Intelligent photovoltaic station unmanned aerial vehicle AI inspection management method and system

Abstract

The invention discloses an intelligent photovoltaic station unmanned aerial vehicle AI inspection management method and system, and relates to the technical field of unmanned aerial vehicle inspection. Marking the photovoltaic module image shot by the unmanned aerial vehicle; and carrying out fault detection on the shot photovoltaic module image. According to the intelligent photovoltaic station unmanned aerial vehicle AI inspection management method provided by the invention, through combination of unmanned aerial vehicle inspection and AI diagnosis technologies, efficient inspection and fault detection of photovoltaic station equipment are realized. The unmanned aerial vehicle automatically flies to shoot visible light and infrared images of the assembly, the AI algorithm synchronously analyzes physical damage and temperature abnormity of the assembly, faults are accurately recognized, and a diagnosis report is generated. And meanwhile, a reinspection route can be automatically planned, a fault area can be polled emphatically, and troubleshooting after maintenance is ensured. According to the invention, the inspection efficiency and the fault identification accuracy are significantly improved, the manual inspection cost and risk are reduced, and the operation and maintenance management level and the power generation benefit of the photovoltaic station are improved.

Description

Technical Field

[0001] The present invention relates to the field of drone inspection technology, and specifically to a smart photovoltaic station drone AI inspection management method and system. Background Art

[0002] In recent years, the photovoltaic power generation industry has achieved remarkable development, and the number and scale of photovoltaic power stations have increased year by year. However, the operation and maintenance management of photovoltaic power stations still faces many challenges. First of all, photovoltaic power stations are usually distributed over a wide range, especially in complex terrains such as mountains, and traditional manual inspection methods have many limitations. When operation and maintenance personnel conduct inspections in complex terrains, not only is the efficiency low, but the inspection coverage is incomplete and there are many safety hazards. This makes it impossible to detect the failure of power station equipment in time, affecting the normal power generation efficiency of the power station.

[0003] Secondly, there are many types of equipment in photovoltaic power stations and they are widely distributed. Traditional monitoring methods are difficult to detect specific equipment or string-level faults in a timely and accurate manner. Existing monitoring systems (such as SCADA) can usually only perform centralized monitoring at the site level. It is difficult to quickly identify common problems such as string open circuits, string short circuits, component hot spots, and shielding, and it is impossible to effectively prevent power generation losses caused by equipment failures. This limitation increases the failure rate of equipment and greatly affects the power generation revenue of the power station.

[0004] To solve the above problems, intelligent inspection and operation and maintenance management systems have begun to be gradually applied. The drone AI inspection system conducts efficient equipment inspections on photovoltaic stations through automatic flight, which can quickly identify equipment anomalies, improve inspection efficiency, and reduce the risk of personnel working in harsh environments. In addition, the introduction of AI diagnostic algorithms and deep learning technology enables the operation and maintenance management system to deeply analyze equipment failures and provide accurate fault warnings and maintenance recommendations. However, there is still a lack of mature solutions on the market that deeply integrate drone inspection systems with intelligent operation and maintenance platforms, making it difficult to link and integrate the systems. Summary of the invention

[0005] In view of the above-mentioned problems, the present invention is proposed.

[0006] Therefore, the technical problem solved by the present invention is that the existing inspection and operation and maintenance management methods of photovoltaic stations have problems such as low manual inspection efficiency, inaccurate fault identification, and delayed fault location when faced with a wide variety of equipment, a wide distribution range, and complex terrain. At the same time, the traditional system cannot achieve refined monitoring of the photovoltaic module string level, making it difficult to detect equipment problems in a timely manner and take measures.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions: a smart photovoltaic station drone AI inspection management method, comprising:

[0008] Drones take photos of photovoltaic stations;

[0009] Annotate the images of photovoltaic modules taken by drones;

[0010] Perform fault detection on the captured images of photovoltaic modules.

[0011] As a preferred solution of the intelligent photovoltaic station drone AI inspection management method described in the present invention, wherein: the drone taking photos of the photovoltaic station includes receiving map data of the photovoltaic station and generating a flight route of the drone based on a preset inspection area, using the equipment and terrain data in the station;

[0012] The ground control system transmits the route planning instructions to the drone, triggering its autonomous takeoff;

[0013] The drone flies along the planned route, using a dual-spectrum camera to capture the surface of the PV module and its temperature characteristics;

[0014] The UAV transmits the collected image data and flight parameters to the ground control system via the wireless network in real time.

