Unmanned aerial vehicle photovoltaic intelligent inspection system and method

Abstract

The invention discloses an unmanned aerial vehicle photovoltaic intelligent inspection system and method, and the system can achieve the building of a three-dimensional model of a whole photovoltaic station through a three-dimensional modeling system. The intelligent routing inspection path optimization system can formulate a routing inspection closed-loop path based on a routing inspection period and a key area fed back by the dynamic optimization intelligent decision-making system and according to three-dimensional modeling data information, and the unmanned aerial vehicle module is used for executing a routing inspection task and transmitting image information. The fault and defect real-time detection system carries out processing and matching identification according to images shot by the unmanned aerial vehicle to obtain fault and defect types and feeds the fault and defect types back to the dynamic optimization intelligent decision-making system, and the dynamic optimization intelligent decision-making system makes a decision according to the fault types to obtain regional health indexes and feeds data information back to the three-dimensional modeling system; and the inspection period is further adjusted and optimized, so that an intelligent optimized inspection closed loop is formed, the resource waste of long-term large-area inspection is reduced, the inspection cost is reduced, and the inspection efficiency is improved.

Description

Technical Field

[0001] This invention belongs to the field of photovoltaic inspection technology, and in particular relates to an unmanned aerial vehicle (UAV) intelligent photovoltaic inspection system and method. Background Technology

[0002] Photovoltaic power plants are mostly located in harsh environments such as mountains, deserts, and Gobi, covering large areas with scattered plots and numerous photovoltaic modules. These modules are susceptible to the effects of the natural environment, climate change, and external pollution, leading to malfunctions such as hot spots, shading, and cracking. This results in decreased power generation efficiency and even serious failures like equipment burnout, impacting the safety and economic benefits of the power plant. However, manual inspections are costly, time-consuming, inefficient, and pose safety hazards, making them unsuitable for the operation and maintenance needs of large-scale photovoltaic power plants. With the continuous development and optimization of drone technology, and the increasing scale and operation and maintenance demands of photovoltaic power plants, drone inspections have become a powerful tool for photovoltaic power plant operation and maintenance, playing an increasingly important role in significantly reducing operation and maintenance costs and improving efficiency.

[0003] In existing technologies, the inspection strategy of drones for large-scale photovoltaic (PV) fields often involves inspecting each PV field individually, which leads to a sharp increase in inspection costs and low inspection efficiency. Furthermore, while PV modules in some PV areas may operate well for a long time, PV modules in other PV areas may frequently malfunction. This results in a lot of redundancy in the inspection and photography of some PV areas, while it is difficult to detect malfunctioning PV areas in a timely manner. Existing technologies cannot optimize the inspection of PV fields, reduce the inspection frequency of areas in good operating condition, and increase the number of inspections of areas prone to malfunction.

[0004] Patent application CN117873162A discloses a drone-based intelligent photovoltaic inspection device, method, electronic device, and storage medium. This invention addresses the technical problems in related technologies where a single inspection strategy leads to low inspection efficiency and makes it difficult to balance inspection efficiency and accuracy. It combines the advantages of high-altitude, high-speed inspection with low-altitude, low-speed inspection, optimizing drone flight control, improving inspection efficiency, and avoiding resource waste or complex, repetitive scanning across the entire plant. However, it does not address optimizing and adjusting the inspection cycle to significantly improve inspection efficiency and reduce inspection costs by minimizing redundant, repetitive inspections.

[0005] Patent application CN119396172A discloses a photovoltaic inspection management method and system based on unmanned aerial vehicles (UAVs). This system primarily plans flight paths and endurance strategies using an inspection planning model, then analyzes and evaluates the real-time transmitted data using a data analysis and evaluation model. Based on the evaluation results, corresponding operation and maintenance measures are formulated, thereby addressing the reliance on manual inspection in existing methods. However, this system does not include a technical solution to reduce inspection costs by optimizing the inspection cycle. Summary of the Invention

[0006] To address the aforementioned technical problems, this invention provides an unmanned aerial vehicle (UAV) intelligent photovoltaic inspection system and method.

