Photovoltaic module hot spot diagnosis and positioning method

Abstract

The invention relates to a photovoltaic module hot spot diagnosis and positioning method. The method comprises the following steps: planning a flight path of an unmanned aerial vehicle; acquiring the string number of the photovoltaic module and the position information of the string center point, and associating the string number and the position information with the geographical position information; collecting a visible light image and an infrared thermal imaging image of the photovoltaic module; processing the visible light image and the infrared thermal imaging image through a machine vision technology, and judging whether the photovoltaic module has a hot spot fault or not; positioning the position of the photovoltaic module with the hot spot fault through an image segmentation technology; the unmanned aerial vehicle sends the position of the photovoltaic module with the hot spot fault to the mobile terminal; and the mobile terminal navigates and marks the position of the photovoltaic module with the hot spot fault in the mobile terminal by combining the map, the three-dimensional model of the photovoltaic power station and the AR module. According to the method, the hot spots of the photovoltaic module can be automatically diagnosed and positioned, the accuracy is high, the operation and maintenance efficiency is improved, and the method belongs to the field of photovoltaic technologies.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic power generation technology, and specifically to a method for diagnosing and locating hot spots in photovoltaic modules. Background Technology

[0002] With the rapid development of photovoltaic (PV) power generation technology, the scale of PV power plants is expanding daily, making efficient operation and maintenance management a crucial link in ensuring PV power generation efficiency and system stability. Hot spot effect in PV modules is a common and serious failure mode, usually caused by partial shading, internal defects, or aging. It not only leads to a decrease in module power generation efficiency but can also cause module burnout, jeopardizing the safe operation of the power plant. Therefore, rapid diagnosis and accurate location of hot spot faults in PV modules are of great significance.

[0003] Drones, due to their flexibility and efficiency, have been widely used in the inspection of photovoltaic power plants. Equipped with infrared thermal imaging equipment, drones can collect real-time thermal imaging data of photovoltaic modules during flight and mark the location of faults on the power plant's layout map. However, existing fault diagnosis systems still have significant shortcomings in practical applications. For example, after maintenance personnel arrive at the designated location based on the marking information provided by the drone inspection system, it is difficult to visually identify the specific faulty module due to the high similarity in appearance of photovoltaic modules. This inaccurate positioning not only prolongs fault diagnosis time but also increases the workload of maintenance personnel and reduces overall efficiency.

[0004] Another shortcoming of existing technologies is the lack of effective guidance for maintenance personnel in troubleshooting. While drones can locate the general area of ​​a fault, the specific identification and location of components still rely on manual judgment, which may lead to misjudgments or omissions. In addition, current systems for hot spot fault diagnosis typically only involve simple thermal imaging processing and fail to conduct deeper analysis in conjunction with the actual on-site environment, making it difficult to meet the needs of refined operation and maintenance of photovoltaic power plants. Summary of the Invention

[0005] To address the technical problems existing in the prior art, the purpose of this invention is to provide a method for diagnosing and locating hot spots in photovoltaic modules, which can automatically diagnose and locate hot spots in photovoltaic modules with high accuracy and improve operation and maintenance efficiency.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] A method for diagnosing and locating hot spots in photovoltaic modules includes the following steps:

[0008] Plan the drone flight path based on the layout information of the photovoltaic power station, so that the drone flight path covers the entire photovoltaic power station area;

[0009] Obtain the string number and the location information of the center point of the photovoltaic module in the photovoltaic power station;

[0010] Associate the string number of the photovoltaic module with its geographical location information;

[0011] The drone flies according to the planned flight path, and collects visible light images and infrared thermal images of the photovoltaic modules during the flight.

[0012] Machine vision technology is used to process visible light images and infrared thermal imaging images to determine whether there are hot spot faults in photovoltaic modules.

[0013] The location of photovoltaic modules with hot spot faults was located using image segmentation technology;

[0014] The drone sends the location of photovoltaic modules with hot spot faults to a mobile device.

[0015] The mobile app combines maps, 3D models of photovoltaic power plants, and AR modules to navigate and mark the locations of photovoltaic modules with hot spot faults.

