Photovoltaic panel defect degree evaluation method and system

Abstract

The invention discloses a photovoltaic panel defect degree evaluation method and system, and the method comprises the steps: obtaining and analyzing an infrared thermal image of a photovoltaic panel during working, and determining a defect region and hot spot defect image features; evaluating the internal defect of the defect area based on the hot spot defect image features to obtain an internal defect degree evaluation value; obtaining and analyzing an appearance image of the defect area, and determining an appearance defect type and an appearance defect image feature in the defect area; performing appearance defect evaluation on the defect area based on the appearance defect type and the appearance defect image features to obtain an appearance defect degree evaluation value; and determining a defect degree evaluation value of the photovoltaic panel based on the internal and appearance defect degree evaluation values of each defect area to evaluate the defect degree of the photovoltaic panel. According to the method, the internal electrical performance defect and the external physical damage of the panel are combined, quantitative extraction and fusion analysis are carried out on defect characteristics from two dimensions, and a comprehensive, quantifiable and accurate defect degree evaluation value is obtained.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic technology, and in particular to a method and system for assessing the degree of defects in photovoltaic panels. Background Technology

[0002] As the global energy structure accelerates its transition towards cleaner and renewable energy, solar photovoltaic (PV) power generation, as the mainstay, has seen its installed capacity grow rapidly. However, during its 25-year or even longer operating cycle, PV panels inevitably expose themselves to various harsh environments, including ultraviolet radiation, drastic temperature and humidity fluctuations, wind and snow loads, sand and dust erosion, and even hail impacts. These factors work together to cause various defects in the panels, such as microcracks, hot spots, PID effects, backsheet yellowing, and aging of encapsulation materials. These defects not only directly cause a decrease in the panel's output power, affecting the overall power generation revenue of the power plant, but may also trigger serious safety accidents such as localized overheating or even fires, directly threatening the asset safety and stable operation of the power plant.

[0003] Traditional defect detection methods, such as manual visual inspection, heavily rely on the experience of inspectors and have inherent limitations such as low efficiency, strong subjectivity, and difficulty in detecting internal latent defects. While portable EL (Elastic Optical Detection) testing offers high accuracy, it requires nighttime or darkroom operation and cannot perform online diagnostics on panels in operation, making it unsuitable for routine inspections in large-scale power plants. Therefore, the industry urgently needs an online, rapid, non-destructive, and quantitative defect assessment method. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a method and system for assessing the defect level of photovoltaic panels, comprising:

[0005] Infrared thermal images of photovoltaic panels during operation are obtained, and image analysis is performed on the infrared thermal images to determine defect areas and corresponding hot spot defect image features.

[0006] The degree of internal defects in the defect region is evaluated based on the features of hot spot defect images, and the evaluation value of the degree of internal defects in the defect region is obtained.

[0007] Acquire the appearance image of the defect area, perform image analysis on the appearance image, and determine the appearance defect type and corresponding appearance defect image features in the defect area;

[0008] The degree of appearance defect in the defect area is evaluated based on the type of appearance defect and the features of the appearance defect image, and the evaluation value of the degree of appearance defect in the defect area is obtained.

[0009] The defect severity assessment value of the photovoltaic panel is determined based on the internal defect severity assessment value and the appearance defect severity assessment value of each defect area, so as to assess the defect severity of the photovoltaic panel.

[0010] Furthermore, the step of acquiring an infrared thermal image of the photovoltaic panel during operation and performing image analysis on the infrared thermal image to determine the defect area and corresponding hot spot defect image features includes:

[0011] The infrared thermal image of the photovoltaic panel during operation is obtained, and the infrared thermal image is preprocessed to obtain a preprocessed infrared thermal image.

[0012] The preprocessed infrared thermal image is converted to grayscale to obtain a grayscale image, and the grayscale value of all pixels in the grayscale image is determined.

[0013] A grayscale dataset is established based on the grayscale values ​​of pixels in a grayscale image, and k initial cluster centers are randomly selected from the grayscale dataset.

[0014] Calculate the Euclidean distance from the gray values ​​in the grayscale dataset to the initial cluster centers, and then assign each pixel to its corresponding cluster based on the Euclidean distance from the gray values ​​in the grayscale dataset to the initial cluster centers.

