A complete system and method for unmanned aerial vehicle (UAV)-based voltage detection and grounding wire attachment / removal.

Abstract

The present invention relates to the technical field of automated maintenance operations in power systems. Specifically, it relates to a complete set of operation systems and methods for power verification and grounding wire hanging and removal based on an unmanned aerial vehicle (UAV). The method includes: determining whether the transmission wire is charged, and generating a permission signal after confirming that there is no electricity; under the trigger of the permission signal, controlling the UAV to perform the operation of hanging or removing the grounding wire, and identifying and discriminating the state of the grounding wire device during the operation; monitoring the operation process in the steps of hanging and removing the grounding wire in real time, and transmitting the operation status and associated operation site images to the ground terminal. Through the present invention, a systematic process integrating power verification, state identification, and operation monitoring is achieved, realizing full-process automation and intelligence from power verification to hanging and removing operations and then to process monitoring, improving operation safety and efficiency.

Description

Technical Field

[0001] This invention relates to the field of automated maintenance technology for power systems, and more specifically, to a complete system and method for power system automation inspection and grounding wire installation and removal based on unmanned aerial vehicles (UAVs). Background Technology

[0002] Power outage maintenance work on high-voltage transmission lines must strictly adhere to the safety procedures of "voltage testing and grounding wire installation." Currently, this process is entirely manual, requiring workers to climb towering towers and perform voltage testing and grounding wire installation / removal operations in a strong electric field environment. This method has three major drawbacks: first, it carries extremely high safety risks, with personnel facing the dual threats of falls from heights and electric shock; second, it is inefficient, time-consuming, and heavily reliant on individual skills and experience; and third, it lacks objective and traceable monitoring methods, making it difficult to guarantee safety and quality.

[0003] With the development of drone technology, its application in power line inspection is becoming increasingly widespread, such as channel inspection and defect identification. However, existing technologies mostly use drones as mobile image acquisition platforms, and the returned videos or images still require significant manual observation and interpretation by ground personnel. For the dynamic operation process of "grounding wire installation and removal," existing solutions lack real-time, automated intelligent analysis capabilities. Furthermore, current attempts to mount voltage detectors on drones face severe interference from strong electric fields on high-frequency wireless communication, resulting in unreliable voltage detection signal transmission and an inability to effectively coordinate with visual monitoring, thus failing to form a complete automated operation loop. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a complete system and method for voltage testing and grounding wire installation and removal based on unmanned aerial vehicles (UAVs). It integrates voltage testing, status identification, and operation monitoring into a systematic process, achieving full automation and intelligence from voltage testing to installation and removal operations and process monitoring, thereby improving operational safety and efficiency.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: A complete set of operations for mounting voltage testing and grounding wire installation / removal based on a drone includes: A method for collecting and processing voltage detection information based on low-frequency wireless transmission is used to determine whether the transmission line is energized, and a permission signal is generated after confirming that it is de-energized. Triggered by a permission signal, the drone is controlled to attach or remove the grounding wire, and visual recognition and image enhancement methods for attaching and removing the grounding wire are used to identify and judge the status of the grounding wire device during the operation. A real-time monitoring system for grounding wire operations based on UAV vision is adopted to monitor the operation process of hanging and removing grounding wires in real time, and transmit the operation status and related on-site images to the ground.

[0006] This invention achieves several key improvements. First, it enables remote and reliable voltage testing via interference-resistant low-frequency wireless transmission, ensuring safe access for operations from the outset. Second, during operations, it utilizes a visual method that integrates scenario prior knowledge and adaptive enhancement to accurately identify the real-time status of the grounding device. Finally, a machine-ground collaborative monitoring system provides remote visual supervision and recording of the entire process. This method fundamentally changes the traditional model that relies on manual operation and judgment, realizing the digital and intelligent transformation of power transmission line grounding operations.

[0007] As a preferred method, a voltage detection information acquisition and processing method based on low-frequency wireless transmission is adopted to determine whether the transmission line is energized, and a permission signal is generated after confirming that it is de-energized. Specifically, this includes: The original induced signals around the power transmission line are obtained by using an electrical detection sensor mounted on a drone. The original induced signal is conditioned and its features are extracted to obtain characteristic values ​​representing the electric field strength; The characteristic value is transmitted to the ground via a low-frequency wireless communication module; The energized state of the transmission line is determined based on the characteristic values ​​received from the ground end.

[0008] This invention enables non-contact signal acquisition via a drone platform, avoiding the risks associated with personnel climbing to heights. By conditioning and extracting features from the original induction signals, complex electric field signals are transformed into stable and quantifiable feature values. The use of a low-frequency wireless communication module for transmission effectively avoids interference from high-voltage electric fields on conventional high-frequency communication, ensuring reliable remote decision-making.

[0009] Preferably, the original induced signal is conditioned and its features are extracted to obtain characteristic values ​​representing the electric field strength, including: The original induced signal is amplified and bandpass filtered to obtain the power frequency signal; The power frequency signal is converted from analog to digital to obtain a digital signal; Based on the digital signal, its effective value within at least one power frequency cycle is calculated, and the effective value is used as the characteristic value.

[0010] This invention amplifies and enhances weak signals; the bandpass filtering process is specifically designed to purify power frequency (50Hz / 60Hz) signals, effectively suppressing non-power frequency noise and high-frequency interference in the environment, thus creating conditions for subsequent extraction of accurate feature values. Calculating the effective value within the power frequency period robustly reflects the electric field strength and has a certain tolerance for waveform distortion, making the extracted feature values ​​more resistant to interference and more representative.

[0011] Preferably, the center frequency of the bandpass filter corresponds to the power frequency of the power system, and is used to suppress non-power frequency noise to enhance the signal-to-noise ratio of the characteristic value.

[0012] This invention precisely aligns the center frequency of the bandpass filter with the power grid frequency, maximizing the passage of power frequency signals reflecting the energized state of the line while significantly attenuating or filtering out non-power frequency interference components such as corona discharge and radio noise. This improves signal quality from the source, resulting in a significantly higher signal-to-noise ratio for the final eigenvalues, providing high-quality data input for accurate subsequent decision-making.

[0013] Preferably, the low-frequency wireless communication module operates in the frequency band of 405MHz to 435MHz.

[0014] This invention selects the relatively low wireless frequency band of 405MHz to 435MHz (especially the 433MHz ISM band) for data transmission because low-frequency radio waves have stronger diffraction and penetration capabilities, resulting in better transmission stability in complex electromagnetic environments compared to higher frequency bands such as 2.4GHz / 5.8GHz. Interference such as corona discharge around high-voltage lines is mainly high-frequency, having a relatively small impact on the low-frequency band. This ensures that voltage detection information can be reliably transmitted to the ground terminal with a low bit error rate even in harsh electromagnetic environments.

[0015] Preferably, the energized state of the transmission line is determined based on the characteristic values ​​received at the ground end, including: The received feature values ​​are compared with a preset decision threshold to obtain the comparison result; Based on the comparison results, the energized state of the transmission line is determined, including whether it is energized or de-energized. If the energized state is determined to be de-energized, a permission signal is generated to allow the grounding work to be performed.

[0016] This invention achieves automatic and objective determination of the energized state of a line by comparing quantified feature values ​​with preset objective thresholds, replacing subjective manual judgment. Specifically, the determination result (de-energized state) is transformed into a permissive signal that can be recognized by subsequent systems, providing crucial logical input for automated safety interlocking of the work process and enabling the voltage detection result to directly participate in control decisions.

[0017] Preferably, the preset decision threshold includes a first threshold and a second threshold, wherein the second threshold is less than the first threshold; Based on the comparison results, the energized state of the transmission line is determined, including whether it is energized or de-energized, including: If the eigenvalue is greater than or equal to the first threshold, then the charged state is determined to be a charged state. If the eigenvalue is less than or equal to the second threshold, then the charged state is determined to be an uncharged state.

[0018] This invention establishes a higher first threshold as a definitive criterion for "energized" and a lower second threshold as a definitive criterion for "de-energized," creating a "buffer zone" or "uncertainty zone" between them. When the characteristic value falls within this range, the system is determined to be in an "unclear state," prompting the operator to pay attention or retest for energization. This design effectively avoids repeated jumps in decision results caused by minor signal fluctuations or critical states, improving the reliability and trustworthiness of the system under complex operating conditions.

[0019] Preferably, upon being triggered by a permission signal, the drone is controlled to perform the connection or removal of the grounding wire, and a visual recognition and image enhancement method for grounding wire connection and removal is employed to identify and determine the status of the grounding wire device during the operation, specifically including: Obtain raw operational images containing the grounding wire device; Based on prior knowledge of the power scenario, dynamic regions of interest are determined from the original operation images; Within a dynamically defined region of interest, adaptive image enhancement processing is performed on the image to obtain an enhanced image. Extract visual features from the enhanced image to identify the grounding device; Based on visual features, the grounding wire device is identified, and its hanging / removal status is determined according to the morphological characteristics of the grounding wire device.

[0020] This system organically combines target detection, image enhancement, and state understanding. First, it acquires the original image as input. Second, it introduces "prior knowledge of the power scene" to intelligently determine a dynamic and focused "region of interest," effectively eliminating interference from irrelevant information such as the sky and complex backgrounds, significantly improving the processing efficiency and targeting of subsequent steps. Then, adaptive image enhancement is performed within this region to mitigate image quality degradation caused by factors such as lighting and weather. Next, visual features characterizing the grounding wire device are extracted from the enhanced image. Finally, based on these features, device identification is completed, and its specific morphological characteristics are further analyzed to achieve intelligent discrimination of operational states such as "reliably connected" and "connected but not locked."

[0021] Preferably, based on prior knowledge of the power scenario, a dynamic region of interest is determined from the original operation image, including: Edge detection is performed on the original image to obtain an edge image; From the edge image, candidate line segments belonging to conductor and tower structures are selected based on length and orientation features; Based on the geometric relationship of the candidate line segments, at least one candidate anchor point is determined, which corresponds to the connection area between the conductor and the tower. The region of interest is determined dynamically based on the candidate anchor point and the preset size rules.

[0022] This invention transforms abstract "prior knowledge of power scenarios" into executable algorithmic steps. Through edge detection and line segment filtering based on geometric features (such as length and direction), key conductor and tower structures are located from a complex background. Furthermore, "candidate anchor points" are determined based on the geometric relationships of these structures (such as finding intersections or nearest points), essentially locating the conductor and tower connection points most likely to have grounding wires installed. Finally, a "dynamic region of interest" is defined centered on this anchor point, ensuring that the target device has a high probability of being located within this region, achieving precise focusing from the entire image to the potential target area.

[0023] Preferably, adaptive image enhancement processing is performed on the image within the dynamically changing region of interest to obtain an enhanced image, including: Illumination conditions are evaluated for images within a dynamic region of interest, and evaluation results are obtained. Based on the evaluation results, adaptive image enhancement processing is performed on the image to obtain the enhanced image, including: Image enhancement is achieved by using an adaptive histogram equalization algorithm that limits contrast.

[0024] This invention first assesses the lighting conditions of the focused area, and then initiates the enhancement process when the scene is determined to be low-light and low-contrast. Using a local enhancement algorithm, the Adaptive Histogram Equalization (CLAHE) algorithm, which limits contrast, it effectively improves local contrast while suppressing excessive noise amplification. This is particularly suitable for improving the visibility of small targets such as grounding wires under adverse lighting conditions.

[0025] Preferably, the illumination conditions of the image within the dynamically active region of interest are evaluated to obtain the evaluation results, including: Calculate the average gray value and gray standard deviation of the image within the dynamic region of interest, and evaluate them to obtain the evaluation result; Based on the evaluation results, adaptive image enhancement processing is performed on the image to obtain an enhanced image, specifically including: When the average gray value is lower than the first preset threshold and the gray standard deviation is lower than the second preset threshold, adaptive image enhancement processing is performed on the image to obtain an enhanced image.

[0026] This invention uses "average grayscale value" to evaluate overall brightness and "grayscale standard deviation" to evaluate contrast. When both are below a preset threshold, it can be clearly determined that the current image quality is poor, belonging to a typical low-light, low-contrast scene (such as cloudy days, dusk, or backlighting), at which point enhancement processing is automatically triggered. This statistical feature-based judgment method is simple and effective, meeting the real-time requirements of embedded systems.

[0027] Preferably, image enhancement is performed using an adaptive histogram equalization algorithm that limits contrast, including: The image within a dynamically generated region of interest is segmented into multiple consecutive and non-overlapping sub-blocks; For each sub-block, calculate its grayscale histogram and crop the grayscale histogram using a preset cropping limit value; Perform equalization processing on the cropped grayscale histogram; Perform bilinear interpolation on the processed image of each sub-block; The interpolated image is then smoothed to obtain the enhanced image.

[0028] This invention employs an adaptive enhancement mechanism to address the variable lighting and noise susceptibility of power field operations. It utilizes a contrast-limited adaptive histogram equalization algorithm to handle low-light and high-noise scenarios. Through block segmentation, local processing, and interpolation fusion, adaptive image enhancement is achieved, ensuring the algorithm's robustness when dealing with complex backgrounds and foregrounds of power transmission lines. Specifically, cropping limits prevent over-enhancement of local areas and the resulting artifacts; bilinear interpolation eliminates boundary effects between sub-blocks, ensuring a smooth and natural enhanced image.

[0029] Preferably, the grounding wire device is identified based on visual characteristics, and its hanging / removal operation status is determined according to the morphological characteristics of the grounding wire device, including: Extract the morphological features of the grounding wire device from the identified locations; The extracted morphological features are matched with multiple preset state discrimination rules to determine the hanging and dismantling status of the grounding wire device.

[0030] Through this invention, after successfully locating the wiring device, its "morphological features" directly related to operational safety are further extracted. These features are closely related to the physical state of the device (such as the opening and closing of the grippers and the overall posture). By matching these quantified morphological features with predefined "state discrimination rules", image information can be transformed into semantic states with clear operational guidance significance, such as "reliably connected" and "connected but not locked", thus achieving a deep understanding of the operation process.

[0031] Preferably, the grounding wire device's attachment and removal operation states include: Reliably connected; connected but not locked; not connected.

[0032] Through this invention, "reliably connected" indicates that the grounding device has been correctly and securely installed on the conductor, which is a permissible state for safe operation; "connected but not locked" indicates that the device has contacted the conductor but is not fully locked, posing a risk of detachment or poor contact, which is an intermediate state requiring warning; "not connected" indicates that the device is not in effective contact with the conductor, which is a typical dangerous state. The ability to distinguish between these three states is the core of ensuring operational safety.

[0033] Preferably, the grounding wire device has the following morphological features: The aspect ratio of the rectangle surrounding the grounding device; the pixel opening degree of the gripper area of ​​the grounding device.

[0034] Preferably, the extracted morphological features are matched with multiple preset state discrimination rules to determine the hanging and removing status of the grounding wire device, which also includes: The timing logic is verified by combining the discrimination results of multiple consecutive frames of images prior to the current moment.

[0035] This invention analyzes the discrimination results of multiple consecutive frames of images and smooths or corrects the discrimination result of the current frame. This effectively filters out accidental recognition errors, ensuring that the output work status has temporal continuity and consistency, which better reflects the continuous characteristics of actual work processes.

[0036] As a preferred option, a real-time monitoring system for grounding wire operations based on UAV vision includes an airborne subsystem and a ground subsystem that communicate with each other: The airborne subsystems include: The image acquisition module is used to acquire video streams or image sequences of the grounding wire operation area of ​​transmission lines; The processing and recognition module is used to process image frames in video streams or image sequences, identify grounding devices and determine their operating status, and output the recognition results. The first wireless communication module is used to send information containing the identification results to the ground subsystem; The ground subsystem includes: The second wireless communication module is used to receive information from the airborne subsystem; The information processing module is used to parse information and generate monitoring instructions; The monitoring and display module is used to display the operation status and related on-site images based on monitoring commands.

