An unmanned aerial vehicle system for substation inspection

Abstract

The present invention discloses a drone system for substation inspection, which relates to the technical field of substation inspection. The inspection task unit is set to obtain inspection tasks, the panoramic image unit is used to obtain panoramic images of the substation, the grid map unit is set to construct a grid map, and the inspection path unit is used to plan the inspection path on the grid map; the drone unit executes inspection tasks based on the inspection path; the image processing unit performs real-time processing on the collected image data. If it is determined according to the matching result that there is a fault in the corresponding area, the adjustment unit is used to mark the fault in the corresponding area. If it is determined according to the matching result that there is an undetected area in the inspected area, the drone is adjusted to reinspect the undetected area. The present invention not only optimizes the inspection path to improve the overall inspection path of the drone, but also performs real-time processing on the collected data, and can timely mark the fault area and reinspect the undetected area.

Description

Technical Field

[0001] The present invention relates to the technical field of substation inspection, and specifically, to a drone system for substation inspection. Background Art

[0002] The drone system for substation inspection is a solution that uses drone technology to perform automated and intelligent inspections on substations. In the prior art, the principle of using drones to inspect substations is as follows: The operation and maintenance personnel pre-plan the flight path and inspection tasks of the drone through the background software, and set the flight altitude, speed, and shooting parameters; the drone flies according to the preset path, and uses a variety of sensors to collect images and data of substation equipment; the collected data is transmitted to the ground control station or cloud server through a wireless communication module; the background software processes and analyzes the collected data to generate a high-precision 3D model and inspection report; the operation and maintenance personnel perform on-site processing according to the inspection report to ensure the normal operation of the equipment.

[0003] However, the above solutions have the following drawbacks: 1) In path planning, the common method is to set multiple straight flight paths, and the adjacent straight flight paths are connected end to end. Although this path planning can cover all areas of the substation, its inspection path is not the optimal path, resulting in an overall low inspection efficiency; 2) In the prior art, the drone transmits the collected data to the ground control station or cloud server in real time, but for the collected data, it needs to be uniformly processed and analyzed by the background after the inspection task is completed to generate an inspection report. Considering that the amount of collected data is often large, when a problem occurs in the substation, it is often impossible to discover and handle it immediately through drone inspection; 3) During the process of drone inspection, there may be missed inspections. The current handling method is to process the collected data after the inspection task is completed, discover the missed inspection area, and use the method of drone re-flight or manual inspection of the area to solve it, which further reduces the inspection efficiency of the drone.

[0004] To solve the above problems, there is an urgent need for a drone system for substation inspection. Summary of the Invention

[0005] To solve the deficiencies of the above prior art, the present invention provides a drone system for substation inspection, and the system includes:

[0006] An inspection task unit: obtaining an inspection task;

[0007] A panoramic image unit: taking partial images of the substation to be inspected from multiple angles by using multiple cameras, and processing all the partial images to obtain a panoramic image of the substation to be inspected;

[0008] Grid map unit: Based on the 3D lidar-scanned substation point cloud data, a grid map for planning the UAV inspection path is constructed.

[0009] Inspection path unit: Based on the inspection task, the inspection start point, multiple inspection nodes, and inspection end point are marked on the grid map. Based on the marked grid map, the first method is used to plan the UAV inspection path.

[0010] UAV unit: Based on the planned UAV inspection path, the inspection task is executed, and the collected image data is transmitted back to the image processing unit in real time.

[0011] Image processing unit: Used to process the collected image data to obtain a real-time image of the substation to be inspected, perform a first match between the real-time image and the panoramic image to obtain a first matching result; use the second method to perform stitching processing on all real-time images to obtain a stitched image of the substation to be inspected, and perform a second match between the stitched image and the panoramic image to obtain a second matching result.

[0012] Adjustment unit: Based on the first matching result, determine whether there is a fault in the corresponding area of the real-time image. If so, mark the corresponding area for the fault; based on the second matching result, determine whether there is an uninspected area in the area that the UAV has inspected. If so, use the third method to adjust the UAV.

