Defect detection method for photovoltaic power station and related device
By combining infrared thermal imaging image preprocessing and convolutional neural networks with latitude and longitude data, the problem of low defect detection accuracy in photovoltaic power plants has been solved, achieving efficient defect detection and location positioning, and improving the efficiency of photovoltaic power plant operation and maintenance management.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- 华能(临高)新能源有限公司
- Filing Date
- 2024-11-22
- Publication Date
- 2026-05-22
AI Technical Summary
Existing technologies for detecting defects in photovoltaic power plants have low accuracy, and manual on-site confirmation is required after drone inspections, which is inefficient and costly.
By employing infrared thermal imaging image preprocessing, key feature extraction, and a convolutional neural network model, combined with latitude and longitude data, defect detection and location can be achieved.
It improves the accuracy and efficiency of defect detection, reduces manual intervention, is applicable to the detection of static and dynamic data, and enhances the operation and maintenance management level of photovoltaic power plants.
Smart Images
Figure CN122072955A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of photovoltaic power plant inspection technology, specifically relating to a defect detection method and related device for photovoltaic power plants. Background Technology
[0002] With the continuous growth of global demand for renewable energy, the installed capacity of photovoltaic (PV) power plants is rapidly expanding, highlighting the increasing importance of their operation and maintenance (O&M) management. Since PV power plants are mostly built in harsh environments such as deserts, grasslands, and Gobi, PV modules are exposed to these conditions for extended periods, making them prone to defects such as hot spots, diode failures, and inefficient operation. These problems not only reduce the power generation efficiency of PV power plants but may also pose safety hazards. Traditional PV power plant O&M primarily relies on manual inspections, requiring maintenance personnel to use handheld testing equipment to check each PV module individually. However, this method is not only inefficient and unable to meet the O&M needs of large-scale PV power plants, but also costly and requires highly skilled O&M personnel.
[0003] In recent years, the rapid development of drone technology has provided new solutions to the challenges of photovoltaic power plant operation and maintenance. Drones, with their small size, high degree of automation, and ability to carry various sensors, have become powerful tools for photovoltaic power plant operation and maintenance management. By equipping drones with infrared thermal imaging cameras, rapid detection of defects in photovoltaic modules can be achieved. However, despite the enormous potential of drone inspection technology in photovoltaic power plant operation and maintenance, it still faces many challenges. On the one hand, the geographical environment of photovoltaic power plants is complex and diverse. Different lighting conditions and complex backgrounds generate additional noise in infrared thermal imaging images, leading to a higher probability of misidentification in pre-trained general target detection models. On the other hand, due to the small pixel distance between adjacent photovoltaic modules and the delay in aerial flight data, the accuracy of the latitude and longitude coordinates of defective modules returned by drones is limited. Maintenance personnel still need to conduct on-site inspections using handheld latitude and longitude positioning devices, consuming significant manpower and time costs.
[0004] In practice, the identification of photovoltaic panels through infrared photography often relies on human visual recognition or traditional, simple image recognition algorithms, such as feature engineering and frequency domain processing. However, both of these approaches suffer from low accuracy and high workload, making it difficult to achieve ideal prediction results. Therefore, in image detection, it is necessary to select appropriate models and methods to detect images transmitted back by drones. Summary of the Invention
[0005] The purpose of this invention is to provide a defect detection method and related apparatus for photovoltaic power plants, so as to solve the problem of low accuracy in the existing technology for defect identification of photovoltaic modules.
[0006] To achieve the above objectives, the present invention adopts the following technical solution: Firstly, a defect detection method for photovoltaic power plants includes the following steps: Collect infrared thermal images and latitude and longitude data of photovoltaic modules; The infrared thermal imaging image is preprocessed; Key features are extracted from the preprocessed infrared thermal imaging image, and the location information of the key features in the photovoltaic string is obtained based on the latitude and longitude data; The key features of the infrared thermal imaging image are input into the defect detection model for identification to obtain the defect detection result. Based on the location information of the key features in the photovoltaic string, the defect location information is obtained.
[0007] In some embodiments, the step of preprocessing the infrared thermal imaging image specifically includes: using a filtering algorithm to remove noise from the infrared thermal imaging image; Image enhancement is performed on the infrared thermal imaging image; The infrared thermal imaging image is then registered.
[0008] In some implementations, scale-invariant feature transformation is used to extract key features from the preprocessed infrared thermal image.
[0009] In some implementations, the key features include temperature features, texture features, and shape features.
