Power grid cable inspection method and system based on unmanned aerial vehicle
By using drones to construct digital twin models of cables and perform finite element analysis, the problem of inaccurate cable load assessment has been solved, enabling precise monitoring of cable icing status and decision support for de-icing, thus improving the stable operation of the power grid under severe weather conditions.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-23
- Publication Date
- 2026-04-07
AI Technical Summary
Existing cable inspection methods cannot accurately assess cable load in icy and snowy weather, which can easily lead to misjudgments and result in untimely or excessive de-icing operations.
The UAV-based cable inspection method generates inspection routes by constructing a digital twin model of the cable, acquiring images, extracting cable and icicle contours, constructing an icing model, and performing finite element analysis to assess the cable stress and determine whether de-icing operations are necessary.
It enables precise, efficient, and intelligent monitoring of cable icing conditions, improves the accuracy and timeliness of early warnings, reduces the safety risks and maintenance costs of manual inspections, and can quickly assess whether de-icing is necessary.
Smart Images
Figure CN121813680A_ABST
Abstract
Description
Technical Field
[0002] This invention belongs to the field of cable inspection technology, and in particular relates to a method and system for power grid cable inspection based on unmanned aerial vehicles (UAVs). Background Technology
[0003] Power grid cable inspection refers to the systematic inspection of the cable body, joints, terminals, channels and ancillary facilities by power operation and maintenance personnel in accordance with national and industry standards on a regular or as-needed basis. Through visual inspection, infrared thermography, partial discharge detection, insulation resistance testing and other means, potential hazards such as outer sheath damage, joint overheating, insulation aging, tunnel water accumulation and chemical corrosion can be detected in a timely manner, thereby preventing faults and ensuring power supply safety.
[0004] In the current cable inspection process, inspections are mainly carried out manually. In icy and snowy weather, ice and icicles are easily formed on cables. When assessing cables, the safety of the cables is mainly evaluated based on the thickness of the ice layer. When the thickness of the ice layer reaches a dangerous value, the cable is judged to be at risk of failure and de-icing operations are required. The current assessment method cannot accurately assess the load of the cables and is prone to misjudgment. Summary of the Invention
[0005] The purpose of this invention is to provide a method for inspecting power grid cables based on unmanned aerial vehicles (UAVs), which aims to solve the problem that current assessment methods cannot accurately assess the load of cables and are prone to misjudgment.
[0006] This invention is implemented as follows: a method for inspecting power grid cables based on unmanned aerial vehicles (UAVs), the method comprising: A digital twin model of the cable is constructed based on the installation location of the towers and the cable span. The digital twin model of the cable includes at least two towers and the cable between the towers. An inspection route is generated based on the installation location of the tower. The drone is controlled to fly based on the inspection route. During the flight, images of the cable are acquired to obtain local cable images. The inspection route is parallel to the cable between the tower. The cables in the local cable images are extracted and stitched together. The outline of the cable in the stitched image is extracted and divided into cable outline and icicle outline. Based on the cable outline and icicle outline, a cable icing model is constructed. The cable icing model is meshed to construct a finite element analysis model, and the stress on the cable at various points is analyzed. The analysis results are output, and the decision on whether to carry out de-icing operations is made based on the analysis results.
[0007] Preferably, the steps of generating an inspection route based on the installation location of the tower, controlling the drone to fly based on the inspection route, and acquiring images of the cable during flight to obtain local cable images specifically include: The coordinates of the tower and the coordinates of each point on the cable are extracted from the digital twin model of the cable. An inspection route is generated based on the coordinates of the cable, and the inspection route is kept parallel to the cable. The drone flight is controlled based on the inspection route, and image acquisition points are set at preset flight distance intervals during the drone flight. The drone is controlled to capture images of the cable at the image acquisition point, thus obtaining a local cable image.
[0008] Preferably, the steps of extracting and stitching cables from local cable images, extracting the cable contours from the stitched images, dividing the contours into cable contours and icicle contours, and constructing a cable icing model based on the cable contours and icicle contours specifically include: Edge recognition is performed on local cable images to extract the edges of ice blocks on the cable, and then the edges of the cable edges are constructed by stitching them together. The contours in the cable edge image are divided to obtain the cable contour and the icicle contour. The horizontally distributed contours are the contours of the ice surrounding the cable, and the vertically distributed contours are the contours of the icicles. The cable icing structure is constructed based on the cable outline, and the ice ridge structure is constructed based on the ice ridge outline. The icing structure and the ice ridge structure are then fused to obtain the cable icing model.