[0016] Preferably, the method for planning the flight path of the drone is as follows: based on the overall layout information of the photovoltaic power station, the photovoltaic power station area is divided into zones using GIS technology, and combined with the arrangement of the photovoltaic modules, the location information of the string center point and the string number of each photovoltaic module are generated; wherein, the location information of the string center point of each photovoltaic module includes latitude and longitude coordinates, and the relative position between the string center point of each photovoltaic module and the geographical reference point of the photovoltaic power station;

[0017] By using the string number, the latitude and longitude coordinates of the string center point, and the relative position information between the string center point of each photovoltaic module and the geographical reference point of the photovoltaic power station, the flight path of the drone is generated, enabling the drone to fly sequentially to all the photovoltaic module strings according to the path, covering the entire photovoltaic power station area.

[0018] Preferably, the method for associating the string number of photovoltaic modules with geographical location information is as follows:

[0019] GNSS receiving equipment is used to collect geographic coordinates at the boundaries and key nodes of the photovoltaic power station to construct an in-station coordinate reference system;

[0020] The photovoltaic module string arrangement information in the photovoltaic power plant design drawings is matched with the GNSS data collected on site to generate a digital map of the actual photovoltaic module string distribution.

[0021] The center point of each photovoltaic module string is marked on a digital map, and the string number is bound to the location information for storage, forming a unique mapping relationship between the number and the location.

[0022] Preferably, the method for acquiring visible light images and infrared thermal images of photovoltaic modules is as follows:

[0023] Based on the center point of the photovoltaic module string and the arrangement information of the photovoltaic modules, the waypoint and altitude of the drone are set so that the camera range of the drone covers the entire photovoltaic module string.

[0024] By adjusting the drone's flight altitude and camera angle according to the environmental conditions of the photovoltaic power station, the captured images are clear and without any missed shots.

[0025] The drone is equipped with a visible light camera and an infrared thermal imaging camera to collect visible light images and thermal imaging images in real time.

[0026] Set the image acquisition frequency and resolution parameters so that the pixel density of each photovoltaic module in the string image can be diagnosed;

[0027] The drone flies over each string of photovoltaic modules, follows a set flight path, flies at a constant speed, and collects images.

[0028] After completing the image acquisition of the photovoltaic module string, it automatically switches to the next waypoint until the entire photovoltaic power station area is covered.

[0029] Preferably, the method for processing visible light images and infrared thermal imaging images using machine vision technology is as follows:

[0030] Transmit images captured by the drone to the ground station;

[0031] The acquired visible light and thermal imaging images are calibrated and registered;

[0032] Denoising and enhancement algorithms are applied to improve image quality and optimize the visibility of hotspot features;

[0033] Detect temperature anomaly regions in infrared thermal imaging images and extract relevant features;

[0034] By combining the texture information of the visible light image, non-faulty areas are eliminated.

[0035] Preferably, the method for determining whether a photovoltaic module has a hot spot fault is as follows: a deep learning algorithm is used to detect hot spots in the acquired infrared thermal imaging image, and false hot spots are removed by combining the texture information of the visible light image, and different types of hot spot faults are classified.

[0036] Preferably, hot spot failure types include blocked hot spots, open-circuit hot spots, and aging hot spots.

[0037] Preferably, the method for locating the photovoltaic module with hot spot faults using image segmentation technology is as follows:

[0038] Based on the string arrangement information of photovoltaic modules, the spatial coordinates of hot spot faults are converted into row and column coordinates within the string of photovoltaic modules;

[0039] Photovoltaic modules are segmented using image segmentation technology to identify the module numbers with hot spot faults;

[0040] Mark faulty modules in the string image of photovoltaic modules and add information on hot spot location and fault type;

[0041] The marked images are used to generate fault reports, which record information including string number, row and column coordinates, fault type, and hot spot characteristics.

[0042] Preferred methods for integrating maps into mobile devices:

[0043] Load the plan layout of the photovoltaic power station onto the mobile device, and mark the photovoltaic module string number and distribution location on the plan layout of the photovoltaic power station;

[0044] Methods for integrating 3D models into mobile devices:

[0045] Based on the design data and scanning information of the photovoltaic power station, a 3D model of the photovoltaic power station is created and uploaded to the cloud.

[0046] The 3D model of the photovoltaic power station includes the arrangement of photovoltaic modules and the terrain where the photovoltaic modules are located;

[0047] By accessing the 3D model of a photovoltaic power station from the cloud via a mobile device, users can rotate and zoom the 3D model of the photovoltaic power station through touch or gesture interaction on the mobile device, view the spatial location of faulty photovoltaic modules, display the faulty modules in the 3D model of the photovoltaic power station, and label their number, row and column coordinates, and fault type information.