[0037] Through this invention, the airborne subsystem carried by the UAV is responsible for real-time acquisition and intelligent analysis of on-site images. It integrates image acquisition, embedded intelligent processing and wireless transmission functions, and can complete visual analysis in real time at the work site, reducing the bandwidth pressure of data backhaul and reducing processing latency. The ground subsystem serves as a remote monitoring terminal, responsible for receiving, parsing and visualizing information. This allows ground monitoring personnel to clearly and in real time grasp the precise status of the grounding device (such as whether it is connected and whether the connection is reliable) at the remote monitoring center without having to go to the high-altitude risk point. This achieves penetrating and non-contact safety supervision of high-risk operation processes.

[0038] Preferably, the image frames in the video stream or image sequence are processed to identify the grounding device and determine its operating status, and the identification result is output, including: Adaptive image enhancement is performed on the input image frame to obtain the enhanced image; Based on enhanced images, identify and locate conductor and tower structures; Based on the conductor and tower structure, the target area containing the grounding wire device is determined; Visual features of the target area are extracted, and the operating status of the grounding wire device is determined based on a preset state classification model to obtain the recognition result.

[0039] This invention deeply integrates general visual recognition tasks with prior knowledge of power operation scenarios: First, adaptive image enhancement improves the quality of the original image, overcoming interference from lighting and weather. Second, the inherent conductor and tower structures of the power transmission scenario serve as "visual landmarks," enabling scene understanding and coarse localization. Then, based on prior knowledge (the grounding wire is always near the connection point between the conductor and tower), it quickly focuses on a small area that may contain the target, greatly improving processing efficiency and eliminating irrelevant background interference. Finally, robust features are extracted within the precisely located area, and a pre-trained classification model makes the final state decision. This streamlined design ensures the clarity of the system's processing logic and the repeatability of the results.

[0040] Preferably, adaptive image enhancement is performed on the input image frame to obtain an enhanced image, including: Calculate at least one global statistical feature of the input image frame; The environment category to which the image frame belongs is determined by comparing global statistical features with a preset threshold. Based on the environment category, the corresponding preset image processing algorithm is invoked to process the image frame.

[0041] Preferably, based on the environment category, the corresponding preset image processing algorithm is invoked to process the image frame, including: When the environment is classified as low-light or high-noise, the preset image processing algorithm is a contrast-limited adaptive histogram equalization algorithm, and the specific steps include: The input image frame is divided into multiple consecutive and non-overlapping sub-blocks; For each sub-block, calculate its grayscale histogram and crop the grayscale histogram using a preset cropping limit value; Perform equalization processing on the cropped grayscale histogram; Perform bilinear interpolation on the processed image of each sub-block; The interpolated image is then smoothed to obtain the enhanced image.

[0042] This invention employs an adaptive enhancement mechanism to address the variable lighting and noise susceptibility of power field operations. It utilizes a contrast-limited adaptive histogram equalization algorithm to handle low-light and high-noise scenarios. Through block segmentation, local processing, and interpolation fusion, adaptive image enhancement is achieved, ensuring the algorithm's robustness when dealing with complex backgrounds and foregrounds of power transmission lines. Specifically, cropping limits prevent over-enhancement of local areas and the resulting artifacts; bilinear interpolation eliminates boundary effects between sub-blocks, ensuring a smooth and natural enhanced image.

[0043] Preferably, based on the conductor and tower structure, a target area containing a grounding wire device is determined, including: Edge detection is performed on the enhanced image frame to obtain the edge image; In the edge image, candidate line segments belonging to conductors and towers are selected based on the length and direction features of the line segments; Based on the relative positions and geometric relationships of candidate line segments, the positions of the conductor structure and the tower structure in the image are determined respectively.

[0044] Preferably, based on the relative positions and geometric relationships of the candidate line segments, the positions of the conductor structure and the tower structure in the image are determined, including: Based on the intersection or nearest neighbor of the identified conductor edge and tower edge, at least one candidate anchor point is determined; Candidate boxes are dynamically generated based on preset size rules, with the candidate anchor point as the center, and the candidate box containing the most edge features is determined as the target region.

[0045] This invention utilizes the rigid geometry of power transmission lines as navigation beacons for locating grounding wire devices. Through edge detection and line segment filtering, conductors and towers are first identified in the image. Then, the algorithm finds the intersection or nearest point of these two types of structures, which is highly likely to be the installation location of grounding wires such as insulator strings or hanging points. Using this point as an "anchor," a pre-defined rectangular area is dynamically generated as the target area for focused analysis. This scene-based prior knowledge-based positioning method avoids the huge computational overhead of traditional sliding window or full-image search methods, enabling the system to quickly and accurately lock visual attention on key areas, significantly improving the real-time performance of monitoring.

[0046] As a preferred method, visual features of the target area are extracted, and the operating status of the grounding wire device is determined based on a preset state classification model to obtain the recognition result: The extraction of visual features of the target region specifically involves calculating the gradient direction histogram features of the target region; The preset state classification model is a support vector machine model; The preset state classification model is a convolutional neural network model.

[0047] Through this invention, the gradient orientation histogram feature can effectively describe the edge and contour shape of the grounding wire device, exhibiting a certain degree of stability to changes in illumination, making it highly suitable for characterizing the geometric shape of metal components such as grippers and hooks. The support vector machine model performs excellently on small-sample, high-dimensional feature classification problems, with high computational efficiency, making it suitable for deployment on embedded platforms with limited computing power. The convolutional neural network model possesses stronger feature self-learning and abstraction capabilities, able to learn more complex patterns from original images, and is suitable for larger and more diverse datasets. These two models provide the system with optional technical paths to adapt to different resource constraints and accuracy requirements.

[0048] As a preferred option, the ground subsystem also includes: The storage module is used to store the received job status, associated image frames and system logs in chronological order, and supports retrieval based on time or status.

[0049] This invention provides a complete digital archive of the work process through a storage module. All timestamped work statuses, on-site images, and system operation logs are stored systematically, forming a traceable chain of evidence. In the event of an anomaly or when a review analysis is required, supervisors can quickly retrieve and replay the work situation at any point in time, providing a solid data foundation for accurately locating problems, assigning responsibilities, and optimizing work procedures, thus achieving a closed loop in safety management.

[0050] A complete set of operations for attaching and removing grounding wires using drones is provided, which adopts the aforementioned complete set of operations for attaching and removing grounding wires using drones. Attached Figure Description

[0051] Figure 1 This is a flowchart illustrating the voltage detection information acquisition and processing method in Example 1.

[0052] Figure 2 This is a flowchart illustrating the visual recognition and image enhancement method in Example 1.

[0053] Figure 3 This is a schematic diagram of bilinear interpolation in the adaptive histogram equalization algorithm of the embodiment.

[0054] Figure 4 This is a schematic diagram of the real-time monitoring system for operations in Example 1.

[0055] Figure 5 This is a flowchart illustrating the real-time monitoring method for operations in Example 1.

[0056] Figure 6 This is a schematic diagram of the voltage testing device in Embodiment 2 of the present invention.

[0057] Figure 7 yes Figure 6 A schematic diagram of the structure of the voltage testing mechanism.

[0058] Figure 8 yes Figure 7 Enlarged view of part A in the middle.

[0059] Figure 9 yes Figure 7 Enlarged view of part B in the middle section.

[0060] Figure 10 yes Figure 7 Enlarged view of part C in the middle.

[0061] Figure 11 yes Figure 9 Exploded view of the photosensitive receiver.

[0062] Figure 12 yes Figure 11 Another perspective view.

[0063] Figure 13 This is a control principle diagram of the voltage detection device in Embodiment 2 of the present invention.

[0064] Figure 14 This is a schematic diagram of the hanging device in Example 2.

[0065] Figure 15 for Figure 14 Enlarged view of part A in the middle.

[0066] Figure 16 for Figure 14A schematic diagram of the structure of the mounting frame.

[0067] Figure 17 for Figure 16 A magnified view of part A in the middle.

[0068] Figure 18 for Figure 16 A magnified view of part B in the middle section.

[0069] Figure 19 This is a schematic diagram of the angle steel end hanging device in Example 2.

[0070] Figure 20 This is a side view of the angle steel end hanging device of the present invention.

[0071] Figure 21 for Figure 19 A schematic diagram showing the connection between the middle gripper drive assembly and the movable gripper.

[0072] Figure 22 for Figure 20 A schematic diagram of the structure of the frame that is hung at the end of the central angle steel.

[0073] Figure 23 This is a schematic diagram of the movable gripper structure of the present invention.

[0074] Figure 24 for Figure 23 Schematic diagram of the gripper frame structure.

[0075] Figure 25 for Figure 23 A schematic diagram of the structure of the active block.

[0076] Figure 26 Schematic diagram of the working state of the device for hanging the end of the angle steel Figure 1 .

[0077] Figure 27 Schematic diagram of the working state of the device for hanging the end of the angle steel Figure 2 .

[0078] Figure 28 Schematic diagram of the working state of the device for hanging the end of the angle steel Figure 3 .

[0079] Figure 29 A schematic diagram of the device for hanging the conductor end.

[0080] Figure 30 for Figure 29 A schematic diagram of the structure after removing the lifting ring at the end of the conductor.

[0081] Figure 31 This is a schematic diagram of the internal structure of the wire end hanging device of the present invention.

[0082] Figure 32 for Figure 31 Schematic diagram of the locking mechanism.

[0083] Figure 33 for Figure 32 Another perspective view.

[0084] Figure 34 for Figure 31 A schematic diagram of the unlocking mechanism.

[0085] Figure 35 for Figure 34 Cross-sectional view.

[0086] Figure 36 This is a schematic diagram of the structure of the wire end hanging frame of the present invention.

[0087] Figure 37 for Figure 36 The front view. Detailed Implementation

[0088] To further understand the content of this invention, the invention will be described in detail with reference to the embodiments. It should be understood that the embodiments are merely illustrative and not limiting of the invention.

[0089] Example 1 This embodiment provides a complete set of operations for voltage testing and grounding wire installation and removal based on a drone, which includes: A method for collecting and processing voltage detection information based on low-frequency wireless transmission is used to determine whether the transmission line is energized, and a permission signal is generated after confirming that it is de-energized. Triggered by a permission signal, the drone is controlled to attach or remove the grounding wire, and visual recognition and image enhancement methods for attaching and removing the grounding wire are used to identify and judge the status of the grounding wire device during the operation. A real-time monitoring system for grounding wire operations based on UAV vision is adopted to monitor the operation process of hanging and removing grounding wires in real time, and transmit the operation status and related on-site images to the ground.

[0090] This embodiment achieves several key improvements. First, it enables remote and reliable voltage testing via interference-resistant low-frequency wireless transmission, ensuring safe access for operations from the outset. Second, during the operation, it utilizes a visual method that integrates scenario prior knowledge and adaptive enhancement to accurately identify the real-time status of the grounding device. Finally, a machine-ground collaborative monitoring system provides remote visual supervision and recording of the entire process. This method fundamentally changes the traditional model that relies on manual operation and judgment, realizing the digital and intelligent transformation of transmission line grounding operations.

[0091] like Figure 1As shown, this embodiment provides a method for collecting and processing voltage detection information, specifically a method for collecting and processing voltage detection information based on low-frequency wireless transmission. This method is used to determine whether a power transmission line is energized and generates a permission signal after confirming that it is de-energized. Specifically, it includes: The original induced signals around the power transmission line are obtained by using an electrical detection sensor mounted on a drone. The original induced signal is conditioned and its features are extracted to obtain characteristic values ​​representing the electric field strength; The characteristic value is transmitted to the ground via a low-frequency wireless communication module; The energized state of the transmission line is determined based on the characteristic values ​​received from the ground end.

[0092] Specifically, once the drone approaches the target power line, its onboard voltage detection sensors activate. The entire system's processing flow is spatially distributed between the airborne unit and the ground station. The airborne unit is responsible for signal sensing, initial processing, and transmission; the ground station is responsible for data reception, final decision-making, and coordinated operational control.

[0093] This embodiment achieves non-contact signal acquisition through a drone platform, avoiding the risks of personnel climbing to heights; by conditioning and extracting features from the original induction signal, the complex electric field signal is transformed into stable and quantifiable feature values; and by using a low-frequency wireless communication module for transmission, interference from high-voltage electric fields on conventional high-frequency communication is effectively avoided, ensuring reliable remote decision-making.

[0094] In this embodiment, the original induced signal is conditioned and its features are extracted to obtain feature values ​​characterizing the electric field strength, including: The original induced signal is amplified and bandpass filtered to obtain the power frequency signal; The power frequency signal is converted from analog to digital to obtain a digital signal; Based on the digital signal, its effective value within at least one power frequency cycle is calculated, and the effective value is used as the characteristic value.

[0095] In this step, the signal conditioning circuit first amplifies the microvolt or millivolt-level raw signal output by the sensor to meet the level requirements of subsequent circuit processing. Then, the amplified signal enters a bandpass filter with a center frequency precisely calibrated at the power grid frequency. This filter acts as a "frequency screen," allowing only narrowband signals centered at the power frequency to pass through, thus "stripping" the power frequency induced component, which reflects whether the conductor is energized, from the raw signal containing a large amount of high-frequency noise, resulting in a relatively pure power frequency signal. This power frequency signal is then sampled and quantized by an analog-to-digital converter, and the processor calculates its root mean square value over one or more complete cycles. This value is the core "eigenvalue" characterizing the electric field strength at that point.

[0096] This embodiment amplifies and enhances weak signals; bandpass filtering is specifically designed to purify power frequency (50Hz / 60Hz) signals, effectively suppressing non-power frequency noise and high-frequency interference in the environment, thus creating conditions for subsequent extraction of accurate feature values. Calculating the effective value within the power frequency period robustly reflects the electric field strength and has a certain tolerance for waveform distortion, making the extracted feature values ​​more resistant to interference and more representative.

[0097] In this embodiment, the center frequency of the bandpass filter corresponds to the power frequency of the power system and is used to suppress non-power frequency noise to enhance the signal-to-noise ratio of the characteristic value.

[0098] The center frequency of the bandpass filter is determined based on the actual power grid frequency (e.g., 50Hz or 60Hz) and remains stable. Its bandwidth setting requires a trade-off between selectivity and response speed: too narrow a bandwidth may be sensitive to small frequency shifts, affecting signal transmission; too wide a bandwidth may include more nearby noise. In this embodiment, while ensuring the power frequency signal passes through without distortion, the bandwidth is set as narrow as possible to maximize the suppression of various interferences outside the power frequency range, such as switching operation harmonics and radio broadcasts, thereby fundamentally improving the quality of the subsequently extracted feature values.

[0099] This embodiment precisely aligns the center frequency of the bandpass filter with the power grid frequency, maximizing the passage of power frequency signals reflecting the energized state of the line while significantly attenuating or filtering out non-power frequency interference components such as corona discharge and radio noise. This improves signal quality from the source, resulting in a significantly higher signal-to-noise ratio for the final eigenvalues, providing high-quality data input for accurate subsequent decision-making.

[0100] In this embodiment, the operating frequency band of the low-frequency wireless communication module is 405MHz to 435MHz.

[0101] In this embodiment, the relatively low wireless frequency band of 405MHz to 435MHz (especially the 433MHz ISM band) is selected for data transmission because low-frequency radio waves have stronger diffraction and penetration capabilities, resulting in better transmission stability in complex electromagnetic environments compared to higher frequency bands such as 2.4GHz / 5.8GHz. Interference such as corona discharge around high-voltage lines is mainly high-frequency, having a relatively small impact on the low-frequency band. This ensures that voltage detection information can be reliably transmitted to the ground terminal with a low bit error rate even in harsh electromagnetic environments.