[0015] As a preferred solution of the smart photovoltaic station drone AI inspection management method described in the present invention, wherein: the dual-spectrum camera includes a visible light camera and an infrared light camera;

[0016] The visible light camera collects images of the physical state of the component, and the infrared camera collects thermal imaging data.

[0017] As a preferred solution of the intelligent photovoltaic station drone AI inspection management method described in the present invention, wherein: the labeling of the photovoltaic component image taken by the drone includes performing image preprocessing, cropping, rotating and geometrically correcting the image after receiving the image data transmitted by the drone;

[0018] For the preprocessed image, the boundary of each photovoltaic module is extracted through the image segmentation algorithm based on the connected domain;

[0019] Each identified component is automatically assigned a unique number, and based on the flight data transmitted by the drone, each component number is bound to its actual location coordinates, and all components are regionally grouped according to their geographic location.

[0020] As a preferred solution of the smart photovoltaic station drone AI inspection management method described in the present invention, the fault detection includes identifying defects of photovoltaic components based on pre-processed visible light and infrared light images;

[0021] The identified defect information is associated with the serial number and GPS coordinate data of each component, each defect is classified, and the location of the defect is marked;

[0022] Generate and write EXIF ​​information containing component number, defect type and GPS coordinates to generate a diagnostic report.

[0023] As a preferred solution of the smart photovoltaic station drone AI inspection management method described in the present invention, the defect identification of photovoltaic components includes synchronously analyzing visible light images and infrared light images, extracting physical characteristics and temperature distribution characteristics of the surface of the photovoltaic components, and classifying defects according to the matching relationship between different feature areas in the two images.

[0024] As a preferred solution of the smart photovoltaic station drone AI inspection management method of the present invention, wherein: the generating of the diagnostic report includes analyzing the fault condition of each component according to the EXIF ​​information of each component combined with the fault data collected during the inspection;

[0025] By comparing the current inspection data with the historical operation data, the data analysis algorithm is used to further confirm the fault type, evaluate the health status of the component, and determine the severity of the fault;

[0026] Automatically generate a complete fault report containing each component number, fault type, fault location and recommended repair plan based on fault severity and component geographical distribution;

[0027] For components marked as faulty in the diagnostic report, the drone re-inspection route is automatically planned based on the severity of the fault and the geographic location of the component.

[0028] Another object of the present invention is to provide a smart photovoltaic station drone AI inspection and management system, which can improve the operation and maintenance efficiency, fault identification accuracy, and fault warning capabilities of photovoltaic stations by building a smart photovoltaic station drone AI inspection and management system, integrating AI analysis and diagnosis technologies, and realizing the automation and intelligence of the inspection and operation and maintenance processes.

[0029] To solve the above technical problems, the present invention provides the following technical solutions: a smart photovoltaic station drone AI inspection and management system, comprising: an image acquisition module, an image annotation module and an inspection and diagnosis module; the image acquisition module is used for drones to take photos of photovoltaic stations; the image annotation module is used to annotate the images of photovoltaic components taken by drones; the inspection and diagnosis module is used to perform fault detection on the taken images of photovoltaic components.

[0030] A computer device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the above-mentioned smart photovoltaic station drone AI inspection management method are implemented.

[0031] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the above-mentioned smart photovoltaic station drone AI inspection management method.

[0032] Beneficial effects of the present invention: The smart photovoltaic station drone AI inspection management method provided by the present invention realizes efficient inspection and fault detection of photovoltaic station equipment by combining drone inspection and AI diagnostic technology. The drone automatically flies to shoot visible light and infrared images of the components, and the AI ​​algorithm simultaneously analyzes the physical damage and temperature anomalies of the components, accurately identifies faults and generates a diagnostic report with component number, fault type, and GPS location. At the same time, it can also automatically plan re-inspection routes, focus on inspecting faulty areas, and ensure troubleshooting after maintenance. The present invention significantly improves inspection efficiency and fault identification accuracy, reduces manual inspection costs and risks, and improves the operation and maintenance management level and power generation efficiency of photovoltaic stations. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0034] Figure 1 An overall flow chart of a smart photovoltaic station drone AI inspection management method provided by an embodiment of the present invention.

[0035] Figure 2 A regional defect map of a smart photovoltaic station drone AI inspection management method provided by an embodiment of the present invention.