[0013] The present invention is realized through the following technical solutions: Firstly, the inspection task unit obtains corresponding inspection tasks. In addition to daily inspections, the substation also includes key inspections for certain equipment. Therefore, corresponding inspection paths need to be set for different inspection tasks; through the panoramic image unit, a panoramic image of the substation is obtained; a grid map unit is constructed, and point cloud data of the substation is obtained through a 3D lidar, and the equipment distribution data of the substation is converted onto the grid map; the inspection path unit is used to plan the inspection path of the drone on the grid map. Firstly, the inspection start point, inspection nodes, and inspection end point that the drone needs to traverse are determined through the inspection task, and then the shortest inspection path of the drone is planned on the grid map through the first method; the drone unit is used to execute the inspection task based on the inspection path and transmit the collected image data back to the image processing unit in real time during the inspection process; the image processing unit is used to perform real-time processing on the collected image data. Firstly, real-time images are extracted from the transmitted video stream data, and a first matching result is obtained through a first match between the real-time images and the panoramic image. Then, all the extracted real-time images are stitched together through the second method to obtain a stitched image, and the stitched image is then subjected to a second match with the panoramic image to obtain a second matching result; the adjustment unit is used to judge whether a fault occurs in the corresponding area of the real-time image based on the first matching result. If so, the corresponding area is marked for the fault. Based on the second matching result, it is judged whether there is an undetected area in the area that the drone has inspected. If so, the third method is used to adjust the drone so that the drone can re-inspect the undetected area in time during the inspection process.

[0014] As an optional technical solution, processing all local images to obtain the panoramic image of the substation to be inspected includes:

[0015] Obtain all local images and place them in a preset image set;

[0016] Obtain any two local images with overlapping areas from the preset image set and process them using a preset method. Any two local images with overlapping areas are denoted as the first image and the second image. The preset method includes;

[0017] Use the SIFT algorithm to extract multiple first key feature points in the first image and multiple second key feature points in the second image;

[0018] Match all the first key feature points in the first image with all the second key feature points in the second image in sequence to obtain a third matching result;

[0019] Based on the third matching result, calculate the first homography matrix;

[0020] Based on the first homography matrix, transform the second image into the coordinate system of the first image, perform a perspective transformation on the second image, and then splice it with the first image;

[0021] Remove the black area after splicing the first image and the second image, and output the spliced image of the first image and the second image;

[0022] Execute the preset method on all local images with overlapping areas in the preset image set until the panoramic image of the substation to be inspected is generated by splicing.

[0023] As an alternative technical solution, constructing a grid map for planning the UAV inspection path based on the substation point cloud data scanned by a 3D lidar includes:

[0024] Obtain all substation point cloud data scanned by the 3D lidar and place them in a preset data set;

[0025] Judge whether the height value of any substation point cloud data in the preset data set is less than a first threshold. If so, determine the corresponding substation point cloud data as ground points and perform a first filtering. If not, do nothing;

[0026] Set the effective value range of the substation point cloud data, and based on the effective value range, perform a second filtering on the preset data set after the first filtering;

[0027] Create a 3D grid space based on all substation point cloud data in the preset data set after the second filtering;

[0028] Construct an initial grid map, project the points corresponding to all substation point cloud data in the 3D grid space onto the initial grid map, and set an expansion coefficient for the points corresponding to substation equipment on the initial grid map to form a grid map for planning the UAV inspection path.