[0102] In this embodiment, the energization state of the transmission line is determined based on the feature values ​​received at the ground end, including: The received feature values ​​are compared with a preset decision threshold to obtain the comparison result; Based on the comparison results, the energized state of the transmission line is determined, including whether it is energized or de-energized. If the energized state is determined to be de-energized, a permission signal is generated to allow the grounding work to be performed.

[0103] This embodiment achieves automatic and objective determination of the energized state of a line by comparing quantified feature values ​​with preset objective thresholds, replacing subjective manual judgment. The determination result (de-energized state) is transformed into a permissive signal that can be recognized by subsequent systems, providing crucial logical input for automated safety interlocking of the work process and enabling the voltage detection result to directly participate in control decisions.

[0104] In this embodiment, the preset decision threshold includes a first threshold and a second threshold, wherein the second threshold is less than the first threshold; Based on the comparison results, the energized state of the transmission line is determined, including whether it is energized or de-energized, including: If the eigenvalue is greater than or equal to the first threshold, then the charged state is determined to be a charged state. If the eigenvalue is less than or equal to the second threshold, then the charged state is determined to be an uncharged state.

[0105] This embodiment sets a higher first threshold as a clear criterion for "energized" and a lower second threshold as a clear criterion for "de-energized," creating a "buffer zone" or "uncertainty zone" between them. When the characteristic value falls within this range, the status is determined to be "unclear," prompting the operator to pay attention or retest for energization. This design effectively avoids repeated jumps in the judgment result caused by minor signal fluctuations or critical states, improving the reliability and trustworthiness of the system under complex operating conditions.

[0106] In this embodiment, the voltage detection sensor is a capacitively coupled voltage detection probe.

[0107] This embodiment demonstrates that the capacitively coupled voltage detector is an ideal sensor for non-contact voltage detection. Its working principle is based on the stray capacitance between the probe and the high-voltage conductor. When the probe approaches the conductor, the power frequency electric field couples through this capacitance, inducing a weak power frequency current or voltage signal in the probe circuit. This method eliminates the need for physical contact with the conductor, ensuring the safety of voltage detection and simplifying the complexity of the UAV mounting mechanism. The probe tip can typically be designed with a "V" shape or guide fork structure to facilitate more reliable conductor entry into the effective sensing area when the UAV approaches the conductor.

[0108] like Figure 2 , Figure 3As shown, in this embodiment, triggered by a permission signal, the drone is controlled to perform the connection or removal of the grounding wire. A visual recognition and image enhancement method for grounding wire connection and removal is employed to identify and determine the status of the grounding wire device during the operation. Specifically, this includes: Obtain raw operational images containing the grounding wire device; Based on prior knowledge of the power scenario, dynamic regions of interest are determined from the original operation images; Within a dynamically defined region of interest, adaptive image enhancement processing is performed on the image to obtain an enhanced image. Extract visual features from the enhanced image to identify the grounding device; Based on visual features, the grounding wire device is identified, and its hanging / removal status is determined according to the morphological characteristics of the grounding wire device.

[0109] This embodiment captures a sequence of raw images of the work site using an airborne camera, and performs visual analysis layer by layer, from coarse to fine. It organically combines target detection, image enhancement, and state understanding. First, the raw images are used as input. Second, "prior knowledge of the power scene" is introduced to intelligently determine a dynamic and focused "region of interest," effectively eliminating interference from irrelevant information such as the sky and complex backgrounds, significantly improving the processing efficiency and targeting of subsequent steps. Then, adaptive image enhancement is performed within this region to improve image quality degradation caused by factors such as lighting and weather. Next, visual features characterizing the grounding wire device are extracted from the enhanced image. Finally, based on these features, device identification is completed, and its specific morphological features are further analyzed to achieve intelligent judgment of work states such as "reliably connected" and "connected but not locked."

[0110] In this embodiment, based on prior knowledge of the power scenario, a dynamic region of interest is determined from the original operation image, including: Edge detection is performed on the original image to obtain an edge image; From the edge image, candidate line segments belonging to conductor and tower structures are selected based on length and orientation features; Based on the geometric relationship of the candidate line segments, at least one candidate anchor point is determined, which corresponds to the connection area between the conductor and the tower. The region of interest is determined dynamically based on the candidate anchor point and the preset size rules.

[0111] In this embodiment, edge detection can be performed using operators such as Canny to highlight structural information in the image. A large number of straight line segments are detected from the edge image using Hough transform. Subsequently, selection is performed based on the stable geometric features of the conductors and towers in the image: transmission conductors typically appear as long line segments in a near-horizontal direction, so length thresholds and directional angle tolerance ranges (e.g., close to 0 degrees or 180 degrees) are set to select "conductor candidates"; the main or diagonal members of towers typically appear as line segments in a near-vertical direction or at a specific tilt angle, so corresponding length and direction thresholds are set to select "tower candidates". Anchor point determination relies on geometric reasoning. One implementation calculates the intersection points of the selected conductor and tower candidate line segments. Considering that line segments in the actual image may not intersect due to occlusion or detection errors, the nearest points between line segments can be calculated, and the spatially adjacent nearest points are clustered. The cluster centers can then serve as candidate anchor points, with each anchor point representing a potential "conductor-tower connection area". The preset size rules are determined based on the projection range of the physical dimensions of the grounding device and its working space in the image, ensuring that the defined rectangular area is sufficient to cover the device and its necessary operating space, while avoiding the inclusion of too much irrelevant background.

[0112] This embodiment transforms abstract "prior knowledge of power scenarios" into executable algorithmic steps. Through edge detection and line segment filtering based on geometric features (such as length and direction), key conductor and tower structures are located from a complex background. Furthermore, "candidate anchor points" are determined based on the geometric relationships of these structures (such as finding intersections or nearest points), essentially locating the conductor and tower connection points most likely to have grounding wires installed. Finally, a "dynamic region of interest" is defined centered on this anchor point, ensuring that the target device has a high probability of being located within this region, achieving precise focusing from the entire map to the potential target area.

[0113] In this embodiment, adaptive image enhancement processing is performed on the image within the dynamically defined region of interest to obtain an enhanced image, including: Illumination conditions are evaluated for images within a dynamic region of interest, and evaluation results are obtained. Based on the evaluation results, adaptive image enhancement processing is performed on the image to obtain the enhanced image, including: Image enhancement is achieved by using an adaptive histogram equalization algorithm that limits contrast.

[0114] In this embodiment, the illumination condition assessment is a decision-making process that determines whether enhancement is needed and to what extent. The contrast-limited adaptive histogram equalization (CLAHE) algorithm is a local contrast enhancement technique suitable for this scenario. Unlike global histogram equalization, CLAHE divides the image into small blocks and performs histogram equalization independently within each block, avoiding the excessive amplification of background noise and oversaturation of target areas that can occur with global processing. Simultaneously, the "contrast-limited" mechanism prevents over-enhancement artifacts caused by a concentration of a few pixel grayscale values ​​in local areas by cropping the histogram. This method is particularly beneficial for enhancing the edge and texture details of small, potentially shadowed or backlit metallic targets such as grounding devices, while maintaining the natural appearance of other parts of the image.

[0115] In this embodiment, the lighting conditions of the focused area are first assessed, and the enhancement process is initiated when the scene is determined to be a low-light, low-contrast scene. Using a local enhancement algorithm such as the Contrast-Limited Adaptive Histogram Equalization (CLAHE), local contrast can be effectively improved while suppressing excessive noise amplification. This is particularly suitable for improving the visibility of small targets such as grounding wires under adverse lighting conditions.

[0116] In this embodiment, the illumination conditions of the image within the dynamically dynamic region of interest are evaluated to obtain the evaluation results, including: Calculate the average gray value and gray standard deviation of the image within the dynamic region of interest, and evaluate them to obtain the evaluation result; Based on the evaluation results, adaptive image enhancement processing is performed on the image to obtain an enhanced image, specifically including: When the average gray value is lower than the first preset threshold and the gray standard deviation is lower than the second preset threshold, adaptive image enhancement processing is performed on the image to obtain an enhanced image.

[0117] In this embodiment, the "average gray value" is used to evaluate overall brightness, and the "gray value standard deviation" is used to evaluate contrast. When both are below a preset threshold, it can be clearly determined that the current image quality is poor, belonging to a typical low-light, low-contrast scene (such as cloudy days, dusk, or backlight), and enhancement processing is automatically triggered. This statistical feature-based judgment method is simple and effective, meeting the real-time requirements of embedded systems.

[0118] In this embodiment, image enhancement is performed on the image using an adaptive histogram equalization algorithm that limits contrast, including: The image within a dynamically generated region of interest is segmented into multiple consecutive and non-overlapping sub-blocks; For each sub-block, calculate its grayscale histogram and crop the grayscale histogram using a preset cropping limit value; Perform equalization processing on the cropped grayscale histogram; Perform bilinear interpolation on the processed image of each sub-block; The interpolated image is then smoothed to obtain the enhanced image.

[0119] The calculation of the global average gray level and gray level standard deviation of the image is used as criteria. If the average gray level is low and the gray level distribution is flat, it can be identified as a low-light scene. If the gray level standard deviation is extremely large, it indicates significant noise or excessive local contrast, and can be identified as a high-noise or high-dynamic-range scene. For scenes with normal lighting, standard histogram equalization or gamma correction is used; while for low-light or high-noise scenes, a more advanced contrast-limited adaptive histogram equalization algorithm is used.

[0120] By dividing the image into blocks and performing histogram equalization independently within each block, local detail contrast can be effectively enhanced, revealing the edges of wires and towers hidden in dark or bright areas. Specifically, a cropping limit is used to restrict the maximum number of pixels that can be obtained at each gray level during equalization, preventing unnatural artifacts or amplified noise in local areas due to over-enhancement. After processing all sub-blocks, bilinear interpolation is used to smoothly fuse the results, eliminating potential boundary artifacts between blocks. Finally, a smoothing filter is applied to suppress residual noise, outputting an enhanced image with moderate overall contrast, clear local details, and better suited for further analysis by machine vision algorithms.

[0121] The specific implementation method is as follows: (1) Image segmentation: Divide the image into M equal blocks, assuming that the size of each region is (a, b); (2) Calculate the grayscale threshold: Divide the pixels in the region evenly into the corresponding grayscale distribution to obtain the pixel mean M of each level. a The expression is:

[0122] m x ×m y M represents the number of pixels within a small area. g This represents the number of gray levels contained within a small area.

[0123] T=CM a (3-2) In the formula: C is the truncation limit factor; T is the threshold.

[0124] (3) Limiting sub-block pixels: Each sub-block is limited by a grayscale threshold, and the excess pixels are evenly distributed according to grayscale levels to obtain a histogram distribution:

[0125] In the formula: S C —Total number of pixels; M e —Number of pixels at each level.

[0126] (4) Equalization: Use histograms to equalize each region.

[0127] (5) Interpolation: To avoid blockiness, bilinear interpolation is used for the points within the region, so that the mapping of each point is determined by the mapping relationship of the corresponding regions of the four nearest points. The gray values ​​corresponding to A1, A2, A3, and A4 are V. 1(i) V 2(i) V 3(i) V 4(i) The expression is as follows: V(i)=a[bV 1(i) +(1−b).V 2(i) ]+(1−a).[bV 3(i) +(1−b).V 4(i) ] In the formula:

[0128] (6) Image smoothing: Use median filtering to smooth the image.

[0129] The histogram smoothing method can smooth out random fluctuations in the gray-level distribution of an original image, making peaks and troughs more pronounced. This makes it easier to select a threshold for the histogram and perform connected component segmentation. The processing flow is as follows: 1) Select the initial threshold T k =(μ1+μ2) / 2, where μ1 is the maximum gray value and μ2 is the minimum gray value; 2) Initial threshold T k The image is segmented into foreground and background regions, and their average gray values ​​μ0 and μ are calculated. B

[0130]

[0131] 3) Obtain the new threshold T k+1 T k+1 =(μ0+μ B ) / 2, using the new threshold as the initial threshold, iterate through steps 2) and 3) until |T k +1−T k <|ε, where ε is a set value.

[0132] This embodiment employs an adaptive enhancement mechanism to address the variable lighting and noise susceptibility of power field operations. A contrast-limited adaptive histogram equalization algorithm is used to handle low-light and high-noise scenarios. Adaptive image enhancement is achieved through block segmentation, local processing, and interpolation fusion, ensuring the algorithm's robustness when dealing with complex backgrounds and foregrounds of power transmission lines. Specifically, cropping limits prevent over-enhancement of local areas and the resulting artifacts; bilinear interpolation eliminates boundary effects between sub-blocks, ensuring a smooth and natural enhanced image.

[0133] In this embodiment, the grounding wire device is identified based on visual features, and its hanging / removal operation status is determined according to the morphological characteristics of the grounding wire device, including: Extract the morphological features of the grounding wire device from the identified locations; The extracted morphological features are matched with multiple preset state discrimination rules to determine the hanging and dismantling status of the grounding wire device.

[0134] In this embodiment, after successfully locating the wiring device, its "morphological features" directly related to operational safety are further extracted. These features are closely related to the physical state of the device (such as the opening and closing of the grippers and the overall posture). By matching these quantified morphological features with predefined "state discrimination rules", image information can be transformed into semantic states with clear operational guidance significance, such as "reliably connected" and "connected but not locked", thus achieving a deep understanding of the operation process.

[0135] In this embodiment, the hanging and removing operation status of the grounding wire device includes: Reliably connected; connected but not locked; not connected.

[0136] In this embodiment, "reliably connected" indicates that the grounding device has been correctly and securely installed on the conductor, which is a permissible state for safe operation; "connected but not locked" indicates that the device has contacted the conductor but is not fully locked, posing a risk of detachment or poor contact, which is an intermediate state requiring warning; "not connected" indicates that the device is not in effective contact with the conductor, which is a typical dangerous state. The ability to distinguish between these three states is the core of ensuring operational safety.

[0137] In this embodiment, the morphological features of the grounding wire device include: The aspect ratio of the rectangle surrounding the grounding device; the pixel opening degree of the gripper area of ​​the grounding device.

[0138] In this embodiment, the extracted morphological features are matched with multiple preset state discrimination rules to determine the hanging and removing status of the grounding wire device, and the process further includes: The timing logic is verified by combining the discrimination results of multiple consecutive frames of images prior to the current moment.

[0139] This embodiment analyzes the discrimination results of multiple consecutive frames of images and smooths or corrects the discrimination result of the current frame. This effectively filters out accidental recognition errors, ensuring that the output work status has temporal continuity and consistency, which better reflects the continuous characteristics of the actual work process.

[0140] In this embodiment, the visual recognition and image enhancement method for grounding wire hanging and detaching further includes: Based on the determination of the hanging and removing status of the grounding wire device, the corresponding operation control signal or status alarm information is generated and output.

[0141] In this embodiment, corresponding instructions are generated based on the determined status. For example, after determining that the connection is "reliable", a "work permission" signal can be output, allowing the drone to perform the next work task (such as flying away or performing work at the next connection point); when the connection is determined to be "not locked" or "not connected", an "alarm" signal is immediately generated to notify the operator through sound, light, text, etc., and the current work process may be automatically paused, waiting for manual intervention and inspection, thereby directly transforming visual monitoring into an active control force to ensure work safety.

[0142] In this embodiment, as Figure 4 , Figure 5 As shown, the real-time monitoring system for grounding wire operations based on UAV vision includes an airborne subsystem and a ground subsystem that communicate with each other: The airborne subsystems include: The image acquisition module is used to acquire video streams or image sequences of the grounding wire operation area of ​​transmission lines; The processing and recognition module is used to process image frames in video streams or image sequences, identify grounding devices and determine their operating status, and output the recognition results. The first wireless communication module is used to send information containing the identification results to the ground subsystem; The ground subsystem includes: The second wireless communication module is used to receive information from the airborne subsystem; The information processing module is used to parse information and generate monitoring instructions; The monitoring and display module is used to display the operation status and related on-site images based on monitoring commands.

