Maintenance methods, devices, storage media, and computer equipment for power equipment

Abstract

This invention discloses a method, apparatus, storage medium, and computer equipment for maintaining power equipment, primarily aimed at improving the efficiency of power equipment maintenance and reducing personnel costs. The method includes: acquiring inspection images collected by a drone; inputting the inspection images into a preset target recognition model for target recognition to obtain a recognition result corresponding to the inspection images; if the recognition result determines that the inspection images contain power equipment, then performing image matting on the inspection images to obtain power equipment images; based on the power equipment images, determining whether the power equipment is faulty; if the power equipment is faulty, then generating work order information corresponding to the power equipment and sending the work order information to maintenance personnel.

Description

Technical Field

[0001] This invention relates to the field of power technology, and in particular to a method, apparatus, storage medium, and computer equipment for maintaining power equipment. Background Technology

[0002] To ensure the normal operation of the power transmission lines on the towers, it is necessary to conduct real-time inspections and tests on the power line equipment. Once any abnormality is found, maintenance personnel will promptly repair it.

[0003] Currently, traditional power equipment maintenance typically involves manual inspection. If a fault is detected, patrol personnel will locate on-duty maintenance staff for repair. However, this method requires dedicated personnel for inspection, significantly increasing labor costs. Furthermore, if patrol personnel fail to detect the fault promptly, maintenance staff cannot arrive on-site immediately, resulting in low maintenance efficiency. Summary of the Invention

[0004] This invention provides a method, apparatus, storage medium, and computer equipment for maintaining power equipment, which mainly improves the maintenance efficiency of power equipment and reduces personnel costs.

[0005] According to a first aspect of the present invention, a method for maintaining electrical equipment is provided, comprising:

[0006] Acquire inspection images collected by drones;

[0007] The inspection image is input into a preset target recognition model for target recognition, and the recognition result corresponding to the inspection image is obtained.

[0008] If the inspection image is determined to contain power equipment based on the recognition result, the inspection image is then processed to obtain an image of the power equipment.

[0009] Based on the image of the power equipment, determine whether the power equipment is faulty;

[0010] If the power equipment malfunctions, a work order is generated corresponding to the power equipment, and the work order is sent to the maintenance personnel.

[0011] According to a second aspect of the present invention, a maintenance device for electrical equipment is provided, comprising:

[0012] The acquisition unit is used to acquire inspection images collected by the drone;

[0013] The recognition unit is used to input the inspection image into a preset target recognition model for target recognition and obtain the recognition result corresponding to the inspection image.

[0014] The image matting unit is used to perform image matting processing on the inspection image to obtain an image of the power equipment if it is determined from the recognition result that the inspection image contains power equipment.

[0015] The determination unit is used to determine whether the power equipment has a fault based on the image of the power equipment;

[0016] The generation unit is used to generate work order information corresponding to the power equipment if the power equipment has a fault, and send the work order information to the operation and maintenance personnel.

[0017] According to a third aspect of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0018] Acquire inspection images collected by drones;

[0019] The inspection image is input into a preset target recognition model for target recognition, and the recognition result corresponding to the inspection image is obtained.

[0020] If the inspection image is determined to contain power equipment based on the recognition result, the inspection image is then processed to obtain an image of the power equipment.

[0021] Based on the image of the power equipment, determine whether the power equipment is faulty;

[0022] If the power equipment malfunctions, a work order is generated corresponding to the power equipment, and the work order is sent to the maintenance personnel.

[0023] According to a fourth aspect of the present invention, a computer device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to perform the following steps:

[0024] Acquire inspection images collected by drones;

[0025] The inspection image is input into a preset target recognition model for target recognition, and the recognition result corresponding to the inspection image is obtained.

[0026] If the inspection image is determined to contain power equipment based on the recognition result, the inspection image is then processed to obtain an image of the power equipment.

[0027] Based on the image of the power equipment, determine whether the power equipment is faulty;

[0028] If the power equipment malfunctions, a work order is generated corresponding to the power equipment, and the work order is sent to the maintenance personnel.