[0009] Preferably, the steps of meshing the cable icing model, constructing a finite element analysis model, analyzing the stress on the cable at various points, outputting the analysis results, and determining whether to perform de-icing operations based on the analysis results include: The cable icing model is divided into multiple model elements based on the total length of the cable. The icing weight of each model element is calculated to generate the corresponding finite element analysis model of the cable. Finite element analysis is performed based on the forces at each point to determine the forces at any point on the cable and output the analysis results. If the tension value at a single point on the cable exceeds the preset value, it is determined that the tension is excessive and de-icing operation is required.
[0010] Preferably, the method further includes extracting the curvature of each point of the cable from the cable icing model, comparing it with the cable digital twin model, and determining whether there is any tensile abnormality.
[0011] Another object of the present invention is to provide a power grid cable inspection system based on unmanned aerial vehicles (UAVs), the system comprising: The twin model building module is used to build a cable digital twin model based on the installation location of the towers and the cable span. The cable digital twin model includes at least two towers and the cable between the towers. The image acquisition module is used to generate an inspection route based on the installation location of the tower, control the drone to fly based on the inspection route, and acquire images of the cable during the flight to obtain local cable images. The inspection route is parallel to the cable between the tower. The contour recognition module is used to extract and stitch cables in local cable images, extract the contours of cables in the stitched images, divide the contours into cable contours and icicle contours, and construct a cable icing model based on the cable contours and icicle contours. The de-icing analysis module is used to mesh the cable icing model, construct a finite element analysis model, analyze the stress on the cable at various points, output the analysis results, and determine whether de-icing operations should be carried out based on the analysis results.
[0012] Preferably, the image acquisition module includes: The inspection route construction unit is used to extract the coordinates of the tower and the coordinates of each point on the cable from the cable digital twin model, and generate the inspection route based on the cable coordinates. The inspection route is parallel to the cable. The data acquisition point generation unit is used to control the flight of the UAV based on the inspection route and to set image acquisition points at preset flight distance intervals during the flight of the UAV. The cable imaging unit is used to control the drone to capture images of the cable at the image acquisition point and obtain local cable images.
[0013] Preferably, the contour recognition module includes: The edge image stitching unit is used to perform edge recognition on local cable images, extract the edges of ice blocks on the cable, and construct a cable edge image by stitching them together. The contour segmentation unit is used to segment the contours in the cable edge image to obtain the cable contour and the icicle contour. The horizontally distributed contours are the contours of the ice surrounding the cable, and the vertically distributed contours are the contours of the icicles. The icing model construction unit is used to construct the cable icing structure based on the cable outline, construct the rotating body based on the ice ridge outline to obtain the ice ridge structure, and fuse the icing structure and the ice ridge structure to obtain the cable icing model.
[0014] Preferably, the de-icing analysis module includes: The analysis model building unit is used to divide the cable icing model into multiple model units based on the total length of the cable, calculate the icing weight of each model unit, and generate the corresponding finite element analysis model of the cable. The analysis result output unit is used to perform finite element analysis based on the forces at each point, determine the forces at any point on the cable, and output the analysis results. The tension determination unit is used to determine that there is excessive tension when the tension value at a single point on the cable exceeds the preset value, and that de-icing operation is required.
[0015] Preferably, the system's operation also includes extracting the curvature of each point on the cable from the cable icing model, comparing it with the cable digital twin model, and determining whether there is any tensile abnormality.
