A power distribution line pole state detection method, system, device and medium

Abstract

This invention provides a method, system, device, and medium for detecting the state of power distribution line poles. The method includes: controlling a climbing robot to climb along the power distribution line pole to be tested; triggering acquisition operations at preset height intervals to acquire a full-circumference image of the power distribution line pole at each height section using an image acquisition device, measuring the distance to the pole via a time-of-flight distance sensor, and acquiring attitude data via an inertial measurement sensor; binding the full-circumference image, pole distance, attitude data, and current height value into structured data groups corresponding to each height section using a unified timestamp; identifying the apparent defects and geometric state of the power distribution line pole to be tested, and constructing a mapping relationship between the apparent defect identification results and the pole geometric state identification results in a preset unified pole coordinate system, thereby realizing the state detection of the power distribution line pole to be tested. This invention can effectively improve the accuracy of power distribution line pole state detection.

Description

Technical Field

[0001] This invention relates to the field of power line pole condition detection, and more particularly to a method, system, device, and medium for power line pole condition detection. Background Technology

[0002] Concrete power poles used for power distribution lines are susceptible to structural damage during long-term service due to environmental factors such as wind and sun exposure, freeze-thaw cycles, acid rain erosion, and external impacts. This damage can lead to surface cracks, spalling, honeycomb-like pitting, and exposed rebar. Furthermore, loose foundations can cause the entire pole to tilt. Regular inspection of concrete poles typically requires acquiring two types of information: first, information on surface defects, which relies on image detection; and second, information on the pole's geometric state (e.g., pole tilt detection). However, because surface inspection and geometric measurement differ fundamentally in their working principles, sensor types, and data acquisition methods, existing inspection methods generally face the technical challenge of inconsistent data acquisition benchmarks and difficulties in data registration.

[0003] Specifically, image acquisition is typically accomplished through manual pole climbing for visual inspection, ground-based telephoto camera photography, or drone hovering for image capture. Tilt assessment, on the other hand, usually relies on fitting a centerline after measuring points at different heights. Even when integrating both types of detection onto the same mobile platform (e.g., mounting a camera and tilt sensor on a climbing robot), current practices often involve operating the image acquisition module and attitude measurement module independently. In this case, the image acquisition module, such as the camera, continuously records or takes photos at a fixed frame rate, while tilt detection devices, such as Time-of-Flight (ToF) ranging sensors and Inertial Measurement Units (IMUs), output data according to their respective sampling periods. There is no unified sampling triggering mechanism between them, nor is there a method to map multi-source data to the same spatial coordinate system. This results in the following: even if multi-sensor data acquisition is completed on the same pole, the final output of the apparent image information and tilt information remain two independent information streams. The lack of spatial consistency between the two directly affects the accuracy of power line pole condition detection. Summary of the Invention

[0004] This invention application provides a method, system, device, and medium for detecting the condition of power distribution line poles, in order to solve the technical problem of how to improve the accuracy of detecting the condition of power distribution line poles.

[0005] To address the aforementioned technical problems, this invention provides a method for detecting the condition of power distribution line poles, comprising: The climbing robot is controlled to climb along the pole of the power distribution line to be tested; wherein, the climbing robot is equipped with an image acquisition device, a time-of-flight ranging sensor and an inertial measurement sensor; During the climbing process of the climbing robot, the acquisition operation is triggered at preset height intervals, so that at each height section, the image acquisition device can acquire the full-circumference image of the power line pole under test, the time-of-flight distance sensor can measure the distance of the pole, and the inertial measurement sensor can collect attitude data. Based on the full-circle image, the ring rod distance, the attitude data, and the current height value, a structured data group corresponding to each height section is bound with a unified timestamp; Based on the structured data set, the apparent defects and geometric state of the poles of the power distribution line under test are identified. A mapping relationship between the apparent defect identification results and the pole geometric state identification results is constructed in a preset unified pole coordinate system, thereby realizing the state detection of the poles of the power distribution line under test.

[0006] As a preferred embodiment, the identification of apparent defects and pole geometry of the power distribution line pole under test based on the structured data set includes: Based on the index corresponding to the height section, query the structured data group to obtain the full-circle image corresponding to the index, the ring rod distance corresponding to the index, and the attitude data corresponding to the index; The full-circle image corresponding to the index is input into a pre-trained deep learning model. Based on the output of the deep learning model, the defect type of the apparent defect of the power distribution line pole under test and the pixel coordinates of the apparent defect in the full-circle image are obtained, thereby realizing the identification of the apparent defect of the power distribution line pole under test. Obtain the installation angle of the time-of-flight ranging sensor; determine the center position of the cross-section of the height section corresponding to the index based on the ring rod distance corresponding to the index and the installation angle; Based on the center position of the cross section and the posture data corresponding to the index, the mapping relationship between the height and tilt state of the power distribution line pole under test is analyzed, thereby realizing the identification of the pole geometry state of the power distribution line pole under test.

[0007] As a preferred embodiment, after querying the structured data group based on the index corresponding to the height section, the method further includes: obtaining the height value corresponding to the index; The unified pole coordinate system uses the height value as the height coordinate, the installation circumferential angle of the image acquisition device and the time-of-flight ranging sensor as the circumferential coordinate, and the distance between the ring poles as the radial coordinate; The step of constructing a mapping relationship between the apparent defect identification results and the bar geometric state identification results in a preset unified bar coordinate system includes: Based on the installation circumferential angle of the image acquisition device corresponding to the full-circumference image of the index, the ring rod distance corresponding to the index, and the height value corresponding to the index, the pixel coordinates are converted into the spatial position of the apparent defect in the unified rod coordinate system. Based on the mapping relationship between the spatial location, height, and tilt state, as well as the height coordinates and circumferential coordinates, a mapping relationship is constructed between the appearance defect identification results and the pole geometry identification results.

[0008] As a preferred embodiment, determining the center position of the cross-section corresponding to the height of the index based on the ring rod distance and the installation angle includes: Based on the installation angle and structural dimensions of each time-of-flight ranging sensor, the distance of the ring rod corresponding to the index is converted into multiple contour points of the height section; The center position of the height section is obtained by fitting the multiple contour points using circle fitting, ellipse fitting, or least squares fitting.

[0009] As a preferred embodiment, after identifying the apparent defects of the pole of the power distribution line under test, the method further includes: Obtain the confidence level for identifying the defect type. When the confidence level is higher than a first confidence threshold and lower than a second confidence threshold, enter the local re-inspection mode; wherein, the first confidence threshold is lower than the second confidence threshold. In the local re-inspection mode, the climbing robot is controlled to reduce its climbing speed or stop at the current height, and the data acquisition operation is re-triggered to achieve re-inspection.