[0010] In some implementations, the step of obtaining the location information of key features in the photovoltaic string based on the latitude and longitude data specifically includes: The corresponding number of the photovoltaic module in the photovoltaic string is obtained based on the latitude and longitude data; The semantic segmentation algorithm is used to extract the region pixels of the photovoltaic module involved in the infrared thermal imaging image; The location information of key features in the photovoltaic string is obtained based on the number and region pixels.
[0011] In some implementations, the defect detection model is trained by inputting defect infrared thermal imaging images and normal infrared thermal imaging images into a convolutional neural network.
[0012] Secondly, a defect detection system for photovoltaic power plants includes: Infrared thermal imaging acquisition module, used to acquire infrared thermal imaging images and latitude and longitude data of photovoltaic modules; An image preprocessing module is used to preprocess the infrared thermal imaging image; The feature and location acquisition module is used to extract key features from the preprocessed infrared thermal imaging image and obtain the location information of the key features in the photovoltaic string based on the latitude and longitude data. The defect detection module is used to input the key features of the infrared thermal imaging image into the defect detection model for identification, obtain the defect detection result, and obtain the defect location information based on the location information of the key features in the photovoltaic string.
[0013] Thirdly, a computer device includes a memory, a processor, and a computer program stored in the memory and executable in the processor, wherein the processor, when executing the computer program, implements the steps of the defect detection method for a photovoltaic power plant.
[0014] Fourthly, a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the defect detection method for a photovoltaic power station.
[0015] Fifthly, a computer program product comprising a computer program, characterized in that, when executed by a processor, the computer program implements the steps of the defect detection method for a photovoltaic power station.
[0016] Compared with the prior art, the present invention has the following beneficial effects: This invention provides a defect detection method for photovoltaic power plants, comprising the following steps: acquiring infrared thermal imaging images and latitude and longitude data of photovoltaic modules; preprocessing the infrared thermal imaging images; extracting key features from the preprocessed infrared thermal imaging images and obtaining the location information of the key features in the photovoltaic string based on the latitude and longitude data; inputting the key features of the infrared thermal imaging images into a defect detection model for identification, obtaining defect detection results, and obtaining defect location information based on the location information of the key features in the photovoltaic string. This method, by acquiring infrared thermal imaging images of photovoltaic modules, can intuitively reflect the temperature distribution of the photovoltaic modules, helping to promptly detect potential defects such as hot spots and fractures, solving the problem of low accuracy in defect identification of photovoltaic modules in existing technologies; combined with latitude and longitude data, the specific location information of key features in the photovoltaic string can be obtained, which is crucial for quickly locating and repairing defects. This method is not only applicable to the detection of static images, but can also be extended to the real-time detection of dynamic data such as video streams, improving the operation and maintenance management level of photovoltaic power plants.
[0017] Furthermore, the present invention preprocesses infrared thermal imaging images, such as filtering and denoising, image enhancement, and image registration, which can further improve image quality, reduce interference factors, and thus improve the accuracy of defect detection.
[0018] Furthermore, this invention employs Scale-Invariant Feature Transform (SIFT) to extract key features from images, which can capture subtle changes in photovoltaic modules and improve detection sensitivity.
[0019] Furthermore, this invention utilizes a semantic segmentation algorithm to extract regional pixels of photovoltaic modules, which can further refine the location of defects and reduce the possibility of false alarms and missed alarms.
[0020] Furthermore, this invention constructs a defect detection model by training a Convolutional Neural Network (CNN), which can realize intelligent and automated defect detection, reduce manual intervention, and improve detection efficiency. Attached Figure Description
[0021] Figure 1 A flowchart of a defect detection method for a photovoltaic power station provided in Example 1; Figure 2 This is a schematic diagram of a defect detection system for a photovoltaic power station provided in Example 2. Detailed Implementation
[0022] To enable those skilled in the art to better understand the present invention, the technical solution of the present invention will be further described in detail below with reference to the accompanying drawings. The content described herein is for explanation rather than limitation of the present invention.
[0023] It should be noted that the terms "comprising" and "having" and any variations thereof in the specification and claims of this invention are intended to cover a non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to such processes, methods, systems, products, or devices.
[0024] Example 1 This embodiment provides a defect detection method for photovoltaic power plants, including the following steps: S1, collects infrared thermal imaging images and latitude and longitude data of photovoltaic modules; This step primarily utilizes drone inspection technology to collect infrared thermal imaging images of the photovoltaic power station. The drone, equipped with a high-resolution infrared thermal imaging camera, flies along a pre-set flight path to comprehensively photograph the photovoltaic power station. Simultaneously, the drone is also equipped with a GPS positioning system, capable of recording latitude and longitude data in real time during shooting. These infrared thermal images and latitude and longitude data will be used for subsequent analysis and processing. To ensure the quality of the acquired images, the drone and infrared thermal imaging camera need to be calibrated before shooting to ensure image clarity and accuracy. Furthermore, it is necessary to set an appropriate flight altitude and speed based on the actual conditions of the photovoltaic power station to ensure that detailed thermal imaging information of the photovoltaic modules can be captured.