[0016] This invention provides a method for inspecting power grid cables based on unmanned aerial vehicles (UAVs). By combining the construction of a digital twin model of the cable with autonomous UAV inspection, it achieves accurate, efficient, and intelligent monitoring of the cable icing status. Through image extraction and stitching technology, it accurately distinguishes the cable from the ice ridge outline, and performs stress simulation on the icy cable based on finite element analysis. This enables a scientific assessment of icing risk and stress status, significantly improving the accuracy and timeliness of early warnings, providing a reliable basis for de-icing decisions, effectively ensuring the stable operation of the power grid under severe weather conditions, reducing the safety risks and maintenance costs of manual inspection, and enabling rapid assessment of whether each cable section needs de-icing with higher accuracy than manual inspection. Attached Figure Description
[0017] Figure 1 A flowchart of a power grid cable inspection method based on unmanned aerial vehicles (UAVs) provided in an embodiment of the present invention; Figure 2 A flowchart illustrating the steps of generating an inspection route based on the installation location of the tower, controlling the flight of a drone based on the inspection route, acquiring images of the cable during the flight, and obtaining local cable images, is provided in this embodiment of the invention. Figure 3 The flowchart illustrates the steps of extracting and stitching cables from local cable images, extracting the cable contours from the stitched images, dividing the contours into cable contours and icicle contours, and constructing a cable icing model based on the cable contours and icicle contours, as provided in this embodiment of the invention. Figure 4 The flowchart illustrates the steps of meshing a cable icing model, constructing a finite element analysis model, analyzing the stress on the cable at various points, outputting the analysis results, and determining whether to perform de-icing operations based on the analysis results, provided in this embodiment of the invention. Figure 5 An architecture diagram of a UAV-based power grid cable inspection system is provided for an embodiment of the present invention; Figure 6 This is an architectural diagram of an image acquisition module provided in an embodiment of the present invention; Figure 7 An architecture diagram of a contour recognition module provided in an embodiment of the present invention; Figure 8 This is an architecture diagram of a de-icing analysis module provided in an embodiment of the present invention. Detailed Implementation
[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.
[0019] like Figure 1 The diagram shows a flowchart of a UAV-based power grid cable inspection method according to an embodiment of the present invention. The method includes: S100, based on the installation location of the towers and the cable span, constructs a cable digital twin model, which includes at least two towers and the cable between the towers.
[0020] In this step, a digital twin model of the cable is constructed based on the installation location of the tower and the cable span. The towers that need to be monitored are modeled, and the 3D models of the towers are generated in 3D software. The parameter information of the towers is recorded, including the tower height and GPS coordinates. The total span, length and diameter of the cables between adjacent towers are also recorded to construct a digital twin model of the cable based on the actual installation conditions. Based on the digital twin model of the cable, the parameters of any tower and the parameters of the cables between towers can be quickly obtained.
[0021] S200 generates an inspection route based on the installation location of the tower, controls the flight of the UAV based on the inspection route, and acquires images of the cable during the flight to obtain local cable images. The inspection route is parallel to the cable between the tower.
[0022] In this step, an inspection route is generated based on the installation location of the tower. The parameters of the cable to be monitored are retrieved from the cable digital twin model, including the cable span, total length, and GPS coordinates of the endpoints. The cable's position in space is determined based on the cable span, total length, and endpoint height. The coordinates of each point on the cable are then determined, generating a series of coordinates corresponding to the cable. This series of coordinates is offset as a whole to generate a set of inspection routes parallel to the cable. The drone is controlled to fly based on this inspection route, and images of the cable are acquired during the flight. Each image acquisition yields a set of local cable images.
[0023] S300 extracts and stitches the cable in the local cable image, extracts the outline of the cable in the stitched image, divides the outline into cable outline and icicle outline, and constructs a cable icing model based on the cable outline and icicle outline.
[0024] In this step, the cable in the local cable image is extracted and stitched together. The local cable image contains the cable image. By stitching together two adjacent sets of local cable images, a continuous and complete spliced cable image is obtained. By performing edge recognition, the edge positions of the cable and ice layer in the spliced image are determined, and the overall edge positions of the cable and ice layer are extracted. The ice layer on the cable is mainly divided into two parts, including the columnar ice layer surrounding the cable and the conical ice ridges located below the cable. Based on this, the ice layer on the cable is divided into two parts to obtain the cable outline and the ice ridge outline. The cable outline is the outline corresponding to the columnar ice layer, and the ice ridge outline is the outline corresponding to the ice ridge. Based on the structure of the two, the corresponding model is constructed in three-dimensional space to obtain the cable icing model.