[0010] As a preferred embodiment, the climbing robot includes a ring detection component, the image acquisition device, the time-of-flight ranging sensor and the inertial measurement sensor are disposed on the ring detection component, and the ring detection component is sleeved on the power pole of the power distribution line to be tested; After measuring the distance to the ring rod using the time-of-flight ranging sensor, the process further includes: The distances between the ring rods measured by multiple time-of-flight ranging sensors at different positions along the circumference of the ring detection component are selected to obtain a filtered dataset; Calculate the difference between the distances of each ring pole in the filtered dataset, and calculate the eccentricity of the center of the ring detection component relative to the center of the pole of the power distribution line under test; When the maximum value of the difference exceeds the preset eccentricity threshold, the ring detection component is determined to be in an eccentric state, and the pixel scale, circumferential unfolding position, and defect size calculation results based on the full-circumference image acquired by the image acquisition device are corrected according to the eccentricity amount.

[0011] As a preferred embodiment, the climbing robot further includes a mechanical fall protection mechanism; the process of controlling the climbing robot to climb along the power line pole to be tested also includes: The system acquires the motor current, drive wheel encoder speed, height measurement value, and real-time climbing attitude output by the inertial measurement sensor of the climbing robot. Based on the combined relationship between the motor current, encoder speed, height measurement value, and real-time climbing posture, abnormal working conditions are identified; among which, the types of abnormal working conditions include wheel slippage, mechanical jamming, and uncontrolled descent. Upon detecting any abnormal operating condition, the power supply to the climbing robot is cut off and the mechanical anti-fall mechanism is activated to lock it in place.

[0012] Accordingly, this invention application also provides a status detection system for power distribution line poles, including a control module, a data acquisition module, a data group construction module, and a status detection module; wherein, The control module is used to control the climbing robot to climb along the power pole of the power distribution line to be tested; wherein, the climbing robot is equipped with an image acquisition device, a time-of-flight ranging sensor and an inertial measurement sensor; The data acquisition module is used to trigger acquisition operations at preset height intervals during the climbing robot's ascent, thereby acquiring a full-circumference image of the power line pole under test through the image acquisition device at each height section, measuring the distance of the pole through the time-of-flight ranging sensor, and acquiring attitude data through the inertial measurement sensor. The data group construction module is used to bind the full-circle image, the ring rod distance, the attitude data and the current height value into structured data groups corresponding to each height section with a unified timestamp; The state detection module is used to identify the apparent defects and geometric state of the power distribution line pole under test based on the structured data group, and to construct a mapping relationship between the apparent defect identification results and the pole geometric state identification results in a preset unified pole coordinate system, thereby realizing the state detection of the power distribution line pole under test.

[0013] As a preferred embodiment, the condition detection module identifies the apparent defects and geometric state of the pole of the power distribution line under test based on the structured data set, including: The state detection module queries the structured data group according to the index corresponding to the height section to obtain the full-circle image, the ring rod distance, and the attitude data corresponding to the index. The full-circle image corresponding to the index is input into a pre-trained deep learning model. Based on the output of the deep learning model, the defect type of the apparent defect of the power distribution line pole under test and the pixel coordinates of the apparent defect in the full-circle image are obtained, thereby realizing the identification of the apparent defect of the power distribution line pole under test. Obtain the installation angle of the time-of-flight ranging sensor; determine the center position of the cross-section of the height section corresponding to the index based on the ring rod distance corresponding to the index and the installation angle; Based on the center position of the cross section and the posture data corresponding to the index, the mapping relationship between the height and tilt state of the power distribution line pole under test is analyzed, thereby realizing the identification of the pole geometry state of the power distribution line pole under test.

[0014] As a preferred embodiment, the state detection module is further configured to: obtain the height value corresponding to the index; The unified pole coordinate system uses the height value as the height coordinate, the installation circumferential angle of the image acquisition device and the time-of-flight ranging sensor as the circumferential coordinate, and the distance between the ring poles as the radial coordinate; The state detection module constructs a mapping relationship between the apparent defect identification results and the geometric state identification results of the rod in a preset unified rod coordinate system, including: The state detection module converts the pixel coordinates into the spatial position of the apparent defect in the unified rod coordinate system based on the installation circumferential angle of the image acquisition device of the full-circumference image corresponding to the index, the ring rod distance corresponding to the index, and the height value corresponding to the index. Based on the mapping relationship between the spatial location, height, and tilt state, as well as the height coordinates and circumferential coordinates, a mapping relationship is constructed between the appearance defect identification results and the pole geometry identification results.

[0015] As a preferred embodiment, the state detection module determines the center position of the cross-section corresponding to the height of the index based on the ring rod distance and the installation angle, including: The status detection module converts the distance of the ring rod corresponding to the index into multiple contour points of the height section based on the installation angle and structural installation dimensions of each time-of-flight ranging sensor. The center position of the height section is obtained by fitting the multiple contour points using circle fitting, ellipse fitting, or least squares fitting.

[0016] As a preferred embodiment, the condition detection system further includes a re-inspection module, which is used after the identification of apparent defects in the power distribution line pole under test is achieved: Obtain the confidence level for identifying the defect type. When the confidence level is higher than a first confidence threshold and lower than a second confidence threshold, enter the local re-inspection mode; wherein, the first confidence threshold is lower than the second confidence threshold. In the local re-inspection mode, the climbing robot is controlled to reduce its climbing speed or stop at the current height, and the data acquisition operation is re-triggered to achieve re-inspection.

[0017] As a preferred embodiment, the climbing robot includes a ring detection component, the image acquisition device, the time-of-flight ranging sensor and the inertial measurement sensor are disposed on the ring detection component, and the ring detection component is sleeved on the power pole of the power distribution line to be tested; The state detection system also includes a correction module, which is used after the time-of-flight ranging sensor measures the distance to the ring rod: The distances between the ring rods measured by multiple time-of-flight ranging sensors at different positions along the circumference of the ring detection component are selected to obtain a filtered dataset; Calculate the difference between the distances of each ring pole in the filtered dataset, and calculate the eccentricity of the center of the ring detection component relative to the center of the pole of the power distribution line under test; When the maximum value of the difference exceeds the preset eccentricity threshold, the ring detection component is determined to be in an eccentric state, and the pixel scale, circumferential unfolding position, and defect size calculation results based on the full-circumference image acquired by the image acquisition device are corrected according to the eccentricity amount.

[0018] As a preferred embodiment, the climbing robot further includes a mechanical fall protection mechanism; during the process of controlling the climbing robot to climb along the pole of the power distribution line to be tested, the control module is also used for: The system acquires the motor current, drive wheel encoder speed, height measurement value, and real-time climbing attitude output by the inertial measurement sensor of the climbing robot. Based on the combined relationship between the motor current, encoder speed, height measurement value, and real-time climbing posture, abnormal working conditions are identified; among which, the types of abnormal working conditions include wheel slippage, mechanical jamming, and uncontrolled descent. Upon detecting any abnormal operating condition, the power supply to the climbing robot is cut off and the mechanical anti-fall mechanism is activated to lock it in place.