[0029] This invention provides a method, apparatus, storage medium, and computer device for maintaining power equipment. Compared to current manual inspection methods, this method acquires inspection images collected by drones and inputs these images into a preset target recognition model for target recognition. If the recognition result indicates that the inspection image contains power equipment, the image is processed to obtain a power equipment image. Simultaneously, based on the power equipment image, it is determined whether the power equipment is faulty. If a fault is found, a work order is generated and sent to maintenance personnel. Thus, by using inspection images collected by drones to determine the presence of faults in power equipment and automatically generating work order information when a fault is found, automated inspection of power equipment is achieved. This simplifies the overall maintenance process of power equipment, reducing manpower consumption, lowering maintenance costs, and improving maintenance efficiency. Furthermore, using drones for high-altitude inspections increases operational safety. Attached Figure Description

[0030] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0031] Figure 1 This invention provides a schematic flowchart of a method for maintaining power equipment according to an embodiment of the present invention.

[0032] Figure 2 This invention provides a schematic flowchart of another method for maintaining power equipment according to an embodiment of the present invention.

[0033] Figure 3 This diagram illustrates the structure of a maintenance device for power equipment according to an embodiment of the present invention.

[0034] Figure 4 This invention provides a schematic diagram of the structure of another power equipment maintenance device according to an embodiment of the invention.

[0035] Figure 5 A schematic diagram of the physical structure of a computer device provided in an embodiment of the present invention is shown. Detailed Implementation

[0036] The present invention will be described in detail below with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in the present application can be combined with each other.

[0037] Traditional power equipment maintenance requires dedicated personnel for inspection and testing, which greatly increases personnel costs. In addition, if the inspection personnel fail to detect equipment failures in time, the maintenance personnel will not be able to reach the site for maintenance immediately, resulting in low maintenance efficiency.

[0038] To address the aforementioned problems, embodiments of the present invention provide a method for maintaining power equipment, applied to a production process control and management system, such as... Figure 1 As shown, the method includes:

[0039] 101. Acquire inspection images collected by drones.

[0040] Among them, the inspection images are images of power lines taken by drones during flight inspections. These images may include power equipment, conductors, trees and buildings around the conductors, etc.

[0041] This invention is primarily applicable to scenarios involving automated inspection of power equipment and intelligentization of the overall operation and maintenance process. The executing entity of this invention is a device or equipment capable of intelligently maintaining power equipment, which can be specifically located on a server side.

[0042] In this embodiment of the invention, the drone's control and management system is integrated with the production process control and management system. Staff can pre-set the drone's inspection route in the production process control and management system and then send this route to the drone's control and management system. The drone can then fly along this route, photographing the power lines during its inspection. Specifically, it can photograph the power lines according to the flight direction, following the principle of continuous coverage: from general to specific, from top to bottom, from left to right, and from front to back. During the photographing process, the drone can take pictures at preset time intervals, such as one image every 0.1 seconds. When the drone finishes its inspection, it has acquired multiple inspection images. Furthermore, the drone needs to record the time of image capture and the inspection location information at the time of image capture. Thus, the drone's control and management system can obtain the capture time and inspection location information corresponding to each of the multiple inspection images and send them to the production process control and management system.

[0043] Furthermore, the drone can also capture video during flight and use the GPS system to record the drone's inspection location information at different times. Then, the inspection video and the drone's inspection location information at different times are sent to the generation process control management system. The generation process control management system will process the inspection video into frames to obtain multiple inspection images. At the same time, based on the time information corresponding to the multiple inspection images and the drone's inspection location information at different times, the inspection location information corresponding to each inspection image can be determined.

[0044] It should be noted that the UAV control and management system can not only send the captured inspection images or videos to the generation process control and management system after the inspection is completed, but also send the inspection images or videos back to the generation process control and management system in real time during the inspection. This embodiment of the invention does not make specific limitations on this.

[0045] 102. Input the inspection image into a preset target recognition model for target recognition to obtain the recognition result corresponding to the inspection image.

[0046] The preset target recognition model can specifically be a Faster R-CNN model, an SSD model, a YOLO model, etc. Taking the Faster R-CNN model as an example, the target recognition process is explained as follows: After the generation process control management system acquires multiple inspection images and their corresponding inspection location information and shooting time, it can input any one of the inspection images into the pre-trained Faster R-CNN model for target recognition, obtaining the target subject contained in the inspection image and the corresponding bounding box information. The target subject can be power equipment, wires, trees, buildings, etc. It should be noted that the training method of the Faster R-CNN model is known to those skilled in the art and will not be specifically described in this embodiment of the invention.