[0036] Figure 3 An overall structural diagram of a smart photovoltaic station drone AI inspection management system provided for one embodiment of the present invention. DETAILED DESCRIPTION

[0037] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0038] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0039] Example 1, reference Figure 1-Figure 2 , is an embodiment of the present invention, and provides a smart photovoltaic station drone AI inspection management method, comprising:

[0040] Drones take photos of photovoltaic stations;

[0041] Annotate the images of photovoltaic modules taken by drones;

[0042] Perform fault detection on the captured images of photovoltaic modules.

[0043] The drone taking the photo of the photovoltaic station includes receiving the map data of the photovoltaic station and generating the flight route of the drone based on the preset inspection area and using the equipment and terrain data in the station;

[0044] The ground control system transmits the route planning instructions to the drone, triggering its autonomous takeoff;

[0045] The drone flies along the planned route, using a dual-spectrum camera to capture the surface of the PV module and its temperature characteristics;

[0046] The UAV transmits the collected image data and flight parameters to the ground control system via the wireless network in real time.

[0047] The dual-spectrum camera includes a visible light camera and an infrared light camera;

[0048] The visible light camera collects images of the physical state of the component, and the infrared camera collects thermal imaging data.

[0049] The labeling of the photovoltaic component image taken by the drone includes performing image preprocessing, cropping, rotating and geometrically correcting the image after receiving the image data transmitted by the drone, so that the image is aligned with the actual component position and the error generated during the shooting process is corrected;

[0050] For the preprocessed image, the boundary of each photovoltaic module is extracted through the image segmentation algorithm based on the connected domain;

[0051] Each identified component is automatically assigned a unique number, and based on the flight data transmitted by the drone, each component number is bound to its actual location coordinates, and all components are regionally grouped according to their geographic location.

[0052] The fault detection includes identifying defects of photovoltaic modules based on pre-processed visible light and infrared light images;

[0053] The identified defect information is associated with the serial number and GPS coordinate data of each component, and each defect is classified and the location of the defect is marked, such as Figure 2 shown.

[0054] Generate and write EXIF ​​information containing component number, defect type and GPS coordinates to generate a diagnostic report.

[0055] The defect identification of photovoltaic modules includes synchronously analyzing visible light images and infrared light images, extracting physical characteristics and temperature distribution characteristics of the photovoltaic module surface, and classifying defects according to the matching relationship between different feature areas in the two images.

[0056] Texture analysis is performed on visible light images to identify the surface physical state of photovoltaic modules. The morphology and location of cracks on the module surface are identified through edge detection algorithms, and crack feature areas are generated. For surface dirt, dust or external obstructions, color and texture inhomogeneity analysis is used to extract the obstructed area and record it as a physical defect. In addition, the missing of components or brackets is determined by checking the integrity of the component outline, and the missing area is also marked as a physical defect. These physical features are bound to the corresponding component number and recorded to form a set of physical features of the component.

[0057] Infrared images are used to detect the temperature distribution characteristics of components. First, detect whether there are local temperature anomalies and mark the hot spot area. Through temperature gradient analysis, further identify whether there are linear or banded high-temperature areas on the component surface. For overall temperature anomalies, such as when the overall temperature of the component is lower than other normal components, it is marked as an overall low-temperature area and associated with the electrical connection status of the component. These temperature characteristics are also bound to the number and spatial position of the corresponding component to ensure that the temperature data corresponds to the physical characteristics one by one.

[0058] A comprehensive analysis is performed based on the features of visible light images and infrared images. First, the crack area detected in the visible light image is matched with the corresponding temperature distribution area in the infrared image. If there is a local temperature rise in the crack area and the temperature difference reaches the set threshold, the area is marked as a diode fault. If a dirt-blocked area is detected in the visible light image, it is matched and analyzed with the temperature anomaly of the corresponding area in the infrared image. If the two match, it is marked as a spot fault. For the strip-shaped high-temperature area, if the infrared image detects a linear temperature rise and no external obstruction is detected in the visible light image, it is judged as a strip-shaped hot spot fault. The overall low temperature is judged in combination with the electrical connection problem of the component. If the electrical connection of the component is normal and there is no external obstruction, it is marked as a no-load fault.