[0007] The present invention is achieved through the following technical solutions.

[0008] The present invention provides an unmanned aerial vehicle (UAV) photovoltaic intelligent inspection system, comprising a 3D modeling system, an intelligent inspection path optimization system, an UAV module, a real-time fault and defect detection system, and a dynamic optimization intelligent decision-making system;

[0009] The 3D modeling system is used to generate 3D models of photovoltaic power plants and can generate fault cloud maps of photovoltaic power plant areas based on the fault conditions of photovoltaic units in different regions.

[0010] The intelligent inspection path optimization system can acquire data information from the 3D modeling system, generate a closed-loop inspection path, and accurately reach the inspection destination based on positioning and navigation to generate inspection tasks.

[0011] The drone module is used to perform inspection tasks of the intelligent inspection path optimization system and generate real-time images of photovoltaic modules;

[0012] The real-time fault and defect detection system includes a photovoltaic module fault database and stores historical fault records of photovoltaic power station modules. After acquiring images of the photovoltaic modules and identifying the fault and defect types, the information is fed back to the dynamic optimization intelligent decision-making system.

[0013] The dynamic optimization intelligent decision-making system generates a health index for each area of ​​the photovoltaic system based on the information transmitted by the real-time fault and defect detection system. The health index and fault and defect information are then transmitted to the 3D modeling system for marking and recording, forming an intelligent inspection closed loop to achieve intelligent optimization of photovoltaic inspection.

[0014] An inspection method for a drone-based intelligent photovoltaic inspection system includes the following steps:

[0015] S1: The three-dimensional modeling system uses laser scanning equipment and other methods to collect data from the entire photovoltaic field area in an all-round way, obtain information such as its geographical coordinates and array arrangement, construct a three-dimensional model, and generate a photovoltaic field area fault cloud map based on the health index and fault status of different areas, so as to urge the operation and maintenance personnel to rectify in a timely manner. On the other hand, the relevant information of the photovoltaic field station is fed back to the intelligent inspection path optimization system.

[0016] S2: Based on the optimal path principle, the intelligent inspection path optimization system comprehensively considers the drone's endurance and other factors, and formulates a closed-loop inspection path from the take-off point to the key area and then to the return point according to information such as the photovoltaic power station's positioning. At the same time, it also takes into account the necessary inspection of other areas, and sends the planned path to the drone control system for execution according to the inspection path.

[0017] S3: The UAV module includes the UAV body, power system, control system, infrared camera and transmission system. After receiving the instruction from the intelligent inspection path optimization system, the UAV begins inspection. After reaching the designated area, it turns on the infrared camera to take pictures and transmits the pictures to the real-time fault and defect detection system in real time through the transmission system.

[0018] S4: The real-time fault and defect detection system will process the received images and match them with its database to identify the fault and defect type of the photovoltaic module, such as module damage, hot spot effect, etc., and feed the results back to the dynamic optimization intelligent decision-making system.

[0019] S5: After receiving information transmitted by the real-time fault and defect detection system, the dynamic optimization intelligent decision-making system makes decisions based on the fault level and priority, calculates the regional health index, and sends it to the 3D modeling system to update the fault distribution of the photovoltaic field. On the other hand, it adjusts and optimizes the inspection cycle and key areas based on the photovoltaic regional health index and feeds it back to the intelligent inspection path optimization system to execute the next inspection task, forming a new inspection closed loop. This realizes intelligent optimization inspection of the photovoltaic field, thereby reducing inspection resource waste, lowering inspection costs, and improving inspection efficiency.

[0020] Preferably, the 3D modeling system can update the photovoltaic field fault cloud map in a timely manner based on the photovoltaic area health index and fault status, and urge operation and maintenance personnel to rectify and repair in a timely manner.

[0021] Preferably, the intelligent inspection path optimization system receives data from the 3D modeling system to obtain information such as the location of the entire photovoltaic power station, and receives the inspection cycle and key areas fed back by the dynamic optimization intelligent decision-making system to plan the path.