[0015] Calculate the average gray value within each cluster, and redetermine the cluster centers based on the average gray value within each cluster;

[0016] Repeat the above steps until the cluster centers no longer change or the number of iterations reaches the preset iteration threshold, to obtain k clusters;

[0017] The preprocessed infrared thermal image is divided into regions based on the position of pixels in each cluster, resulting in multiple image regions. The average gray value of all pixels in each image region is then calculated.

[0018] Image regions with a grayscale average value greater than a preset threshold are identified as defect regions, and the grayscale average value, area, and location are identified as hot spot defect image features of the defect regions.

[0019] Furthermore, the assessment of the internal defect degree of the defect region based on hot spot defect image features to obtain an assessment value of the internal defect degree of the defect region includes:

[0020] Determine the preset baseline grayscale average value and baseline area, and calculate the difference between the grayscale average value of the defect area and the baseline grayscale average value, as well as the difference between the area and the baseline area.

[0021] The calculated differences are normalized to obtain the weights corresponding to the average gray value and area. The average gray value and area of ​​the defect area are evaluated to obtain the evaluation values ​​corresponding to the average gray value and area.

[0022] Determine the pre-set preset position-position coefficient mapping relationship, and determine the position coefficient of the defect area from the preset position-position coefficient mapping relationship based on the location of the defect area;

[0023] The internal defect level assessment value of the defect area is obtained by calculating the location coefficient, average gray value, area assessment value and weight of the defect area.

[0024] Furthermore, the formula for calculating the internal defect severity assessment value of the defective region is as follows:

[0025] ,

[0026] Where P is the flame combustion state assessment value of the flame region, α is the weight of the average gray value, H is the assessment value of the average gray value, β is the weight of the area, and M is the assessment value of the area.

[0027] Furthermore, the step of acquiring the appearance image of the defective region and performing image analysis on the appearance image to determine the type of appearance defect and the corresponding appearance defect image features in the defective region includes:

[0028] Obtain the appearance image of the defect area, and preprocess the appearance image of the defect area to obtain the preprocessed appearance image of the defect area;

[0029] Image recognition is performed on the appearance image of the preprocessed defect area to identify damage, burn marks, and discoloration as appearance defect types. The number and area of ​​damage, the average brightness value and area of ​​burn marks, and the average RGB component value and area of ​​discoloration are extracted as appearance defect image features of the defect area.

[0030] Furthermore, the evaluation of the degree of appearance defect in the defect area based on the type of appearance defect and the image features of the appearance defect, to obtain an evaluation value of the degree of appearance defect in the defect area, includes:

[0031] The number of damages, the average brightness value of burn marks, and the average RGB component value of discoloration are evaluated and values ​​are obtained to obtain evaluation values ​​for damage, burn marks, and discoloration. The areas of damage, burn marks, and discoloration are normalized to obtain the weights of damage, burn marks, and discoloration.

[0032] The degree of appearance defect in the defective area is calculated based on the evaluation values ​​of damage, burn marks, and discoloration and their corresponding weights.

[0033] Furthermore, the formula for calculating the degree of appearance defect in the defective area is as follows:

[0034] ,

[0035] Where E is the assessment value of the degree of appearance defect in the defect area, γ is the weight of damage, P is the assessment value of damage, σ is the weight of burn mark, R is the assessment value of burn mark, δ is the weight of discoloration, and W is the assessment value of discoloration.

[0036] Furthermore, the determination of the defect severity assessment value of the photovoltaic panel based on the internal defect severity assessment value and the appearance defect severity assessment value of each defect area, in order to assess the defect severity of the photovoltaic panel, includes:

[0037] The sub-defect degree assessment value of each defect area is calculated based on the internal defect degree assessment value and the external defect degree assessment value of each defect area.

[0038] Calculate the ratio of the area of ​​each defect region to the total area of ​​the photovoltaic panel, and normalize these ratios to obtain the weight of each defect region;

[0039] The sub-defect severity assessment values ​​of each defective region are weighted and added together with their corresponding weights to obtain the defect severity assessment value of the photovoltaic panel. The defect severity of the photovoltaic panel is then assessed based on the defect severity assessment value.

[0040] Furthermore, the formula for calculating the sub-defect severity assessment value of each defect region is as follows:

[0041] ,

[0042] Where S is the sub-defect severity assessment value of each defect area, a is the first preset weight, I is the internal defect severity assessment value of the defect area, b is the second preset weight, and E is the appearance defect severity assessment value of the defect area.