[0025] The S400 algorithm meshes the cable icing model, constructs a finite element analysis model, analyzes the stress on the cable at various points, outputs the analysis results, and determines whether de-icing operations should be carried out based on the analysis results.
[0026] In this step, the cable icing model is meshed, and the parameters of the cable icing model are set according to the preset ice density. The weight of the ice block and the weight of the cable corresponding to each grid position on the cable icing model are determined. The finite element analysis model can output the tensile force borne by each point on the cable to determine whether there is a local tensile force exceeding the limit. If so, it is determined that de-icing operation is required.
[0027] like Figure 2 As shown in the preferred embodiment of the present invention, the steps of generating an inspection route based on the installation location of the tower, controlling the drone to fly based on the inspection route, and acquiring images of the cable during flight to obtain local cable images specifically include: S201 extracts the coordinates of the tower and the coordinates of various points on the cable from the cable digital twin model, and generates an inspection route based on the cable coordinates, with the inspection route kept parallel to the cable.
[0028] In this step, the coordinates of the tower and the coordinates of each point on the cable are extracted from the cable digital twin model. Since the corresponding endpoint coordinates are set for the cable endpoints in the cable digital twin model, the spatial distribution can be determined based on the total length of the cable, the height of the endpoints, and the span of the cable, thereby outputting the coordinates of each point on the cable. The cable coordinates are translated to construct the inspection route, which is parallel to the cable.
[0029] S202 controls the flight of the drone based on the inspection route, and sets image acquisition points at preset flight distance intervals during the drone's flight.
[0030] In this step, the drone is controlled to fly based on the inspection route, and image acquisition points are set according to the preset flight interval. Specifically, based on the cable length included in the images acquired by the drone, adjacent images are made to have overlapping areas to facilitate subsequent image stitching.
[0031] S203 controls the drone to acquire images of the cable at the image acquisition point, obtaining a local cable image.
[0032] In this step, the drone is controlled to collect images of the cable at the image acquisition point. Whenever the drone arrives at a new image acquisition point, it is controlled to hover there and collect an image, obtaining a set of local cable images. Then, it flies along the inspection route to the next image sampling point.
[0033] like Figure 3 As shown, in a preferred embodiment of the present invention, the steps of extracting and stitching the cable in a local cable image, extracting the cable contour in the stitched image, dividing the contour into cable contour and icicle contour, and constructing a cable icing model based on the cable contour and icicle contour specifically include: S301 performs edge recognition on a local cable image, extracts the edges of ice blocks on the cable, and constructs a cable edge image by stitching them together.
[0034] In this step, edge recognition is performed on local cable images. Specifically, distortion correction and illumination normalization preprocessing are performed on the local cable images. A multi-scale edge detection method is used to identify the edge contour of cable icing in each local image and extract SIFT feature points. Based on feature matching and homography matrix estimation, these local images are seamlessly stitched and fused to construct a continuous cable edge image.
[0035] S302, the contours in the cable edge image are divided to obtain the cable contour and the icicle contour. The horizontally distributed contours are the contours of the ice surrounding the cable, and the vertically distributed contours are the contours of the icicles.
[0036] In this step, the contour in the cable edge image is divided. The lines in the contour are segmented according to a preset length, and the contour is divided into multiple curves of a preset length. The two endpoints of the curves are connected to obtain multiple line segments. The angle between each line segment and the horizontal plane is calculated. Based on the angle range, the line segments with an angle smaller than the predicted angle (e.g., 15°) are assigned to the cable contour, while other line segments are assigned to the icicle contour, so as to construct a complete cable contour and icicle contour.
[0037] S303: Construct a cable icing structure based on the cable outline, construct a rotating body based on the ice ridge outline to obtain the ice ridge structure, and fuse the icing structure and the ice ridge structure to obtain the cable icing model.
[0038] In this step, an ice-covered cable structure is constructed based on the cable profile, which includes an upper and lower profile. The ice layer at this point is close to a curved cylinder. A ring is constructed based on the upper and lower profiles to obtain an ice ridge structure. Similarly, for the ice ridge profile, which includes a left and right profile, the symmetrical line of the two is used as the axis of rotation to construct the ice ridge structure. The two are then fused to obtain the cable ice-covered model.