[0019] Accordingly, this application also provides a terminal device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor executes the computer program to implement the power distribution line pole status detection method.

[0020] Accordingly, this application also provides a computer-readable storage medium, which includes a stored computer program, wherein the computer program, when running, controls the device where the computer-readable storage medium is located to execute the power distribution line pole status detection method.

[0021] Compared with the prior art, this invention application has the following beneficial effects: This invention provides a method, system, device, and medium for detecting the condition of power distribution line poles. By uniformly triggering image acquisition equipment, time-of-flight ranging sensors, and inertial measurement sensors at preset height intervals during the climbing process, and binding the full-circumference images, pole distance, and attitude data at the same height section into a structured data set with a unified timestamp, this invention ensures that the multi-source data at each height section accurately correspond in terms of acquisition time and spatial location. This fundamentally avoids the problems in existing technologies where different data streams are asynchronous in time and inconsistent in spatial reference due to the independent operation of each sensor and the lack of a unified triggering mechanism. This guarantees that subsequent defect identification results and pole geometric state identification results can be correlated under the same spatial reference, effectively improving the accuracy of power distribution line pole condition detection.

[0022] Furthermore, since the full-circumference images, ring rod distances, attitude data, and height values ​​of each height section are all bound to a structured data group with a unified timestamp, each apparent defect and each geometric anomaly in the condition detection results can be retrieved through a unified spatial index to retrieve the corresponding full-circumference image, distance data, and attitude data original records. This allows subsequent analysis to be strictly performed at the same height, avoiding the verification difficulties caused by scattered data and lack of a unified index in existing technologies. Attached Figure Description

[0023] Figure 1 This is a flowchart illustrating an embodiment of the power distribution line pole status detection method provided in this application.

[0024] Figure 2 This is a schematic diagram of an embodiment of the power distribution line pole condition detection system provided in this application. Detailed Implementation

[0025] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] Example 1 Please refer to Figure 1 , Figure 1 This is a flowchart illustrating an embodiment of the power distribution line pole status detection method provided in this application.

[0027] During the preparation phase of the inspection operation, the operator first moves the climbing robot to the bottom of the target concrete utility pole (power line pole), and then inserts the wheeled pole climbing mechanism from the side or separates it around the pole.

[0028] In this embodiment, the main structure of the climbing robot includes an approximately hexagonal outer frame, which is assembled from multiple sections of industrial aluminum profiles through corner brackets and connecting plates to form an approximately hexagonal closed structure on a plane, with a through hole in the middle for accommodating power line poles.

[0029] The frame has crawling module mounting bases on three sides spaced approximately 120° apart, while the remaining sides are used to install the electrical control box, battery, and detection component interfaces. By arranging the heavier components as close as possible to the pole center, the center of gravity of the entire machine is made close to the pole axis, thus improving its anti-overturning stability.

[0030] In the specific structure of the crawling module, each crawling module includes a pair of drive wheels and driven wheels arranged vertically, corresponding support rods, and elastic clamping components.

[0031] One end of the upper and lower support rods is hinged to the outer frame via a pivot, and the other end is respectively equipped with a drive wheel and a driven wheel covered with rubber treads. The drive wheel is connected to the output shaft of the DC geared motor via a synchronous belt or sprocket, and the driven wheel rotates freely with the support rod via a bearing. A tension spring is arranged between the upper and lower support rods so that the two wheels generate a continuous clamping force on the surface of the power distribution line pole to buffer the surface roughness and assembly errors and maintain stable attachment.

[0032] The three crawling modules are connected to the outer frame via hinged joints. Three lead screw-type telescopic rods distributed at different 120° angles allow for adjustment of the radial position of the crawling modules, enabling the robot to adapt to a certain range of concrete pole diameters (such as the aforementioned power line poles). After adjustment, the telescopic mechanism is secured by locking nuts or pins, and anti-loosening washers or pins are used to prevent loosening due to vibration, thus ensuring stable and reliable clamping force during long-term climbing.

[0033] To ensure sufficient anti-slip redundancy for the robot when designing the wheel diameter and clamping force, the maximum working mass of the entire machine (including the detection components) is set to m, the coefficient of friction between the rubber wheel and the concrete surface is μ, and the single clamping force is F. N The total friction of the three crawling modules is 3μF. N And introduce a safety factor S f To take into account environmental and aging effects, the following anti-slip conditions should be met during climbing: 3μF N ≥S f W; This formula is used to back-calculate the spring preload and the telescopic locking force, ensuring sufficient anti-slip redundancy under the design quality and friction coefficient. Compared with the traditional two-point or multi-point rigid clamping method, the three-point symmetrical elastic clamping not only ensures stable adhesion but also improves adaptability to different rod diameters and rod surface defects.

[0034] Figure 1 The method for detecting the status of power distribution line poles shown includes steps S101 to S104; each step is described in detail below: Step S101: Control the climbing robot to climb along the pole of the power distribution line to be tested; wherein, the climbing robot is equipped with an image acquisition device, a time-of-flight ranging sensor and an inertial measurement sensor.

[0035] Specifically, after completing the above preparation phase and confirming that the self-test is normal, the operator issues an "ascend" command from the ground. The motion control unit (e.g., MCU) controls the three drive motors to drive the drive wheel at synchronous speed, causing the robot to ascend at a constant speed along the vertical direction of the pole. During the climbing process, the robot maintains stable attachment of the three sets of climbing modules to the pole surface through the elastic clamping mechanism, without relying on magnetic attraction or negative pressure, and can achieve reliable continuous climbing even on rough circular cross-section concrete poles.

[0036] In some preferred technical solutions, the climbing robot further includes a mechanical fall protection mechanism; the process of controlling the climbing robot to climb along the power line pole to be tested also includes: The system acquires the motor current, drive wheel encoder speed, height measurement value, and real-time climbing attitude output by the inertial measurement sensor of the climbing robot. Based on the combined relationship between the motor current, encoder speed, height measurement value, and real-time climbing posture, abnormal working conditions are identified; among which, the types of abnormal working conditions include wheel slippage, mechanical jamming, and uncontrolled descent. Upon detecting any abnormal operating condition, the power supply to the climbing robot is cut off and the mechanical anti-fall mechanism is activated to lock it in place.