[0047] 103. If the inspection image contains power equipment according to the recognition result, the inspection image is processed to obtain an image of the power equipment.

[0048] In this embodiment of the invention, since the drone cannot guarantee that every inspection image includes power equipment during continuous shooting, for any inspection image, it is necessary to determine whether the inspection image contains power equipment based on the image recognition result. If it does not contain power equipment, no further processing is performed. If it is determined that the image contains power equipment, the power equipment image is extracted from the inspection image based on the border information corresponding to the power equipment, so as to determine whether the power equipment has any faults or defects based on the power equipment image.

[0049] 104. Based on the image of the power equipment, determine whether the power equipment is faulty.

[0050] In this embodiment of the invention, after acquiring an image of the power equipment, a preset fault detection model can be used to detect whether the power equipment in the image has any faults or defects. Based on this, step 104 includes: inputting the power image into the corresponding preset fault detection model for fault detection, obtaining a first probability value indicating that the power equipment has a fault, and a second probability value indicating that the power equipment does not have a fault; if the first probability value is greater than the second probability value, then it is determined that the power equipment has a fault. The preset fault detection model is essentially an image classification model, specifically such as the AlexNet model or the VGG model.

[0051] The following uses the AlexNet model to illustrate the fault detection process for power equipment. Specifically, the power equipment image is input into the pre-trained AlexNet model to obtain the probability values ​​of the power equipment image belonging to different categories. That is, the first probability value is that the power equipment in the power equipment image has a fault, and the second probability value is that the power equipment in the power equipment image does not have a fault. If the first probability value is greater than the second probability value, it is determined that the power equipment in the power equipment image has a fault and needs to be detected; if the second probability value is greater than the first probability value, it is determined that the power equipment in the power equipment image does not have a fault, and the next power equipment image is processed.

[0052] Furthermore, the AlexNet model can be used to detect the specific fault type of the power equipment. For example, the probability value of the power equipment having a fault is the first probability value, the probability value of the power equipment having a type 1 fault is the second probability value, and the probability value of the power equipment having a type 2 fault is the third probability value. If the third probability value is the largest, it can be determined that the power equipment has a fault defect and the fault type is fault type 2.

[0053] Furthermore, before fault detection, sample images of this type of power equipment can be collected, and the sample images can be labeled according to whether there is a defect or fault in the power equipment in the image. For example, if the power equipment in the sample image has a defect or fault, the sample image is labeled as 1; if the power equipment in the sample image does not exist, the sample image is labeled as 0. This yields a sample image set, which is then used to train the AlexNet model.

[0054] 105. If the power equipment is faulty, a work order information corresponding to the power equipment is generated and the work order information is sent to the maintenance personnel.

[0055] In this embodiment of the invention, once a fault or defect is identified in the power equipment, a corresponding work order needs to be generated and sent to the maintenance personnel. Upon receiving the work order, the maintenance personnel will immediately repair the power equipment. As an optional implementation, the method for generating work order information includes: acquiring the inspection location information of the UAV when the inspection image is captured; matching the inspection location information with the equipment location information of each power device in a preset power equipment database; determining the identification information corresponding to the power device based on the matching result; and generating the work order information based on the identification information corresponding to the power device. The preset power equipment database records the identification information and equipment location information of each power device. Specifically, the identification information can be the power device's number, and the equipment location information can be the longitude and latitude information of the power device.

[0056] Specifically, if there is a defect or fault in the power equipment in the power equipment image, the inspection location information corresponding to the inspection image where the power equipment is located is determined, and the inspection location information is matched with the equipment location information corresponding to each power equipment in the preset power equipment database. Specifically, the positional distance between the inspection location information and multiple equipment location information can be calculated separately, and the minimum positional distance is selected from the multiple calculated positional distances. The power equipment corresponding to the minimum positional distance is the power equipment in the power equipment image, thereby determining the identification information of the power equipment in the power equipment image.