[0143] The image acquisition module includes a visible light camera with optical image stabilization, mounted on the drone's gimbal to ensure stable and clear image capture of the work area even when the drone is hovering or affected by wind. The processing and recognition module, integrating an embedded processor and algorithm software, is the intelligent core of the system. The first and second wireless communication modules preferably employ digital transmission protocols with strong anti-interference capabilities, such as a COFDM-based image transmission system, to adapt to the complex electromagnetic environment surrounding high-voltage transmission lines and ensure the stability and low latency of the monitoring data link. The monitoring display module's interface design integrates video display, status overlay, alarm prompts, and basic control functions.

[0144] In this embodiment, the airborne subsystem carried by the UAV is responsible for real-time acquisition and intelligent analysis of on-site images. It integrates image acquisition, embedded intelligent processing, and wireless transmission functions, enabling real-time visual analysis at the work site, reducing bandwidth pressure on data transmission and lowering processing latency. The ground subsystem serves as a remote monitoring terminal, responsible for receiving, parsing, and visualizing information. This allows ground monitoring personnel to clearly and in real-time grasp the precise status of the grounding device (such as whether it is connected and whether the connection is reliable) from the remote monitoring center without having to be physically present at high-altitude risk points. This achieves penetrating and non-contact safety supervision of high-risk operations.

[0145] In this embodiment, image frames in a video stream or image sequence are processed to identify grounding devices and determine their operating status, and the identification results are output, including: Adaptive image enhancement is performed on the input image frame to obtain the enhanced image; Based on enhanced images, identify and locate conductor and tower structures; Based on the conductor and tower structure, the target area containing the grounding wire device is determined; Visual features of the target area are extracted, and the operating status of the grounding wire device is determined based on a preset state classification model to obtain the recognition result.

[0146] This embodiment deeply integrates general visual recognition tasks with prior knowledge of power operation scenarios: First, adaptive image enhancement improves the quality of the original image, overcoming interference from lighting and weather. Second, the inherent conductor and tower structures of the power transmission scenario serve as "visual landmarks," enabling scene understanding and coarse localization. Then, based on prior knowledge (grounding wires are always near the connection points between conductors and towers), it quickly focuses on a small area that may contain the target, greatly improving processing efficiency and eliminating irrelevant background interference. Finally, robust features are extracted within the precisely located area, and a pre-trained classification model makes the final state decision. This streamlined design ensures the clarity of the system's processing logic and the repeatability of the results.

[0147] In this embodiment, adaptive image enhancement is performed on the input image frame to obtain an enhanced image, including: Calculate at least one global statistical feature of the input image frame; The environment category to which the image frame belongs is determined by comparing global statistical features with a preset threshold. Based on the environment category, the corresponding preset image processing algorithm is invoked to process the image frame.

[0148] In this embodiment, based on the environment category, the corresponding preset image processing algorithm is invoked to process the image frame, including: When the environment is classified as low-light or high-noise, the preset image processing algorithm is a contrast-limited adaptive histogram equalization algorithm, and the specific steps include: The input image frame is divided into multiple consecutive and non-overlapping sub-blocks; For each sub-block, calculate its grayscale histogram and crop the grayscale histogram using a preset cropping limit value; Perform equalization processing on the cropped grayscale histogram; Perform bilinear interpolation on the processed image of each sub-block; The interpolated image is then smoothed to obtain the enhanced image.

[0149] This embodiment employs an adaptive enhancement mechanism to address the variable lighting and noise susceptibility of power field operations. A contrast-limited adaptive histogram equalization algorithm is used to handle low-light and high-noise scenarios. Adaptive image enhancement is achieved through block segmentation, local processing, and interpolation fusion, ensuring the algorithm's robustness when dealing with complex backgrounds and foregrounds of power transmission lines. Specifically, cropping limits prevent over-enhancement of local areas and the resulting artifacts; bilinear interpolation eliminates boundary effects between sub-blocks, ensuring a smooth and natural enhanced image.

[0150] In this embodiment, based on the conductor and tower structure, the target area containing the grounding wire device is determined, including: Edge detection is performed on the enhanced image frame to obtain the edge image; In the edge image, candidate line segments belonging to conductors and towers are selected based on the length and direction features of the line segments; Based on the relative positions and geometric relationships of candidate line segments, the positions of the conductor structure and the tower structure in the image are determined respectively.

[0151] In this embodiment, based on the relative positions and geometric relationships of candidate line segments, the positions of the conductor structure and the tower structure in the image are determined, including: Based on the intersection or nearest neighbor of the identified conductor edge and tower edge, at least one candidate anchor point is determined; Candidate boxes are dynamically generated based on preset size rules, with the candidate anchor point as the center, and the candidate box containing the most edge features is determined as the target region.

[0152] This embodiment utilizes the rigid geometry of transmission lines as navigation beacons for locating grounding wire devices. Through edge detection and line segment filtering, conductors and towers are first identified in the image. Then, the algorithm finds the intersection or nearest point of these two types of structures, which is highly likely to be the installation location of grounding wires such as insulator strings or hanging points. Using this point as an "anchor," a pre-defined rectangular area is dynamically generated as the target area for focused analysis. This scene-based prior knowledge-based positioning method avoids the huge computational overhead of traditional sliding window or full-image search methods, enabling the system to quickly and accurately lock visual attention on key areas, significantly improving the real-time performance of monitoring.

[0153] In this embodiment, visual features of the target area are extracted, and the operating status of the grounding wire device is determined based on a preset state classification model to obtain the recognition result: The extraction of visual features of the target region specifically involves calculating the gradient direction histogram features of the target region; The preset state classification model is a support vector machine model; The preset state classification model is a convolutional neural network model.

[0154] Within a precise target area, features that effectively characterize the grounding device and its state need to be extracted. Histogram of Gradient Orientations (HOG) features, by statistically analyzing the distribution of gradient directions within a local image region, can effectively describe the edge and shape information of the target and exhibits some invariance to changes in illumination, making it ideal for depicting the metal outline and jaw shape of the grounding clamp. After extracting HOG features, they are input into a pre-trained classification model for state discrimination. Support Vector Machines (SVMs) are a classic machine learning classifier with good generalization ability in small sample sizes. The model is relatively lightweight and has fast inference speed, making it ideal for deployment on computationally limited embedded platforms (such as UAV onboard computers) to achieve real-time state classification. Convolutional Neural Networks (CNNs) are representative models of deep learning, possessing powerful automatic feature extraction and learning capabilities. Using CNNs, image patches of the target region can be directly used as input, and the network automatically learns feature representations from low-level edges to high-level semantics, typically achieving higher recognition accuracy when data is abundant. Users can choose or switch between SVMs and CNNs based on their actual requirements for system real-time performance, accuracy, and hardware resource configuration. Both models require offline training using a large number of labeled grounding device images (covering various states, angles, and lighting conditions). The trained model files are then loaded into the processing and recognition module for use.

[0155] In this embodiment, the gradient orientation histogram feature can effectively describe the edge and contour shape of the grounding wire device, exhibiting a certain degree of stability to changes in illumination, making it highly suitable for characterizing the geometry of metal components such as grippers and hooks. The support vector machine model performs excellently on small-sample, high-dimensional feature classification problems, offering high computational efficiency and making it suitable for deployment on embedded platforms with limited computing power. The convolutional neural network model possesses stronger feature self-learning and abstraction capabilities, enabling it to learn more complex patterns from original images, and is suitable for larger and more diverse datasets. These two models provide the system with optional technical paths to adapt to different resource constraints and accuracy requirements.

[0156] In this embodiment, the ground subsystem further includes: The storage module is used to store the received job status, associated image frames and system logs in chronological order, and supports retrieval based on time or status.

[0157] The storage module, based on a database, enables traceability and closed-loop management. All data received through the second wireless communication module, including timestamps, UAV numbers, tower numbers, identified operational statuses (e.g., "not attached," "reliable attachment," "abnormal attachment"), status confidence levels, and corresponding on-site image snapshots or keyframes, are structured and stored in the database. Simultaneously, system operation logs and alarm events are also recorded. This module provides retrieval and playback functions. Ground management personnel can perform combined queries based on task date, tower number, and operational status type to quickly locate historical operational records and view the on-site footage and system judgment results. This not only provides intuitive and reliable electronic evidence for verifying operational quality and tracing accident causes but also accumulates valuable on-site data for subsequent optimization of identification algorithms and analysis of operational behavior. This embodiment provides a complete digital archive of the work process through a storage module. All timestamped work statuses, on-site images, and system operation logs are stored systematically, forming a traceable chain of evidence. In the event of an anomaly or when a review analysis is required, supervisors can quickly retrieve and replay the work situation at any point in time, providing a solid data foundation for accurately locating problems, assigning responsibilities, and optimizing work procedures, thus achieving a closed loop in safety management.

[0158] In this embodiment, a real-time monitoring system for grounding wire operations based on UAV vision is used to monitor the operation process of hanging and removing grounding wires in real time, and transmit the operation status and associated on-site images to the ground terminal. The specific operation steps are as follows: Controlling drones relies on their onboard subsystems: The image acquisition module acquires video streams or image sequences of the power transmission line grounding wire operation area. Based on the processing and recognition module, the image frames in the video stream or image sequence are processed to identify the grounding wire device and determine its working status, and the recognition result is output. The first wireless communication module transmits information containing the identification results to the ground subsystem. At the ground station via the ground subsystem: The second wireless communication module receives information from the airborne subsystem. The information processing module parses the information and generates monitoring instructions. The monitoring and display module displays the operation status and related on-site images based on monitoring commands.

[0159] The specific operating procedures for power transmission line maintenance are as follows: Before the operation begins, the ground operator plans the drone's flight path, guiding it to the vicinity of the grounding wire work point on the tower to be inspected and hovering stably. Once the drone is in position, the system enters automatic monitoring mode. The onboard camera continuously captures images of the work point, and the onboard processing unit analyzes each frame or key frame in real time according to the aforementioned process. For example, during the grounding wire installation process, the system continuously outputs a "not installed" status; when the operator uses the mechanical device to install the grounding wire, the system recognizes that the grounding clamp is closed on the conductor, and the status updates to "reliable installation," this result along with the real-time image is quickly transmitted back to the ground station. Ground monitoring personnel can clearly see the grounding wire device marked with a green box on the monitoring screen, displaying the words "reliable installation" and the confidence level. Throughout the process, ground personnel do not need to climb the tower to confirm and supervise the results of the high-altitude operation from a remote, safe location.

[0160] This embodiment seamlessly integrates UAV flight control, onboard intelligent analysis, and ground monitoring and command, forming a repeatable and verifiable operating procedure. Operators simply need to follow the procedure to guide the UAV to the designated location, and the system automatically executes the entire chain of tasks from image acquisition to status feedback, presenting the results intuitively to ground personnel. This method encapsulates a complex technical system into simple operating steps, lowering the barrier to entry and facilitating the promotion and application of this technology in maintenance teams.

[0161] Example 2 This embodiment provides a complete set of grounding wire operation system based on UAV for maintenance work in high-voltage transmission lines, which uses the complete set of operation method in Embodiment 1 for operation.

[0162] like Figures 6-36As shown, the system includes a drone 1, a mounting device, and a voltage testing device. The mounting device includes an angle steel end mounting device and a conductor end mounting device that can be mounted on the drone 1. The drone 1 transports a pair of grounding wire mounting devices to the angle steel tower and the transmission line respectively, and the devices are mounted on the angle steel of the angle steel tower and the transmission line respectively, so as to achieve grounding of the transmission line, release residual voltage, ensure the safety of workers on the angle steel tower, and prevent workers from being electrocuted.

[0163] After the operation is completed, the grounding wire installation device is still removed from the angle steel and power transmission line by the drone 1. In this embodiment, the grounding wire installation and removal operation is carried out by the drone 1. A single person can complete the voltage testing and grounding wire installation and removal operation, which saves operation time, improves operation efficiency, and reduces the safety risks of the operation.

[0164] This embodiment provides an electrical testing device, the specific structure of which is as follows: Figures 6-13 As shown, The voltage detection device includes a voltage detection mechanism 3 installed on a drone 1. The drone 1 includes a body and a handheld terminal. The body and the handheld terminal are respectively provided with a body communication module and a handheld terminal communication module that can communicate with each other. The body communication module and the handheld terminal communication module communicate wirelessly with each other. The voltage detection mechanism 3 includes an insulating rod 301 with its end installed on the drone 1. A photosensitive receiver 302 and an voltage detector 303 connected to the photosensitive receiver 302 are installed on the end of the insulating rod 301 away from the drone 1. The electroscope 303 includes a probe 307 and an electroscope audible and visual alarm device for emitting an audible and visual alarm signal based on the detection result of the probe 307; the photosensitive receiver 302 includes a photosensitive sensor for detecting the audible and visual alarm signal and sending the detected signal to the body communication module; the insulating rod 301 is a hollow rod, and wires for electrically connecting the UAV 1 and the photosensitive receiver 302 are arranged inside it; the handheld terminal also includes a handheld audible and visual alarm device for emitting an audible and visual alarm signal based on the signal received by the handheld terminal communication module.

[0165] In this embodiment, the electroscope 303 is a capacitive electroscope 303, which can detect whether the line is energized. If the line is energized, the electroscope 303 will emit an audible and visual alarm signal. At this time, the photosensitive receiver 302 can collect the light signal emitted by the electroscope 303 through the photosensitive sensor, and then transmit the collected information to the machine body through the machine body communication module. The machine body transmits the information to the handheld terminal communication module through wireless communication. At this time, the handheld terminal will simultaneously emit an audible and visual alarm signal so that ground personnel can judge in real time whether the line to be tested is energized.

[0166] By using this drone control method, workers no longer need to climb high into the air to test electricity, effectively reducing the workload of operators, improving work efficiency, and reducing work risks.

[0167] The voltage detector 303 includes a voltage detector body 304 and a guide frame 305 disposed at the end of the voltage detector body 304. The guide frame 305 consists of two guide arms 306, one end of which is connected to each other and fixed to the voltage detector body 304, and the other end extends outward to form a V-shaped frame. A voltage detection area is formed between the two guide arms 306. The probe 307 extends from the voltage detector body 304 into the voltage detection area through the connection of the two guide arms 306. In addition, the outer surfaces of the two guide arms 306 are provided with reinforcing arms 309 arranged along the length direction of the guide arms 306. The setting of the reinforcing arms 309 can effectively increase the structural strength of the guide frame 305 and prevent it from bending or deforming, which would affect the voltage detection effect.

[0168] The V-shaped guide frame 305 can guide the wire, allowing the wire to accurately enter the voltage detection position, effectively solving the problem that the voltage detector 303 cannot make accurate contact with the wire due to the inability of the UAV 1 to perform precise operation.

[0169] By setting a V-shaped guide frame 305 at the front end of the voltage detector 303, once the wire enters the opening range of the guide frame 305, it can enter the voltage detection area along the guide arm 306. The angle formed by the guide arm 306 and the probe 307 of the voltage detector 303 can restrict the position of the wire, ensuring that the wire can always contact the probe 307 of the voltage detector 303 and will not swing randomly in the guide fork.

[0170] By setting a guide frame 305 at the front end of the voltage detector 303 to form a contact guiding mechanism for the wire, the fault tolerance rate of the operation can be effectively increased, the complexity of the operation of the UAV 1 can be reduced, the operation process can be simplified, the operation time can be shortened, and the practicality of the device can be improved.

[0171] In this embodiment, a conductive wire 308 is installed on the inner side of each of the two guide arms 306. One end of the conductive wire 308 is fixed to the extension end of the guide arm 306, and the other end is fixed to the side of the probe 307, forming a conductive connection with the probe 307. The conductive wire 308 forms a conductive connection with the probe 307, and the other end is fixed to the end of the guide arm 306 away from the probe 307, so that the conductive wire 308 forms an outward-facing detection area. This allows the wire to enter the detection area and contact the conductive wire 308 to achieve guidance detection, effectively increasing the detection contact range, further improving the operational error tolerance, and reducing operational difficulty.