[0037] In this embodiment, the joint judgment criteria for the above-mentioned abnormal operating conditions can be as follows: when the encoder speed is detected to be too high, the height measurement value remains almost unchanged, and the IMU attitude is stable, it is judged as wheel slippage, i.e., the wheel is spinning freely; when the motor current is detected to rise sharply, the encoder speed to drop sharply, and the height measurement value to stagnate, it is judged as mechanical jamming, i.e., mechanical blockage; when the height measurement value is detected to be continuously decreasing and the IMU detects a downward angular velocity exceeding a preset threshold, it is judged as uncontrolled sliding, i.e., an abnormal slide has occurred. The above-mentioned multi-sensor joint judgment mechanism synchronously judges the combination relationship of current, encoder, height, and IMU data to ensure accurate identification of various abnormal operating conditions.

[0038] In the specific implementation of the mechanical fall arrest mechanism, this embodiment arranges a basic guide rail with ratchet teeth on one side of the outer frame. The pawl engages with the guide rail to form a one-way self-locking mechanism. During normal ascent, the pawl, under the action of the spring, sequentially passes over the teeth along the inclined surface of the ratchet teeth; when the robot accidentally slides down, the pawl wedges tightly with the straight edge of the ratchet teeth, self-locking at the nearest tooth position, preventing the robot from sliding down the freefall distance h. maxThe constraint is limited to a tooth pitch p, i.e., it satisfies: h max ≤p; When manual descent is required, the pawl can be temporarily lifted using an electromagnet or mechanical rope, allowing only a slow, preset descent to prevent accidental slippage. Furthermore, emergency stop buttons are prominently located on both the robot body and the ground control unit. Upon triggering, the system immediately cuts off the motor drive power while maintaining the pawl engagement, achieving self-locking upon power failure. Further, when the IMU detects that the robot's overall tilt angle relative to the rod axis exceeds a preset threshold (e.g., 8° to 10°), the system enters a fault state, prohibiting further ascent and executing controlled descent based on the current height, or remaining in place awaiting manual intervention. The entire safety design follows this principle: prioritizing intrinsic fall protection through mechanical structures, supplemented by electrical emergency stops and operational status monitoring, to achieve a multi-layered protection system combining mechanical self-locking, emergency power failure, and intelligent protection.

[0039] The entire safety design follows the basic principles of GB / T 15706 "General Rules for Safety Design of Machinery" and GB / T 5226.1: prioritizing inherent fall protection through mechanical structure, supplemented by electrical emergency stop and operation status monitoring, to achieve a multi-layered protection system that combines mechanical self-locking, emergency power-off and intelligent protection, significantly reducing the risk of equipment falling and misoperation.

[0040] In step S102, during the climbing robot's ascent, acquisition operations are triggered at preset height intervals, thereby acquiring a full-circumference image of the power distribution line pole under test through the image acquisition device at each height section, measuring the distance of the pole through the time-of-flight ranging sensor, and acquiring attitude data through the inertial measurement sensor.

[0041] Specifically, as the climbing robot ascends the utility pole, the controller triggers a synchronous data acquisition operation at fixed height intervals (e.g., every 100mm or 200mm). Each time it is triggered, the controller synchronously records the current height value h, using the section number j as a unified index. j IMU attitude data (including roll angle, pitch angle, and angular velocity) and ToF ranging data are bound together and stored in the same data set G. j When reading data, the full-circumference image, ring distance, and attitude data under that section can be retrieved directly through index j, achieving one-to-one matching.

[0042] In some preferred embodiments, the climbing robot includes a ring detection component, the image acquisition device, the time-of-flight ranging sensor and the inertial measurement sensor are disposed on the ring detection component, and the ring detection component is sleeved on the power line pole to be tested.

[0043] In this embodiment, the ring detection component specifically comprises a split ring bracket, a multi-channel industrial camera and supplementary lighting unit, a time-of-flight (ToF) ranging module, and a support structure. The ring bracket consists of two halves of aluminum alloy or carbon fiber arc-shaped parts, which are assembled into a complete ring on site using quick-release latches. Its inner diameter is slightly larger than the diameter of the power line pole. The ring is connected to the outer frame of the wheeled column climbing mechanism via multiple support arms. During installation, fine adjustments are made using the holes or telescopic structure of the support arms to ensure that the center of the ring is basically coaxial with the pole axis, thus avoiding interference with the climbing module while still covering the target cross-section.

[0044] Three to six industrial cameras, each equipped with a wide-angle lens, are installed at equal intervals along the circumference of the ring to ensure a certain degree of overlap between adjacent fields of view, providing feature redundancy for subsequent image stitching. A white LED strip is installed inside the ring, with additional infrared illumination available for nighttime or low-light scenarios to achieve stable lighting and consistent exposure control. The cameras are connected to the edge computing module via USB or GigE interfaces, and a unified trigger timestamp mechanism is used to achieve synchronous acquisition by multiple cameras, reducing stitching misalignment caused by climbing movements. To minimize the impact of lens distortion on measurements, camera intrinsic parameter calibration and distortion correction can be performed at the factory or during initial deployment, and distortion removal and brightness uniformity processing is applied to the images from each camera before stitching.

[0045] To meet the requirement of quantitative detection of the width of small surface defects such as cracks, let the effective field of view arc length of a single camera on the pole surface be L, and the number of pixels in the horizontal direction be N. x Then the actual size s corresponding to the circumferential pixel is: s=L / N x ; One or more Time-of-Flight (ToF) ranging sensors are placed near each camera to measure the instantaneous distance from the inner wall of the ring to the surface of the rod. Assume the ToF module performs the time difference measurement internally, and the output distance is d. means If the structural dimension from the sensor to the inner edge of the ring is calibrated to l0 during installation, then the radius r of the rod surface at the corresponding height section can be approximated as: r=l0+d means ; After measuring the distance to the ring rod using the time-of-flight ranging sensor, the process further includes: The distances between the ring rods measured by multiple time-of-flight ranging sensors at different positions along the circumference of the ring detection component are selected to obtain a filtered dataset; Calculate the difference between the distances of each ring pole in the filtered dataset, and calculate the eccentricity of the center of the ring detection component relative to the center of the pole of the power distribution line under test; When the maximum value of the difference exceeds the preset eccentricity threshold, the ring detection component is determined to be in an eccentric state, and the pixel scale, circumferential unfolding position, and defect size calculation results based on the full-circumference image acquired by the image acquisition device are corrected according to the eccentricity amount.

[0046] Specifically, for the j-th height section, the controller acquires distance measurements at each circumferential position using multiple ToF ranging sensors arranged in a ring. When there is an eccentricity between the center of the detection ring and the center of the pole, the measured distances between the ring and the pole in each circumferential direction will exhibit a non-uniform distribution. This embodiment calculates the eccentricity in the following way: substituting the installation angle and structural dimensions of each ToF sensor, the measured distances between the ring and the pole are converted into multiple contour points of the height section; by fitting these contour points, the center position of the section is obtained; the center position of the section is compared with the geometric center position of the detection ring to obtain the eccentricity d and the eccentricity direction angle θ.