[0025] S2, preprocess the infrared thermal imaging image; Preprocessing is an important step in infrared thermal imaging image analysis. It can improve image quality and provide an accurate data foundation for subsequent feature extraction and defect detection.
[0026] First, filtering algorithms are used to remove noise from the infrared thermal imaging image. This noise may originate from factors such as the camera sensor and environmental interference, which can interfere with image clarity and affect subsequent analysis results. Therefore, Gaussian filtering or mean filtering is chosen to remove this noise.
[0027] Next, image enhancement is performed on the infrared thermal imaging images. Image enhancement can improve the visual effect of the images and make key information in the images stand out more. Histogram equalization and contrast stretching are used to enhance the contrast of the images, making temperature differences more obvious.
[0028] Finally, the infrared thermal imaging images are registered. Image registration is the process of aligning images captured at different times, from different viewpoints, or by different sensors. Because drones may be affected by factors such as airflow and wind direction during filming, slight shifts or rotations may occur between the captured images. Therefore, image registration technology is used to align these images to ensure the accuracy of subsequent analysis.
[0029] S3, extract the key features of the preprocessed infrared thermal imaging image, and obtain the location information of the key features in the photovoltaic string based on the latitude and longitude data; In this step, a scale-invariant feature transform algorithm is used to extract key features from the infrared thermal imaging image. These key features include temperature features, texture features, and shape features, reflecting the thermal distribution of the photovoltaic module, thereby helping to identify potential defects.
[0030] Simultaneously, the corresponding photovoltaic (PV) module's number within the PV string is obtained based on latitude and longitude data. This is achieved by matching the latitude and longitude data with the layout map of the PV power plant. Based on the PV module's number, semantic segmentation algorithms are used to extract the relevant pixel areas from the infrared thermal imaging image. Then, based on the number and the pixel areas, the location information of key features within the PV string is obtained, providing accurate location information for subsequent defect detection.
[0031] S4, input the key features of the infrared thermal imaging image into the defect detection model for identification, obtain the defect detection result, and obtain the defect location information based on the location information of the key features in the photovoltaic string; In this step, a convolutional neural network (CNN) is used to build a defect detection model. This defect detection model is trained by inputting a large number of defective infrared thermal imaging images and normal infrared thermal imaging images into the CNN. During training, the CNN automatically learns the features in the images and establishes a mapping relationship between features and defects.
[0032] CNN-based object detection algorithms fall into two main categories. One category uses a two-stage deep convolutional neural network based on candidate regions. First, it generates candidate regions that may contain the target object. Then, it classifies and regresses the location of these regions to obtain bounding boxes. The other category uses a single-stage deep convolutional neural network based on regression calculation. This method integrates target localization into a single CNN network, simultaneously predicting the target's category and location. This embodiment employs a deep convolutional network based on candidate regions, using weighted non-maximum suppression (NMS) in the target detection prediction stage. This enhances the recognition ability for multiple targets and occluded targets, obtaining the optimal bounding box.
[0033] After the defect detection model is trained, key features from preprocessed infrared thermal imaging images are input into the model for identification. The model outputs a defect detection result based on the input features, including information such as the type, size, and location of the defect. Then, based on the location information of these key features within the photovoltaic string, these defect locations are mapped onto the layout map of the photovoltaic power plant to obtain the specific defect locations.
[0034] This defect location information can be used to guide the operation and maintenance of photovoltaic power plants. For example, targeted maintenance plans can be developed based on the location and type of defects to reduce power generation losses caused by defects. Simultaneously, statistical analysis of defect location information can identify potential problems in photovoltaic power plants, providing data support for plant optimization and upgrades.
[0035] Example 2 A defect detection system for photovoltaic power plants includes: an infrared thermal imaging acquisition module, an image preprocessing module, a feature and location acquisition module, and a defect detection module; Infrared thermal imaging acquisition module, used to acquire infrared thermal imaging images and latitude and longitude data of photovoltaic modules; An image preprocessing module is used to preprocess the infrared thermal imaging image; The feature and location acquisition module is used to extract key features from the preprocessed infrared thermal imaging image and obtain the location information of the key features in the photovoltaic string based on the latitude and longitude data. The defect detection module is used to input the key features of the infrared thermal imaging image into the defect detection model for identification, obtain the defect detection result, and obtain the defect location information based on the location information of the key features in the photovoltaic string.