[0043] The present invention also provides a photovoltaic panel defect assessment system, comprising:

[0044] The first acquisition module is used to acquire infrared thermal images of the photovoltaic panel when it is working, and to perform image analysis on the infrared thermal images to determine the defect areas and the corresponding hot spot defect image features.

[0045] The first evaluation module is used to evaluate the degree of internal defects in the defect region based on the features of hot spot defect images, and obtain the evaluation value of the degree of internal defects in the defect region.

[0046] The second acquisition module is used to acquire the appearance image of the defect area, and perform image analysis on the appearance image to determine the appearance defect type and the corresponding appearance defect image features in the defect area.

[0047] The second evaluation module is used to evaluate the degree of appearance defects in the defect area based on the type of appearance defect and the features of the appearance defect image, and to obtain the evaluation value of the degree of appearance defects in the defect area.

[0048] The calculation module is used to determine the defect level assessment value of the photovoltaic panel based on the internal defect level assessment value and the appearance defect level assessment value of each defect area, so as to assess the defect level of the photovoltaic panel.

[0049] Compared with existing technologies, the photovoltaic panel defect assessment method and system of this invention have the following advantages:

[0050] This invention can quantitatively assess the severity of defects by analyzing the image features of hot spot regions. At the same time, by identifying and classifying appearance image features, it can assess structural integrity and aging status. This diagnostic mechanism, which combines internal and external aspects, reveals the causal relationship between appearance defects and performance degradation, and achieves a profound insight into the root causes of defects and their impact.

[0051] This invention abandons the qualitative judgment that relies on human experience in traditional operation and maintenance. By extracting and analyzing the image features of hot spot defects and appearance defects, the defect status is transformed into a series of calculable internal and appearance defect severity assessment values, making the severity of defects comparable. This provides an accurate data foundation for formulating unified operation and maintenance standards and priorities, and greatly reduces misjudgments and omissions caused by subjective differences among personnel.

[0052] This invention unifies internal electrical performance loss and external structural risks into a single evaluation system through a data fusion algorithm. This system can scientifically guide operation and maintenance decisions, and upgrade the management from simply dealing with a single defect point to a global and strategic management of the health status of the entire panel, thereby achieving optimal allocation of operation and maintenance resources. Attached Figure Description

[0053] Figure 1 This is a schematic diagram of the process structure of the photovoltaic panel defect degree assessment method in an embodiment of the present invention;

[0054] Figure 2 This is a schematic diagram of the composition of the photovoltaic panel defect assessment system in an embodiment of the present invention. Detailed Implementation

[0055] The specific embodiments of this application will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and are not intended to limit the scope of the invention.

[0056] In the description of this application, it should be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the platform or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0057] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0058] like Figure 1 As shown in the embodiments of this application, a method for evaluating the defect level of a photovoltaic panel is provided, including: S100: acquiring an infrared thermal image of the photovoltaic panel during operation, and performing image analysis on the infrared thermal image to determine the defect area and the corresponding hot spot defect image features; S200: evaluating the internal defect level of the defect area based on the hot spot defect image features to obtain an internal defect level evaluation value for the defect area; S300: acquiring an external image of the defect area, and performing image analysis on the external image to determine the external defect type and the corresponding external defect image features for the defect area; S400: evaluating the external defect level of the defect area based on the external defect type and external defect image features to obtain an external defect level evaluation value for the defect area; S500: determining the defect level evaluation value of the photovoltaic panel based on the internal defect level evaluation value and the external defect level evaluation value of each defect area, so as to evaluate the defect level of the photovoltaic panel.

[0059] Furthermore, this invention can quantitatively assess the severity of defects by analyzing the image features of hot spot areas. Simultaneously, by identifying and classifying appearance image features, it can assess structural integrity and aging status. This combined internal and external diagnostic mechanism reveals the causal relationship between appearance defects and performance degradation, achieving a profound insight into the root causes and impacts of defects. This invention abandons the qualitative judgments relying on human experience in traditional operations and maintenance. By extracting and analyzing hot spot defect image features and appearance defect image features, it transforms the defect status into a series of calculable internal and external defect severity assessment values, making defect severity comparable. This provides a precise data foundation for formulating unified operations and maintenance standards and priorities, greatly reducing misjudgments and omissions caused by subjective differences in personnel. This invention unifies internal electrical performance losses and external structural risks into a single evaluation system through data fusion algorithms, scientifically guiding operations and maintenance decisions. It upgrades the management from simple handling of individual defect points to a global and strategic management of the entire panel's health status, achieving optimal allocation of operations and maintenance resources.