[0022] Preferably, the intelligent inspection path optimization system can not only plan inspection paths for key areas, but also plan inspection paths for other areas according to the inspection cycle, and can assign appropriate drones for inspection based on the planned path and the drone's endurance performance.

[0023] Preferably, the drone module can be solar charged on the take-off and landing platform, and the take-off and landing platform can be set in key areas according to the photovoltaic power station situation to save inspection costs and improve inspection efficiency.

[0024] Preferably, the real-time fault and defect detection system can classify faults and defects into levels so that they can be processed according to their priority.

[0025] Preferably, the real-time fault and defect detection system can expand the fault and defect database with images captured by the UAV, further improving accuracy.

[0026] Preferably, the dynamic optimization intelligent decision-making system can not only make decisions on faults and defects, but also optimize the next inspection based on information such as defects and key areas of the photovoltaic field. At the same time, it can formulate the weight of the corresponding fault based on the fault level in order to calculate the regional health index.

[0027] Preferably, the regional health index of the photovoltaic field is determined by the number of faulty modules in the area and the severity of the fault (scored from 0 to 10 based on the type of fault, with higher scores indicating more severe faults), denoted as S0, S1, ..., S... i ,...,S 10 The total number of regional components constitutes the regional health index, and the formula for calculating the regional health index is as follows:

[0028]

[0029] Where H is the regional health index of the photovoltaic field; N is the number of modules that failed in the region; and M is the total number of modules in the region. This indicates the failure rate of the component.

[0030] Preferably, the T i Let T0 be the intelligent optimized inspection cycle for the i-th inspection in the photovoltaic intelligent inspection system. The basic inspection cycle T0 is the preset standard inspection time interval in the photovoltaic intelligent inspection system, representing the cycle for a comprehensive inspection of the photovoltaic area under ideal conditions (no faults, no abnormalities, etc.). T1 is the inspection cycle corresponding to the first inspection, i.e., the initial inspection cycle T1 = T0. The optimized inspection cycle for this area is expressed as:

[0031]

[0032] Where α is the adjustment coefficient, 0 < α < 1, which represents the weight of the health index on the inspection cycle and can be adjusted according to actual operation and maintenance experience and power plant conditions.

[0033] Preferably, if the photovoltaic area continues to be faulty, the inspection cycle is gradually shortened and the inspection frequency is increased. If there is no fault after the fault is repaired and there is no fault in the next inspection cycle, the inspection cycle is gradually extended and the inspection frequency is reduced. This realizes intelligent optimization of the inspection cycle based on the photovoltaic area health index, which can both focus on key areas for key inspections and reduce the inspection frequency of fault-free areas, thereby reducing inspection costs and improving inspection efficiency.

[0034] The beneficial effects of this invention are as follows:

[0035] 1. Improve inspection efficiency: The intelligent inspection path optimization system plans the optimal inspection path, avoiding the blindness of traditional one-by-one inspections, reducing the invalid flight distance and time of drones. At the same time, combined with the dynamically adjusted inspection cycle, it can quickly and efficiently inspect key areas and areas prone to failure, greatly improving inspection efficiency.

[0036] 2. Reduce inspection costs: By reducing redundant inspections of areas in good operating condition and rationally scheduling UAV flight missions and inspection cycles, the frequency of UAV use, fuel consumption, and maintenance costs are reduced. At the same time, manpower input is reduced, effectively lowering the overall inspection cost.

[0037] 3. Precise Fault Detection: The real-time fault and defect detection system can receive images captured by drones in real time and perform image matching and recognition. It can quickly and accurately identify various fault and defect types of photovoltaic modules, providing a reliable basis for timely maintenance, improving the timeliness and accuracy of fault handling, and ensuring the stable operation of photovoltaic power plants.

[0038] 4. Precise Component Location: Based on the location information of each photovoltaic component in the 3D model, precise location is achieved, which helps to accurately find the faulty component after a fault occurs, enabling maintenance personnel to handle the faulty component in a timely manner and improving maintenance efficiency.