[0039] like Figure 4 As shown, in a preferred embodiment of the present invention, the steps of meshing the cable icing model, constructing a finite element analysis model, analyzing the stress on the cable at various points, outputting the analysis results, and determining whether to perform de-icing operations based on the analysis results include: S401 divides the cable icing model into multiple model elements based on the total length of the cable, calculates the icing weight of each model element, and generates the corresponding finite element analysis model of the cable.
[0040] In this step, the cable icing model is segmented based on the total length of the cable. Specifically, the segmentation is set according to the analysis accuracy of the finite element analysis model. The larger the number of model elements obtained from the segmentation, the higher the analysis accuracy. For example, 1 centimeter is used as a model element for segmentation. The weight of the ice block at that location is determined based on the volume and density of the ice block, thereby completing the finite element segmentation of the cable icing model and obtaining the finite element analysis model.
[0041] S402 performs finite element analysis based on the forces at each point, determines the forces at any point on the cable, and outputs the analysis results.
[0042] In this step, finite element analysis is performed based on the forces at each point. Finite element analysis software is used to extract the tension at any point on the cable to output the analysis results. The analysis results can extract the force analysis results at any position.
[0043] S403: If the tension value at a single point on the cable exceeds the preset value, it is determined that the tension is excessive and de-icing operation is required.
[0044] In this step, a preset number of stress analysis points are set on the cable, and the tensile force value of the stress analysis points is extracted and compared with the current maximum load-bearing capacity of the cable. To ensure the safety of the cable, the maximum load-bearing capacity is N, and 0.8N is used as the upper limit of the tensile force. If the tensile force at any point on the cable exceeds 0.8N, it is determined that de-icing operation is required. When de-icing operation is not performed, the maximum tensile force on the cable can be continuously recorded to determine the increase of the tensile force value. The faster the increase of the tensile force value, the more frequent the inspection of the cable should be in the subsequent process to ensure the safety of the cable.
[0045] In one embodiment of the present invention, the curvature of each point of the cable is extracted from the cable icing model and compared with the cable digital twin model to determine whether there is any tensile abnormality.
[0046] like Figure 5 As shown, an embodiment of the present invention provides a power grid cable inspection system based on a drone, the system comprising: The twin model building module 100 is used to build a cable digital twin model based on the installation location of the tower and the cable span. The cable digital twin model includes at least two towers and the cable between the towers.
[0047] In this system, the twin model construction module 100 constructs a cable digital twin model based on the installation location of the tower and the cable span. It models the towers that need to be monitored, marks and generates a 3D model of the tower in 3D software, records the tower's parameter information, including the tower height and GPS coordinates, and records the total span, cable length, and cable diameter of the cables between adjacent towers. This allows for the construction of a cable digital twin model based on the actual installation conditions. Based on the cable digital twin model, the parameters of any tower and the parameters of the cables between towers can be quickly obtained.
[0048] The image acquisition module 200 is used to generate an inspection route based on the installation location of the tower, control the drone to fly based on the inspection route, and acquire images of the cable during the flight to obtain local cable images. The inspection route is parallel to the cable between the tower.
[0049] In this system, the image acquisition module 200 generates an inspection route based on the installation location of the tower. It retrieves the parameters of the cable to be monitored from the cable digital twin model, including the cable span, total length, and GPS coordinates of the endpoints. Based on the cable span, total length, and height of the endpoints, it determines the cable's position in space and thus determines the coordinates of each point on the cable, thereby generating a series of coordinates corresponding to the cable. The entire sequence of coordinates is offset to generate a set of inspection routes parallel to the cable. Based on this inspection route, the UAV is controlled to fly and acquire images of the cable during flight. Each image acquisition yields a set of local cable images.
[0050] The contour recognition module 300 is used to extract and stitch cables in local cable images, extract the contours of cables in the stitched images, divide the contours into cable contours and icicle contours, and construct a cable icing model based on the cable contours and icicle contours.