[0172] In use, by operating the insulating rod 301, the conductive wire 308 is fitted onto the outer periphery of the overhead conductor to form a detection area until the conductive wire 308 comes into contact with the outer periphery of the overhead conductor. Since the outer end of the conductive wire 308 is fixed to the end of the guide arm 306, the detection area is located inside the voltage detection area. During the process of the guide arm 306 being fitted onto the outer periphery of the overhead conductor, the conductor can also avoid the outer end of the conductive wire 308.

[0173] Meanwhile, this method, which uses the guide arm 306 and the electroscope 303 probe 307 to form an angle for detection, and further improves the detection contact range through the conductive wire 308, can effectively avoid the problem of the probe 307 getting stuck in the wire compared to the traditional form of voltage detection where the probe 307 contacts the wire.

[0174] In use, simply operate the drone 1 to adjust the angle of the insulating rod 301 so that the length of the insulating rod 301 points towards the conductor. Then, slide the insulating rod 301 along its own axis to cover the detection area around the cable, which improves the ease of operation of the device.

[0175] In this embodiment, the photosensitive receiver 302 includes a receiver housing 310, and the receiver housing 310 is provided with a battery mounting slot 334. The battery mounting slot 334 is provided with a power supply battery that can supply power to the photosensitive receiver 302.

[0176] One end of the receiver housing 310 forms a voltage detector connection end, and the other end forms an insulating rod connection end. The voltage detector connection end and the insulating rod connection end are respectively provided with a voltage detector connector 310 and an insulating rod connector 311 for the voltage detector 303 to be inserted and the insulating rod 301 to be inserted.

[0177] The electroscope connector 310 includes a plug body 312 and a rotating disk 313 disposed at one end of the plug body 312. A rotating disk locking block 314 is provided on the inner side wall of the rotating disk 313. A rotating head 315 is provided at the end of the receiver housing 310. A rotating head locking block 316 that cooperates with the rotating disk locking block 314 is provided on the outer side wall of the rotating head 315. The rotating disk 313 is rotatably locked onto the rotating head 315. An electroscope slot 317 for the end of the electroscope 303 to be inserted is provided at the other end of the plug body 312.

[0178] The insulating rod connector 311 includes a fixing plate 318 fixed to the end face of the receiver housing 310. The fixing plate 318 has an arc-shaped plate 319 on its surface. An arc groove with an upward opening is formed in the middle of the arc-shaped plate 319. An upper limit plate 333 is fixed on the upper end face of the arc plate 319. The lower end face of the upper limit plate 333 and the arc groove together form an insulating rod insertion groove 320. The end of the insulating rod 301 is inserted into the insulating rod insertion groove 320 and its end face is fixed on the fixing plate 318.

[0179] The insulating rod 301 and the electroscope 303 are installed at both ends of the photosensitive receiver 302 by plugging them in. During installation, they can be plugged in and pre-fixed first, and then tightened with bolts to achieve detachable and stable fixing. This ensures sufficient connection stability and allows for convenient replacement of parts.

[0180] When installing the voltage detector 303, the rotating disc 313 can be connected to the rotating head 315 of the receiver housing 310 via the rotating disc 313 wire sleeve. Then, rotate the rotating disc 313 so that the rotating disc locking block 314 and the rotating head locking block 316 are locked together, thus completing the installation of the voltage detector connector 310 at the end of the optical receiver housing. After it is accommodated, insert the end of the voltage detector 303 into the voltage detector 303 plug to complete the installation of the voltage detector 303. After the insertion is completed, bolts can be added for further tightening to avoid separation due to unstable insertion.

[0181] In this embodiment, to enable the installation of the voltage testing mechanism 3 on the drone 1, a voltage testing mechanism mounting frame 321 is also provided on the top surface of the drone 1. The voltage testing mechanism mounting frame 321 includes a mounting frame base plate 322 fixed to the top surface of the drone 1. The mounting base plate 322 is provided with a rod end connecting seat 323 for connecting the end of the insulating rod 301. The insulating rod 301 can rotate around the rod end connecting seat 323 to be parallel or perpendicular to the mounting base plate 322. The mounting base plate 322 is provided with a limiting component for keeping the insulating rod 301 parallel or perpendicular to the mounting base plate 322.

[0182] In this embodiment, the installation position of the voltage testing mechanism mounting bracket 321 and its ability to limit the insulation rod 301 in both horizontal and vertical directions enable the switching of the voltage testing mechanism 3 position and the rapid replacement of horizontal and vertical voltage testing positions to meet the voltage testing needs of conductors with different arrangements.

[0183] Due to the complexity of overhead power lines, multiple conductors are often distributed in parallel intervals, including horizontally spaced, vertically spaced, or randomly distributed configurations. In these different conductor distribution states, a single fixed angle between the insulating rod 301 and the drone 1 is clearly insufficient. For example, if the conductors are horizontally spaced, and the insulating rod 301 is horizontally mounted on the drone 1, the conductors on the innermost side cannot be detected. Similarly, if the conductors are vertically spaced, and the insulating rod 301 is vertically mounted on the drone 1, the conductors on the innermost side also cannot be detected. Furthermore, even if the insulating rod 301 is installed at an angle on the drone 1, interference with adjacent conductors may occur. Therefore, this embodiment employs a voltage detection mechanism mounting bracket 321 with a specific structure, allowing the insulating rod 301 to rotate and adjust around a horizontal axis perpendicular to the axis of the insulating rod 301. This enables adjustment of the mounting angle of the insulating rod 301 on the drone 1.

[0184] Since the insulating rod 301 can be adjusted to a specific angle relative to the drone 1, it can avoid other wires when the drone 1 is hoisted. This means that the voltage detector 303 arranged at the mounting end of the insulating rod 301 does not have to be a traditional contact voltage detector 303 with a probe 307. By arranging a non-contact voltage detector 303 at the mounting end of the insulating rod 301, accurate live testing of the wires can also be achieved.

[0185] The mounting bracket 321 for the voltage testing mechanism and its adjustment method are as follows: The rod end connector 323 includes a connector base plate 324 fixed to one end of the mounting bracket base plate 322. The connector base plate 324 has connecting ear plates 325 arranged opposite to each other on both sides. The end of the insulating rod 301 is provided with a connector head 326 connected to the two connecting ear plates 325 by a pin. The connector head 326 includes a U-shaped rotating plate 327. The end of the insulating rod 301 is fixed to the outer end face of the U-shaped rotating plate 327. The two side walls of the U-shaped rotating plate 327 are connected to the two connecting ear plates 325 by a pin. The two side walls of the U-shaped rotating plate 327 and the two connecting ear plates 325 are also provided with limiting holes 328 that can be bolted to keep the insulating rod 301 and the mounting bracket base plate 322 in a vertical state. The other end of the mounting base plate 322 is also fixed with a rod fixing seat 329. The rod fixing seat 329 includes an upper fixing part 330 and a lower fixing part 331 that are connected to each other by bolts. A limiting circular hole 332 is formed between the upper fixing part 330 and the lower fixing part 331 for the insulating rod 301 to pass through, which is used to keep the insulating rod 301 parallel to the mounting base plate 322.

[0186] The connector 326 is rotatably connected to the rod end connector 323 via a pin. During the adjustment process, when the insulating rod 301 needs to be in a horizontal state parallel to the top surface of the UAV 1, the rod body of the insulating rod 301 is horizontally inserted into the lower fixing part 331, and then the upper fixing part 330 limits the rod body so as to keep the rod body in a horizontal state through the axially horizontal limiting round hole 332.

[0187] When the insulating rod 301 needs to be parallel to the top surface of the UAV 1, the connector 326 rotates upward around the pin axis. When the connector 326 rotates to the point where its opening is directly opposite the rod end connector 323, the insulating rod 301 is in a vertical state and the limiting hole 328 on the connector 326 is opposite to the limiting hole 328 on the connecting ear plate 325 above the pin axis. At this time, the side walls of the connecting ear plate 325 and the U-shaped rotating plate 327 can be further tightened by bolts to keep the insulating rod 301 in this vertical state.

[0188] In this embodiment, a conductive wire 308 bent into a V-shape, W-shape, or semi-circular arc shape can also be installed between the two guide arms 306. The opening direction of the conductive wire 308 is consistent with the opening direction of the guide frame 305. The middle part of the conductive wire 308 is electrically connected to the probe 307, and the two ends are respectively fixed to the inner side of the two guide arms 306.

[0189] The conductive wire 308 is constructed with open detection areas of different shapes, which can make contact with the inner wall of the conductive wire 308 when wires of different diameters slide into the inner cavity of the detection area, thus ensuring adaptability to different wire detection operations.

[0190] like Figures 14-28 As shown, in this embodiment, the mounting device is mounted on the UAV via a mounting frame 100. Angle steel end hanging ring 210 and wire end hanging ring 09 are respectively installed on the upper ends of the angle steel end mounting frame 201 and the wire end mounting frame 011. The bottom of the mounting frame 100 is provided with two hooks 200 for hanging the angle steel end hanging ring 210 and the wire end hanging ring 09.

[0191] The mounting frame 100 includes two vertically arranged and parallel rod assemblies 300. One end of each rod assembly 300 is mounted on the UAV 1, and the other end is used to mount a hook 200. A connecting crossbar 400 is provided between the two rod assemblies 300 and at the hook mounting end. Each rod assembly 300 includes multiple sequentially connected vertical connecting rods 500, and adjacent vertical connecting rods 500 are connected by hinges 600. The structural form of the rod assembly 300 allows its length to be adjusted according to actual needs. By assembling or disassembling the vertical connecting rods 500, the length of the rod assembly 300 can be changed to suit different requirements.

[0192] The mounting bracket 100 also includes a mounting frame 700 installed on the bottom surface of the drone. The mounting frame 700 includes two long connecting plates 800 bolted to the side of the drone 1. The two long connecting plates 800 are arranged parallel to each other, and a connecting frame 900 at the same horizontal plane as the long connecting plates 800 is provided between the two long connecting plates 800. A hinge seat 1000 is installed on the connecting frame 900, and the end of the rod assembly 300 is hinged to the hinge seat 1000. The mounting bracket 100 can be detachably installed on the drone body via the mounting frame 700, and installation and removal are both convenient.

[0193] This embodiment provides a device for hanging angle steel ends, the specific structure of which is as follows: The mounting device includes a device body 20, which includes an angle steel end mounting frame 201 and a gripper drive assembly 202.

[0194] The angle steel end hanging frame 201 includes a fixed clamp 203 and a movable clamp 204. The angle steel end hanging frame 201 includes a U-shaped frame 209 with an opening facing downward. The U-shaped frame 209 includes a horizontal arm and two vertically extending vertical arms 221 and 222. The vertical arm 221 forms the fixed clamp 203. The movable clamp 204 includes a clamp body 10. The clamp body 10 has a vertical clamping and limiting surface 103 opposite to the fixed clamp 203. The clamp body 10 is arranged opposite to the fixed clamp 203 and can move towards or away from the fixed clamp 203 in the horizontal direction.

[0195] Since the entire device needs to be carried to a high altitude by a drone, the mounting device should not be too heavy. The mounting device is used for grounding wires and needs to be made of metal. In this application, it is basically made of welded steel plates or bolted connections. Therefore, the structure of the mounting device needs to be designed to reduce its weight as much as possible.

[0196] In this embodiment, the hanging structure is designed to consist of two symmetrically arranged frame side plates 227 and multiple connecting rods 228 connecting the two frame side plates 227. The frame side plates 227 include U-shaped plates 229 with downward openings. The U-shaped plates 229 in the two frame side plates 227 and the connecting rods 228 connecting the U-shaped plates 229 together constitute a U-shaped frame 209.

[0197] On the one hand, this type of frame reduces the end face sealing plate and is connected by only two frame side plates 227 and connecting rods 228, which can greatly reduce its weight. On the other hand, the frame side plates 227 are also perforated to further reduce the self-weight of the frame 201 hanging at the angle steel end.

[0198] This embodiment provides a gripper drive assembly 202 for driving the movable gripper 204 to move toward or away from the fixed gripper 203. The gripper drive assembly 202 includes a component one for driving the movable gripper 204 to move toward the fixed gripper 203 and a component two for driving the movable gripper 204 to move away from the fixed gripper 203. The component one includes an elastic member 206 that can pull the movable gripper 204. The elastic member 206 provides a pulling force on the movable gripper 204 by its own elasticity. The elastic element 206 provides tension on the movable gripper 204 by its own elasticity. The elastic element 206 includes a mounting rod 207 and a coil spring 208 mounted on the mounting rod 207. One end of the coil spring 208 is fixedly connected to the movable gripper 204. The mounting rod 207 is fixed on the vertical arm 221. The coil spring 208 includes a coil roller and a spring sheet wound on the coil roller. One end of the spring sheet is fixed on the coil roller, and the other end is fixed on the gripper body 10 of the movable gripper 204. The coil roller can rotate on the mounting rod 207.

[0199] Component 2 includes a rigid rope 205 that can pull the movable gripper 204. The rigid rope 205 provides a pulling force on the movable gripper 204 by the weight of the device body 20. It also includes an angle steel end hanging ring 210 that is set on the top of the angle steel end hanging frame 201 and can move up and down relative to the angle steel end hanging frame 201. The two ends of the rigid rope 205 are respectively connected to the angle steel end hanging ring 210 and the movable gripper 204. When the angle steel end hanging ring 210 moves upward relative to the angle steel end hanging frame 201, it can pull the movable gripper 204 away from the fixed gripper 203 through the rigid rope 205.

[0200] In this embodiment, the gripper frame 104 includes an upper end plate 112, a lower end plate 113, and two side plates 114. Both side plates 114 are U-shaped plates 229 and are arranged opposite to each other. The upper end plate 112 is provided with sliding connecting plates 235 on both sides. Both side plates 227 are fixed with sliding groove plates 234 forming a groove in the middle. The length direction of the groove is set along the movement direction of the movable gripper 204. The sliding connecting plate 235 is provided with rollers 236 that can move in the groove. Based on the groove and the rollers 236, the movable gripper 204 can achieve more stable reciprocating motion.

[0201] In this embodiment, a lifting ring mounting bracket 211 is installed on the top of the angle steel end hanging frame 201. The lifting ring mounting bracket 211 includes a lifting ring mounting arm 212. Both ends of the lifting ring mounting arm 212 are connected to a vertically downward extending guide rod 213. The angle steel end hanging frame 201 is provided with a guide hole 214 for the guide rod 213 to pass through and slide in cooperation with the guide rod 213. The bottom end of the guide rod 213 is provided with a limiting ring 215 with an outer diameter larger than the diameter of the guide hole 214. The setting of the limiting ring 215 can effectively prevent the guide rod 213 from completely detaching from the angle steel end hanging frame 201.

[0202] The lifting ring mounting arm 212 is also fixed with a lifting ring mounting seat 216 for fixing the angle steel end lifting ring 210. The lifting ring mounting seat 216 includes a mounting seat base plate 217 fixed to the lifting ring mounting arm 212 by bolts and a snap-fit ​​block 218 set on the mounting seat base plate 217. The snap-fit ​​block 218 has a snap-fit ​​groove 219 with an upward opening in the middle. The bottom of the angle steel end lifting ring 210 is snapped into the snap-fit ​​groove 219 and fastened by bolts. The axial direction of the angle steel end lifting ring 210 snapped into the snap-fit ​​groove 219 is the same as the movement direction of the movable gripper 204.

[0203] During installation, the angle steel on the outside of the angle steel tower is usually selected for installation. The axial direction of the angle steel end lifting ring 210 is the same as the movement direction of the movable jaw 204, that is, the thickness direction of the angle steel end lifting ring 210 is the same as the length direction of the angle steel. This allows the angle steel end lifting ring 210 to be close to the angle steel tower from the side during installation, which can minimize the interference of the angle steel end lifting ring 210 on the installation process.