[0048] Methods for integrating AR modules into mobile devices:

[0049] On-site, a mobile camera is used to scan the photovoltaic module strings and identify their locations.

[0050] Images of faulty components captured by drones are overlaid onto real-time footage, and virtual markers are displayed at the component locations.

[0051] Clicking the virtual marker displays detailed information such as the component's fault type, abnormal temperature range, and fault number.

[0052] Preferably, the following steps are also included:

[0053] By storing the design data and actual survey information of the photovoltaic power station in the cloud and updating the operating status and equipment change information of the photovoltaic power station in sync, the photovoltaic power station model is consistent with the actual site.

[0054] The cloud and mobile devices are connected, enabling mobile devices to access photovoltaic power plant model data;

[0055] The visible light images and infrared thermal images of photovoltaic modules collected by drones, as well as information on whether there are hot spot faults in the photovoltaic modules, are uploaded to the cloud to form a fault file and store maintenance records.

[0056] By using big data analytics to predict component failure trends, we can optimize the operation and maintenance plans for photovoltaic power plants.

[0057] In summary, the present invention has the following advantages:

[0058] The photovoltaic module hot spot diagnosis and location method of the present invention associates the string number and the location information of the string center point of the photovoltaic module in the photovoltaic power station with the geographic location. It uses a drone to collect visible light images and infrared thermal images of the photovoltaic module, and uses machine vision technology and image segmentation technology to analyze and locate the photovoltaic module with hot spot faults. The operation and maintenance personnel can receive the location information of the photovoltaic module with hot spot faults through a mobile terminal, and can accurately navigate to the corresponding location for maintenance. That is, the method of this application not only improves the accuracy of fault diagnosis, but also provides operation and maintenance personnel with intuitive guidance for identifying and eliminating faulty components, significantly improving the operation and maintenance efficiency of photovoltaic power stations. Attached Figure Description

[0059] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0060] Figure 1 Flowchart of a method for diagnosing and locating hot spots in photovoltaic modules;

[0061] Figure 2 Illustrations for image acquisition and fault diagnosis of component drones;

[0062] Figure 3 Visualize the fault guidance diagram. Detailed Implementation

[0063] Various exemplary embodiments of the present invention will now be described in detail. This detailed description should not be considered as a limitation of the present invention, but rather as a more detailed description of certain aspects, features, and embodiments of the present invention.

[0064] It should be understood that the terminology used in this invention is merely for describing particular embodiments and is not intended to limit the invention. Furthermore, with respect to numerical ranges in this invention, it should be understood that each intermediate value between the upper and lower limits of the range is also specifically disclosed. Every smaller range between any stated value or intermediate value within a stated range, and any other stated value or intermediate value within said range, is also included in this invention. The upper and lower limits of these smaller ranges may be independently included or excluded from the range.

[0065] Unless otherwise stated, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art. While only preferred methods and materials have been described herein, any methods and materials similar or equivalent to those described herein may be used in the implementation or testing of this invention. All references to this specification are incorporated by way of citation to disclose and describe methods and / or materials associated with those references. In the event of any conflict with any incorporated reference, the content of this specification shall prevail.

[0066] Various modifications and variations can be made to the specific embodiments described in this specification without departing from the scope or spirit of the invention, as will be apparent to those skilled in the art. Other embodiments derived from this specification will also be readily apparent to those skilled in the art. This specification and embodiments are merely exemplary.

[0067] The terms “include,” “including,” “have,” “contain,” etc., used in this article are all open-ended terms, meaning that they include but are not limited to.

[0068] The technical solution of the present invention will be further illustrated by the following embodiments.

[0069] Example 1

[0070] This embodiment provides a method for diagnosing and locating hot spots in photovoltaic modules, including the following steps:

[0071] (1) Information collection of the target area;

[0072] Target area information collection includes UAV flight path planning, string position information collection, and photovoltaic power station model matching. High-precision spatial data acquisition and calibration technology provides accurate basic data support for subsequent UAV inspections and fault diagnosis. The specific implementation plan is as follows:

[0073] 1. Unmanned aerial vehicle (UAV) pre-flight path planning;

[0074] Before the drone-based formal inspection, the power station area is divided into zones based on the overall layout information of the photovoltaic power station using Geographic Information System (GIS) technology. Combined with the arrangement of the photovoltaic modules, the center point location information and number of each photovoltaic module string are generated. This information includes:

[0075] The number of each string group (e.g., G1, G2...Gn);

[0076] Latitude and longitude coordinates of the center point of the string;

[0077] The relative position information between each string and the power plant's geographical reference point.