[0057] Furthermore, work order information is generated based on the identification information of the power equipment, fault type, fault detection time, and inspection personnel. Specifically, registered faults and defects can be associated with power equipment to form an equipment history. It can also support the generation of fault orders through alarm information linkage or manual registration. Convenient entry points for defect registration are provided in multiple places such as equipment inspection, equipment maintenance, fault handling, and equipment testing. Manual entry is also allowed to record the source of faults and defects.

[0058] This invention provides a method for maintaining power equipment. By using inspection images collected by drones, the method determines whether there is a fault in the power equipment and automatically generates work order information when a fault is found. This enables automatic inspection of power equipment and makes the overall operation and maintenance process of power equipment intelligent, thereby reducing manpower consumption, lowering operation and maintenance costs, and improving the maintenance efficiency of power equipment. In addition, using drones to replace personnel for high-altitude inspections can increase operational safety.

[0059] Furthermore, to better illustrate the above-described inspection and maintenance process for power equipment, as a refinement and extension of the above embodiments, this invention provides another method for maintaining power equipment, such as... Figure 2 As shown, the method includes:

[0060] 201. Acquire inspection images collected by the drone.

[0061] In this embodiment of the invention, the specific process of obtaining the inspection images collected by the UAV is exactly the same as step 101, and will not be repeated here.

[0062] 202. Input the inspection image into a preset target recognition model for target recognition to obtain the target subject and its corresponding border information contained in the inspection image.

[0063] The border information includes the horizontal coordinate, vertical coordinate, length, and width of the border containing the target entity. The number of target entities can be one, two, or more, and they can be electrical equipment, power lines, trees, buildings, etc. In this embodiment of the invention, the inspection image is input into a preset target recognition model for target recognition, resulting in the target entities contained in the inspection image and their corresponding border information.

[0064] 203. If the target subject includes the power equipment, then the inspection image is cut out according to the horizontal coordinate information, vertical coordinate information, length information and width information in the border information corresponding to the power equipment to obtain the image of the power equipment.

[0065] In this embodiment of the invention, if the identified target subject includes power equipment, subsequent equipment fault detection can be performed. Before formal detection, it is necessary to extract the power equipment image from the inspection image based on the horizontal coordinate information, vertical coordinate information, length information and width information in the border information of the power equipment, so as to determine whether there are defects or faults in the power equipment in the image based on the power equipment image.

[0066] 204. Based on the image of the power equipment, determine whether the power equipment is faulty. If the power equipment is faulty, generate the corresponding work order information for the power equipment and send the work order information to the maintenance personnel.

[0067] For this embodiment of the invention, the specific process of fault detection using the preset fault detection model and the specific process of generating work order information are exactly the same as steps 104 and 105, and will not be repeated here. Further, after generating work order information, the most suitable maintenance personnel can be automatically selected based on the location of the fault and the location of the maintenance personnel, combined with natural resource conditions (wind and solar resources), as well as the professional skills and workload of on-site personnel, and the work order information can be sent to them. The maintenance personnel will then accept and confirm the work order arrangement, issue a work ticket, and submit it for signature. Further, the plant / station is equipped with a positioning system with functions such as personnel walking positioning, trajectory tracking, real-time video intercom, accidental entry interval alarm, proximity alarm, and violation behavior recognition (not within the scope of this contract). Intelligent positioning management can be associated with work orders. After the work order is completed, a navigation map to the work location can be sent to the maintenance personnel, accurately guiding them to the site. The walking route can be displayed on the electronic map of the intelligent maintenance platform, and timely alarm prompts will be given when the route deviates or the wrong interval is reached.

[0068] Furthermore, after maintenance personnel finish their work, they can fill in the defect and fault handling status on handheld terminals or PC clients; if components are replaced, the replacement status can be recorded; if major components are replaced, the component information in the equipment file can be updated with the new spare parts; equipment defects belonging to product batches can be marked. In addition, during fault acceptance, maintenance personnel will explain the defect elimination process, and the on-duty personnel will determine the work completion or return the work to relevant personnel for further processing based on the defect acceptance status. Furthermore, defect and fault statistics and analysis can be performed. Based on information such as defective equipment, defect nature, discovery time, elimination status, equipment type, equipment manufacturer, and equipment model, relevant defect and fault records can be queried. The system can also statistically analyze the number of equipment defects or faults, processing time, etc., and analyze the types of frequently occurring defects based on statistical results and model calculations. Simultaneously, during the processing of defect and fault orders, on-site photo and video recordings are supported, linked to work orders, and automatically saved to the equipment management ledger.