[0204] The lifting ring mounting base 216 is located at the middle position of the lifting ring mounting arm 212. The length direction of the lifting ring mounting arm 212 is set along the length direction of the angle steel end hanging frame 201, so that the angle steel end lifting ring 210 is basically located on the center line of the entire angle steel end hanging frame 201, thereby ensuring the hanging stability of the angle steel end lifting ring 210.

[0205] In addition, the angle steel end hanging frame 201 is provided with at least two tensioning rods 220 for tensioning the rigid rope 205. The rigid rope 205 is wound around the tensioning rods 220. The tensioning rods 220 can tension the rigid rope 205 on the one hand, and directly serve as connecting rods 228 between the two frame side plates 227 on the other hand. One of the tensioning rods 220 is located in the lower part of the angle steel end lifting ring 210, which can serve as the middle connecting rod 228 between the two frame side plates 227. By replacing the connecting rods 228 with tensioning rods 220, the number of connecting rods 228 can be reduced, which can improve the assembly efficiency of the entire device and further control the weight of the device.

[0206] Between the two frame side plates 227, there is also a long strip connecting plate 237 for installing guide rods 213. The long strip connecting plate 237 is set at both ends of the top of the frame side plate 227, and it can directly replace the connecting rod 228 at that position. The middle of the long strip connecting plate 237 is provided with a sliding hole for the guide rod 213 to pass through. The way the connecting plate replaces the connecting rod 228 makes the load-bearing device more stable and has a larger load during operation.

[0207] In this embodiment, the structure of the bracket 201 hanging at the end of the angle steel is fully designed, and the positions of each rod and plate are optimized so that its shape, weight, stress and other aspects are kept in a better state. It has the characteristics of exquisite structure, stable operation and full ability to meet the operation requirements.

[0208] The specific work process is as follows: In the initial state, the drone hooks onto the lifting ring 210 at the angle steel end. At this time, a grounding wire is connected to the angle steel end mounting frame 201. The drone drives the device to the corresponding angle steel tower position. During the drone's operation and before the mounting device is installed, the lifting ring 210 at the angle steel end is at its highest position relative to the angle steel end mounting frame 201, and the guide rod 213 is extended out of the angle steel end mounting frame 201. Under the pulling action of the lifting ring 210, the movable gripper 204 is pulled open to the furthest point from the fixed gripper 203 via the rigid rope 205. At this time, the installation opening is in its maximum state, and the coil spring 208 is in its maximum extended state.

[0209] Continue lowering the angle steel end mounting frame 201 until the angle steel enters through the installation port and the angle steel end mounting frame 201 lands on the angle steel. At this point, continue to control the drone to descend slowly. Since the angle steel end mounting frame 201 is supported, as the drone descends, the rigid rope 205 on the angle steel end lifting ring 210 no longer provides tension. The movable gripper 204 can then gradually move towards the fixed gripper 203 under the elastic tension of the coil spring 208 until the angle steel is completely clamped. This completes the installation of the angle steel end mounting frame 201 on the angle steel. At this point, the hook on the drone can be operated to detach from the angle steel end lifting ring 210. The drone will automatically detach as it descends further, and the drone will return to base while the workers on the angle steel tower carry out their work.

[0210] After the workers finish their work, they leave the angle steel tower. The drone then returns to the designated position and hooks the angle steel end lifting ring 210 again. The drone only needs to move upwards to pull the rigid rope 205 through the angle steel end lifting ring 210. This causes the movable gripper 204 to overcome the spring force of the coil spring 208 and move away from the fixed gripper 203, releasing the clamp on the angle steel. Then, the entire angle steel end hanging frame 201 is directly pulled up, completing the removal of the angle steel end hanging frame 201 from the angle steel tower. This removal method does not require any electric drive; it can be directly pulled up by the drone, eliminating the need to consider the power source. It is also safer and easier to operate.

[0211] The mounting device in this embodiment not only facilitates dismantling after the work is completed, but also facilitates mounting on the angle steel tower.

[0212] Specifically, in this embodiment, the lower ends of vertical arm 1 221 and vertical arm 222 are respectively provided with extension arm 1 223 and extension arm 224, and both extension arm 1 223 and extension arm 224 are inclined towards the outside of the U-shaped frame 209. The first extension arm 223 and the first vertical arm 221 form an angle of 5-10°, and the second extension arm 224 and the second vertical arm 222 form an angle of 45-60°. In this embodiment, the first extension arm 223 and the first vertical arm 221 form an angle of 5°, and the second extension arm 224 and the second vertical arm 222 form an angle of 60°. The lower end of the first extension arm 223 is provided with a terminal block 225 for connecting a grounding wire.

[0213] The extension arm 1 223 and the extension arm 2 224 are tilted outward at different angles. On the one hand, the bottom of the extension arm 1 223 is connected to the ground wire, which has a certain weight. By using the extension arm 1 223, which is basically vertically extended, the extension arm 1 223 maintains a similar traction angle with the ground wire when the device is running upward. This reduces the tilting of the entire device caused by the extension arm 1 223 flipping downward and the extension arm 2 224 flipping upward during the hanging device. On the other hand, the larger outward tilt of the second extension arm 224 makes it easier for the angle steel to enter between the first extension arm 223 and the second extension arm 224 during installation, reducing the difficulty for operators to operate the drone.

[0214] When the operator observes that the angle steel has entered between the first extension arm 223 and the second extension arm 224, the operator moves the drone horizontally so that the first extension arm 223 fits against the side of the angle steel. Then, the operator moves the drone downward along the first extension arm 223. At this time, the first extension arm 223 can play a certain guiding role. Along the end face of the first extension arm 223, the angle steel will inevitably enter the installation opening between the fixed clamp 203 and the movable clamp 204, and the angle steel entering the opening can also basically fit against the side of the fixed clamp 203 until the hanging device is completely mounted on the angle steel, completing this stage of installation.

[0215] To facilitate more accurate drone operation by personnel below the tower, red identification plates 226 are installed on the outer surface of the second extension arm 224. These red identification plates 226 are positioned along the length of the second extension arm 224 and can be made of red plastic. During operation, staff can better determine whether the angle iron is positioned between the first extension arm 223 and the second extension arm 224 based on the position of the red identification plates 226 in the camera view.

[0216] In this embodiment, the frame side plate 227 also includes an inclined plate 230 and an inclined plate 231 extending downward and outward from the lower end of the U-shaped plate 229. The inclined plate 230 in the two frame side plates 227 and the connecting rod 228 connecting the inclined plate 230 together constitute an extension arm 223. The inclined plate 231 in the two frame side plates 227 and the connecting rod 228 connecting the inclined plate 231 together constitute an extension arm 224.

[0217] In addition, to increase the load-bearing surface of the mounting device on the angle steel, an L-shaped top plate 232 is provided on the outer end face of the U-shaped plate 229 and at the top of the opening of the U-shaped plate 229. The lower surface of the L-shaped top plate 232 forms an extension limiting surface 233 that extends horizontally to the outside of the frame side plate 227 for limiting the upper end of the clamped part. By adding two extension limiting surfaces 233 in the length direction of the angle steel, the installation of the angle steel end mounting frame 201 on the angle steel is realized, further improving the installation stability of the angle steel end mounting frame 201.

[0218] In this embodiment 1, the movable gripper 204 is a variable-form gripper, which includes a gripper body 10. The gripper body 10 includes a plurality of movable blocks 101 stacked sequentially at intervals in the height direction, and friction pads 102 are provided between adjacent movable blocks 101. Each movable block 101 can move independently in the horizontal direction under the action of external force. The end face of one end of the movable block 101 along the horizontal movement direction forms a clamping and limiting surface 103, and adjacent clamping and limiting surfaces 103 can be on the same vertical surface or on different vertical surfaces.

[0219] The movable gripper 204, based on the movable block 101, can change shape so that it can adapt to the shape, size and orientation of the angle steel during the clamping process, thereby achieving better clamping of the angle steel. By changing its own shape to always fit the angle steel, it is possible to eliminate the need to specially find a suitable angle steel to hang when installing the hanging device, effectively reducing the operational requirements of the operator.

[0220] In this embodiment, the friction pad 102 can be a diamond friction pad 102, which is designed to be long and strip-shaped and smaller than the movable block 101, so as to avoid it protruding from the clamping and limiting surface 103 formed by the movable block 101.

[0221] Specifically, the gripper body 10 also includes a gripper frame 104. The gripper frame 104 has a movable block channel 105 in the middle. A vertically arranged mounting post 106 is installed in the movable block channel 105 for sliding cooperation with the movable block 101. The movable block 101 and the friction pad 102 are arranged alternately in the axial direction of the mounting post 106. The movable block 101 has an elongated hole 107 for the mounting post 106 to pass through. The length direction of the elongated hole 107 is arranged along the axial direction of the movable block channel 105. The movable block 101 can slide relative to the mounting post 106 along the length direction of the elongated hole 107. The friction pad 102 has a circular hole 108 for the mounting post 106 to pass through and is clearance-fitted with the mounting post 106.

[0222] The gripper frame 104 is constructed as a frame with a U-shaped groove in the middle, and the frame includes a U-shaped inner bottom wall 109, a U-shaped inner side wall 110 and a U-shaped top wall 111; The movable block 101 is constructed in the shape of a rectangular plate, and the clamping and limiting surface 103 of each movable block 101 is a vertical plane. The clamping and limiting surface 103 has three states: one, flush with the inner bottom wall 109 of the U-shape; two, protruding from the inner bottom wall 109 of the U-shape but not protruding from the top wall 111 of the U-shape; and three, protruding from the top wall 111 of the U-shape. Each movable block 101 can individually maintain the clamping limiting surface 103 in state one, state two, or state three under the action of external force.

[0223] In this embodiment, the gripper frame 104 includes an upper end plate 112, a lower end plate 113, and two side plates 114. Both side plates 114 are U-shaped and arranged opposite each other. The movable block channel 105 is formed by the upper end plate 112, the lower end plate 113, and the two side plates 114. Each side plate 114 includes a vertical arm 115 and horizontal arms 116 located at the upper and lower ends of the vertical arm 115. The upper end plate 112 and the lower end plate 113 are respectively fixed to the horizontal arms 116 at the upper and lower ends. The movable block channel 105 is formed by the upper end plate 112, the lower end plate 113, and the vertical arms 115 of the two side plates 114. The upper and lower ends of the mounting column 106 are respectively fixed to the upper end plate 112 and the lower end plate 113. Friction pads 102 are provided between the upper end plate 112 and the uppermost movable block 101, and between the lowermost movable block 101 and the lower end plate 113.

[0224] During operation, the angle steel first enters the mounting opening between the movable gripper 204 and the fixed gripper 203, and the angle steel end hanging frame 201 is mounted on the upper end surface of the angle steel. The lower end surfaces of the horizontal arms 116 at the upper ends of the two side plates 114 in the gripper frame 104 and the lower surface of the L-shaped top plate 232 are on the same plane and are supported together on the upper end surface of the angle steel. The lower end surfaces of the horizontal arms 116 at the upper ends of the two side plates 114 in the gripper frame 104 and the lower surface of the L-shaped top plate 232 together constitute the first horizontal limiting surface for the installation of the hanging device on the angle steel. Then the drone descends and disengages from the angle steel end ring 210. The rigid rope 205 no longer has a pulling force on the hanging device. At this time, the movable gripper 204 can gradually move towards the fixed gripper 203 under the action of the torsion spring.

[0225] During the movement of the movable gripper 204, the movable block 101 on it will automatically fit against the end face of the angle steel based on the shape of the end face of the angle steel, and the angle steel in the installation opening has the following shape; (1) When the angle steel horizontal plate is above or below and the angle steel opening faces the fixed clamp 203, the end face of the angle steel facing the movable clamp 204 is a vertical plane. During the movement of the movable clamp 204 towards the angle steel under the action of the coil spring 208, the movable block 101, which is at the upper part and matches the height of the angle steel, is pushed in the opposite direction by the end face of the angle steel until the clamping limiting surface 103 is flush with the U-shaped inner bottom wall 109. The clamping limiting surface 103 of this part of the movable block 101 is pressed against the end face of the angle steel under the action of the coil spring 208, forming the first vertical limiting surface. The movable block 101, which is at the lower part exceeding the height of the angle steel, is pressed against the friction pad 102. The function is still in the initial state, that is, the clamping and limiting surface 103 protrudes from the U-shaped top wall 111. The uppermost surface of the movable block 101 in this part constitutes the second horizontal limiting surface at the lower end of the angle steel when the hanging device is installed on the angle steel (if the height of the angle steel is exactly the same as the height of the clamping and limiting surface 103 formed by all the movable plates, then the upper end of the horizontal arm 116 at the lower end of the two side plates 114 in the gripper frame 104 constitutes the second horizontal limiting surface). The first vertical limiting surface, the first horizontal limiting surface and the second horizontal limiting surface together constitute the limiting in this state, which has better limiting stability.

[0226] (2) The angle steel horizontal plate is at the top and the angle steel opening faces the movable clamp 204. At this time, the end face of the angle steel facing the movable clamp 204 is a stepped surface. The upper end face of the angle steel horizontal plate and the first horizontal limiting surface are still mutually limiting. The movable block 101, which is at the top and is adapted to the thickness of the middle horizontal plate of the angle steel, is pushed in the opposite direction by the horizontal plate of the angle steel until the clamping limiting surface 103 is flush with the U-shaped inner bottom wall 109. The clamping limiting surface 103 of this part of the movable block 101 is pressed against the end face of the angle steel under the action of the coil spring 208, forming the first vertical limiting surface. The movable block 101, which is located beyond the thickness of the horizontal plate of the angle steel, is still in its initial state under the action of the friction pad 102, that is, the clamping and limiting surface 103 protrudes from the U-shaped top wall 111. The uppermost surface of the movable block 101 constitutes the second horizontal limiting surface at the lower end of the horizontal plate of the angle steel when the hanging device is installed on the angle steel. The clamping and limiting surface 103 of the movable block 101 is in contact with the side of the vertical plate of the angle steel under the action of the friction pad 102, and can also play a certain limiting role.

[0227] (3) The angle steel horizontal plate is at the bottom and the angle steel opening faces the movable clamp 204. At this time, the end face of the angle steel facing the movable clamp 204 is a stepped surface. The upper end face of the angle steel vertical plate is mutually limited with the first horizontal limiting surface. The upper movable block 101 is in the initial state. The lower surface of the lowermost movable block 101 in this part of the movable block 101 forms the second horizontal limiting surface limited to the upper end face of the angle steel horizontal plate. The clamping limiting surface 103 of the movable block 101 opposite to the angle steel horizontal plate is pressed against the end face of the angle steel horizontal plate under the action of the coil spring 208, forming the first vertical limiting surface. The movable block 101 that extends beyond the lower end face of the horizontal plate is still in the initial state under the action of the friction pad 102. The upper end face of this part of the movable plate forms the third horizontal limiting surface mutually limited with the lower end face of the angle steel.

[0228] To further improve the motion stability of the movable block 101, two parallel mounting posts 106 are provided between the upper end plate 112 and the lower end plate 113. Two parallel elongated holes 107 are provided on the movable block 101, which correspond one-to-one with the two mounting posts 106. The friction pad 102 is constructed in the shape of a long strip and has round holes 108 at both ends for the two mounting posts 106 to pass through.

[0229] In this embodiment, the movable block 101 has an opening in the middle that faces away from the U-shaped groove. The arrangement of the clearance groove 117 allows the movable gripper 204 to be at its farthest point relative to the fixed gripper 203 when the device is in operation. At this time, the clearance groove 117 can avoid the guide column, so that the movable gripper 204 is located between the two frame side plates 227, rather than in the U-shaped groove where the angle steel end of the frame 201 is hung, so that the entire U-shaped groove where the angle steel end of the frame 201 is hung constitutes an installation opening.