[0060] In embodiments of this application, a method for assessing the defect level of a photovoltaic panel is provided. The method involves acquiring an infrared thermal image of the photovoltaic panel during operation and performing image analysis on the infrared thermal image to determine defect areas and corresponding hot spot defect image features. The method includes: acquiring an infrared thermal image of the photovoltaic panel during operation and preprocessing the infrared thermal image to obtain a preprocessed infrared thermal image; performing grayscale processing on the preprocessed infrared thermal image to obtain a grayscale image and determining the grayscale values ​​of all pixels within the grayscale image; establishing a grayscale dataset based on the grayscale values ​​of the pixels in the grayscale image and randomly selecting k initial cluster centers from the grayscale dataset; and calculating the Euclidean distance from the grayscale values ​​in the grayscale dataset to the initial cluster centers. The process involves dividing each pixel into its corresponding cluster based on the Euclidean distance from the grayscale value in the grayscale dataset to the initial cluster center; calculating the average grayscale value within each cluster and redetermining the cluster center based on the average grayscale value within each cluster; repeating the above steps until the cluster center no longer changes or the number of iterations reaches a preset iteration threshold, resulting in k clusters; dividing the preprocessed infrared thermal image into regions based on the position of the pixels in each cluster, resulting in multiple image regions, and calculating the average grayscale value of all pixels in each image region; identifying image regions with an average grayscale value greater than a preset threshold as defect regions, and determining the average grayscale value, area, and position as the hot spot defect image features of the defect region.

[0061] Specifically, by preprocessing and grayscale conversion of infrared thermal images of photovoltaic panels in operation, a dataset based on pixel grayscale values ​​was constructed. Using the K-means unsupervised clustering algorithm, cluster centers were iteratively optimized, ultimately intelligently dividing the image pixels into K regions with different temperature characteristics. This replaced the traditional segmentation method relying on fixed thresholds and could adaptively identify abnormally high-temperature regions in the image. After region division, the average grayscale value of each region was calculated and compared with a preset threshold to accurately locate defective regions. Based on this, three core features of each defective region were further extracted: average grayscale value (quantifying the severity of hot spots), area (assessing the impact range of the defect), and location (assisting in determining the cause of the defect and maintenance strategies). This step automates and intelligentizes hot spot detection, significantly improving detection efficiency and reducing reliance on human experience. The adaptive nature of clustering algorithms enhances the accuracy and robustness of hot spot identification under different environments and equipment conditions. Quantitative feature extraction provides a precise and reliable data foundation for subsequent defect assessment, fault diagnosis, and predictive maintenance, ultimately driving the transformation of photovoltaic power plant operation and maintenance from a "passive response" to an intelligent model of "proactive early warning" and "precise decision-making."

[0062] In an embodiment of this application, a method for assessing the degree of defects in a photovoltaic panel is provided. The method involves assessing the internal defect degree of a defect region based on hot spot defect image features to obtain an assessment value for the internal defect degree of the defect region. This includes: determining a pre-set baseline grayscale average value and a baseline area; calculating the differences between the grayscale average value and the baseline grayscale average value, and between the area and the baseline area of ​​the defect region; normalizing the calculated differences to obtain weights corresponding to the grayscale average value and area; evaluating the grayscale average value and area of ​​the defect region to obtain assessment values ​​corresponding to the grayscale average value and area; determining a pre-set preset position-position coefficient mapping relationship; determining the position coefficient of the defect region from the preset position-position coefficient mapping relationship based on the location of the defect region; and calculating the internal defect degree assessment value of the defect region based on the position coefficient, grayscale average value, area assessment value, and weights.