[0036] The module division in this embodiment of the invention is illustrative and represents only one logical functional division. In actual implementation, other division methods may be used. Furthermore, the functional modules in the various embodiments of the invention can be integrated into a single processor, exist as separate physical entities, or be integrated into a single module. The integrated modules described above can be implemented in hardware or as software functional modules.
[0037] This embodiment also provides a computer device, which includes a processor and a memory. The memory is used to store a computer program (in this embodiment, the computer program includes a computing component and an iterative component, capable of model calculation and model updating). The computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a Central Processing Unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, and is suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions in the computer storage medium to realize the corresponding method flow or corresponding function. The processor described in this embodiment can be used for the operation of a defect detection method for photovoltaic power plants.
[0038] This embodiment also provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in a computer device used to store programs and data. It is understood that the computer-readable storage medium here can include both the built-in storage medium in the computer device and extended storage media supported by the computer device. The computer-readable storage medium provides storage space that stores the terminal's operating system. Furthermore, this storage space also stores one or more instructions suitable for loading and execution by a processor. These instructions can be one or more computer programs (including program code). It should be noted that the computer-readable storage medium here can be high-speed RAM or non-volatile memory, such as at least one disk storage device. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the defect detection method for a photovoltaic power station described in the above embodiment.
[0039] This embodiment also provides a computer program product, which includes a computer program that, when executed by a processor, implements the corresponding steps of a defect detection method for photovoltaic power plants described in the above embodiment.
[0040] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0041] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0042] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0043] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0044] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.
Claims
1. A defect detection method for photovoltaic power plants, characterized in that, Includes the following steps: Collect infrared thermal images and latitude and longitude data of photovoltaic modules; The infrared thermal imaging image is preprocessed; Key features of the preprocessed infrared thermal imaging image are extracted, and the location information of the key features in the photovoltaic string is obtained based on the latitude and longitude data; The key features of the infrared thermal imaging image are input into the defect detection model for identification to obtain the defect detection result. Based on the location information of the key features in the photovoltaic string, the defect location information is obtained.
2. The defect detection method for photovoltaic power plants according to claim 1, characterized in that, The step of preprocessing the infrared thermal imaging image specifically includes: using a filtering algorithm to remove noise from the infrared thermal imaging image; Image enhancement is performed on the infrared thermal imaging image; The infrared thermal imaging image is then image registered.
3. The defect detection method for photovoltaic power plants according to claim 1, characterized in that, Key features of the preprocessed infrared thermal image are extracted using scale-invariant feature transformation.
4. The defect detection method for photovoltaic power plants according to claim 3, characterized in that, The key features include temperature features, texture features, and shape features.
5. A defect detection method for photovoltaic power plants according to claim 1, characterized in that, The step of obtaining the location information of key features in the photovoltaic string based on the latitude and longitude data specifically includes: The corresponding number of the photovoltaic module in the photovoltaic string is obtained based on the latitude and longitude data; The semantic segmentation algorithm is used to extract the region pixels of the photovoltaic module involved in the infrared thermal imaging image; The location information of key features in the photovoltaic string is obtained based on the number and region pixels.
6. The defect detection method for photovoltaic power plants according to claim 1, characterized in that, The defect detection model is trained by inputting defect infrared thermal imaging images and normal infrared thermal imaging images into a convolutional neural network.
7. A defect detection system for photovoltaic power plants, characterized in that, include: Infrared thermal imaging acquisition module, used to acquire infrared thermal imaging images and latitude and longitude data of photovoltaic modules; An image preprocessing module is used to preprocess the infrared thermal imaging image; The feature and location acquisition module is used to extract key features from the preprocessed infrared thermal imaging image and obtain the location information of the key features in the photovoltaic string based on the latitude and longitude data. The defect detection module is used to input the key features of the infrared thermal imaging image into the defect detection model for identification, obtain the defect detection result, and obtain the defect location information based on the location information of the key features in the photovoltaic string.
8. A computer device, characterized in that, The method includes a memory, a processor, and a computer program stored in the memory and executable in the processor, wherein the processor executes the computer program to implement the steps of the defect detection method for a photovoltaic power plant as described in any one of claims 1 to 6.
9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the defect detection method for a photovoltaic power station as described in any one of claims 1 to 6.
10. A computer program product, said computer product comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the defect detection method for photovoltaic power plants as described in any one of claims 1 to 6.