[0039] 5. Optimize inspection cycle: Based on the fault type of each photovoltaic area and the fault weight, the regional health index is obtained, and the inspection cycle is optimized in real time. This reduces the inspection frequency of fault-free areas and strengthens the key inspections of key areas, avoiding ineffective inspections, concentrating resources on key areas, shortening the fault detection time, improving the overall inspection efficiency, and ensuring the safe and stable operation of photovoltaic power stations.

[0040] 6. Achieving Intelligent Closed-Loop Operation and Maintenance: The 3D modeling system, intelligent inspection path optimization system, real-time fault and defect detection system, and dynamic optimization intelligent decision-making system collaborate and share data to form a complete intelligent optimization inspection closed loop. The intelligent inspection system can automatically and dynamically adjust inspection strategies based on the actual operating conditions of the photovoltaic power station, achieving intelligent and refined operation and maintenance management of the photovoltaic power station. This reduces human intervention, improves operation and maintenance management levels, and has significant advantages in cost, efficiency, and operation and maintenance, making it suitable for cost reduction and efficiency improvement in large-scale photovoltaic inspections. Attached Figure Description

[0041] Figure 1 This is a flowchart illustrating the inspection method of the present invention;

[0042] Figure 2 This is a flowchart illustrating the inspection system of the present invention. Detailed Implementation

[0043] The technical solution of the present invention is further described below, but the scope of protection is not limited to what is described.

[0044] Example 1:

[0045] like Figure 1 As shown, an inspection method for a drone-based intelligent photovoltaic inspection system includes the following steps:

[0046] S1: Before the first inspection by the drone, the dynamic optimization intelligent decision-making system should preset the basic inspection cycle T0 (also the first inspection cycle T1) based on the photovoltaic power station inspection records, operation and maintenance experience, power station conditions, etc. At the same time, the initial value of the adjustment coefficient should be set according to the influence weight of the health index on the inspection cycle. The adjustment coefficient can also be adjusted and optimized according to the actual inspection situation.

[0047] S2: The intelligent inspection path optimization system combines information from the 3D modeling system and inspection strategies to plan the path, generate a closed-loop inspection path, and initiate the i-th intelligent optimization inspection.

[0048] S3: After receiving instructions from the intelligent inspection path optimization system, the UAV takes off from the take-off and landing platform and arrives at the photovoltaic inspection area according to the planned path instructions (including parameters such as flight altitude, speed, and shooting angle). It turns on the infrared camera to capture images and transmits the images to the real-time fault and defect detection system in real time through the UAV transmission system. The system processes the image data, matches and identifies the fault and defect types, and transmits them to the dynamic optimization intelligent decision-making system. After the UAV finishes shooting, it returns to the take-off and landing platform according to the inspection closed-loop path, completing the closed loop of one inspection task.

[0049] S4: If there is a fault in the inspection area, the dynamic optimization intelligent decision-making system calculates the health index of the photovoltaic area according to the calculation formula of fault weight and health index H, and calculates and adjusts the next inspection cycle Ti+1 according to the optimization inspection cycle formula. It also determines the key areas by combining the feedback information such as the location of the fault area. If there is no fault, the next inspection cycle Ti+1 will be executed according to the inspection cycle Ti-1. If it is the first inspection, it will be executed according to the basic inspection cycle.

[0050] S5: The dynamic optimization intelligent decision-making system feeds back information such as new inspection cycles and key areas to the intelligent inspection path optimization system, generates new inspection tasks, and begins the (i+1)th intelligent optimization inspection. Simultaneously, it sends regional health indices and fault information to the 3D modeling system to update the overall operation of the photovoltaic power station in real time and urges maintenance personnel to promptly address and repair faulty areas.

[0051] Example 2:

[0052] refer to Figure 2 Based on Embodiment 1, this embodiment provides a drone-based intelligent photovoltaic inspection system, comprising:

[0053] The 3D modeling system collects multi-dimensional data such as geographic coordinates, component layout, and topography of photovoltaic power stations through laser scanning and other methods. It constructs a 3D model of the photovoltaic power station, marks the positioning coordinates of the center point of each photovoltaic unit, and numbers each photovoltaic unit according to the actual coordinate coding rules of the photovoltaic field. This provides the intelligent inspection path optimization system with information such as geographic location and component code to assist in path planning. At the same time, it receives regional health index and fault information from the dynamic optimization intelligent decision-making system to update the photovoltaic area status label in the model, realizes a three-dimensional and intuitive display of the photovoltaic power station, and urges operation and maintenance personnel to handle and rectify fault areas in a timely manner.