[0051] In this system, the contour recognition module 300 extracts and stitches the cable in the local cable image. The local cable image contains the cable image. By stitching two adjacent sets of local cable images, a continuous and complete spliced cable image is obtained. By performing edge recognition, the edge positions of the cable and ice layer in the spliced image are determined, and the overall edge positions of the cable and ice layer are extracted. The ice layer on the cable is mainly divided into two parts, including the columnar ice layer surrounding the cable and the conical ice ridges located below the cable. Based on this, the ice layer on the cable is divided into two parts to obtain the cable contour and the ice ridge contour. The cable contour is the contour corresponding to the columnar ice layer, and the ice ridge contour is the contour corresponding to the ice ridge. Based on the structure of the two, the corresponding model is constructed in three-dimensional space to obtain the cable icing model.
[0052] The de-icing analysis module 400 is used to mesh the cable icing model, construct a finite element analysis model, analyze the stress on the cable at various points, output the analysis results, and determine whether to carry out de-icing operations based on the analysis results.
[0053] In this system, the de-icing analysis module 400 performs meshing processing on the cable icing model, sets parameters for the cable icing model according to the preset ice density, determines the weight of the ice block and the cable weight corresponding to each mesh position on the cable icing model, and uses the finite element analysis model to output the tensile force borne by each point on the cable to determine whether there is a local tensile force exceeding the limit. If so, it is determined that de-icing operation is required.
[0054] like Figure 6 As shown, in a preferred embodiment of the present invention, the image acquisition module 200 includes: The inspection route construction unit 201 is used to extract the coordinates of the tower and the coordinates of each point on the cable from the cable digital twin model, and generate an inspection route based on the cable coordinates. The inspection route is parallel to the cable.
[0055] In this module, the inspection route construction unit 201 extracts the coordinates of the tower and the coordinates of each point on the cable from the cable digital twin model. Since the corresponding endpoint coordinates are set for the cable endpoints in the cable digital twin model, the spatial distribution can be determined based on the total length of the cable, the height of the endpoints, and the span of the cable, thereby outputting the coordinates of each point on the cable. The cable coordinates are translated to construct the inspection route, which remains parallel to the cable.
[0056] The acquisition point generation unit 202 is used to control the flight of the UAV based on the inspection route and to set image acquisition points at preset flight distance intervals during the flight of the UAV.
[0057] In this module, the acquisition point generation unit 202 controls the flight of the UAV based on the inspection route and sets image acquisition points according to the preset flight interval. Specifically, based on the cable length contained in the image screen acquired by the UAV, the adjacent images acquired have overlapping areas to facilitate subsequent image stitching.
[0058] The cable imaging unit 203 is used to control the drone to capture images of the cable at the image acquisition point and obtain local cable images.
[0059] In this module, the cable imaging unit 203 controls the drone to collect images of the cable at the image acquisition point. Whenever the drone arrives at a new image acquisition point, it is controlled to hover there to collect an image, obtaining a set of local cable images, and then flying along the inspection route to the next image sampling point.
[0060] like Figure 7 As shown, in a preferred embodiment of the present invention, the contour recognition module 300 includes: The edge image stitching unit 301 is used to perform edge recognition on local cable images, extract the edges of ice blocks on the cable, and construct cable edge images by stitching them together.
[0061] In this module, the edge image stitching unit 301 performs edge recognition on local cable images. Specifically, it performs distortion correction and illumination normalization preprocessing on the local cable images, uses a multi-scale edge detection method to identify the edge contour of cable icing in each local image, and extracts SIFT feature points. Based on feature matching and homography matrix estimation, these local images are seamlessly stitched and fused to construct a continuous cable edge image.
[0062] The contour segmentation unit 302 is used to segment the contours in the cable edge image to obtain the cable contour and the icicle contour. The horizontally distributed contours are the contours of the ice surrounding the cable, and the vertically distributed contours are the contours of the icicles.
[0063] In this module, the contour segmentation unit 302 segments the contour in the cable edge image, divides the lines in the contour into multiple curves of a preset length according to a preset length, connects the two endpoints of the curves to obtain multiple line segments, calculates the angle between each line segment and the horizontal plane, classifies them based on the angle range, and assigns line segments with an angle less than the predicted angle (e.g., 15°) to the cable contour, while other line segments are assigned to the icicle contour, so as to construct a complete cable contour and icicle contour.
[0064] The icing model construction unit 303 is used to construct the cable icing structure based on the cable outline, construct the rotating body based on the ice ridge outline to obtain the ice ridge structure, and fuse the icing structure and the ice ridge structure to obtain the cable icing model.