[0063] Specifically, a baseline grayscale average and a baseline area are set as reference lines for the normal state. By calculating the difference between the actual value of the defect area and these baseline values ​​and normalizing them, features of different dimensions are unified onto a comparable scale, eliminating the influence of dimensions. The dynamic weight of each feature is automatically determined according to the magnitude of the difference; the larger the difference, the higher the weight of the defect severity. The impact of defect location on severity is quantified through a preset mapping relationship (e.g., hot spots near the main busbar at the center of the cell have a higher risk coefficient than hot spots at the edge). By comprehensively calculating the location coefficient with the weighted grayscale evaluation value and area evaluation value, an accurate internal defect severity evaluation value is output. This step achieves objectivity and standardization of assessment criteria, eliminating the subjectivity and inconsistency of human judgment, making the assessment results repeatable and comparable; it achieves comprehensiveness of assessment dimensions, as the model does not view temperature or area in isolation, but comprehensively considers the intensity, range, and criticality of hot spots, thus providing a more comprehensive and accurate characterization of defect risks; and it achieves precision in operation and maintenance decision-making, with the output quantitative assessment values ​​providing direct and reliable data for formulating precise maintenance priorities, thereby optimizing the allocation of operation and maintenance resources and improving power plant safety and power generation efficiency.

[0064] In an embodiment of this application, a method for assessing the defect level of a photovoltaic panel is provided, wherein the formula for calculating the internal defect level assessment value of the defective region is as follows:

[0065] ,

[0066] Where P is the flame combustion state assessment value of the flame region, α is the weight of the average gray value, H is the assessment value of the average gray value, β is the weight of the area, and M is the assessment value of the area.

[0067] In an embodiment of this application, a method for assessing the degree of defects in a photovoltaic panel is provided. The step of acquiring an appearance image of a defective area and performing image analysis on the appearance image to determine the type of appearance defect and the corresponding appearance defect image features in the defective area includes: acquiring an appearance image of a defective area and preprocessing the appearance image of the defective area to obtain a preprocessed appearance image of the defective area; performing image recognition on the preprocessed appearance image of the defective area to identify damage, burn marks, and discoloration in the preprocessed appearance image of the defective area as appearance defect types, and extracting the number and area of ​​damage, the average brightness value and area of ​​burn marks, and the average RGB component value and area of ​​discoloration as appearance defect image features of the defective area.

[0068] Specifically, high-resolution images of defective areas are preprocessed to improve image quality and lay the foundation for accurate identification. Advanced image recognition algorithms (such as deep learning-based semantic segmentation or traditional image processing techniques) are used to accurately identify and classify three key types of appearance defects from the preprocessed images: damage, burn marks, and discoloration. For each type of defect, refined feature quantification is performed: for damage, the number and total area are extracted to assess the degree of structural fragmentation; for burn marks, the average brightness value (reflecting the severity of carbonization) and area (reflecting the extent of impact) are extracted; for discoloration, the average RGB components are extracted. The process transforms subjective, qualitative observations by combining values ​​(precisely quantifying the degree of color deviation, such as yellowing) and area into a series of objective, calculable image feature data. This step automates and intelligently detects cosmetic defects, significantly improving detection efficiency and avoiding the subjectivity and fatigue errors of manual inspection. Through the precise quantification of multiple feature parameters, it achieves accurate characterization of the severity of cosmetic defects, rather than just judging their type. These extracted quantitative features, combined with previous internal defect assessment values, provide solid data support for subsequent defect root cause analysis, comprehensive level assessment, and the development of precise operation and maintenance strategies.

[0069] In an embodiment of this application, a method for evaluating the degree of defects in a photovoltaic panel is provided. The method involves evaluating the degree of appearance defects in a defect area based on the type of appearance defect and the image features of the appearance defect to obtain an evaluation value for the degree of appearance defects in the defect area. This includes: evaluating the number of breaks, the average brightness value of burn marks, and the average RGB component value of discoloration to obtain evaluation values ​​for breakage, burn marks, and discoloration; normalizing the area of ​​breakage, the area of ​​burn marks, and the area of ​​discoloration to obtain weights for breakage, burn marks, and discoloration; and calculating the degree of appearance defects in the defect area based on the evaluation values ​​of breakage, burn marks, and discoloration and their corresponding weights.

[0070] Specifically, the number of defects, the average brightness value of burn marks, and the average RGB component value of discoloration are independently evaluated and assigned values, quantifying the "intensity" or "density" of each defect. For example, the more defects, the darker the burn marks (low brightness value), and the more significant the discoloration (large deviation of RGB components from normal values), the higher the corresponding "evaluation value." The areas of defects, burn marks, and discoloration are normalized and converted into their respective weights, making the impact range of the defect a lever for measuring its overall importance. A slight discoloration covering a large area may have a greater negative impact on the overall panel condition than a severe burn mark covering a small area. By combining the intensity evaluation value with the range weight, the degree of appearance defect in the defect area can be accurately evaluated. This step provides unprecedented clarity for operation and maintenance decisions, enabling resources to be prioritized for addressing the most critical defects with "high intensity and large range," thereby achieving a leap from experience-based fuzzy judgment to data-driven precision operation and maintenance, greatly improving the efficiency and effectiveness of photovoltaic power plant maintenance.