[0054] The 3D modeling system can update the fault cloud map of the photovoltaic field in a timely manner based on the health index and fault status of the photovoltaic area, and urge the operation and maintenance personnel to rectify and repair in a timely manner; the drone module can realize solar charging on the take-off and landing platform, and the take-off and landing platform can be set in key areas according to the photovoltaic station situation.

[0055] The intelligent inspection path optimization system has a built-in optimal path planning algorithm. It obtains data information from the 3D modeling system, clarifies the spatial structure of the photovoltaic power station, and realizes path planning. At the same time, it receives the optimized inspection cycle and key area adjustment instructions from the dynamic optimization intelligent decision-making system, determines the next inspection time and inspection area, and sends the planned path instructions (including parameters such as flight altitude, speed, and shooting angle) to the UAV module when the next inspection begins, driving it to execute the inspection task.

[0056] The intelligent inspection path optimization system receives data from the 3D modeling system to obtain the location information of the entire photovoltaic power station. It can also receive the inspection cycle and key areas fed back by the dynamic optimization intelligent decision-making system to plan the path.

[0057] The intelligent inspection path optimization system can not only plan inspection paths for key areas, but also plan inspection paths for other areas according to the inspection cycle, and can assign appropriate drones for inspection based on the planned path and the drone's endurance.

[0058] After receiving the path instruction from the intelligent inspection path optimization system, the drone module starts the inspection according to its planned path. After reaching the designated photovoltaic area, it takes real-time pictures and transmits them to the real-time fault and defect detection system.

[0059] The real-time fault and defect detection system is used to receive image data transmitted back in real time from the UAV module, process and match it to quickly determine whether there are faults such as hot spots or damage, mark the fault type and location, and feed it back to the dynamic optimization intelligent decision-making system for further decision-making.

[0060] The real-time fault and defect detection system can classify faults and defects according to their levels so that they can be processed according to their priority. It can also expand the fault and defect database with images taken by drones to further improve accuracy.

[0061] The dynamic optimization intelligent decision-making system, based on fault information fed back by the real-time fault and defect detection system, and according to the regional health index and the optimized inspection cycle calculation formula, obtains the regional health index, optimized inspection cycle, and key areas. It then feeds the latest inspection strategy back to the intelligent inspection path optimization system to ensure that the next inspection is executed according to the latest strategy. When determining the optimized inspection cycle, the dynamic optimization intelligent decision-making system can adjust the adjustment coefficient according to the influence weight of the health index on the inspection cycle to better optimize the inspection cycle. Simultaneously, it updates the regional health index to the 3D modeling system, forming a closed loop for intelligent optimized inspection decision-making.

[0062] The dynamic optimization intelligent decision-making system can not only make decisions about faults and defects, but also optimize the next inspection based on the defects and key area information of the photovoltaic field. At the same time, it can formulate the corresponding weight of faults according to the fault level in order to calculate the regional health index of the photovoltaic field.

[0063] After receiving information from the real-time fault and defect detection system, the dynamic optimization intelligent decision-making system makes decisions based on the type and priority of faults and defects, obtains the health index of each photovoltaic area, and then transmits the health index and fault and defect information to the 3D modeling system for marking and recording. Maintenance personnel carry out inspection and rectification based on the fault information, forming an intelligent inspection closed loop. The inspection cycle is then optimized based on the health index to achieve intelligent optimization of photovoltaic inspections. Different inspection plans are adopted for different photovoltaic areas to avoid repeated redundant inspections, greatly improve inspection efficiency, and reduce inspection costs.