[0230] Furthermore, the upper end plate 112 and the lower end plate 113 have the same structure, both including a first plate 118 and a second plate 119 fixed to each other. The first plate 118 is fixed to the side plate 114 by bolts, and the second plate 119 is used to fix the mounting post 106. The side of the second plate 119 is also provided with an extension 120 facing away from the U-shaped groove, and a reinforcing rib 121 is provided between the extension 120 and the first plate 118. The extension 120 and the reinforcing rib 121 make the installation of the movable block 101 on the gripper frame 104 more stable.

[0231] In this embodiment, the movable blocks 101 are all made of steel plates and have a certain weight. Under the premise of reducing the weight of the basic structure of the angle steel end hanging frame 201, since the clamping force of the movable claw 204 is generated by the coil spring 208, and the selection of the coil spring 208 is related to the self-weight of the frame, it is not advisable for the angle steel end hanging frame 201 to be too light. At this time, the movable blocks 101 can play a certain counterweight role. Furthermore, the weight of the movable blocks 101 is related to the friction force provided by the friction pad 102. The movable blocks 101 should not be made of lightweight materials, otherwise it will be difficult to keep the movable blocks 101 in the initial state. Therefore, this method of reducing the weight of the frame structure and increasing the weight of the movable blocks 101 can reasonably and effectively balance the weight of the angle steel end hanging frame 201.

[0232] like Figures 29-37 As shown, this embodiment also provides a device for hanging the end of a conductor, the specific structure of which is as follows: The device includes a wire end hanging frame 011, which includes two oppositely arranged frame connecting plates 0102 of the same shape. The upper and lower ends and the middle position of the two frame connecting plates 0102 are provided with connecting rods 0803 or connecting plates to form an integral frame.

[0233] A locking mechanism for locking the wire is installed between the two frame connecting plates 0102. The frame connecting plate 0102 includes a plate body 0103 and two extension arms 0104 extending downward from both sides of the lower end face of the plate body 0103. The lower end face of the plate body 0103 is provided with an upwardly recessed arc groove 0105. The locking mechanism can lock the wire in the arc groove 0105. The two extension arms 0104 extend at an angle of less than 90°, and the opposite sides of the two extension arms 0104 extend to the inner side of the circular groove 0105. The two extension arms 0104 form an angle between them and their upper ends extend to the circular groove. This allows the UAV to automatically enter the circular groove without needing to be aligned with the power line during the mounting process. This enables the UAV to be guided by the two extension arms 0104 to easily adjust the position and attitude of the grounding device during the flight of the UAV, ensuring its precise connection with the cable.

[0234] The two extension arms 0104 include a vertical arm 0106 and an inclined arm 0107. The vertical arm 0106 includes an outer vertical surface 0109 and an inner first inclined surface 0108. The vertical surface 0109 is parallel to the end face of the wire end ring 09. The inclined arm 0107 includes an outer second inclined surface 0110 and an inner third inclined surface 0111. The first inclined surface 0108 and the third inclined surface 0111 are opposite to each other and extend to the inner side of the arc groove 0105. The first inclined surface 0108 and the third inclined surface 0111 form a V-shaped guide angle of less than 90°. The vertical surface 0109 and the second inclined surface 0110 are respectively connected to the outer side of the plate body 0103. The V-shaped guide angle can effectively guide the power transmission line into the arc groove 0105, simplifying the precise control of the drone and reducing the difficulty of drone operation for operators.

[0235] The extension arm 0104 is equipped with a vertical arm and an inclined arm 0107, which serves two purposes: firstly, to facilitate the guidance of the power transmission line, and secondly, to better pull the ground wire.

[0236] In this embodiment, a connecting plate 0112 for connecting a grounding wire is installed between the vertical arms 0106 of the two frame connecting plates 0102 and at the lower end of the vertical arms 0106. The connecting plate 0112 is set on the vertical arm. The grounding wire has a certain weight. By using this kind of basically vertically extending vertical arm, the vertical arm maintains a roughly the same traction angle with the grounding wire when the device is running upward, reducing the tilt of the entire device caused by the vertical arm side flipping downward and the tilting arm 0107 side flipping upward in the hanging device. When the operator observes the power transmission line entering between the vertical arm and the tilting arm 0107, the operator moves the drone horizontally so that the vertical arm fits against the side of the power transmission line. Then, the operator moves the drone downward along the vertical arm. At this time, the vertical arm can play a certain guiding role. The power transmission line will inevitably enter the arc groove 0105 along the end face corner of the vertical arm. The operation is relatively simple and convenient.

[0237] To facilitate more accurate drone operation by personnel below the tower, an identification plate 0113 is installed on the outer surface of the tilting arm 0107 of the two frame connecting plates 0102, with its length direction aligned with the length of the tilting arm 0107. The identification plate 0113 can be made of red plastic. During operation, staff can better determine whether the conductor is between the tilting arm 0107 and the vertical arm 0106 based on the position of the red identification plate 0113 in the camera view.

[0238] In this embodiment, the locking mechanism includes two fixing parts 01 that limit the upper side of the conductor and a locking claw 02 disposed between the two fixing parts 01. The locking claw 02 can automatically engage with the lower side of the conductor. When the locking claw 02 is engaged with the lower side of the conductor, it can form a conductor locking area together with the two fixing parts 01. The fixing part 01 is provided on the plate body 0103 and located at the intersection of the two extension arms 0104. The fixing part 01 is constructed in the shape of an arc plate with an opening facing downward and is coaxially arranged with the arc groove 0105. The fixing part 01 and the lower end face of the arc groove 0105 together constitute the upper limit surface of the wire locking area.

[0239] The fixing part 01 can increase the force-bearing area of ​​the wire end hanging frame 011 on the wire, increase the stability of the hanging device on the wire, and also increase the force-bearing area on the upper side of the wire itself, reducing damage to the wire.

[0240] The locking mechanism in this embodiment also includes a gripper drive unit. The gripper drive unit includes a torsion spring 03 for driving the locking gripper 02 to move toward the fixed part 01, a pre-tightening handle 06 for pressing the torsion spring 03 to drive the locking gripper 02 to move toward the fixed part 01, and a positioning block 07 for positioning the pre-tightening handle 06. The positioning block 07 is installed on the frame connecting plate 0102 and can move up and down on the frame connecting plate 0102. When one end of the pre-tightening handle 06 presses the torsion spring 03, the other end can press against the positioning block 07 under the action of the torsion spring 03.

[0241] The gripper drive unit also includes a clamping push rod 04 whose end extends to the wire locking area and a positioning hook 05 that can be engaged with the locking gripper 02 and is used to restrict the movement of the locking gripper 02 toward the wire locking area. The positioning hook 05 is connected to the clamping push rod 04. When the clamping push rod 04 moves out of the wire locking area under the action of the wire, it can drive the positioning hook 05 to release the engagement with the locking gripper 02.

[0242] The locking jaw 02 includes an upper engaging portion 0201 and a lower jaw portion 0202. The positioning hook 05 engages with the engaging portion 0201 to restrict the jaw portion 0202 from moving toward the wire locking area. The jaw portion 0202 can engage with the lower side of the wire under the action of the elastic element.

[0243] The gripper drive unit also includes a torsion spring mounting shaft 0301 and a torsion spring 03 pre-tensioning part. The helical part of the torsion spring 03 is sleeved on the torsion spring mounting shaft 0301. One of the torsion arms of the torsion spring 03 can abut against the back of the gripper part 0202 away from the wire locking area. Under the action of the torsion spring 03, the gripper part 0202 can rotate around the torsion spring mounting shaft 0301 to a position opposite to the fixing part 01. The back of the locking gripper 02 away from the wire locking area is provided with a connecting seat 0302 rotatably connected to both ends of the torsion spring mounting shaft 0301. The torsion spring 03 pre-tensioning part includes a pre-tensioning handle 06 rotatably connected to both ends of the torsion spring mounting shaft 0301. The other torsion arm of the torsion spring 03 is located below the pre-tensioning handle 06. When the pre-tensioning handle 06 rotates downward around the torsion spring mounting shaft 0301, it can press down on the other torsion arm of the torsion spring 03 so that one of the torsion arms of the torsion spring 03 abuts against the back of the gripper part 0202 away from the wire locking area.

[0244] The two ends of the torsion spring mounting shaft 0301 are respectively mounted on the two frame connecting plates 0102. During the operation, in the initial state, the locking claw 02 and the pre-tightening handle 06 are both in a movable state on the torsion spring mounting shaft 0301. At this time, the entire wire end hanging frame 011 is in a free state. Before mounting the drone, the operator needs to rotate both the pre-tightening handle 06 and the locking claw 02 around the torsion spring mounting shaft 0301. The pre-tightening handle 06 presses down on the torsion spring 03 arm, and the locking claw 02 engages with the positioning hook 05. The torsion spring 03 is in a torsional state under the combined action of the pre-tightening handle 06 and the locking claw 02. At the same time, the locking claw 02 is positioned by the positioning hook 05, and the pre-tightening handle 06 is positioned by the positioning block 07. At this time, the wire end mounting frame 011 is in a pre-tightened state and is ready for operation. When the drone moves the wire end mounting frame 011 onto the corresponding wire, the power line is guided to the wire locking area by the extension arm 0104. Before fully entering the circular groove, it will first encounter the clamping push rod 04. The downward gravity of the wire end mounting frame 011 will push the clamping push rod 04 upward. During the upward movement of the clamping push rod 04, it will drive the positioning hook 05 to disengage from the locking claw 02's latching hole 0203. The positioning hook 05 will then position the locking claw 02. Under the action of the torsion spring 03, the locking claw 02 will move its claw part 0202 toward the wire locking area until the upper surface of the claw part 0202 forms the lower side of the wire locking area. Together with the fixing part 01, the power line will be locked. At this time, the power line will also be fully entered into the circular groove, completing the locking.

[0245] In this embodiment, the locking gripper 02 includes a fastening plate body constituting the fastening part 0201 and a gripper plate body constituting the gripper part 0202. The fastening plate body has a fastening hole 0203 in the middle for the positioning hook 05 to pass through and fasten to the back of the fastening plate body. The gripper plate body is constructed as a V-shaped plate with an included angle greater than 90°. The fixing part 01 is constructed as an arc-shaped plate with an opening facing downward. When the fastening plate body is fastened to the positioning hook 05, the gripper plate body is outside the wire locking area. When the positioning hook 05 releases the fastening of the fastening plate body, the V-shaped concave surface of the gripper plate body constitutes the lower limit surface of the wire locking area.

[0246] The structural design of the locking jaw 02 enables it to provide a more stable limit for the transmission line when locking it. In the non-locking state, it is located outside the conductor locking area, i.e., off the lower part of the conductor locking area. At this time, the jaw plate is located between the two vertical arms of the frame connecting plate 0102. This prevents the locking jaw 02 from obstructing the transmission line when the line is inserted, ensuring that the upper end of the transmission line contacts the lower end face of the clamping push rod 04 first when the line is inserted. It also ensures that the transmission line can eventually move to the top of the jaw plate, so that the V-shaped concave surface of the jaw plate can form the lower limit surface of the conductor locking area.

[0247] The gripper drive unit also includes a hook mounting unit, which includes a hook mounting shaft 0501 with both ends mounted on two frame connecting plates 0102 respectively. The middle part of the positioning hook 05 is rotatably mounted on the hook mounting shaft 0501. The positioning hook 05 passes through the fastening hole 0203 and fastens to the back of the fastening plate to form a hook hook part 0502. The other end forms a hook connecting part 0503. A push rod mounting plate 0504 is installed at the hook connecting part 0503. The middle part of the push rod mounting plate 0504 is installed on the lower end face of the hook connecting part 0503, and a clamping push rod 04 is installed at both ends respectively.

[0248] The snap-fit ​​hole 0203 is a square through hole with a height greater than the thickness of the hook part 0502. The end face of the hook part 0502 is an arc surface and the upper surface is constructed with an L-shaped hook groove 0505 that snaps onto the back of the snap-fit ​​plate. When the positioning hook 05 rotates around the hook mounting shaft 0501, the hook part 0502 can rotate downward in the snap-fit ​​hole 0203 to separate the L-shaped hook groove 0505 from the back of the snap-fit ​​plate.

[0249] The clamping push rod 04 is constructed in the shape of a V-shaped rod with an included angle greater than 90°. When the hook part 0502 of the hook is engaged with the fastening hole 0203, the upper part of the clamping push rod 04 extends vertically downward from the push rod mounting plate 0504, and the lower part extends obliquely into the position of the wire locking area and extends away from the clamp plate. The structural shape of the clamping push rod 04 allows it to better extend into the wire locking area and can better apply force to the lower oblique end of the clamping push rod 04 during the process of hanging the frame 011 at the end of the wire and lowering it towards the power transmission line. This makes the entire clamping push rod 04 form a lever to rotate the positioning hook 05, which can more smoothly realize the unlocking of the multiple locking claws 02 of the subsequent positioning hook 05.

[0250] The end face of the positioning claw 05 is an arc surface, which allows the locking claw 02 to rotate towards the positioning claw 05. It can move smoothly along the arc surface while pressing down on the positioning claw 05 and moving towards the L-shaped hook groove 0505 of the positioning claw 05, and finally engage in the L-shaped hook groove 0505. Under the action of the torsion spring 03 and the gravity of the claw mounting part, the positioning claw 05 can maintain the locking of the locking claw 02.

[0251] Meanwhile, the height of the latching hole 0203 is greater than the thickness of the hook part 0502, which provides a flipping stroke for the subsequent flipping and unlocking of the positioning hook 05, ensuring that the clamping push rod 04 can drive the hook part 0502 to flip downward in the latching hole 0203, and finally make the locking gripper 02 disengage from the hook part 0502 under the action of the torsion spring 03.

[0252] The mounting device and its locking mechanism enable it to be quickly, accurately and stably mounted on the power transmission line, saving operation time and improving operation efficiency.

[0253] In this embodiment, the wire end mounting bracket 011 is also provided with an unlocking mechanism for automatically unlocking the locking mechanism. The unlocking mechanism includes an unlocking frame 08 that can move up and down on the wire end mounting bracket 011. The unlocking frame 08 is provided with a positioning spring 0801 for keeping the positioning block 07 in a positioning state relative to the pre-tightening handle 06 and a wire end lifting ring 09 that can drive the positioning block 07 to separate from the pre-tightening handle 06. The wire end lifting ring 09 is connected to the unlocking frame 08 and can drive the positioning block 07 to separate from the pre-tightening handle 06 when the wire end lifting ring 09 is pulled upward.

[0254] On both sides of the wire end hanging frame 011, there is a vertically arranged slide groove 0101. The slide groove 0101 is set on the connecting plate 0102 of the two frames. The positioning block 07 is set in the slide groove 0101 and can move up and down along the length of the slide groove 0101.

[0255] Specifically, the positioning block 07 includes a sliding part and a positioning part. The positioning part is located inside the frame connecting plate 0102, and the pre-tightening handle 06 can be pressed against the side of the positioning part. The sliding part is located inside the slide groove 0101, and a limit plate is installed on the sliding part by bolts or screws. The limit plate is located outside the frame connecting plate 0102, thereby realizing the detachable installation of the positioning block 07 on the frame connecting plate 0102. The width of both the limit plate and the positioning part is greater than the width of the slide groove 0101.

[0256] When the positioning block 07 is located at the lower end of the slide groove 0101, the pre-tightening handle 06 is in a positioning state. When the positioning block 07 is located at the upper end of the slide groove 0101, the pre-tightening handle 06 is in an unlocked state. When the unlocking bracket 08 pulls the wire end ring 09 upward, it can drive the positioning block 07 to move upward in the slide groove 0101. The pre-tightening handle 06 is constructed in the shape of a U-shaped plate, and the opening faces the torsion spring 03. Both ends of the pre-tightening handle 06 are provided with extension protrusions 0601. The lower end surface of the extension protrusions 0601 and the lower end surface of the pre-tightening handle 06 are on the same plane and together form a pressing surface that can abut against the positioning block 07.