[0078] The path planning software automatically generates flight paths for the drones based on this data, ensuring that the drones fly to all target strings sequentially along the optimal path, covering the entire power plant area. The planning process must consider factors such as the terrain of the photovoltaic power plant, the drone's flight altitude, obstacle avoidance requirements, and weather conditions to ensure the safety of the drone flight and the integrity of the data collection.

[0079] 2. Marking the center point and number of the string;

[0080] Utilizing existing design drawings and layout models of the photovoltaic power station, and through high-precision GPS positioning technology and on-site geographic coordinate system calibration, the geographical location of each string of photovoltaic cells is accurately determined. The implementation steps are as follows:

[0081] High-precision GNSS receiving equipment is used to collect geographic coordinates at the boundaries and key nodes of the photovoltaic power station to construct an in-station coordinate reference system;

[0082] The string arrangement information in the photovoltaic power station design drawings is matched with the GNSS data collected on site to generate a digital map of the actual string distribution.

[0083] The center point of each string is marked, and the string number is bound to the position information for storage, forming a unique mapping relationship between the number and the position.

[0084] 3. Data verification and consistency confirmation;

[0085] To ensure the accuracy of the collected data, the following methods are used for verification:

[0086] Repeated positioning: Perform multiple GPS positioning operations on key sequences and compare the results to ensure that the error of the sequence center point is within the allowable range;

[0087] Drawing comparison: Compare the generated string distribution map with the design drawings of the photovoltaic power station to confirm that the string numbering is consistent with the actual layout;

[0088] UAV test flight calibration: The UAV flies once according to the preliminary path plan, collects real-time images of each string, compares them with the planning data, and calibrates any possible deviations.

[0089] 4. Data storage is matched with the photovoltaic power plant model;

[0090] The collected string location information is uploaded to the cloud platform (cloud) via a data interface and matched with the power plant's 3D digital model to complete the string model annotation. The annotation content includes:

[0091] String number, center point coordinates;

[0092] The relative positional relationship with surrounding strings;

[0093] Information on the number and arrangement of photovoltaic modules contained in a string.

[0094] In this way, each string and component of the photovoltaic power station has accurate location information and number, forming a digital twin model of the power station, providing high-precision data support for drone fault diagnosis and subsequent operation and maintenance.

[0095] (2) Component-based UAV image acquisition and fault diagnosis;

[0096] By utilizing drone flight path planning and image acquisition, machine vision technology for diagnosis, and fault location and marking, accurate identification and location of hot spot faults in photovoltaic modules can be achieved. The specific implementation plan is as follows:

[0097] 1. Drone flight path planning and image acquisition;

[0098] The drones, based on waypoint planning within the target area (the area designated by the photovoltaic power station, i.e., the area where the drones will fly), fly sequentially to the predetermined string positions and collect visible light and infrared thermal images of the photovoltaic modules. The implementation steps include:

[0099] Flight route planning and parameter setting: Based on the center point location and component arrangement information of the string, set the waypoints and altitude of the drone flight to ensure that the entire string is covered within the drone's camera range.

[0100] By taking into account the environmental characteristics of the photovoltaic power station (such as terrain and module tilt angle), the flight altitude and camera angle of the drone are adjusted to ensure that the captured images are clear and that no images are missed.

[0101] Image acquisition equipment configuration: The drone is equipped with a high-resolution visible light camera and an infrared thermal imaging camera to acquire visible light and thermal images in real time.

[0102] Configure the image acquisition frequency and resolution parameters to ensure that each component has sufficient pixel density in the image string to support subsequent diagnosis.

[0103] Data Acquisition Process: The drone automatically flies over each group of images, follows a set path, flies at a constant speed, and acquires images.

[0104] After completing the image acquisition of the target string, it automatically moves to the next waypoint until all target areas are covered.