[0064] The regional health index of the photovoltaic field is determined by the number of faulty modules in the area and the severity of the fault, denoted as S0, S1, ..., S... i ,...,S 10 The total number of regional components constitutes the regional health index, and the formula for calculating the regional health index is as follows:

[0065]

[0066] Where H is the regional health index of the photovoltaic field; N is the number of modules that failed in the region; and M is the total number of modules in the region. This indicates the failure rate of the component.

[0067] The T i Let T0 be the intelligent optimized inspection cycle for the i-th inspection in the photovoltaic intelligent inspection system. The basic inspection cycle T0 is the preset standard inspection time interval in the photovoltaic intelligent inspection system, representing the cycle for a comprehensive inspection of the photovoltaic area under ideal conditions. T1 is the inspection cycle corresponding to the first inspection, i.e., the initial inspection cycle T1 = T0. The optimized inspection cycle for this area is expressed as:

[0068]

[0069] Where α is the adjustment coefficient, 0 < α < 1, representing the weight of the health index on the inspection cycle.

[0070] If the photovoltaic area continues to experience a fault, the inspection cycle will be gradually shortened and the inspection frequency will be increased. If there is no fault after the fault is repaired and there is no fault in the next inspection cycle, the inspection cycle will be gradually extended and the inspection frequency will be reduced. This achieves intelligent optimization of the inspection cycle based on the photovoltaic area health index, which can both focus on key areas for inspection and reduce the inspection frequency of fault-free areas, thereby reducing inspection costs and improving inspection efficiency.

[0071] In the description of this invention, it should be noted that the embodiments provided are only for the purpose of facilitating the description of this invention and simplifying the description. The accompanying drawings are only for illustrative purposes and do not represent its defects or limitations. The dimensions in the drawings do not represent actual dimensions, and the arrow "→" only indicates direction and has no other actual meaning.

Claims

1. A drone-based intelligent photovoltaic inspection system, characterized in that: It includes a 3D modeling system, an intelligent inspection path optimization system, a drone module, a real-time fault and defect detection system, and a dynamic optimization intelligent decision-making system; The 3D modeling system is used to generate 3D models of photovoltaic power plants and can generate fault cloud maps of photovoltaic power plant areas based on the fault conditions of photovoltaic units in different regions. The intelligent inspection path optimization system can acquire data information from the 3D modeling system, generate a closed-loop inspection path, and accurately reach the inspection destination based on positioning and navigation to generate inspection tasks. The drone module is used to perform inspection tasks of the intelligent inspection path optimization system and generate real-time images of photovoltaic modules; The real-time fault and defect detection system includes a photovoltaic module fault database and stores historical fault records of photovoltaic power station modules. After acquiring images of the photovoltaic modules and identifying the fault and defect types, the information is fed back to the dynamic optimization intelligent decision-making system. The dynamic optimization intelligent decision-making system generates a health index for each area of ​​the photovoltaic system based on the information transmitted by the real-time fault and defect detection system. The health index and fault and defect information are then transmitted to the 3D modeling system for marking and recording, forming an intelligent inspection closed loop to achieve intelligent optimization of photovoltaic inspection.

2. An inspection method for the unmanned aerial vehicle (UAV) photovoltaic intelligent inspection system as described in claim 1, characterized in that, Includes the following steps: S1: The three-dimensional modeling system is used to collect data on the entire photovoltaic field area, obtain its geographical coordinates and array arrangement information, construct a three-dimensional model, and generate a photovoltaic field area fault cloud map based on the health index and fault conditions of different areas. S2: Based on the optimal path principle, the intelligent inspection path optimization system comprehensively considers the drone's endurance and the photovoltaic power station's positioning information to formulate a closed-loop inspection path from the take-off point to the key area and then to the return point, while also taking into account the necessary inspection of other areas. The planned path is then sent to the drone control system for execution according to the inspection path. S3: The UAV module includes the UAV body, power system, control system, infrared camera and transmission system. After receiving the instruction from the intelligent inspection path optimization system, the UAV begins inspection. After reaching the designated area, it takes pictures and transmits the pictures to the real-time fault and defect detection system in real time through the transmission system. S4: The real-time fault and defect detection system will process the received images and match them with its database to determine the fault and defect type of the photovoltaic module, and feed the results back to the dynamic optimization intelligent decision-making system. S5: After receiving information transmitted by the real-time fault and defect detection system, the dynamic optimization intelligent decision-making system makes a decision based on the fault level and priority, calculates the regional health index, and sends it to the three-dimensional modeling system to update the fault distribution of the photovoltaic field. On the other hand, it adjusts and optimizes the inspection cycle and key areas based on the photovoltaic regional health index and feeds it back to the intelligent inspection path optimization system to execute the next inspection task, forming a new inspection closed loop and realizing intelligent inspection of the photovoltaic field.