[0257] The unlocking frame 08 includes a U-shaped unlocking frame body 0802 with the opening facing downwards. The two ends of the side wall of the U-shaped unlocking frame body 0802 are respectively connected to the positioning blocks 07 on both sides of the wire end hanging frame body 011. Specifically, both ends of the side wall of the U-shaped unlocking frame body 0802 are provided with U-shaped slots 0805 with horizontal openings. The side wall of the U-shaped unlocking frame body 0802 is engaged with the positioning block 07 through the U-shaped slots 0805.

[0258] The top wall of the U-shaped unlocking frame 0802 is provided with a connecting rod 0803 that is connected to the wire end hanging ring 09. The connecting rod 0803 is fitted with a sleeve rod 0804 that can drive the wire end hanging frame 011 to move upward. The positioning spring 0801 is fitted on the connecting rod 0803, with one end pressed against the lower end surface of the sleeve rod 0804 and the other end pressed against the upper surface of the top wall of the U-shaped unlocking frame 0802.

[0259] The upper surface of the wire end hanging frame 011 is equipped with a wire end lifting ring 09 and a wire end lifting ring mounting seat 0901 for mounting the wire end lifting ring 09. Specifically, the wire end lifting ring mounting seat 0901 is mounted on the unlocking frame 08, and the wire end lifting ring 09 is fixed on the wire end lifting ring mounting seat 0901. Both ends of the connecting rod 0803 are provided with threaded holes, and both ends of the connecting rod 0803 are fixed to the wire end lifting ring mounting seat 0901 and the top wall of the U-shaped unlocking frame 0802 respectively by bolts.

[0260] An upper end plate is installed between the upper ends of the two frame connecting plates 0102. The upper end of the sleeve rod 0804 passes through the upper end plate and can slide up and down relative to the upper end plate. The lower end of the sleeve rod 0804 is provided with a flange. The flange can be pressed against the lower surface of the upper end plate under the action of the positioning spring 0801. The wire end lifting ring mounting seat 0901 is fixed to the upper end of the connecting rod 0803 by bolts. The wire end lifting ring mounting seat 0901 is provided with a lifting ring locking groove 0902 with an upward opening for locking the wire end lifting ring 09. The lower end of the wire end lifting ring 09 is locked in the lifting ring locking groove 0902 and fastened by bolts.

[0261] During the unlocking process, the drone pulls the wire end ring 09 upward via a rope. As the wire end ring 09 moves upward, it first drives the connecting rod 0803 upward via the wire end ring mounting seat 0901. The upward movement of the connecting rod 0803 will drive the U-shaped unlocking frame 0802 upward. Then, the U-shaped unlocking frame 0802 will drive the positioning block 07 to move to the upper end of the slide groove 0101. At this time, the pressing surface of the pre-tightening handle 06 loses the support of the positioning block 07, and the pre-tightening handle 06 returns to a free state where it can move. The spring force of the torsion spring 03 is released, and the locking claw 02 loses the squeezing force of the torsion spring 03 and also returns to a free state, thus completing the unlocking of the power transmission line by the locking claw 02. During the unlocking process, the lifting ring 09 at the conductor end is further lifted upwards. As the U-shaped unlocking frame 0802 moves upwards, the positioning spring 0801 presses against the sleeve rod 0804. The sleeve rod 0804 moves upwards, causing the upper end plate to move upwards. This allows the entire conductor end mounting frame 011 to be removed from the power transmission line. This means that the device can be removed from the power transmission line simply by the drone making an upward movement, saving operation time and improving operation efficiency.

[0262] It is readily understood that those skilled in the art can combine, split, or reorganize the embodiments provided in this application to obtain other embodiments, all of which do not exceed the protection scope of this application.

[0263] The present invention and its embodiments have been described above illustratively. This description is not restrictive, and the embodiments shown are only part of the embodiments of the present invention. The actual structure is not limited thereto. Therefore, if those skilled in the art are inspired by this description and design similar structures and embodiments without departing from the spirit of the present invention, they should all fall within the protection scope of the present invention.

Claims

1. A complete set of operation methods for unmanned aerial vehicle (UAV)-based voltage testing and grounding wire installation / removal, which includes: A method for collecting and processing voltage detection information based on low-frequency wireless transmission is adopted to determine whether the transmission line is energized, and a permission signal is generated after confirming that it is de-energized. Triggered by a permission signal, the drone is controlled to attach or remove the grounding wire, and visual recognition and image enhancement methods for attaching and removing the grounding wire are used to identify and judge the status of the grounding wire device during the operation. A real-time monitoring system for grounding wire operations based on UAV vision is adopted to monitor the operation process of hanging and removing grounding wires in real time, and transmit the operation status and related on-site images to the ground.

2. The method for complete operation of unmanned aerial vehicle (UAV)-based voltage testing and grounding wire installation / removal as described in claim 1, wherein, The method for collecting and processing voltage detection information based on low-frequency wireless transmission, which determines whether the transmission line is energized and generates a permission signal after confirming that it is de-energized, specifically includes: The original induced signals around the power transmission line are obtained by using an electrical detection sensor mounted on a drone. The original induced signal is conditioned and its features are extracted to obtain characteristic values ​​representing the electric field strength; The characteristic value is transmitted to the ground via a low-frequency wireless communication module; The energized state of the transmission line is determined based on the characteristic values ​​received from the ground end.

3. The complete set of operation methods for mounting voltage testing and grounding wire installation / removal based on UAVs according to claim 2, wherein, The conditioning and feature extraction of the original induced signal to obtain feature values ​​characterizing the electric field strength includes: The original induced signal is amplified and bandpass filtered to obtain the power frequency signal; The power frequency signal is converted from analog to digital to obtain a digital signal; Based on the digital signal, its effective value within at least one power frequency cycle is calculated, and the effective value is used as the characteristic value.

4. The complete set of operation methods for mounting voltage testing and grounding wire installation / removal based on UAV as described in claim 3, wherein: The center frequency of the bandpass filter corresponds to the power frequency of the power system and is used to suppress non-power frequency noise to enhance the signal-to-noise ratio of the characteristic value.

5. The complete set of operation methods for mounting voltage testing and grounding wire installation / removal based on UAV as described in claim 4, wherein: The low-frequency wireless communication module operates in the frequency band of 405MHz to 435MHz.

6. The complete set of operation methods for mounting voltage testing and grounding wire installation / removal based on UAV as described in claim 2, wherein, Determining the energized state of the transmission line based on feature values ​​received from the ground end includes: The received feature values ​​are compared with a preset decision threshold to obtain the comparison result; Based on the comparison results, the energized state of the transmission line is determined, including whether it is energized or de-energized. If the energized state is determined to be de-energized, a permission signal is generated to allow the grounding work to be performed.

7. The method for complete operation of unmanned aerial vehicle (UAV)-based voltage testing and grounding wire installation / removal as described in claim 6, wherein, The preset decision thresholds include a first threshold and a second threshold, where the second threshold is less than the first threshold; determining the energized state of the transmission line based on the comparison results, including an energized state or a de-energized state, includes: If the eigenvalue is greater than or equal to the first threshold, then the charged state is determined to be a charged state. If the characteristic value is less than or equal to the second threshold, the charged state is determined to be an uncharged state.

8. The complete set of operation methods for mounting voltage testing and grounding wire installation / removal based on UAV as described in claim 1, wherein, The process involves controlling a drone to attach or remove a grounding wire upon triggering a permission signal. A visual recognition and image enhancement method for grounding wire attachment and removal is employed to identify and determine the status of the grounding wire device during the operation. Specifically, this includes: Obtain raw operational images containing the grounding wire device; Based on prior knowledge of the power scenario, dynamic regions of interest are determined from the original operation images; Within a dynamically defined region of interest, adaptive image enhancement processing is performed on the image to obtain an enhanced image. Extract visual features from the enhanced image to identify the grounding device; Based on visual features, the grounding wire device is identified, and its hanging / removal status is determined according to the morphological characteristics of the grounding wire device.

9. The method for complete operation of unmanned aerial vehicle (UAV)-based voltage testing and grounding wire installation / removal as described in claim 8, wherein determining the dynamic region of interest from the original operation image based on prior knowledge of the power scenario includes: Edge detection is performed on the original image to obtain an edge image; From the edge image, candidate line segments belonging to conductor and tower structures are selected based on length and orientation features; Based on the geometric relationship of the candidate line segments, at least one candidate anchor point is determined, which corresponds to the connection area between the conductor and the tower. The region of interest is determined dynamically based on the candidate anchor point and the preset size rules.

10. The method for unmanned aerial vehicle (UAV)-based voltage testing and grounding wire installation / removal as described in claim 8, wherein performing adaptive image enhancement processing on the image within the dynamic region of interest to obtain an enhanced image includes: Illumination conditions are evaluated for images within a dynamic region of interest, and evaluation results are obtained. Based on the evaluation results, adaptive image enhancement processing is performed on the image to obtain the enhanced image, including: Image enhancement is achieved by using an adaptive histogram equalization algorithm that limits contrast.

11. The complete set of operation methods for mounting voltage testing and grounding wire installation / removal based on UAV as described in claim 10, wherein: The process of evaluating the illumination conditions of the image within the dynamically dynamic region of interest to obtain the evaluation results includes: Calculate the average gray value and gray standard deviation of the image within the dynamic region of interest, and evaluate them to obtain the evaluation result; Based on the evaluation results, adaptive image enhancement processing is performed on the image to obtain an enhanced image, specifically including: When the average gray value is lower than the first preset threshold and the gray standard deviation is lower than the second preset threshold, adaptive image enhancement processing is performed on the image to obtain an enhanced image.

12. The method for complete operation of mounting voltage testing and grounding wire installation / removal based on a UAV according to claim 10, wherein, The image enhancement algorithm, which limits contrast through adaptive histogram equalization, includes: The image within a dynamically generated region of interest is segmented into multiple consecutive and non-overlapping sub-blocks; For each sub-block, calculate its grayscale histogram and crop the grayscale histogram using a preset cropping limit value; Perform equalization processing on the cropped grayscale histogram; Perform bilinear interpolation on the processed image of each sub-block; The interpolated image is then smoothed to obtain the enhanced image.

13. The method for complete operation of unmanned aerial vehicle (UAV)-based voltage testing and grounding wire installation / removal as described in claim 8, wherein, The method of identifying grounding wire devices based on visual features and determining their hanging / removing status based on their morphological characteristics includes: Extract the morphological features of the grounding wire device from the identified locations; The extracted morphological features are matched with multiple preset state discrimination rules to determine the hanging and dismantling status of the grounding wire device.

14. The complete set of operation methods for mounting voltage testing and grounding wire installation / removal based on a UAV as described in claim 13, wherein, The hanging and removing operation status of the grounding wire device includes: Reliably connected; connected but not locked; not connected.

15. The complete set of operation methods for unmanned aerial vehicle (UAV)-based voltage testing and grounding wire installation / removal as described in claim 13, wherein, The morphological features of the grounding wire device include: The aspect ratio of the rectangle surrounding the grounding device; the pixel opening degree of the gripper area of ​​the grounding device.

16. The method for complete operation of unmanned aerial vehicle (UAV)-based voltage testing and grounding wire installation / removal as described in claim 13, wherein, The step of matching the extracted morphological features with multiple preset state discrimination rules to determine the hanging and dismantling status of the grounding wire device also includes: The timing logic is verified by combining the discrimination results of multiple consecutive frames of images prior to the current moment.

17. The method for complete operation of unmanned aerial vehicle (UAV)-based voltage testing and grounding wire installation / removal as described in claim 1, wherein, The real-time monitoring system for grounding wire operations based on UAV vision includes an airborne subsystem and a ground subsystem that communicate with each other. The airborne subsystems include: The image acquisition module is used to acquire video streams or image sequences of the grounding wire operation area of ​​transmission lines; The processing and recognition module is used to process image frames in video streams or image sequences, identify grounding devices and determine their operating status, and output the recognition results. The first wireless communication module is used to send information containing the identification results to the ground subsystem; The ground subsystem includes: The second wireless communication module is used to receive information from the airborne subsystem; The information processing module is used to parse information and generate monitoring instructions; The monitoring and display module is used to display the operation status and related on-site images based on monitoring commands.

18. The complete set of operation methods for mounting voltage testing and grounding wire installation / removal based on a UAV as described in claim 17, wherein, The process of processing image frames in a video stream or image sequence, identifying grounding devices and determining their operational status, and outputting the identification results includes: Adaptive image enhancement is performed on the input image frame to obtain the enhanced image; Based on enhanced images, identify and locate conductor and tower structures; Based on the conductor and tower structure, the target area containing the grounding wire device is determined; Visual features of the target area are extracted, and the operating status of the grounding wire device is determined based on a preset state classification model to obtain the recognition result.

19. The complete set of operation methods for mounting voltage testing and grounding wire installation / removal based on a UAV as described in claim 18, wherein, The adaptive image enhancement of the input image frame to obtain the enhanced image includes: Calculate at least one global statistical feature of the input image frame; The environment category to which the image frame belongs is determined by comparing global statistical features with a preset threshold. Based on the environment category, the corresponding preset image processing algorithm is invoked to process the image frame.

20. The method for complete operation of mounting voltage testing and grounding wire installation / removal based on a UAV according to claim 19, wherein, The step of processing the image frame by calling the corresponding preset image processing algorithm based on the environment category includes: When the environment is classified as low-light or high-noise, the preset image processing algorithm is a contrast-limited adaptive histogram equalization algorithm, and the specific steps include: The input image frame is divided into multiple consecutive and non-overlapping sub-blocks; For each sub-block, calculate its grayscale histogram and crop the grayscale histogram using a preset cropping limit value; Perform equalization processing on the cropped grayscale histogram; Perform bilinear interpolation on the processed image of each sub-block; The interpolated image is then smoothed to obtain the enhanced image.

21. The method for complete operation of unmanned aerial vehicle (UAV)-based voltage testing and grounding wire installation / removal as described in claim 18, wherein, The determination of the target area containing the grounding wire device based on the conductor and tower structure includes: Edge detection is performed on the enhanced image frame to obtain the edge image; In the edge image, candidate line segments belonging to conductors and towers are selected based on the length and direction features of the line segments; Based on the relative positions and geometric relationships of candidate line segments, the positions of the conductor structure and the tower structure in the image are determined respectively.

22. The method for complete operation of unmanned aerial vehicle (UAV)-based voltage testing and grounding wire installation / removal as described in claim 21, wherein, The determination of the positions of the conductor structure and the tower structure in the image based on the relative positions and geometric relationships of the candidate line segments includes: Based on the intersection or nearest neighbor of the identified conductor edge and tower edge, at least one candidate anchor point is determined; Candidate boxes are dynamically generated based on preset size rules, with the candidate anchor point as the center, and the candidate box containing the most edge features is determined as the target region.

23. The method for complete operation of unmanned aerial vehicle (UAV)-based voltage testing and grounding wire installation / removal as described in claim 18, wherein, The visual features of the target area are extracted, and the operating status of the grounding wire device is determined based on a preset state classification model to obtain the recognition result. The extraction of visual features of the target region specifically involves calculating the gradient direction histogram features of the target region; The preset state classification model is a support vector machine model; The preset state classification model is a convolutional neural network model.

24. The method for complete operation of unmanned aerial vehicle (UAV)-based voltage testing and grounding wire installation / removal as described in claim 19, wherein, The ground subsystem also includes: The storage module is used to store the received job status, associated image frames and system logs in chronological order, and supports retrieval based on time or status.

25. A complete system for unmanned aerial vehicle (UAV)-based voltage testing and grounding wire installation / removal, wherein, It adopts the complete set of operation methods for mounting voltage testing and attaching / removing grounding wires based on any one of claims 1-24.

Images