[0105] 2. Image processing and fault type identification;

[0106] The acquired images are transmitted to a ground station or the cloud for automated processing using deep learning algorithms based on machine vision. The specific process is as follows:

[0107] Image preprocessing: The acquired visible light and thermal imaging images are calibrated and registered to ensure their spatial consistency.

[0108] Denoising and enhancement algorithms are applied to improve image quality and optimize the visibility of hotspot features.

[0109] Hot spot detection and feature extraction: Detecting temperature anomaly regions (hot spots) in infrared thermal imaging images and extracting relevant features (such as area, shape, and temperature value).

[0110] By combining the texture information of the visible light image, possible non-faulty areas (such as false hot spots caused by specular reflection) are eliminated.

[0111] Fault type classification: Based on trained deep learning models (such as convolutional neural networks CNN), the detected hot spots are classified according to their types.

[0112] The classification results include common hot spots caused by shielding, hot spots caused by open circuits, and hot spots caused by aging, supporting more accurate analysis of the causes of failures.

[0113] 3. Fault location and component marking;

[0114] Based on string layout information and image segmentation technology, the detected hot spot faults are accurately located. The implementation process is as follows:

[0115] Fault location: Based on the string arrangement information, the spatial coordinates of the hot spot fault are converted into row and column coordinates within the string.

[0116] Machine vision technology is used to segment photovoltaic modules and determine the specific faulty module number.

[0117] Digital labeling: Faulty components are prominently marked in the string image, and hot spot location and fault type information are added.

[0118] The marked images are used to generate fault reports, which record relevant information, including string number, row and column coordinates, fault type, and hot spot characteristics.

[0119] 4. Data storage and subsequent support;

[0120] The processed fault information is transmitted to cloud storage via the network, linked to the power plant's digital model, and used as a basis for subsequent operation and maintenance work. Cloud storage includes: original images of components and diagnostic results; generated visual reports for operation and maintenance personnel to access; and historical diagnostic records for big data analysis and preventative maintenance.

[0121] (3) Visualized fault guidance;

[0122] To achieve intuitive location and precise guidance of faulty components, this embodiment features a mobile application that combines map navigation, 3D model display, and AR guidance. The implementation plan is as follows:

[0123] 1. Large map navigation module;

[0124] Photovoltaic power plant layout shown:

[0125] Load the power plant's floor plan (2D map) into the mobile application, and mark the string numbers and their distribution locations.

[0126] The location of the faulty group is dynamically displayed, and the row and column coordinates and number of the faulty component are marked.

[0127] Navigation function:

[0128] It provides navigation route planning functionality to guide maintenance personnel from their current location to the specific location of the faulty cluster.

[0129] It supports real-time location updates, using GPS to pinpoint the current location of maintenance personnel and dynamically adjust navigation routes.

[0130] 2. 3D model display module;

[0131] 3D model creation: Using the design data and on-site scanning information of the photovoltaic power station, a 3D model of the power station is created in advance and uploaded to the cloud for mobile devices to access.

[0132] The model includes component arrangement, terrain, and equipment connection relationships.

[0133] Model interaction function: Operation and maintenance personnel can use touch or gestures to rotate and zoom the 3D model in the mobile application to view the specific spatial location of the faulty component.

[0134] The faulty components are dynamically highlighted in the model, and their numbers, row and column coordinates, and fault type information are labeled.

[0135] 3. AR guidance module;

[0136] Augmented reality tagging: Maintenance personnel use the camera of their mobile devices to scan target strings on-site, and the application automatically identifies the location of the strings.

[0137] Images of faulty components captured by drones are overlaid onto the real-time view, and virtual markers (such as red boxes or arrows) are displayed at the component locations.

[0138] Dynamic fault information display: Clicking the virtual marker will display detailed information such as the fault type, abnormal temperature range, and fault number of the component.

[0139] It provides an interactive troubleshooting record function, which makes it easy for maintenance personnel to update the repair progress in real time.

[0140] (4) Cloud technology application and data management (cloud);

[0141] The introduction of cloud technology has enabled centralized management and information-based traceability of operation and maintenance data for photovoltaic power plants. The specific implementation plan is as follows:

[0142] 1. Storage of photovoltaic power plant modeling information;

[0143] Power plant data upload: Store the power plant's design data (such as string number, component arrangement, and equipment connection relationship) and actual surveying information to the cloud platform.

[0144] Regularly synchronize and update the power plant's operating status and equipment change information to ensure that the model is consistent with the actual site.