[0029] As an alternative technical solution, the first method includes:

[0030] Set a first set for storing expanded nodes and a second set for storing nodes to be expanded;

[0031] Place the inspection starting point in the first set, and place all inspection nodes and the inspection end point in the second set;

[0032] Calculate the cost value between the inspection starting point and any inspection node using a first formula, including:

[0033]

[0034] is the cost value from the inspection starting point to any inspection node, represents the actual cost from the inspection starting point to any inspection node, is the weight of the actual cost, represents the estimated cost from the inspection starting point to any inspection node, is the weight representing the estimated cost, represents the search tendency cost from the inspection starting point to any inspection node, is the weight representing the search tendency cost;

[0035] Based on the first formula, adjacent inspection nodes of the inspection starting point are screened out from the second set, and the adjacent inspection nodes are used as new inspection starting points. Then, adjacent inspection nodes of the new inspection starting points are screened out in the second set until the adjacent inspection node is the inspection end point.

[0036] As an alternative technical solution, performing a single matching on the real-time image and the panoramic image to obtain a first matching result includes:

[0037] Extracting a plurality of first feature points of the real-time image and a plurality of second feature points in the panoramic image, and calculating descriptors of all the first feature points, denoted as first descriptors, and descriptors of all the second feature points, denoted as second descriptors;

[0038] Calculating the Euclidean distance between all the first descriptors and all the second descriptors, and finding the nearest neighbor feature points of all the first feature points of the real-time image in the panoramic image;

[0039] Estimating a second homography matrix from all the nearest neighbor feature points, and based on the second homography matrix, fusing the real-time image into the panoramic image to obtain the first matching result.

[0040] As an alternative technical solution, performing stitching processing on all the real-time images by using a second method to obtain a stitched image of the substation to be inspected includes:

[0041] Performing binarization processing on all the real-time images to convert them into grayscale images;

[0042] Dividing each grayscale image into an overlapping region and a non-overlapping region, and calculating the grayscale difference value between the overlapping regions of any two grayscale images to obtain a calculation result;

[0043] Based on the calculation result, adjusting any two grayscale images to make the edges of the corresponding overlapping regions match each other, and stitching any two adjusted grayscale images in all the real-time images to obtain a stitched image of the substation to be inspected.

[0044] As an alternative technical solution, performing a secondary matching on the stitched image and the panoramic image to obtain a second matching result includes:

[0045] Extracting a plurality of third feature points of the stitched image and a plurality of fourth feature points in the panoramic image, and calculating descriptors of all the third feature points, denoted as third descriptors, and descriptors of all the fourth feature points, denoted as fourth descriptors;

[0046] Calculating the Euclidean distance between all the third descriptors and all the fourth descriptors, and finding the nearest neighbor feature points of all the third feature points of the real-time image in the panoramic image;

[0047] Estimating a third homography matrix from all the nearest neighbor feature points, and based on the third homography matrix, fusing the stitched image into the panoramic image to obtain the second matching result.

[0048] As an alternative technical solution, the third method includes:

[0049] When it is determined that there is a missed inspection area in the area already inspected by the drone, denoted as the first missed inspection area;

[0050] Obtaining the current position coordinates of the drone, denoted as the first coordinate point, and the central position coordinates of the first missed inspection area, denoted as the second coordinate point;

[0051] Based on the first coordinate point and the second coordinate point, planning a secondary inspection path for the drone from the current position to the first missed inspection area;

[0052] Obtaining the current flight altitude of the drone, denoted as the first altitude;

[0053] Based on the first altitude, determining whether there are obstacles on the secondary inspection path. If so, adjusting the secondary inspection path using the fourth method. If not, the drone reinspects the first missed inspection area based on the secondary inspection path.

[0054] As an alternative technical solution, the fourth method includes:

[0055] When it is determined that there is an obstacle on the secondary inspection path, denoted as the first obstacle;

[0056] Obtaining and based on the three-dimensional point cloud data of the first obstacle, modeling the first obstacle;

[0057] Using the A* algorithm to calculate the shortest path for the drone to pass through the first obstacle.