[0069] In specific application scenarios, drones can not only collect inspection images during flight, but also collect parameter values ​​of power equipment, and determine whether there are faults or defects in the power lines based on the magnitude of the parameter values. Based on this, the method further includes: acquiring parameter values ​​collected by the drone at different inspection locations; drawing a first power line trend map based on the parameter values ​​collected at the different inspection locations; and determining whether there are faults in the power lines based on the first power line trend map and a preset parameter value range.

[0070] Specifically, the parameter values ​​collected by the drone can be the temperature of the conductor, etc. Furthermore, the preset parameter value range can be set according to actual business needs, and this embodiment of the invention does not impose specific limitations. For example, the conductor temperature values ​​collected by the drone at different inspection locations during flight are obtained. Then, a first conductor trend graph is plotted with the inspection location information as the horizontal axis and the temperature value as the vertical axis. This first conductor trend graph clearly shows the conductor temperature values ​​at different inspection locations. Next, it is determined whether the conductor temperature values ​​in the first conductor trend graph are within the preset temperature range. If the conductor temperature value at a certain inspection location exceeds the preset temperature range, it indicates that the conductor at that inspection location may have a fault or defect.

[0071] Furthermore, in order to accurately identify whether the conductor at the inspection location is faulty, it is necessary to analyze whether the temperature values ​​at other inspection locations adjacent to the inspection location show an upward trend compared to the temperature value at the inspection location. If there is an upward trend, it indicates that the conductor at the inspection location is faulty, and a work order needs to be generated and sent to the relevant maintenance personnel. Conversely, if there is no upward trend, it indicates that only an isolated point has a high temperature value, and it does not necessarily mean that the conductor is faulty.

[0072] In specific application scenarios, the timing of a conductor failure can also be estimated by drawing a conductor trend chart. Based on this, the method includes: acquiring parameter values ​​of the conductor collected by a drone at different time points for the same inspection location; drawing a second conductor trend chart based on the parameter values ​​at different time points; determining the trend of parameter value changes at the inspection location based on the second conductor trend chart; and estimating the timing of a conductor failure at the inspection location based on the trend of changes.

[0073] For example, temperature values ​​collected by drones at different time points can be gathered for the same inspection location. A second conductor trend chart can then be plotted with time points on the x-axis and temperature values ​​on the y-axis. This chart can be used to analyze the temperature trend of the conductor at the inspection location. If the conductor temperature at that location consistently rises, the time point at which the conductor temperature exceeds a preset temperature range can be predicted. This allows maintenance personnel to be contacted as early as possible to prevent conductor defects and failures.

[0074] 205. If the target subject also includes an environmental obstacle subject and a guide wire, then the center coordinates of the border where the environmental obstacle subject is located and the center coordinates of the border where the guide wire is located are calculated respectively based on the border information corresponding to the environmental obstacle subject and the border information corresponding to the guide wire.

[0075] The environmental obstacles include trees, buildings, etc. In this embodiment of the invention, during the target recognition process of the inspection image, the target can be not only power equipment, but also environmental obstacles and power lines. In order to ensure that the power line can operate normally, the environmental obstacles need to maintain a certain distance from the power line, that is, it is necessary to analyze the surrounding environment of the power line. During the environmental analysis, the distance between the environmental obstacles and the power line can be inferred from the recognition results of the inspection image.

[0076] Specifically, the center coordinates of the border where the wire is located can be calculated based on the horizontal coordinate, vertical coordinate, length, and width information of the border information corresponding to the wire in the recognition results. Similarly, the center coordinates of the border where the environmental obstacle is located can be calculated based on the horizontal coordinate, vertical coordinate, length, and width information of the border information corresponding to the environmental obstacle.

[0077] 206. Based on the center coordinates of the frame where the environmental obstacle is located and the center coordinates of the frame where the conductor is located, calculate the distance between the conductor and the environmental obstacle.