3. The inspection method of the UAV photovoltaic intelligent inspection system as described in claim 2, characterized in that: The 3D modeling system can update the fault cloud map of the photovoltaic field in a timely manner based on the health index and fault status of the photovoltaic area, and urge the operation and maintenance personnel to rectify and repair in a timely manner; the drone module can realize solar charging on the take-off and landing platform, and the take-off and landing platform can be set in key areas according to the photovoltaic station situation.

4. The inspection method of the UAV photovoltaic intelligent inspection system as described in claim 2, characterized in that: The intelligent inspection path optimization system receives data from the 3D modeling system to obtain the location information of the entire photovoltaic power station. It can also receive the inspection cycle and key areas fed back by the dynamic optimization intelligent decision-making system to plan the path.

5. The inspection method of the UAV photovoltaic intelligent inspection system as described in claim 2, characterized in that: The intelligent inspection path optimization system can not only plan inspection paths for key areas, but also plan inspection paths for other areas according to the inspection cycle, and can assign appropriate drones for inspection based on the planned path and the drone's endurance.

6. The inspection method of the UAV photovoltaic intelligent inspection system as described in claim 2, characterized in that: The real-time fault and defect detection system can classify faults and defects according to their levels so that they can be processed according to their priority. It can also expand the fault and defect database with images taken by drones to further improve accuracy.

7. The inspection method of the UAV photovoltaic intelligent inspection system as described in claim 2, characterized in that: The dynamic optimization intelligent decision-making system can not only make decisions about faults and defects, but also optimize the next inspection based on the defects and key area information of the photovoltaic field. At the same time, it can formulate the corresponding weight of faults according to the fault level in order to calculate the regional health index of the photovoltaic field.

8. The inspection method of the unmanned aerial vehicle photovoltaic intelligent inspection system as described in claim 7, characterized in that: The regional health index of the photovoltaic field is determined by the number of faulty modules in the area and the severity of the fault, denoted as S0, S1, ..., S... i ,...,S 10 The total number of regional components constitutes the regional health index, and the formula for calculating the regional health index is as follows: Where H is the regional health index of the photovoltaic field; N is the number of modules that failed in the region; and M is the total number of modules in the region. This indicates the failure rate of the component.

9. The inspection method of the unmanned aerial vehicle photovoltaic intelligent inspection system as described in claim 2, characterized in that: The T i Let T0 be the intelligent optimized inspection cycle for the i-th inspection in the photovoltaic intelligent inspection system. The basic inspection cycle T0 is the preset standard inspection time interval in the photovoltaic intelligent inspection system, representing the cycle for a comprehensive inspection of the photovoltaic area under ideal conditions. T1 is the inspection cycle corresponding to the first inspection, i.e., the initial inspection cycle T1 = T0. The optimized inspection cycle for this area is expressed as: Where α is the adjustment coefficient, 0 < α < 1, representing the weight of the health index on the inspection cycle.

10. The inspection method of the unmanned aerial vehicle photovoltaic intelligent inspection system as described in claim 2, characterized in that: If the photovoltaic area continues to experience a fault, the inspection cycle will be gradually shortened and the inspection frequency will be increased. If there is no fault after the fault is repaired and there is no fault in the next inspection cycle, the inspection cycle will be gradually extended and the inspection frequency will be reduced. This achieves intelligent optimization of the inspection cycle based on the photovoltaic area health index, which can both focus on key areas for inspection and reduce the inspection frequency of fault-free areas, thereby reducing inspection costs and improving inspection efficiency.

Images