[0071] In an embodiment of this application, a method for assessing the defect level of a photovoltaic panel is provided, wherein the formula for calculating the appearance defect level assessment value of the defective area is as follows:

[0072] ,

[0073] Where E is the assessment value of the degree of appearance defect in the defect area, γ is the weight of damage, P is the assessment value of damage, σ is the weight of burn mark, R is the assessment value of burn mark, δ is the weight of discoloration, and W is the assessment value of discoloration.

[0074] In an embodiment of this application, a method for assessing the defect level of a photovoltaic panel is provided. The method involves determining a defect level assessment value of the photovoltaic panel based on the internal defect level assessment value and the appearance defect level assessment value of each defect region, in order to assess the defect level of the photovoltaic panel. The method includes: calculating a sub-defect level assessment value for each defect region based on the internal defect level assessment value and the appearance defect level assessment value of each defect region; calculating the ratio of the area of ​​each defect region to the total area of ​​the photovoltaic panel, and normalizing these ratios to obtain the weight of each defect region; weighted summing the sub-defect level assessment value of each defect region with its corresponding weight to obtain the defect level assessment value of the photovoltaic panel, and assessing the defect level of the photovoltaic panel based on the defect level assessment value.

[0075] Specifically, by calculating and combining the internal defect severity assessment value and the external defect severity assessment value of each defect area into a sub-defect severity assessment value, a comprehensive local score is given for each independent defect point. By calculating and normalizing the ratio of the area of ​​each defect area to the total area of ​​the panel, the proportion of the impact of different defects on the overall panel is quantified. A serious defect occupying a large area naturally has a higher weight than an equivalent defect in a small area. The final photovoltaic panel defect severity assessment value is obtained by weighted summing the sub-defect severity assessment value of each defect area with its area weight. This step achieves a shift in evaluation dimensions from a "point" to a "surface," enabling the evaluation conclusions to no longer be limited to the severity of a single defect, but to reflect the comprehensive impact of defect clusters on the overall performance of the panel. This ensures the scientific rigor and fairness of the evaluation results, reasonably amplifying the impact of large-area defects, and effectively capturing the cumulative effect of multiple small defects. The final output of this single, quantitative evaluation value provides a direct and clear decision-making basis for power plant operation and maintenance management. Based on this, thousands of panels can be prioritized, and differentiated strategies such as "replacement, repair, and observation" can be scientifically formulated to achieve optimal allocation of operation and maintenance resources and ensure the overall power generation efficiency and safe operation of the photovoltaic power plant.

[0076] In an embodiment of this application, a method for assessing the defect level of a photovoltaic panel is provided, wherein the calculation formula for the sub-defect level assessment value of each defect region is as follows:

[0077] ,

[0078] Where S is the sub-defect severity assessment value of each defect area, a is the first preset weight, I is the internal defect severity assessment value of the defect area, b is the second preset weight, and E is the appearance defect severity assessment value of the defect area.

[0079] like Figure 2As shown in the embodiments of this application, a photovoltaic panel defect severity assessment system is provided, comprising: a first acquisition module, used to acquire an infrared thermal image of the photovoltaic panel during operation, and perform image analysis on the infrared thermal image to determine the defect area and the corresponding hot spot defect image features; a first assessment module, used to assess the internal defect severity of the defect area based on the hot spot defect image features, and obtain an internal defect severity assessment value for the defect area; a second acquisition module, used to acquire an external image of the defect area, and perform image analysis on the external image to determine the external defect type and the corresponding external defect image features for the defect area; a second assessment module, used to assess the external defect severity of the defect area based on the external defect type and external defect image features, and obtain an external defect severity assessment value for the defect area; and a calculation module, used to determine the defect severity assessment value of the photovoltaic panel based on the internal defect severity assessment value and the external defect severity assessment value of each defect area, so as to assess the defect severity of the photovoltaic panel.