[0078] In this embodiment of the invention, the distance between the guide wire and the main body of the environmental obstacle is calculated based on the horizontal and vertical coordinates of the center point of the frame where the main body of the environmental obstacle is located, and the horizontal and vertical coordinates of the center point of the frame where the guide wire is located.

[0079] 207. Based on the distance, determine whether the environment in which the conductor is located meets the safety requirements.

[0080] In this embodiment of the invention, if the distance is greater than or equal to a preset distance, it indicates that the environment in which the conductor is located is safe, meaning that environmental obstacles will not affect the normal operation of the conductor; if the distance is less than the preset distance, it indicates that the environment in which the conductor is located is unsafe, meaning that environmental obstacles will affect the normal operation of the conductor, and in this case, an alarm message can be sent to relevant personnel. It should be noted that in this embodiment of the invention, the distance between conductors can also be calculated based on the identification result, i.e., determining whether the conductors are within a safe distance.

[0081] Another power equipment maintenance method provided by this invention uses inspection images collected by drones to determine whether there is a fault in the power equipment, and automatically generates work order information when a fault is found. This enables automatic inspection of power equipment and makes the overall operation and maintenance process of power equipment intelligent, thereby reducing manpower consumption, lowering operation and maintenance costs, and improving the maintenance efficiency of power equipment. In addition, using drones to replace personnel for high-altitude inspections can increase operational safety.

[0082] Furthermore, as Figure 1In a specific implementation, embodiments of the present invention provide a maintenance device for power equipment, such as... Figure 3 As shown, the device includes: an acquisition unit 31, an identification unit 32, a cutout unit 33, a judgment unit 34, and a generation unit 35.

[0083] The acquisition unit 31 can be used to acquire inspection images collected by the UAV.

[0084] The recognition unit 32 can be used to input the inspection image into a preset target recognition model for target recognition, and obtain the recognition result corresponding to the inspection image.

[0085] The image matting unit 33 can be used to perform image matting on the inspection image to obtain an image of the power equipment if it is determined from the recognition result that the inspection image contains power equipment.

[0086] The determination unit 34 can be used to determine whether the power equipment has a fault based on the image of the power equipment.

[0087] The generation unit 35 can be used to generate work order information corresponding to the power equipment if the power equipment has a fault, and send the work order information to the operation and maintenance personnel.

[0088] In specific application scenarios, the recognition unit 32 can be used to input the inspection image into a preset target recognition model for target recognition, and obtain the target subject and its corresponding border information contained in the inspection image.

[0089] The image matting unit 33 can be specifically used to perform image matting on the inspection image based on the horizontal coordinate information, vertical coordinate information, length information and width information in the border information corresponding to the power equipment if the target subject includes the power equipment, so as to obtain the image of the power equipment.

[0090] In specific application scenarios, the determination unit 34, such as Figure 4 As shown, it includes: a detection module 341 and a first determination module 342.

[0091] The detection module 341 can be used to input the power image into a corresponding preset fault detection model for fault detection, and obtain a first probability value that the power equipment has a fault and a second probability value that the power equipment does not have a fault.

[0092] The first determining module 342 can be used to determine that the power equipment is faulty if the first probability value is greater than the second probability value.

[0093] In a specific application scenario, the generation unit 35 includes: an acquisition module 351, a matching module 352, a second determination module 353, and a generation module 354.

[0094] The acquisition module 351 can be used to acquire the inspection location information of the UAV when the inspection image is captured.

[0095] The matching module 352 can be used to match the inspection location information with the equipment location information of each power device in the preset power equipment database, wherein the preset power equipment database records the identification information and equipment location information of each power device.

[0096] The second determining module 353 can be used to determine the identification information corresponding to the power equipment based on the matching result.

[0097] The generation module 354 can be used to generate the work order information based on the identification information corresponding to the power equipment.

[0098] In specific application scenarios, the device further includes a computing unit 36.

[0099] The calculation unit 36 ​​can be used to calculate the center coordinates of the border where the environmental obstacle is located, based on the border information corresponding to the environmental obstacle, if the target subject also includes an environmental obstacle and a wire.