[0059] After the preliminary analysis, an in-depth correlation analysis is performed on the failure conditions of each component. If cracks and hot spots are found to exist at the same time, we will first analyze whether the cracks cause local overheating to determine whether it is a diode failure. In the case where obstructions and hot spots coexist, it will be determined whether the hot spot is caused by the obstruction based on the physical characteristics of the obstruction and the area size of the hot spot. If there is a significant difference between the obstruction area caused by dirt and the temperature abnormality area, this area will be marked as a combined fault. Further analysis is performed to determine whether there are chain failures, such as strip hot spots caused by cracks extending to surrounding components, or abnormal temperatures of surrounding components caused by no-load failures. By expanding the detection area, ensure that all potential faults are correctly identified.

[0060] Generating the diagnostic report includes analyzing the fault condition of each component based on the EXIF ​​information of each component combined with the fault data collected during the inspection;

[0061] By comparing the current inspection data with the historical operation data, the data analysis algorithm is used to further confirm the fault type, evaluate the health status of the component, and determine the severity of the fault;

[0062] Automatically generate a complete fault report containing each component number, fault type, fault location and recommended repair plan based on fault severity and component geographical distribution;

[0063] For components marked as faulty in the diagnostic report, the drone re-inspection route is automatically planned based on the severity of the fault and the geographic location of the component.

[0064] Example 2

[0065] Reference Figure 3 , is an embodiment of the present invention, which provides a smart photovoltaic station drone AI inspection management system, including:

[0066] Image acquisition module, image annotation module and inspection and diagnosis module;

[0067] The image acquisition module is used by drones to take photos of photovoltaic stations;

[0068] The image annotation module is used to annotate the photovoltaic module image taken by the drone;

[0069] The inspection and diagnosis module is used to perform fault detection on the photographed photovoltaic component images.

[0070] Embodiment 3, an embodiment of the present invention, is different from the first two embodiments in that:

[0071] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0072] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0073] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0074] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0075] Example 4 is an embodiment of the present invention, which provides a smart photovoltaic station drone AI inspection management method. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0076] The experimental site is a 500kW photovoltaic power station, covering an area of ​​about 10 hectares, including 500 photovoltaic modules, some of which are set as faulty modules. The site has a complex terrain, with both flat areas and a small amount of slopes. The experimental equipment includes a drone equipped with a dual-spectrum camera (visible light + infrared camera), with a flight altitude set at 30 meters, a speed of 5 meters / second, and a flight time of 40 minutes / time. The ground control system receives image data and flight parameters transmitted by the drone through the wireless network, and runs the AI ​​inspection management system for image processing, fault identification, diagnostic report generation, and re-inspection route planning.

[0077] First, the site map data and equipment distribution information are input into the AI ​​inspection system, and the system automatically generates the flight route for the drone. After receiving the route planning, the drone automatically takes off and flies along the preset route. It uses a dual-spectrum camera to capture visible light and infrared images of photovoltaic modules and transmits them to the ground control system in real time. The system pre-processes the image, including image cropping, rotation, and geometric correction, and uses an image segmentation algorithm to mark the component boundaries and generate component numbers to match the GPS location. Next, the system synchronously analyzes visible light and infrared images to identify physical problems (such as cracks, occlusions) and hot spot failures on the surface of the component. After the fault information is associated with the component number and GPS location, the system generates a diagnostic report that records the component fault type, location, and recommended maintenance measures. Subsequently, the system automatically plans a re-inspection route based on the severity of the fault, and the drone performs the re-inspection task, focusing on inspecting the fault area.

[0078] The traditional SCADA system is compared with the conventional drone inspection scheme. The SCADA system collects basic electrical data of inverters and components through sensors and performs site-level monitoring. Conventional drones perform regular inspection tasks, fly along fixed routes, and take images of the appearance of photovoltaic components (without infrared function). After the images are transmitted to the ground station, manual analysts manually identify physical problems of the components, such as cracks or obstructions, based on the images. After the fault data is sorted, an inspection report is generated. During re-inspection, it is necessary to manually plan the re-inspection route and use drones to conduct a second inspection of specific fault areas. The experimental results are shown in Table 1.

[0079] Table 1 Comparison of experimental results

[0080]

[0081] From the comparative analysis of experimental data, the drone AI inspection system of the present invention can complete efficient inspections of a large range of photovoltaic components in a relatively short period of time through preset routes and real-time data transmission. However, the traditional solution relies on the limited data of the SCADA system and the manual analysis of conventional drones, so the inspection efficiency is low and there is a phenomenon of missed detection. The AI ​​algorithm can process visible light and infrared images simultaneously, and identify hot spot faults while detecting surface problems of components. The accuracy of fault identification is higher than that of traditional solutions. In terms of diagnostic report generation, traditional systems rely on manual analysis and take a long time, while the present invention can generate diagnostic reports in real time after the inspection, providing detailed fault data and maintenance suggestions.