[0047] When the eccentricity d exceeds the preset eccentricity threshold (e.g., 5mm or 10mm, set according to the pole diameter and detection accuracy requirements), the system initiates eccentricity compensation. The specific contents of the compensation include: (1) correcting the pixel scale of the full-circumference image to compensate for the local stretching or compression of the image caused by eccentricity; (2) offset compensation for the circumferential unfolding position of the image during circumferential unfolding, so that the circumferential coordinates of the spliced ​​unfolded image are aligned with the actual circumferential position of the pole surface; (3) correcting the defect size calculation result identified based on the full-circumference image according to the distortion correction coefficient corresponding to the eccentricity d.

[0048] As a specific implementation, when the detection loop is off-center, the distortion correction coefficient k corresponding to the circumferential unfolding of the image has the following mapping relationship with the off-center distance d: ; △r = △xcosθ + △ysinθ; ; ; In the formula, R is the radius of the pole. This is the corrected pixel scale. The unbiased reference pixel scale (pixel scale before correction) is used. The eccentric coordinates of the detection ring center are (△x, △y), where △x and △y are the eccentric abscissa and eccentric ordinate of the detection ring center, respectively, and △r is the radial offset in the circumferential θ direction.

[0049] Through the above-mentioned eccentricity detection and compensation mechanism, even if the detection ring inevitably produces a certain degree of eccentricity during the climbing robot's ascent on the surface of a rough concrete pole, the system can still ensure the spatial accuracy of the full-circle image stitching and the reliability of defect size quantification.

[0050] Step S103: Based on the full-circle image, the ring rod distance, the attitude data, and the current height value, bind them with a unified timestamp to form a structured data group corresponding to each height section.

[0051] Specifically, using section number j as a unified index, the full-circumference image, ring rod distance, IMU attitude data, and height value h (after binding) of the same height section synchronously triggered in step S102 are combined. j (A mapping relationship exists between the data and the height section) Data is bound and stored using a unified timestamp. Each data group G contains at least the following fields: section index j, height value h. j The data includes full-circumference image data (or a distortion-corrected unfolded diagram), the distance between the ring rods corresponding to each circumferential angle, the IMU roll and pitch angles, and the coordinates of the center position of the cross-section after eccentricity compensation. This structured data organization allows for direct retrieval of complete multi-source data at this height cross-section via index j during subsequent defect identification, geometric state analysis, and result verification, achieving one-to-one data correspondence and full traceability.

[0052] Step S104: Based on the structured data group, identify the apparent defects and geometric state of the pole of the power distribution line to be tested, and construct the mapping relationship between the apparent defect identification results and the pole geometric state identification results in a preset unified pole coordinate system, thereby realizing the state detection of the pole of the power distribution line to be tested.

[0053] In some preferred embodiments, identifying the apparent defects and pole geometry of the power distribution line pole under test based on the structured data set includes: Based on the index corresponding to the height section, query the structured data group to obtain the full-circle image corresponding to the index, the ring rod distance corresponding to the index, and the attitude data corresponding to the index; The full-circle image corresponding to the index is input into a pre-trained deep learning model. Based on the output of the deep learning model, the defect type of the apparent defect of the power distribution line pole under test and the pixel coordinates of the apparent defect in the full-circle image are obtained, thereby realizing the identification of the apparent defect of the power distribution line pole under test. Obtain the installation angle of the time-of-flight ranging sensor; determine the center position of the cross-section of the height section corresponding to the index based on the ring rod distance corresponding to the index and the installation angle; Based on the center position of the cross section and the posture data corresponding to the index, the mapping relationship between the height and tilt state of the power distribution line pole under test is analyzed, thereby realizing the identification of the pole geometry state of the power distribution line pole under test.

[0054] Specifically, the above-mentioned appearance defect identification process is completed in conjunction with the edge computing unit. The edge computing unit performs distortion correction, brightness equalization, and image stitching on the images acquired by each camera to generate a unfolded diagram of the pole corresponding to the height section. A pre-trained deep learning model (such as a semantic segmentation or object detection model based on a convolutional neural network) analyzes the unfolded diagram, identifies and labels appearance defect types such as longitudinal cracks, transverse cracks, honeycombing, pitting, spalling, and exposed reinforcement, as well as their pixel coordinates in the image.

[0055] In this embodiment, the edge computing unit can be a low-power x86 industrial computer or an ARM core embedded board to perform tasks such as industrial camera image reception, distortion correction and stitching, crack and defect identification, tilt calculation and data compression.

[0056] By combining pre-labeled pole defect samples (up to several thousand images), a lightweight target detection or segmentation network is used to train the model, achieving a detection accuracy of over 0.9 mAP@0.5, and inference is completed at a near real-time frame rate. Compared to traditional manual image review, the edge computing module can output defect type, location, and size online, significantly improving detection efficiency and consistency.

[0057] The power supply section uses a 24V lithium battery pack in conjunction with a BMS management system to provide main power to the three drive motors, and distributes 12V and 5V voltages to the camera, IMU, ToF, and edge computing boards via a DC-DC module. Battery capacity C Ah You can set it as follows: ; In the formula, P m This refers to the rated power of a single drive motor. P s For edge computing and sensor power consumption, T work The expected continuous working time for a single task. S E This is the energy safety factor (which can range from 1.2 to 1.3). This formula allows for the control of battery size and weight while ensuring battery life, thus balancing portability and reliability.

[0058] In terms of communication and human-machine interaction, the climbing robot connects to a ground-based tablet, laptop, or dedicated control terminal via Wi-Fi or wired Ethernet. Operators can view the operating status, images, and alarm information in real time on the host computer interface and issue commands such as ascending, descending, and stopping. It is also equipped with a simple handheld remote control or a wired control box to meet the needs of close-range debugging and emergency operation. The electrical design generally conforms to GB / T 5226.1 "Electrical Safety of Machinery - General Technical Conditions for Electrical Equipment," meeting general requirements in terms of emergency stop circuits, self-holding and reset logic, protection against electric shock, and wiring.

[0059] The controller is configured with a unified data acquisition triggering mechanism. Whenever the height measurement device detects that the robot has risen to a preset height interval Δh, or the encoder's accumulated pulses reach a preset value, the controller simultaneously triggers data recording from the camera, ToF module, and IMU, and writes the same timestamp and the same height number to this set of data.