[0065] In this module, the icing model construction unit 303 constructs a cable icing structure based on the cable profile. The cable profile includes an upper profile line and a lower profile line. The ice layer at this point is close to a curved cylinder. Based on the upper and lower profile lines, a ring body is constructed to obtain an ice ridge structure. Similarly, for the ice ridge profile, it includes a left profile line and a right profile line. The symmetrical line of the two is used as the axis of rotation to construct the ice ridge structure. The two are then fused to obtain the cable icing model.
[0066] like Figure 8 As shown, in a preferred embodiment of the present invention, the de-icing analysis module 400 includes: The analysis model building unit 401 is used to divide the cable icing model into multiple model units based on the total length of the cable, calculate the icing weight of each model unit, and generate the corresponding finite element analysis model of the cable.
[0067] In this module, the analysis model construction unit 401 divides the cable icing model based on the total length of the cable. Specifically, the settings are based on the analysis accuracy of the finite element analysis model. The larger the number of model units obtained from the division, the higher the analysis accuracy. For example, dividing the cable icing model by 1 centimeter as a model unit completes the finite element division of the cable icing model and obtains the finite element analysis model.
[0068] Analysis result output unit 402 is used to perform finite element analysis based on the force at each point, determine the force at any point on the cable, and output the analysis results.
[0069] In this module, the analysis result output unit 402 performs finite element analysis based on the force at each point, and uses finite element analysis software to extract the tension at any point on the cable to output the analysis results. The analysis results can extract the force analysis results at any position.
[0070] The tension determination unit 403 is used to determine that there is excessive tension and that de-icing operation is required when there is a single point tension value on the cable that exceeds the preset value.
[0071] In this module, the tension determination unit 403 sets a preset number of stress analysis points on the cable, extracts the tension value of the stress analysis points, and compares it with the current maximum bearing capacity of the cable. To ensure the safety of the cable, the maximum bearing capacity is N, and 0.8N is used as the upper limit of the tension. If the tension at any point on the cable exceeds 0.8N, it is determined that de-icing operation is required. When de-icing operation is not performed, the maximum tension on the cable can be continuously recorded to determine the increase of the tension value. The faster the increase of the tension value, the more frequent the inspection of the cable will be in the subsequent process to ensure the safety of the cable.
[0072] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A method for inspecting power grid cables based on unmanned aerial vehicles (UAVs), characterized in that, The method includes: A digital twin model of the cable is constructed based on the installation location of the towers and the cable span. The digital twin model of the cable includes at least two towers and the cable between the towers. An inspection route is generated based on the installation location of the tower. The drone is controlled to fly based on the inspection route. During the flight, images of the cable are acquired to obtain local cable images. The inspection route is parallel to the cable between the tower. The cables in the local cable images are extracted and stitched together. The outline of the cable in the stitched image is extracted and divided into cable outline and icicle outline. Based on the cable outline and icicle outline, a cable icing model is constructed. The cable icing model is meshed to construct a finite element analysis model, and the stress on the cable at various points is analyzed. The analysis results are output, and the decision on whether to carry out de-icing operations is made based on the analysis results.
2. The method for inspecting power grid cables based on unmanned aerial vehicles (UAVs) according to claim 1, characterized in that, The steps of generating an inspection route based on the installation location of the tower, controlling the drone to fly based on the inspection route, and acquiring images of the cable during flight to obtain local cable images specifically include: The coordinates of the tower and the coordinates of each point on the cable are extracted from the digital twin model of the cable. An inspection route is generated based on the coordinates of the cable, and the inspection route is kept parallel to the cable. The drone flight is controlled based on the inspection route, and image acquisition points are set at preset flight distance intervals during the drone flight. The drone is controlled to capture images of the cable at the image acquisition point, thus obtaining a local cable image.