[0080] In summary, this invention provides a method and system for assessing the defect severity of a photovoltaic panel, comprising: acquiring and analyzing an infrared thermal image of the photovoltaic panel during operation to determine the defect region and hot spot defect image features; assessing the internal defects of the defect region based on the hot spot defect image features to obtain an internal defect severity assessment value; acquiring and analyzing an external image of the defect region to determine the external defect type and external defect image features; assessing the external defects of the defect region based on the external defect type and external defect image features to obtain an external defect severity assessment value; and determining the defect severity assessment value of the photovoltaic panel based on the internal and external defect severity assessment values ​​of each defect region to assess the defect severity of the photovoltaic panel. This invention combines internal electrical performance defects and external physical damage to quantitatively extract and fuse defect features from two dimensions, resulting in a comprehensive, quantifiable, and accurate defect severity assessment value.

[0081] Finally, it should be noted that those skilled in the art can obviously make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims and their equivalents, this invention also intends to include these modifications and variations.

[0082] The above description is merely one embodiment of the present invention, and should not be construed as limiting the scope of the invention. Any structural changes made based on the present invention, as long as they do not depart from the essence of the invention, should be considered as falling within the protection scope of the present invention and subject to its restrictions. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process and related descriptions of the platform described above can be referred to the corresponding processes in the foregoing platform embodiments, and will not be repeated here.

[0083] The term "comprising" or any other similar term is intended to cover non-exclusive inclusion, such that a process, platform, article, or device / platform that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to those processes, platforms, articles, or devices / platforms.

[0084] The technical solutions of the present invention have been described in conjunction with the accompanying drawings and further embodiments. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to closely related technical features, and the technical solutions resulting from such changes or substitutions will all fall within the scope of protection of the present invention.

[0085] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention.

Claims

1. A method for assessing the defect level of a photovoltaic panel, characterized in that, include: Infrared thermal images of photovoltaic panels during operation are obtained, and image analysis is performed on the infrared thermal images to determine defect areas and corresponding hot spot defect image features. The degree of internal defects in the defect region is evaluated based on the features of hot spot defect images, and the evaluation value of the degree of internal defects in the defect region is obtained. Acquire the appearance image of the defect area, perform image analysis on the appearance image, and determine the appearance defect type and corresponding appearance defect image features in the defect area; The degree of appearance defect in the defect area is evaluated based on the type of appearance defect and the features of the appearance defect image, and the evaluation value of the degree of appearance defect in the defect area is obtained. The defect severity assessment value of the photovoltaic panel is determined based on the internal defect severity assessment value and the appearance defect severity assessment value of each defect area, so as to assess the defect severity of the photovoltaic panel.

2. The method for assessing the defect level of a photovoltaic panel according to claim 1, characterized in that, The process of acquiring an infrared thermal image of the photovoltaic panel during operation and performing image analysis on the infrared thermal image to determine the defect area and corresponding hot spot defect image features includes: The infrared thermal image of the photovoltaic panel during operation is obtained, and the infrared thermal image is preprocessed to obtain a preprocessed infrared thermal image. The preprocessed infrared thermal image is converted to grayscale to obtain a grayscale image, and the grayscale value of all pixels in the grayscale image is determined. A grayscale dataset is established based on the grayscale values ​​of pixels in a grayscale image, and k initial cluster centers are randomly selected from the grayscale dataset. Calculate the Euclidean distance from the gray values ​​in the grayscale dataset to the initial cluster centers, and then assign each pixel to its corresponding cluster based on the Euclidean distance from the gray values ​​in the grayscale dataset to the initial cluster centers. Calculate the average gray value within each cluster, and redetermine the cluster centers based on the average gray value within each cluster; Repeat the above steps until the cluster centers no longer change or the number of iterations reaches the preset iteration threshold, to obtain k clusters; The preprocessed infrared thermal image is divided into regions based on the position of pixels in each cluster, resulting in multiple image regions. The average gray value of all pixels in each image region is then calculated. Image regions with a grayscale average value greater than a preset threshold are identified as defect regions, and the grayscale average value, area, and location are identified as hot spot defect image features of the defect regions.