[0100] The calculation unit 36 ​​can also be used to calculate the center coordinates of the border where the wire is located based on the border information corresponding to the wire.

[0101] The calculation unit 36 ​​can also be used to calculate the distance between the conductor and the environmental obstacle based on the center coordinates of the frame where the environmental obstacle is located and the center coordinates of the frame where the conductor is located.

[0102] The determination unit 34 can also be used to determine whether the environment in which the conductor is located meets the safety requirements based on the distance.

[0103] In specific application scenarios, the device further includes a drawing unit 37.

[0104] The acquisition unit 31 can also be used to acquire parameter values ​​collected by the UAV at different inspection locations.

[0105] The drawing unit 37 can be used to draw a first traverse trend diagram based on the parameter values ​​collected at the different inspection locations.

[0106] The determination unit 34 can also be used to determine whether there is a fault in the conductor based on the first conductor trend diagram and the preset parameter value range.

[0107] In specific application scenarios, the device further includes: a determination unit 38 and a prediction unit 39.

[0108] The acquisition unit 31 can also be used to acquire parameter values ​​of the conductor collected by the UAV at different time nodes for the same inspection location.

[0109] The drawing unit 37 can also be used to draw a second conductor trend graph based on parameter values ​​at different time points.

[0110] The determining unit 38 can be used to determine the changing trend of parameter values ​​at the inspection location based on the second traverse trend diagram.

[0111] The prediction unit 39 can be used to predict the time point when the conductor at the inspection location will fail based on the changing trend.

[0112] It should be noted that other corresponding descriptions of the functional modules involved in the power equipment maintenance device provided in this embodiment of the invention can be found in [reference]. Figure 1 The corresponding description of the method shown will not be repeated here.

[0113] Based on the above, Figure 1 Accordingly, this embodiment of the invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, performs the following steps: acquiring inspection images collected by a drone; inputting the inspection images into a preset target recognition model for target recognition to obtain a recognition result corresponding to the inspection images; if the recognition result determines that the inspection images contain power equipment, then performing image matting processing on the inspection images to obtain power equipment images; based on the power equipment images, determining whether the power equipment has a fault; if the power equipment has a fault, then generating work order information corresponding to the power equipment and sending the work order information to maintenance personnel.

[0114] Based on the above, Figure 1 The method shown and as Figure 3 The embodiment of the device shown in the invention also provides a physical structure diagram of a computer device, such as... Figure 5As shown, the computer device includes: a processor 41, a memory 42, and a computer program stored in the memory 42 and executable on the processor. Both the memory 42 and the processor 41 are mounted on a bus 43. When the processor 41 executes the program, it performs the following steps: acquiring inspection images collected by a drone; inputting the inspection images into a preset target recognition model for target recognition to obtain a recognition result corresponding to the inspection images; if the recognition result determines that the inspection images contain power equipment, then performing image matting on the inspection images to obtain power equipment images; based on the power equipment images, determining whether the power equipment is faulty; if the power equipment is faulty, then generating work order information corresponding to the power equipment and sending the work order information to maintenance personnel.

[0115] This invention uses inspection images collected by drones to determine whether there is a fault in power equipment, and automatically generates work order information when a fault is found. This enables automatic inspection of power equipment and makes the overall operation and maintenance process of power equipment intelligent, thereby reducing manpower consumption, lowering operation and maintenance costs, and improving the maintenance efficiency of power equipment. In addition, using drones to replace personnel for high-altitude inspections can increase operational safety.

[0116] It is obvious to those skilled in the art that the modules or steps of the present invention described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computer-executable program code, thereby storing them in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those presented herein, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, the present invention is not limited to any particular combination of hardware and software.

[0117] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. 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.