[0145] Global access and real-time updates: Supports accessing modeling data anytime on mobile and PC devices.

[0146] Data is updated in real time to ensure the accuracy of navigation and guidance information.

[0147] 2. Fault information management;

[0148] Fault data storage: Fault results from UAV diagnostics, including fault component number, location, type, and diagnosis time, are uploaded to the cloud in real time to form a fault file.

[0149] Fault records can be retrieved by time, component number, or fault type, facilitating tracing and statistical analysis.

[0150] Visualization: Cloud-based fault statistics charts (such as distribution by type, time trend, etc.) provide intuitive decision support for operations and maintenance personnel.

[0151] 3. Track maintenance records;

[0152] Maintenance data recording: Maintenance personnel record the fault handling process through a mobile application, including repair methods, time taken, and materials used.

[0153] Recorded data is uploaded to the cloud in real time, forming a closed-loop management system with fault information.

[0154] Standardized maintenance: Based on historical data analysis, standardized operation and maintenance processes are formed to optimize resource allocation.

[0155] 4. Data mining and intelligent analysis;

[0156] Big data analytics: Long-term stored operation and maintenance data is used to uncover fault patterns through machine learning algorithms to optimize maintenance strategies.

[0157] Analyze the temporal and spatial distribution patterns of hot spot failures, predict high-risk areas and components, and carry out preventative maintenance.

[0158] Intelligent reminders and optimization: The system automatically generates maintenance plans and sends reminders based on fault prediction results.

[0159] Provide optimization suggestions for power plant operation status to improve power plant power generation efficiency and operation and maintenance efficiency.

[0160] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments. Any changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be considered equivalent substitutions and shall be included within the protection scope of the present invention.

Claims

1. A method for diagnosing and locating hot spots in photovoltaic modules, characterized in that, Includes the following steps: Plan the drone flight path based on the layout information of the photovoltaic power station, so that the drone flight path covers the entire photovoltaic power station area; Obtain the string number and the location information of the center point of the photovoltaic module in the photovoltaic power station; Associate the string number of the photovoltaic module with its geographical location information; The drone flies according to the planned flight path, and collects visible light images and infrared thermal images of the photovoltaic modules during the flight. Machine vision technology is used to process visible light images and infrared thermal imaging images to determine whether there are hot spot faults in photovoltaic modules. The location of photovoltaic modules with hot spot faults was located using image segmentation technology; The drone sends the location of photovoltaic modules with hot spot faults to a mobile device. The mobile app combines maps, 3D models of photovoltaic power plants, and AR modules to navigate and mark the locations of photovoltaic modules with hot spot faults.

2. The method for diagnosing and locating hot spots in a photovoltaic module according to claim 1, characterized in that, The method for planning the flight path of drones is as follows: Based on the overall layout information of the photovoltaic power station, the photovoltaic power station area is divided into zones using GIS technology. Combined with the arrangement of photovoltaic modules, the location information of the string center point and the string number of each photovoltaic module are generated. Among them, the location information of the string center point of each photovoltaic module includes latitude and longitude coordinates and the relative position between the string center point of each photovoltaic module and the geographical reference point of the photovoltaic power station. By using the string number, the latitude and longitude coordinates of the string center point, and the relative position information between the string center point of each photovoltaic module and the geographical reference point of the photovoltaic power station, the flight path of the drone is generated, enabling the drone to fly sequentially to all the photovoltaic module strings according to the path, covering the entire photovoltaic power station area.

3. The method for diagnosing and locating hot spots in a photovoltaic module according to claim 1, characterized in that, The method for associating the string number of photovoltaic modules with geographical location information is as follows: GNSS receiving equipment is used to collect geographic coordinates at the boundaries and key nodes of the photovoltaic power station to construct an in-station coordinate reference system; The photovoltaic module string arrangement information in the photovoltaic power plant design drawings is matched with the GNSS data collected on site to generate a digital map of the actual photovoltaic module string distribution. The center point of each photovoltaic module string is marked on a digital map, and the string number is bound to the location information for storage, forming a unique mapping relationship between the number and the location.