[0060] In the pole geometry state recognition, this embodiment further considers the scenario where the ToF ranging data changes abruptly when the robot passes through local protrusions on the pole (such as junction boxes, honeycomb protrusions, or local bulges) during the climbing process. Since such abrupt changes in the ToF cross-section center caused by local protrusions are not the true overall tilt of the pole, if they are not eliminated, they will introduce "pseudo-tilt" interference, affecting the accuracy of tilt assessment. To this end, this embodiment adopts a dynamic weight allocation strategy that integrates ToF and IMU: the system compares the center coordinates of the cross-section obtained by ToF fitting of the current height cross-section with the ToF center coordinates of the historical cross-sections within the sliding window in real time. If the deviation exceeds a preset threshold, it is determined that a data abrupt change caused by a local protrusion has occurred in the current cross-section. At this time, the weight of ToF data in the tilt calculation is immediately reduced, and the tilt calculation is based mainly on IMU attitude data. At the same time, a second verification is performed by combining the mean of the sliding window to eliminate single-point outliers and retain the continuous and stable true tilt trend of the pole, thereby effectively eliminating "pseudo-tilt" interference.

[0061] Preferably, determining the center position of the cross-section corresponding to the height of the index based on the distance between the ring rods corresponding to the index and the installation angle includes: Based on the installation angle and structural dimensions of each time-of-flight ranging sensor, the distance of the ring rod corresponding to the index is converted into multiple contour points of the height section; The center position of the height section is obtained by fitting the multiple contour points using circle fitting, ellipse fitting, or least squares fitting.

[0062] Specifically, based on the circumferential distances to the poles obtained by the ToF array, and combined with the pre-calibrated sensor installation angles and structural dimensions, each distance value is reconstructed into the coordinates of discrete contour points on the pole surface at that height. Since the cross-section of power distribution line poles is not an ideal circle, any one of three methods can be used to fit the discrete contour points to obtain a more accurate cross-sectional center position: circle fitting (suitable for scenarios with good cross-sectional roundness), ellipse fitting (suitable for scenarios with a certain degree of ellipticity), or least squares fitting (suitable for scenarios where contour points have noise or outliers). By fitting the center positions of each cross-section along the height direction, the spatial centerline of the pole can be obtained; further, by combining the IMU attitude data of each cross-section, a height-tilt angle mapping relationship can be established.

[0063] In some preferred embodiments, after querying the structured data group based on the index corresponding to the height section, the method further includes: obtaining the height value corresponding to the index. The unified pole coordinate system uses the height value as the height coordinate, the installation circumferential angle of the image acquisition device and the time-of-flight ranging sensor as the circumferential coordinate, and the distance between the ring poles as the radial coordinate.

[0064] The step of constructing a mapping relationship between the apparent defect identification results and the bar geometric state identification results in a preset unified bar coordinate system includes: Based on the installation circumferential angle of the image acquisition device corresponding to the full-circumference image of the index, the ring rod distance corresponding to the index, and the height value corresponding to the index, the pixel coordinates are converted into the spatial position of the apparent defect in the unified rod coordinate system. Based on the mapping relationship between the spatial location, height, and tilt state, as well as the height coordinates and circumferential coordinates, a mapping relationship is constructed between the appearance defect identification results and the pole geometry identification results.

[0065] Through the aforementioned unified pole coordinate system (using height h as the height coordinate, the circumferential angle θ of each camera and sensor installation as the circumferential coordinate, and the radial distance r calculated from the distance between the pole rings as the radial coordinate), the apparent defect identification results at each height section (including defect type, defect size, and defect location information) are uniformly bound to the cross-sectional geometry (including cross-sectional radius changes and eccentricity) and tilt state at the same height and circumferential position. Therefore, when outputting the detection results, not only can the type of defect present at a certain location on the pole be determined, but the trend of pole radius change at the height of the defect, the tilt angle of the pole relative to the vertical direction at that location, and the correlation between the two can also be simultaneously provided, thus offering maintenance personnel a more complete basis for judging the structural condition of the pole.

[0066] Preferably, after identifying the apparent defects of the power distribution line pole under test, the method further includes: Obtain the confidence level for identifying the defect type. When the confidence level is higher than a first confidence threshold and lower than a second confidence threshold, enter the local re-inspection mode; wherein, the first confidence threshold is lower than the second confidence threshold. In the local re-inspection mode, the climbing robot is controlled to reduce its climbing speed or stop at the current height, and the data acquisition operation is re-triggered to achieve re-inspection.

[0067] Specifically, when the confidence level of a suspected defect identified by the deep learning model falls within the range between the first and second confidence thresholds, the system automatically enters a local re-inspection mode. In this mode, the motion control unit controls the robot to reduce its climbing speed or stop near its current height; the camera exposure control module readjusts the exposure time based on the current image brightness; the supplementary lighting control module increases or changes the supplementary lighting brightness to ensure image quality; and the ToF ranging module synchronously re-acquires the distance data of the ring rod at that height. If necessary, the motion control unit controls the robot to retreat a preset distance and then ascend again to acquire data, ensuring that at least two sets of images and distance data for the same suspected defect area are obtained at different times or in different postures. The system performs position registration between the initial acquisition results and the re-inspection results. If the two acquisition results meet the consistency conditions in defect height, circumferential position, defect direction, and defect size, the output is a confirmed defect; if the two results do not meet the consistency conditions, the area is marked as an area awaiting manual review, and the inspection report indicates that the area has insufficient image quality, abnormal ranging, or unstable identification results.

[0068] After completing all climbing inspections, the edge computing unit stores the raw images, defect identification results, and tilt data generated during the inspection task locally, and transmits them to the host computer software via a wired or wireless network interface. The host computer then compiles a standardized inspection report based on the pole number and height information, including defect type, location, size, corresponding cross-sectional geometry, and tilt. This report can be exported to formats such as PDF and supports archiving, querying, and trend analysis on the power transmission and distribution network operation and maintenance platform, thus providing data support for the full lifecycle status management of concrete poles in power distribution lines.

[0069] Accordingly, such as Figure 2 As shown, this invention application also provides a status detection system 200 for power distribution line poles, including a control module 201, a data acquisition module 202, a data group construction module 203, and a status detection module 204; wherein, The control module 201 is used to control the climbing robot to climb along the pole of the power distribution line to be tested; wherein, the climbing robot is equipped with an image acquisition device, a time-of-flight ranging sensor and an inertial measurement sensor; The data acquisition module 202 is used to trigger acquisition operations at preset height intervals during the climbing robot's ascent, thereby acquiring a full-circumference image of the power line pole under test through the image acquisition device at each height section, measuring the distance of the pole through the time-of-flight ranging sensor, and acquiring attitude data through the inertial measurement sensor. The data group construction module 203 is used to bind the full-circle image, the ring rod distance, the attitude data and the current height value into structured data groups corresponding to each height section with a unified timestamp. The state detection module 204 is used to identify the apparent defects and geometric state of the power distribution line pole under test based on the structured data group, and to construct a mapping relationship between the apparent defect identification results and the pole geometric state identification results in a preset unified pole coordinate system, thereby realizing the state detection of the power distribution line pole under test.