Claims

1. A method for maintaining electrical equipment, characterized in that, include: Acquire inspection images collected by drones; The process of inputting the inspection image into a preset target recognition model for target recognition and obtaining the recognition result corresponding to the inspection image includes: inputting the inspection image into a preset target recognition model for target recognition and obtaining the target subject contained in the inspection image and its corresponding border information; If the inspection image is determined to contain power equipment based on the recognition result, the inspection image is processed to obtain a power equipment image, including: if the target subject includes the power equipment, the inspection image is processed to obtain the power equipment image based on the horizontal coordinate information, vertical coordinate information, length information and width information in the border information corresponding to the power equipment; Based on the image of the power equipment, determine whether the power equipment is faulty; If the power equipment malfunctions, a work order is generated corresponding to the power equipment, and the work order is sent to the maintenance personnel. If the target subject also includes an environmental obstacle and a conductor, then the center coordinates of the frame where the environmental obstacle is located are calculated based on the frame information corresponding to the environmental obstacle; the center coordinates of the frame where the conductor is located are calculated based on the frame information corresponding to the conductor; the distance between the conductor and the environmental obstacle is calculated based on the center coordinates of the frame where the environmental obstacle is located and the center coordinates of the frame where the conductor is located; and the distance is used to determine whether the environment in which the conductor is located meets the safety requirements. The parameter values ​​collected by the UAV at different inspection locations are obtained; a first traverse trend map is drawn based on the parameter values ​​collected at the different inspection locations; and the presence of a fault in the traverse is determined based on the first traverse trend map and a preset parameter value range. For the same inspection location, the parameter values ​​of the conductor collected by the UAV at different time points are obtained; based on the parameter values ​​at different time points, a second conductor trend map is drawn; according to the second conductor trend map, the changing trend of the parameter values ​​at the inspection location is determined; based on the changing trend, the time point at which the conductor at the inspection location will fail is estimated.

2. The method according to claim 1, characterized in that, The step of determining whether the power equipment is faulty based on the image of the power equipment includes: The image of the power equipment is input into a corresponding preset fault detection model for fault detection, and a first probability value of the power equipment having a fault and a second probability value of the power equipment not having a fault are obtained. If the first probability value is greater than the second probability value, then it is determined that the power equipment is faulty.

3. The method according to claim 1, characterized in that, The generation of work order information corresponding to the power equipment includes: Acquire the inspection location information of the UAV when capturing the inspection image; The inspection location information is matched with the equipment location information of each power device in the preset power equipment database, wherein the preset power equipment database records the identification information and equipment location information of each power device; Based on the matching results, the identification information corresponding to the power equipment is determined; The work order information is generated based on the identification information corresponding to the power equipment.

4. A maintenance device for power equipment, characterized in that, include: The acquisition unit is used to acquire inspection images collected by the drone; The recognition unit is used to input the inspection image into a preset target recognition model for target recognition and obtain the recognition result corresponding to the inspection image, including: inputting the inspection image into the preset target recognition model for target recognition and obtaining the target subject and its corresponding border information contained in the inspection image; The image matting unit is used to perform image matting processing on the inspection image to obtain an image of the power equipment if it is determined from the recognition result that the inspection image contains power equipment. The process includes: if the target subject includes the power equipment, then performing image matting on the inspection image based on the horizontal coordinate information, vertical coordinate information, length information, and width information in the border information corresponding to the power equipment to obtain the image of the power equipment. The determination unit is used to determine whether the power equipment has a fault based on the image of the power equipment; The generation unit is used to generate work order information corresponding to the power equipment if the power equipment has a fault, and send the work order information to the operation and maintenance personnel. If the target subject also includes an environmental obstacle and a conductor, then the center coordinates of the frame where the environmental obstacle is located are calculated based on the frame information corresponding to the environmental obstacle; the center coordinates of the frame where the conductor is located are calculated based on the frame information corresponding to the conductor; the distance between the conductor and the environmental obstacle is calculated based on the center coordinates of the frame where the environmental obstacle is located and the center coordinates of the frame where the conductor is located; and the distance is used to determine whether the environment in which the conductor is located meets the safety requirements. The parameter values ​​collected by the UAV at different inspection locations are obtained; a first traverse trend map is drawn based on the parameter values ​​collected at the different inspection locations; and the presence of a fault in the traverse is determined based on the first traverse trend map and a preset parameter value range. For the same inspection location, the parameter values ​​of the conductor collected by the UAV at different time points are obtained; based on the parameter values ​​at different time points, a second conductor trend map is drawn; according to the second conductor trend map, the changing trend of the parameter values ​​at the inspection location is determined; based on the changing trend, the time point at which the conductor at the inspection location will fail is estimated.

5. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 3.

6. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 3.

Images