[0058] One or more technical solutions provided by the present invention have at least the following technical effects or advantages:

[0059] The present invention discloses a drone system for substation inspection. The inspection task unit is set to obtain inspection tasks, the panoramic image unit is used to obtain the panoramic image of the substation, the grid map unit is set to convert the distribution data of the substation onto the grid map, and then the inspection path unit plans the inspection path on the grid map to obtain the shortest inspection path of the drone. The drone unit then executes corresponding inspection tasks based on the planned inspection path and collects image data during the inspection process. The image processing unit is used to perform real-time processing on the collected image data. First, it performs a primary match between the real-time image and the panoramic image. If it is determined according to the matching result that there is a fault in the corresponding area, the adjustment unit is used to mark the fault in the corresponding area. Then, the real-time images are stitched to obtain a stitched image, and the stitched image is subjected to a secondary match with the panoramic image. If it is determined according to the matching result that there is an undetected area in the inspected area, the drone is adjusted to reinspect the undetected area. The present invention not only optimizes the inspection path to improve the overall inspection path of the drone, but also performs real-time processing on the collected data, enabling timely marking of the fault area and reinspection of the undetected area. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] The drawings described herein are used to provide a further understanding of the embodiments of the present invention and form a part of the present invention, but do not limit the embodiments of the present invention.

[0061] Figure 1 It is a schematic diagram of the composition of a drone system for substation inspection in the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0062] In order to more clearly understand the above objects, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the drawings and specific embodiments. It should be noted that, without conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other.

[0063] In the following description, many specific details are set forth to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described within the scope hereof. Therefore, the protection scope of the present invention is not limited by the specific embodiments disclosed below.

[0064] Embodiment

[0065] Please refer to Figure 1 , Figure 1 which is a drone system for substation inspection in the present invention. The system includes:

[0066] An inspection task unit: obtaining inspection tasks;

[0067] Panoramic image unit: Multiple cameras are used to capture partial images of the substation to be inspected from multiple angles, and all partial images are processed to obtain the panoramic image of the substation to be inspected;

[0068] Grid map unit: Based on the point cloud data of the substation scanned by a 3D lidar, a grid map for planning the inspection path of the UAV is constructed;

[0069] Inspection path unit: Based on the inspection task, the inspection starting point, multiple inspection nodes, and inspection end point are marked on the grid map. Based on the marked grid map, the first method is used to plan the inspection path of the UAV;

[0070] UAV unit: Based on the planned inspection path of the UAV, the inspection task is executed, and the collected image data is transmitted back to the image processing unit in real time;

[0071] Image processing unit: Used to process the collected image data to obtain a real-time image of the substation to be inspected, perform a first match between the real-time image and the panoramic image to obtain a first matching result; use a second method to perform stitching processing on all real-time images to obtain a stitched image of the substation to be inspected, and perform a second match between the stitched image and the panoramic image to obtain a second matching result;

[0072] Adjustment unit: Based on the first matching result, determine whether a fault has occurred in the corresponding area of the real-time image. If so, mark the corresponding area for the fault; based on the second matching result, determine whether there is an area that has been missed in the area inspected by the UAV. If so, use a third method to adjust the UAV.

[0073] The specific embodiments of the present invention are as follows:

[0074] Inspection task unit:

[0075] Used to obtain the inspection task. In addition to the daily inspection of the substation, it also includes regular inspections, key inspections of special equipment, etc. Therefore, in order to ensure the inspection efficiency of the UAV, the corresponding inspection path is planned for each inspection task;

[0076] Panoramic image unit:

[0077] Processing all partial images to obtain the panoramic image of the substation to be inspected includes:

[0078] Obtain all partial images and place them in a preset image set;

[0079] Obtain any two partial images with an overlapping area from the preset image set and process them using a preset method. Any two partial images with an overlapping area are denoted as the first image and the second image, and the preset method includes;

[0080] Extract multiple first key feature points in the first image and multiple second key feature points in the second image using the SIFT algorithm;

[0081] Match all the first key feature points in the first image with all the second key feature points in the second image in sequence to obtain a third matching result;

[0082] Calculate a first homography matrix based on the third matching result;

[0083] Based on the first homography matrix, transform the second image into the coordinate system of the first image, and perform perspective transformation on the second image and then splice it with the first image;

[0084] Remove the black area after splicing the first image and the second image, and output the spliced image of the first image and the second image;

[0085] Execute the preset method on all local images with overlapping areas in the preset image set until the panoramic image of the substation to be inspected is generated by splicing.