3. The method for inspecting power grid cables based on unmanned aerial vehicles (UAVs) according to claim 1, characterized in that, The steps of extracting and stitching cables from local cable images, extracting the cable contours from the stitched images, dividing the contours into cable contours and icicle contours, and constructing a cable icing model based on the cable contours and icicle contours specifically include: Edge recognition is performed on local cable images to extract the edges of ice blocks on the cable, and then the edges of the cable edges are constructed by stitching them together. The contours in the cable edge image are divided to obtain the cable contour and the icicle contour. The horizontally distributed contours are the contours of the ice surrounding the cable, and the vertically distributed contours are the contours of the icicles. The cable icing structure is constructed based on the cable outline, and the ice ridge structure is constructed based on the ice ridge outline. The icing structure and the ice ridge structure are then fused to obtain the cable icing model.
4. The method for inspecting power grid cables based on unmanned aerial vehicles (UAVs) according to claim 1, characterized in that, The steps of meshing the cable icing model, constructing a finite element analysis model, analyzing the stress on the cable at various points, outputting the analysis results, and determining whether to carry out de-icing operations based on the analysis results include: The cable icing model is divided into multiple model elements based on the total length of the cable. The icing weight of each model element is calculated to generate the corresponding finite element analysis model of the cable. Finite element analysis is performed based on the forces at each point to determine the forces at any point on the cable and output the analysis results. If the tension value at a single point on the cable exceeds the preset value, it is determined that the tension is excessive and de-icing operation is required.
5. The method for inspecting power grid cables based on unmanned aerial vehicles (UAVs) according to claim 1, characterized in that, The method also includes extracting the curvature of each point of the cable from the cable icing model, comparing it with the cable digital twin model, and determining whether there is any tensile abnormality.
6. A power grid cable inspection system based on unmanned aerial vehicles (UAVs), characterized in that, The system includes: The twin model building module is used to build a cable digital twin model based on the installation location of the towers and the cable span. The cable digital twin model includes at least two towers and the cable between the towers. The image acquisition module is used to generate an inspection route based on the installation location of the tower, control the drone to fly based on the inspection route, and acquire images of the cable during the flight to obtain local cable images. The inspection route is parallel to the cable between the tower. The contour recognition module is used to extract and stitch cables in local cable images, extract the contours of cables in the stitched images, divide the contours into cable contours and icicle contours, and construct a cable icing model based on the cable contours and icicle contours. The de-icing analysis module is used to mesh the cable icing model, construct a finite element analysis model, analyze the stress on the cable at various points, output the analysis results, and determine whether de-icing operations should be carried out based on the analysis results.
7. The UAV-based power grid cable inspection system according to claim 6, characterized in that, The image acquisition module includes: The inspection route construction unit is used to extract the coordinates of the tower and the coordinates of each point on the cable from the cable digital twin model, and generate the inspection route based on the cable coordinates. The inspection route is parallel to the cable. The data acquisition point generation unit is used to control the flight of the UAV based on the inspection route and to set image acquisition points at preset flight distance intervals during the flight of the UAV. The cable imaging unit is used to control the drone to capture images of the cable at the image acquisition point and obtain local cable images.
8. The UAV-based power grid cable inspection system according to claim 6, characterized in that, The contour recognition module includes: The edge image stitching unit is used to perform edge recognition on local cable images, extract the edges of ice blocks on the cable, and construct a cable edge image by stitching them together. The contour segmentation unit is used to segment the contours in the cable edge image to obtain the cable contour and the icicle contour. The horizontally distributed contours are the contours of the ice surrounding the cable, and the vertically distributed contours are the contours of the icicles. The icing model construction unit is used to construct the cable icing structure based on the cable outline, construct the rotating body based on the ice ridge outline to obtain the ice ridge structure, and fuse the icing structure and the ice ridge structure to obtain the cable icing model.
9. The UAV-based power grid cable inspection system according to claim 6, characterized in that, The de-icing analysis module includes: The analysis model building unit is used to divide the cable icing model into multiple model units based on the total length of the cable, calculate the icing weight of each model unit, and generate the corresponding finite element analysis model of the cable. The analysis result output unit is used to perform finite element analysis based on the forces at each point, determine the forces at any point on the cable, and output the analysis results. The tension determination unit is used to determine that there is excessive tension when the tension value at a single point on the cable exceeds the preset value, and that de-icing operation is required.
10. The UAV-based power grid cable inspection system according to claim 6, characterized in that, The system's operation also includes extracting the curvature of each point on the cable from the cable icing model, comparing it with the cable's digital twin model, and determining whether there is any tensile abnormality.