize the automated inspection of the power facilities in the area to be inspected.

6. An artificial intelligence-based flight inspection system, characterized in that: include: The data acquisition module is used to collect data from the power lines in real time by using the data acquisition equipment carried by the drone to obtain the corresponding image data; A data processing module, used to process the image data using a neural network model, and perform flight path planning based on the data processing result to obtain a flight path planning result; A model library construction module is used to obtain equipment information of all power line equipment in the distribution network project, and to construct a power line equipment model library based on a parameterized modeling algorithm and the equipment information; The flight inspection module is used to perform flight inspection on the area to be inspected by utilizing the front-end line-of-sight algorithm, the flight path planning result and the equipment model library, so as to realize the automated inspection of the power facilities in the area to be inspected.

7. The artificial intelligence-based flight inspection system according to claim 6, characterized in that: The data acquisition module comprises: A data acquisition unit, used to collect data from the power line in real time by using a drone equipped with a data acquisition device to obtain image data corresponding to the power line; The wire tracking unit is used to automatically identify and track the wires in the image data using a front-end sight distance algorithm to obtain corresponding image data.

8. The artificial intelligence-based flight inspection system according to claim 6, characterized in that: The data processing module comprises: A feature processing unit, used to perform target detection and feature extraction on the image data using a neural network model, and perform image preprocessing and optimization in combination with environmental conditions to obtain optimized image data; The path processing unit is used to judge the wire path of the optimized image data through an artificial intelligence algorithm, and to perform dynamic tracking through a front-end line-of-sight camera to correct the flight trajectory of the UAV in real time to ensure that the UAV always patrols on the correct path of the wire path.

9. The artificial intelligence-based flight inspection system according to claim 6, characterized in that: The model library building module includes: A parameter modeling unit is used to model the device information of common devices in the distribution network using a parameterized modeling algorithm, and modularize the physical parameters, working status and environmental adaptability of each type of the common devices to build an initial device model library; The model optimization unit is used to import the geometric shape, material characteristics and spatial relationship with other power facilities of each of the common equipment into the initial equipment model library to obtain a power line equipment model library.

10. The artificial intelligence-based flight inspection system according to claim 6, characterized in that: The flight inspection module includes: A flight identification unit, used to perform flight identification on the area to be detected by using the front-end line-of-sight algorithm, the flight path planning result and the device model library, so as to identify the identification results of all power line devices in the area to be detected in real time; The flight inspection unit is used to autonomously adjust the flight path and shooting parameters of the drone according to the recognition result to realize the automated inspection of the power facilities in the area to be inspected.

Images