[0086] Grid map unit:

[0087] Constructing a grid map for planning the UAV inspection path based on the substation point cloud data scanned by a 3D lidar includes:

[0088] Obtain all the substation point cloud data scanned by the 3D lidar and place them in a preset data set;

[0089] Judge whether the height value of any substation point cloud data in the preset data set is less than a first threshold. If so, determine the corresponding substation point cloud data as ground points and perform a first filtering. If not, do nothing;

[0090] Set the effective value range of the substation point cloud data, and based on the effective value range, perform a second filtering on the preset data set after the first filtering;

[0091] Create a 3D grid space based on all the substation point cloud data in the preset data set after the second filtering;

[0092] Construct an initial grid map, project the points corresponding to all the substation point cloud data in the 3D grid space onto the initial grid map, and set an expansion coefficient for the points corresponding to the substation equipment on the initial grid map to form a grid map for planning the UAV inspection path.

[0093] Among them, a grid map is constructed to reflect the equipment distribution of the substation. During the construction of the grid map, other data that affect the accuracy of UAV path planning are filtered out by filtering the ground point cloud and screening the effective values of the point cloud. And the expansion coefficient of the substation equipment is introduced into the grid map. First, it can ensure the safety of the UAV during flight as much as possible. Second, it can be more reasonable when planning the inspection path.

[0094] Inspection path unit:

[0095] Based on the inspection task, the inspection start point, multiple inspection nodes and the inspection end point are marked on the grid map. Based on the marked grid map, the first method is used to plan the UAV inspection path;

[0096] The first method includes:

[0097] Set a first set to store the expanded nodes, and a second set to store the nodes to be expanded;

[0098] Put the inspection start point into the first set, and put all inspection nodes and the inspection end point into the second set;

[0099] Use the first formula to calculate the cost value from the inspection start point to any inspection node, including:

[0100]

[0101] is the cost value from the inspection start point to any inspection node, represents the actual cost from the inspection start point to any inspection node, is the weight of the actual cost, represents the estimated cost from the inspection start point to any inspection node, represents the weight of the estimated cost, represents the search tendency cost from the inspection start point to any inspection node, represents the weight of the search tendency cost;

[0102] Based on the first formula, the adjacent inspection nodes of the inspection start point are screened out in the second set, and the adjacent inspection nodes are used as the new inspection start points. The adjacent inspection nodes of the new inspection start points are screened out in the second set until the adjacent inspection node is the inspection end point.

[0103] Among them, in this embodiment, a first set is set to store the expanded nodes, a second set is set to store the nodes to be expanded, the inspection starting point is placed in the first set, and then a first formula is used to calculate the inspection node with the lowest cost value in the second set relative to the inspection starting point as the adjacent node of the inspection starting point, and the adjacent node is used as the new inspection starting point, and the next adjacent node is searched for in the second set until the adjacent node is the inspection end point. It should be noted that the successfully screened inspection nodes in the second set will all be placed in the first set and will not be retained in the second set.

[0104] UAV unit:

[0105] Based on the planned UAV inspection path, perform the inspection task and transmit the collected image data back to the image processing unit in real time; for the size and type of the UAV, adjustments can be made according to the actual situation, and this embodiment does not make specific limitations.