[0070] As a preferred embodiment, the state detection module 204 identifies the apparent defects and geometric state of the pole of the power distribution line under test based on the structured data set, including: The state detection module 204 queries the structured data group according to the index corresponding to the height section to obtain the full-circle image, the ring rod distance, and the attitude data corresponding to the index. The full-circle image corresponding to the index is input into a pre-trained deep learning model. Based on the output of the deep learning model, the defect type of the apparent defect of the power distribution line pole under test and the pixel coordinates of the apparent defect in the full-circle image are obtained, thereby realizing the identification of the apparent defect of the power distribution line pole under test. Obtain the installation angle of the time-of-flight ranging sensor; determine the center position of the cross-section of the height section corresponding to the index based on the ring rod distance corresponding to the index and the installation angle; Based on the center position of the cross section and the posture data corresponding to the index, the mapping relationship between the height and tilt state of the power distribution line pole under test is analyzed, thereby realizing the identification of the pole geometry state of the power distribution line pole under test.

[0071] As a preferred embodiment, the state detection module 204 is further configured to: obtain the height value corresponding to the index; The unified pole coordinate system uses the height value as the height coordinate, the installation circumferential angle of the image acquisition device and the time-of-flight ranging sensor as the circumferential coordinate, and the distance between the ring poles as the radial coordinate; The state detection module 204 constructs a mapping relationship between the apparent defect identification results and the geometric state identification results of the rod in a preset unified rod coordinate system, including: The state detection module 204 converts the pixel coordinates into the spatial position of the apparent defect in the unified rod coordinate system based on the installation circumferential angle of the image acquisition device of the full-circumference image corresponding to the index, the ring rod distance corresponding to the index, and the height value corresponding to the index. Based on the mapping relationship between the spatial location, height, and tilt state, as well as the height coordinates and circumferential coordinates, a mapping relationship is constructed between the appearance defect identification results and the pole geometry identification results.

[0072] As a preferred embodiment, the state detection module 204 determines the center position of the cross-section corresponding to the height of the index based on the ring rod distance and the installation angle, including: The state detection module 204 converts the distance of the ring rod corresponding to the index into multiple contour points of the height section according to the installation angle and structural installation dimensions of each time-of-flight ranging sensor. The center position of the height section is obtained by fitting the multiple contour points using circle fitting, ellipse fitting, or least squares fitting.

[0073] As a preferred embodiment, the condition detection system 200 further includes a re-inspection module, which is used after the identification of apparent defects in the power distribution line pole under test is achieved: Obtain the confidence level for identifying the defect type. When the confidence level is higher than a first confidence threshold and lower than a second confidence threshold, enter the local re-inspection mode; wherein, the first confidence threshold is lower than the second confidence threshold. In the local re-inspection mode, the climbing robot is controlled to reduce its climbing speed or stop at the current height, and the data acquisition operation is re-triggered to achieve re-inspection.

[0074] As a preferred embodiment, the climbing robot includes a ring detection component, the image acquisition device, the time-of-flight ranging sensor and the inertial measurement sensor are disposed on the ring detection component, and the ring detection component is sleeved on the power pole of the power distribution line to be tested; The state detection system 200 further includes a correction module, which is used after the time-of-flight ranging sensor measures the distance to the ring rod: The distances between the ring rods measured by multiple time-of-flight ranging sensors at different positions along the circumference of the ring detection component are selected to obtain a filtered dataset; Calculate the difference between the distances of each ring pole in the filtered dataset, and calculate the eccentricity of the center of the ring detection component relative to the center of the pole of the power distribution line under test; When the maximum value of the difference exceeds the preset eccentricity threshold, the ring detection component is determined to be in an eccentric state, and the pixel scale, circumferential unfolding position, and defect size calculation results based on the full-circumference image acquired by the image acquisition device are corrected according to the eccentricity amount.

[0075] As a preferred embodiment, the climbing robot further includes a mechanical fall protection mechanism; during the process of controlling the climbing robot to climb along the power line pole to be tested, the control module 201 is also used for: The system acquires the motor current, drive wheel encoder speed, height measurement value, and real-time climbing attitude output by the inertial measurement sensor of the climbing robot. Based on the combined relationship between the motor current, encoder speed, height measurement value, and real-time climbing posture, abnormal working conditions are identified; among which, the types of abnormal working conditions include wheel slippage, mechanical jamming, and uncontrolled descent. Upon detecting any abnormal operating condition, the power supply to the climbing robot is cut off and the mechanical anti-fall mechanism is activated to lock it in place.

[0076] Accordingly, this application also provides a terminal device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor executes the computer program to implement the power distribution line pole status detection method.

[0077] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the terminal, connecting various parts of the terminal via various interfaces and lines.

[0078] The memory can be used to store the computer program. The processor implements various functions of the terminal by running or executing the computer program stored in the memory and calling the data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, phonebook, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0079] Accordingly, this application also provides a computer-readable storage medium, which includes a stored computer program, wherein the computer program, when running, controls the device where the computer-readable storage medium is located to execute the power distribution line pole status detection method.

[0080] Wherein, if the modules of the device / terminal equipment / system integration are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when the computer program is executed by a processor, it can implement the steps of the various method embodiments described above. Wherein, the computer program includes computer program code, which can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc.

[0081] Compared with the prior art, this invention application has the following beneficial effects: This invention provides a method, system, device, and medium for detecting the condition of power distribution line poles. By uniformly triggering image acquisition equipment, time-of-flight ranging sensors, and inertial measurement sensors at preset height intervals during the climbing process, and binding the full-circumference images, pole distance, and attitude data at the same height section into a structured data set with a unified timestamp, this invention ensures that the multi-source data at each height section accurately correspond in terms of acquisition time and spatial location. This fundamentally avoids the problems in existing technologies where different data streams are asynchronous in time and inconsistent in spatial reference due to the independent operation of each sensor and the lack of a unified triggering mechanism. This guarantees that subsequent defect identification results and pole geometric state identification results can be correlated under the same spatial reference, effectively improving the accuracy of power distribution line pole condition detection.

[0082] Furthermore, since the full-circumference images, ring rod distances, attitude data, and height values ​​of each height section are all bound to a structured data group with a unified timestamp, each apparent defect and each geometric anomaly in the condition detection results can be retrieved through a unified spatial index to retrieve the corresponding full-circumference image, distance data, and attitude data original records. This allows subsequent analysis to be strictly performed at the same height, avoiding the verification difficulties caused by scattered data and lack of a unified index in existing technologies.