4. The method for diagnosing and locating hot spots in a photovoltaic module according to claim 1, characterized in that, The specific methods for acquiring visible light images and infrared thermal images of photovoltaic modules are as follows: Based on the center point of the photovoltaic module string and the arrangement information of the photovoltaic modules, the waypoint and altitude of the drone are set so that the camera range of the drone covers the entire photovoltaic module string. By adjusting the drone's flight altitude and camera angle according to the environmental conditions of the photovoltaic power station, the captured images are clear and without any missed shots. The drone is equipped with a visible light camera and an infrared thermal imaging camera to collect visible light images and thermal imaging images in real time. Set the image acquisition frequency and resolution parameters so that the pixel density of each photovoltaic module in the string image can be diagnosed; The drone flies over each string of photovoltaic modules, follows a set flight path, flies at a constant speed, and collects images. After completing the image acquisition of the photovoltaic module string, it automatically switches to the next waypoint until the entire photovoltaic power station area is covered.

5. A method for diagnosing and locating hot spots in a photovoltaic module according to claim 4, characterized in that, The method for processing visible light images and infrared thermal imaging images using machine vision technology is as follows: Transmit images captured by the drone to the ground station; The acquired visible light and thermal imaging images are calibrated and registered; Denoising and enhancement algorithms are applied to improve image quality and optimize the visibility of hotspot features; Detect temperature anomaly regions in infrared thermal imaging images and extract relevant features; By combining the texture information of the visible light image, non-faulty areas are eliminated.

6. A method for diagnosing and locating hot spots in a photovoltaic module according to claim 1, characterized in that, The method for determining whether a photovoltaic module has a hot spot fault is as follows: a deep learning algorithm is used to detect hot spots in the acquired infrared thermal imaging images, and false hot spots are removed by combining the texture information of the visible light images. Different types of hot spot faults are then classified.

7. A method for diagnosing and locating hot spots in a photovoltaic module according to claim 6, characterized in that: Hot spot failure types include blocked hot spots, open-circuit hot spots, and aging hot spots.

8. A method for diagnosing and locating hot spots in a photovoltaic module according to claim 1, characterized in that, The method for locating photovoltaic modules with hot spot faults using image segmentation technology is as follows: Based on the string arrangement information of photovoltaic modules, the spatial coordinates of hot spot faults are converted into row and column coordinates within the string of photovoltaic modules; Photovoltaic modules are segmented using image segmentation technology to identify the module numbers with hot spot faults; Mark faulty modules in the string image of photovoltaic modules and add information on hot spot location and fault type; The marked images are used to generate fault reports, which record information including string number, row and column coordinates, fault type, and hot spot characteristics.

9. A method for diagnosing and locating hot spots in a photovoltaic module according to claim 1, characterized in that: Methods for integrating maps into mobile devices: Load the plan layout of the photovoltaic power station onto the mobile device, and mark the photovoltaic module string number and distribution location on the plan layout of the photovoltaic power station; Methods for integrating 3D models into mobile devices: Based on the design data and scanning information of the photovoltaic power station, a 3D model of the photovoltaic power station is created and uploaded to the cloud. The 3D model of the photovoltaic power station includes the arrangement of photovoltaic modules and the terrain where the photovoltaic modules are located; By accessing the 3D model of a photovoltaic power station from the cloud via a mobile device, users can rotate and zoom the 3D model of the photovoltaic power station through touch or gesture interaction on the mobile device, view the spatial location of faulty photovoltaic modules, display the faulty modules in the 3D model of the photovoltaic power station, and label their number, row and column coordinates, and fault type information. Methods for integrating AR modules into mobile devices: On-site, a mobile camera is used to scan the photovoltaic module strings and identify their locations. Images of faulty components captured by drones are overlaid onto real-time footage, and virtual markers are displayed at the component locations. Clicking the virtual marker displays detailed information such as the component's fault type, abnormal temperature range, and fault number.

10. A method for diagnosing and locating hot spots in a photovoltaic module according to claim 1, characterized in that: It also includes the following steps: By storing the design data and actual survey information of the photovoltaic power station in the cloud and updating the operating status and equipment change information of the photovoltaic power station in sync, the photovoltaic power station model is consistent with the actual site. The cloud and mobile devices are connected, enabling mobile devices to access photovoltaic power plant model data; The visible light images and infrared thermal images of photovoltaic modules collected by drones, as well as information on whether there are hot spot faults in the photovoltaic modules, are uploaded to the cloud to form a fault file and store maintenance records. By using big data analytics to predict component failure trends, we can optimize the operation and maintenance plans for photovoltaic power plants.

Images