[0106] Image processing unit:

[0107] It is used to process the collected image data to obtain a real-time image of the substation to be inspected, perform a first matching between the real-time image and the panoramic image to obtain a first matching result; use a second method to perform stitching processing on all real-time images to obtain a stitched image of the substation to be inspected, and perform a second matching between the stitched image and the panoramic image to obtain a second matching result;

[0108] Performing a first matching between the real-time image and the panoramic image to obtain a first matching result includes:

[0109] Extract multiple first feature points of the real-time image and multiple second feature points in the panoramic image, and calculate the descriptors of all first feature points, denoted as the first descriptors, and the descriptors of all second feature points, denoted as the second descriptors;

[0110] Calculate the Euclidean distance between all first descriptors and all second descriptors, and search for the nearest neighbor feature points of all first feature points of the real-time image in the panoramic image;

[0111] Estimate a second homography matrix from all the nearest neighbor feature points, and based on the second homography matrix, fuse the real-time image into the panoramic image to obtain the first matching result.

[0112] Using a second method to perform stitching processing on all real-time images to obtain a stitched image of the substation to be inspected includes:

[0113] Perform binarization processing on all real-time images to convert them into grayscale images;

[0114] Divide each grayscale image into overlapping regions and non-overlapping regions, and calculate the grayscale difference value between the overlapping regions in any two grayscale images to obtain the calculation result;

[0115] Based on the calculation result, adjust any two grayscale images so that the edges of the corresponding overlapping regions match each other, and splice any two adjusted grayscale images in all real-time images to obtain a spliced image of the substation to be inspected.

[0116] Perform a secondary match between the spliced image and the panoramic image to obtain the second match result, including:

[0117] Extract multiple third feature points of the spliced image and multiple fourth feature points of the panoramic image, and calculate the descriptors of all third feature points, denoted as the third descriptor, and the descriptors of all fourth feature points, denoted as the fourth descriptor;

[0118] Calculate the Euclidean distance between all third descriptors and all fourth descriptors, and find the nearest neighbor feature points of all third feature points of the real-time image in the panoramic image;

[0119] Estimate the third homography matrix from all the nearest neighbor feature points, and based on the third homography matrix, fuse the spliced image into the panoramic image to obtain the second match result.

[0120] Among them, the image processing unit is used to perform real-time processing on the collected image data. First, extract the real-time images from the returned video stream data, perform a primary match between the real-time images and the panoramic image to obtain the first match result, then splice all the extracted real-time images using the second method to obtain a spliced image, and then perform a secondary match between the spliced image and the panoramic image to obtain the second match result.

[0121] Adjustment unit:

[0122] Based on the first match result, determine whether a fault has occurred in the corresponding area of the real-time image. If so, mark the corresponding area with a fault; based on the second match result, determine whether there is an undetected area in the area that the drone has inspected. If so, adjust the drone using the third method.

[0123] The third method includes:

[0124] When it is determined that there is an undetected area in the area that the drone has inspected, denoted as the first undetected area;

[0125] Obtain the current position coordinates of the drone, denoted as the first coordinate point, and the central position coordinates of the first undetected area, denoted as the second coordinate point;

[0126] Based on the first coordinate point and the second coordinate point, plan a secondary inspection path for the UAV from the current position to the first missed inspection area;

[0127] Obtain the flight altitude of the current UAV, denoted as the first altitude;

[0128] Based on the first altitude, determine whether there are obstacles on the secondary inspection path. If so, adjust the secondary inspection path using the fourth method. If not, the UAV reinspects the first missed inspection area based on the secondary inspection path.

[0129] The fourth method includes:

[0130] When it is determined that there is an obstacle on the secondary inspection path, denoted as the first obstacle;

[0131] Obtain and, based on the three-dimensional point cloud data of the first obstacle, model the first obstacle;

[0132] Use the A* algorithm to calculate the shortest path for the UAV to pass through the first obstacle.