[0082] In addition, the present invention automatically plans re-inspection routes through AI, effectively reducing manual intervention and improving the automation and intelligence level of inspections. The re-inspection route planning is accurate, avoiding missed inspections or repeated inspections due to human misjudgment. Compared with traditional technical solutions, the present invention greatly reduces operation and maintenance costs and reduces manual participation, especially in complex terrain environments such as mountains. Inspection safety has been significantly improved. Overall, the technical solution of the present invention improves the efficiency and reliability of photovoltaic station operation and maintenance management through automated and intelligent inspection processes, and can effectively ensure the continuity and stability of photovoltaic power generation.

[0083] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A smart photovoltaic station drone AI inspection management method, characterized in that: include: Drones take photos of photovoltaic stations; Annotate the images of photovoltaic modules taken by drones; Perform fault detection on the captured images of photovoltaic modules.

2. The smart photovoltaic station drone AI inspection management method according to claim 1 is characterized by: The drone taking the photo of the photovoltaic station includes receiving the map data of the photovoltaic station and generating the flight route of the drone based on the preset inspection area and using the equipment and terrain data in the station; The route planning instructions are transmitted to the UAV through the ground control system, triggering its autonomous takeoff; The drone flies along the planned route, using a dual-spectrum camera to capture the surface of the PV module and its temperature characteristics; The drone transmits the collected image data and flight parameters to the ground control system via the wireless network in real time.

3. The smart photovoltaic station drone AI inspection management method according to claim 2 is characterized by: The dual-spectrum camera includes a visible light camera and an infrared light camera; The visible light camera collects images of the physical state of the component, and the infrared camera collects thermal imaging data.

4. The smart photovoltaic station drone AI inspection management method according to claim 3 is characterized by: The labeling of the photovoltaic module image taken by the drone includes performing image preprocessing, cropping, rotating and geometrically correcting the image after receiving the image data transmitted by the drone; For the preprocessed image, the boundary of each photovoltaic module is extracted through the image segmentation algorithm based on the connected domain; Each identified component is automatically assigned a unique number, and based on the flight data transmitted by the drone, each component number is bound to its actual location coordinates, and all components are regionally grouped according to their geographic location.

5. The smart photovoltaic station drone AI inspection management method according to claim 4 is characterized by: The fault detection includes identifying defects of photovoltaic modules based on pre-processed visible light and infrared light images; The identified defect information is associated with the serial number and GPS coordinate data of each component, each defect is classified, and the location of the defect is marked; Generate and write EXIF ​​information containing component number, defect type and GPS coordinates to generate a diagnostic report.

6. The smart photovoltaic station drone AI inspection management method according to claim 5 is characterized by: The defect identification of photovoltaic modules includes synchronously analyzing visible light images and infrared light images, extracting physical characteristics and temperature distribution characteristics of the photovoltaic module surface, and classifying defects according to the matching relationship between different feature areas in the two images.

7. The smart photovoltaic station drone AI inspection management method according to claim 6 is characterized by: Generating the diagnostic report includes analyzing the fault condition of each component based on the EXIF ​​information of each component combined with the fault data collected during the inspection; By comparing the current inspection data with the historical operation data, the data analysis algorithm is used to further confirm the fault type, evaluate the health status of the component, and determine the severity of the fault; Automatically generate a complete fault report containing each component number, fault type, fault location and recommended repair plan based on fault severity and component geographical distribution; For components marked as faulty in the diagnostic report, the drone re-inspection route is automatically planned based on the severity of the fault and the geographic location of the component.

8. A system using the smart photovoltaic station drone AI inspection management method as described in any one of claims 1 to 7, characterized in that: include: Image acquisition module, image annotation module and inspection and diagnosis module; The image acquisition module is used by drones to take photos of photovoltaic stations; The image annotation module is used to annotate the photovoltaic module image taken by the drone; The inspection and diagnosis module is used to perform fault detection on the photographed photovoltaic component images.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the smart photovoltaic station drone AI inspection management method described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the smart photovoltaic station drone AI inspection management method described in any one of claims 1 to 7 are implemented.

Images