[0083] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A method for detecting the condition of power distribution line poles, characterized in that, include: The climbing robot is controlled to climb along the pole of the power distribution line to be tested; wherein, the climbing robot is equipped with an image acquisition device, a time-of-flight ranging sensor and an inertial measurement sensor; During the climbing process of the climbing robot, the acquisition operation is triggered at preset height intervals, so that at each height section, the image acquisition device can acquire the full-circumference image of the power line pole under test, the time-of-flight distance sensor can measure the distance of the pole, and the inertial measurement sensor can collect attitude data. Based on the full-circle image, the ring rod distance, the attitude data, and the current height value, a structured data group corresponding to each height section is bound with a unified timestamp; Based on the structured data set, the apparent defects and geometric state of the poles of the power distribution line under test are identified. A mapping relationship between the apparent defect identification results and the pole geometric state identification results is constructed in a preset unified pole coordinate system, thereby realizing the state detection of the poles of the power distribution line under test.

2. The method for detecting the condition of power distribution line poles as described in claim 1, characterized in that, The identification of apparent defects and pole geometry of the power distribution line poles under test based on the structured data set includes: Based on the index corresponding to the height section, query the structured data group to obtain the full-circle image corresponding to the index, the ring rod distance corresponding to the index, and the attitude data corresponding to the index; The full-circle image corresponding to the index is input into a pre-trained deep learning model. Based on the output of the deep learning model, the defect type of the apparent defect of the power distribution line pole under test and the pixel coordinates of the apparent defect in the full-circle image are obtained, thereby realizing the identification of the apparent defect of the power distribution line pole under test. Obtain the installation angle of the time-of-flight ranging sensor; determine the center position of the cross-section of the height section corresponding to the index based on the ring rod distance corresponding to the index and the installation angle; Based on the center position of the cross section and the posture data corresponding to the index, the mapping relationship between the height and tilt state of the power distribution line pole under test is analyzed, thereby realizing the identification of the pole geometry state of the power distribution line pole under test.

3. The method for detecting the condition of power distribution line poles as described in claim 2, characterized in that, The step of querying the structured data group based on the index corresponding to the height section further includes: obtaining the height value corresponding to the index; The unified pole coordinate system uses the height value as the height coordinate, the installation circumferential angle of the image acquisition device and the time-of-flight ranging sensor as the circumferential coordinate, and the distance between the ring poles as the radial coordinate; The step of constructing a mapping relationship between the apparent defect identification results and the bar geometric state identification results in a preset unified bar coordinate system includes: Based on the installation circumferential angle of the image acquisition device corresponding to the full-circumference image of the index, the ring rod distance corresponding to the index, and the height value corresponding to the index, the pixel coordinates are converted into the spatial position of the apparent defect in the unified rod coordinate system. Based on the mapping relationship between the spatial location, height, and tilt state, as well as the height coordinates and circumferential coordinates, a mapping relationship is constructed between the appearance defect identification results and the pole geometry identification results.

4. The method for detecting the condition of power distribution line poles as described in claim 2, characterized in that, The step of determining the center position of the cross-section corresponding to the height of the index based on the ring rod distance and the installation angle includes: Based on the installation angle and structural dimensions of each time-of-flight ranging sensor, the distance of the ring rod corresponding to the index is converted into multiple contour points of the height section; The center position of the height section is obtained by fitting the multiple contour points using circle fitting, ellipse fitting, or least squares fitting.

5. The method for detecting the condition of power distribution line poles as described in claim 2, characterized in that, After identifying the apparent defects of the poles of the power distribution line under test, the method further includes: Obtain the confidence level for identifying the defect type. When the confidence level is higher than a first confidence threshold and lower than a second confidence threshold, enter the local re-inspection mode; wherein, the first confidence threshold is lower than the second confidence threshold. In the local re-inspection mode, the climbing robot is controlled to reduce its climbing speed or stop at the current height, and the data acquisition operation is re-triggered to achieve re-inspection.

6. The method for detecting the condition of power distribution line poles as described in claim 1, characterized in that, The climbing robot includes a ring detection component. The image acquisition device, time-of-flight ranging sensor, and inertial measurement sensor are mounted on the ring detection component, which is sleeved on the power pole of the power distribution line to be tested. After measuring the distance to the ring rod using the time-of-flight ranging sensor, the process further includes: The distances between the ring rods measured by multiple time-of-flight ranging sensors at different positions along the circumference of the ring detection component are selected to obtain a filtered dataset; Calculate the difference between the distances of each ring pole in the filtered dataset, and calculate the eccentricity of the center of the ring detection component relative to the center of the pole of the power distribution line under test; When the maximum value of the difference exceeds the preset eccentricity threshold, the ring detection component is determined to be in an eccentric state, and the pixel scale, circumferential unfolding position, and defect size calculation results based on the full-circumference image acquired by the image acquisition device are corrected according to the eccentricity amount.

7. A method for detecting the condition of power distribution line poles as described in any one of claims 1 to 6, characterized in that, The climbing robot also includes a mechanical fall protection mechanism; the process of controlling the climbing robot to climb along the power line pole to be tested also includes: The system acquires the motor current, drive wheel encoder speed, height measurement value, and real-time climbing attitude output by the inertial measurement sensor of the climbing robot. Based on the combined relationship between the motor current, encoder speed, height measurement value, and real-time climbing posture, abnormal working conditions are identified; among which, the types of abnormal working conditions include wheel slippage, mechanical jamming, and uncontrolled descent. Upon detecting any abnormal operating condition, the power supply to the climbing robot is cut off and the mechanical anti-fall mechanism is activated to lock it in place.

8. A condition monitoring system for power distribution line poles, characterized in that, It includes a control module, a data acquisition module, a data group construction module, and a status detection module; among which, The control module is used to control the climbing robot to climb along the power pole of the power distribution line to be tested; wherein, the climbing robot is equipped with an image acquisition device, a time-of-flight ranging sensor and an inertial measurement sensor; The data acquisition module is used to trigger acquisition operations at preset height intervals during the climbing robot's ascent, thereby acquiring a full-circumference image of the power line pole under test through the image acquisition device at each height section, measuring the distance of the pole through the time-of-flight ranging sensor, and acquiring attitude data through the inertial measurement sensor. The data group construction module is used to bind the full-circle image, the ring rod distance, the attitude data and the current height value into structured data groups corresponding to each height section with a unified timestamp; The state detection module is used to identify the apparent defects and geometric state of the power distribution line pole under test based on the structured data group, and to construct a mapping relationship between the apparent defect identification results and the pole geometric state identification results in a preset unified pole coordinate system, thereby realizing the state detection of the power distribution line pole under test.

9. A terminal device, characterized in that, The method includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor, when executing the computer program, implements the state detection method for power distribution line poles as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program, wherein, when the computer program is executed, it controls the device containing the computer-readable storage medium to perform the state detection method for power distribution line poles as described in any one of claims 1 to 7.

Images