[0133] Among them, the adjustment unit first, based on the first matching result, determines whether the corresponding area of the real-time image has a fault. If so, it marks the corresponding area for the fault. Then, based on the second matching result, it determines whether there is a missed inspection area in the area that the UAV has inspected. If so, by obtaining the position coordinates of the current UAV, denoted as the first coordinate point, and the position coordinates of the missed inspection area, denoted as the second coordinate point, based on the first coordinate point and the second coordinate point, plan a secondary inspection path for the UAV from the current position to the missed inspection area, and by obtaining the flight altitude of the UAV, determine whether there are obstacles (i.e., substation equipment) on this secondary inspection path. If so, it is also necessary to adjust this secondary inspection path. By modeling the obstacle and using the A* algorithm to calculate the shortest path for the UAV to pass through the obstacle. If not, the UAV reinspects based on this reinspection path. It should be noted that after the UAV reinspects the missed inspection area, it returns to the current position along the original route and executes the subsequent inspection tasks.

[0134] Although the preferred embodiments of the present invention have been described, those skilled in the art can make additional changes and modifications to these embodiments once they learn the basic creative concepts. Therefore, the appended claims are intended to be construed to include the preferred embodiments as well as all changes and modifications that fall within the scope of the present invention.

[0135] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. A drone system for substation inspection, characterized in that: The system comprises: Inspection task unit: obtain inspection tasks; Panoramic image unit: using multiple cameras to capture local images of the substation to be inspected from multiple angles, and processing all local images to obtain a panoramic image of the substation to be inspected; Grid map unit: Based on the substation cloud data scanned by 3D LiDAR, a grid map is constructed for planning the inspection path of drones. Inspection path unit: based on the inspection task, marking the inspection starting point, multiple inspection nodes and the inspection end point in the grid map, and based on the marked grid map, adopting the first method to perform UAV inspection path planning; UAV unit: based on the planned UAV inspection path, executes the inspection task and transmits the collected image data back to the image processing unit in real time; An image processing unit is used to process the collected image data to obtain a real-time image of the substation to be inspected, match the real-time image with the panoramic image once, and obtain a first matching result; use a second method to stitch all the real-time images to obtain a stitched image of the substation to be inspected, and match the stitched image with the panoramic image twice to obtain a second matching result; Adjustment unit: based on the first matching result, determine whether a fault occurs in the corresponding area of ​​the real-time image, and if so, mark the corresponding area for fault; based on the second matching result, determine whether there is a missed inspection area in the area inspected by the drone, and if so, adjust the drone using a third method; The second method comprises: All real-time images are binarized and converted into grayscale images; Each grayscale image is divided into an overlapping area and a non-overlapping area, and the grayscale difference value between the overlapping areas in any two grayscale images is calculated to obtain a calculation result; Based on the calculation result, any two grayscale images are adjusted so that the edges of the corresponding overlapping areas match each other, and any two adjusted grayscale images in all real-time images are spliced ​​to obtain a spliced ​​image of the substation to be inspected; The third method comprises: When it is determined that there is a missed inspection area in the area that the drone has inspected, it is recorded as the first missed inspection area; Obtain the coordinates of the current location of the drone, recorded as the first coordinate point, and the coordinates of the center position of the first missed detection area, recorded as the second coordinate point; Based on the first coordinate point and the second coordinate point, planning a secondary inspection path of the drone from the current position to the first missed inspection area; Get the current flight altitude of the drone, recorded as the first altitude; Based on the first height, determine whether there are obstacles on the secondary inspection path. If so, use the fourth method to adjust the secondary inspection path. If not, the drone re-inspects the first missed inspection area based on the secondary inspection path.

2. The UAV system for substation inspection according to claim 1 is characterized in that: Processing all local images to obtain a panoramic image of the substation to be inspected includes: Get all partial images and put them into the preset image set; Acquire any two partial images with overlapping areas from the preset image set and process them using a preset method, wherein the any two partial images with overlapping areas are recorded as a first image and a second image, and the preset method includes: Extracting a plurality of first key feature points in the first image and a plurality of second key feature points in the second image using a SIFT algorithm; Matching all first key feature points in the first image with all second key feature points in the second image in sequence to obtain a third matching result; Based on the third matching result, a first homography matrix is ​​calculated; Based on the first homography matrix, transform the second image into the coordinate system of the first image, and perform perspective transformation on the second image and then splice it with the first image; removing a black area after the first image and the second image are spliced ​​together, and outputting a spliced ​​image of the first image and the second image; The preset method is executed on all the local images with overlapping areas in the preset image set until a panoramic image of the substation to be inspected is generated by splicing.

3. The UAV system for substation inspection according to claim 1 is characterized in that: Based on the substation cloud data scanned by 3D LiDAR, a raster map for planning drone inspection paths is constructed, including: Obtain all substation cloud data based on 3D LiDAR scanning and put it into the preset data set; Determine whether the height value of any substation cloud data in the preset data set is less than a first threshold value, if so, determine that the corresponding substation cloud data is a ground point and filter it out once, if not, do nothing; Setting a valid value range for the substation site cloud data, and performing a second filtering on the preset data set after the first filtering based on the valid value range; Creating a three-dimensional grid space based on all substation cloud data in the preset data set after secondary filtering; An initial grid map is constructed, and the points corresponding to the cloud data of all substations in the three-dimensional grid space are projected onto the initial grid map. An expansion coefficient is set for the points corresponding to the substation equipment on the initial grid map to form a grid map for planning the inspection path of drones.

4. The UAV system for substation inspection according to claim 1 is characterized in that: The first method comprises: Setting a first set to store expanded nodes, and a second set to store nodes to be expanded; Put the inspection starting point into the first set, and put all inspection nodes and the inspection end point into the second set; The first formula is used to calculate the cost value of the inspection starting point and any inspection node, including: is the cost from the inspection starting point to any inspection node, Indicates the actual cost from the inspection starting point to any inspection node. is the weight of the actual cost, It represents the estimated cost from the inspection starting point to any inspection node. represents the weight of the estimated cost, Indicates the search tendency cost from the inspection starting point to any inspection node. The weight representing the search tendency cost; Based on the first formula, the adjacent inspection nodes of the inspection starting point are screened in the second set, and the adjacent inspection nodes are used as new inspection starting points, and the adjacent inspection nodes of the new inspection starting point are screened in the second set until the adjacent inspection nodes are the inspection end points.

5. The UAV system for substation inspection according to claim 1 is characterized in that: Matching the real-time image with the panoramic image once to obtain a first matching result includes: Extracting a plurality of first feature points of the real-time image and a plurality of second feature points in the panoramic image, and calculating descriptors of all first feature points, recorded as first descriptors, and descriptors of all second feature points, recorded as second descriptors; Calculating the Euclidean distances between all first descriptors and all second descriptors, and searching for the nearest neighbor feature points of all first feature points of the real-time image in the panoramic image; A second homography matrix is ​​estimated from all nearest neighbor feature points, and based on the second homography matrix, the real-time image is fused into the panoramic image to obtain the first matching result.

6. The UAV system for substation inspection according to claim 1, characterized in that: Performing secondary matching between the stitched image and the panoramic image to obtain a second matching result includes: Extracting a plurality of third feature points from the stitched image and a plurality of fourth feature points from the panoramic image, and calculating descriptors of all third feature points, recorded as third descriptors, and descriptors of all fourth feature points, recorded as fourth descriptors; Calculating the Euclidean distances between all third descriptors and all fourth descriptors, and searching for the nearest neighbor feature points of all third feature points of the real-time image in the panoramic image; A third homography matrix is ​​estimated from all nearest neighbor feature points, and based on the third homography matrix, the stitched image is fused into the panoramic image to obtain the second matching result.

7. The UAV system for substation inspection according to claim 1 is characterized in that: The fourth method comprises: When it is determined that there is an obstacle on the secondary inspection path, it is recorded as the first obstacle; Acquire and model the first obstacle based on the three-dimensional point cloud data of the first obstacle; The A* algorithm is used to calculate the shortest path for the drone to pass through the first obstacle.

Images