3. The method for assessing the defect level of a photovoltaic panel according to claim 2, characterized in that, The assessment of the internal defect degree of the defect region based on hot spot defect image features, to obtain an assessment value of the internal defect degree of the defect region, includes: Determine the preset baseline grayscale average value and baseline area, and calculate the difference between the grayscale average value of the defect area and the baseline grayscale average value, as well as the difference between the area and the baseline area. The calculated differences are normalized to obtain the weights corresponding to the average gray value and area. The average gray value and area of ​​the defect area are evaluated to obtain the evaluation values ​​corresponding to the average gray value and area. Determine the pre-set preset position-position coefficient mapping relationship, and determine the position coefficient of the defect area from the preset position-position coefficient mapping relationship based on the location of the defect area; The internal defect level assessment value of the defect area is obtained by calculating the location coefficient, average gray value, area assessment value and weight of the defect area.

4. The method for assessing the defect level of a photovoltaic panel according to claim 3, characterized in that, The formula for calculating the internal defect severity assessment value of the defective region is as follows: ， Where P is the flame combustion state assessment value of the flame region, α is the weight of the average gray value, H is the assessment value of the average gray value, β is the weight of the area, and M is the assessment value of the area.

5. The method for assessing the defect level of a photovoltaic panel according to claim 3, characterized in that, The process of acquiring an appearance image of the defective region and performing image analysis on the appearance image to determine the type of appearance defect and the corresponding appearance defect image features in the defective region includes: Obtain the appearance image of the defect area, and preprocess the appearance image of the defect area to obtain the preprocessed appearance image of the defect area; Image recognition is performed on the appearance image of the preprocessed defect area to identify damage, burn marks, and discoloration as appearance defect types. The number and area of ​​damage, the average brightness value and area of ​​burn marks, and the average RGB component value and area of ​​discoloration are extracted as appearance defect image features of the defect area.

6. The method for assessing the defect level of a photovoltaic panel according to claim 5, characterized in that, The assessment of the degree of appearance defect in the defect area based on the type of appearance defect and the image features of the appearance defect, to obtain an assessment value of the degree of appearance defect in the defect area, includes: The number of damages, the average brightness value of burn marks, and the average RGB component value of discoloration are evaluated and values ​​are obtained to obtain evaluation values ​​for damage, burn marks, and discoloration. The areas of damage, burn marks, and discoloration are normalized to obtain the weights of damage, burn marks, and discoloration. The degree of appearance defect in the defective area is calculated based on the evaluation values ​​of damage, burn marks, and discoloration and their corresponding weights.

7. The method for assessing the defect level of a photovoltaic panel according to claim 6, characterized in that, The formula for calculating the assessment value of the appearance defect degree of the defective area is as follows: ， Where E is the assessment value of the degree of appearance defect in the defect area, γ is the weight of damage, P is the assessment value of damage, σ is the weight of burn mark, R is the assessment value of burn mark, δ is the weight of discoloration, and W is the assessment value of discoloration.

8. The method for assessing the defect level of a photovoltaic panel according to claim 6, characterized in that, The method of determining the defect severity assessment value of the photovoltaic panel based on the internal defect severity assessment value and the appearance defect severity assessment value of each defect area, in order to assess the defect severity of the photovoltaic panel, includes: The sub-defect degree assessment value of each defect area is calculated based on the internal defect degree assessment value and the external defect degree assessment value of each defect area. Calculate the ratio of the area of ​​each defect region to the total area of ​​the photovoltaic panel, and normalize these ratios to obtain the weight of each defect region; The sub-defect severity assessment values ​​of each defective region are weighted and added together with their corresponding weights to obtain the defect severity assessment value of the photovoltaic panel. The defect severity of the photovoltaic panel is then assessed based on the defect severity assessment value.

9. The method for assessing the defect level of a photovoltaic panel according to claim 8, characterized in that, The formula for calculating the sub-defect severity assessment value of each defect region is as follows: ， Where S is the sub-defect severity assessment value of each defect area, a is the first preset weight, I is the internal defect severity assessment value of the defect area, b is the second preset weight, and E is the appearance defect severity assessment value of the defect area.

10. A photovoltaic panel defect assessment system, characterized in that, include: The first acquisition module is used to acquire infrared thermal images of the photovoltaic panel when it is working, and to perform image analysis on the infrared thermal images to determine the defect areas and the corresponding hot spot defect image features. The first evaluation module is used to evaluate the degree of internal defects in the defect region based on the features of hot spot defect images, and obtain the evaluation value of the degree of internal defects in the defect region. The second acquisition module is used to acquire the appearance image of the defect area, and perform image analysis on the appearance image to determine the appearance defect type and the corresponding appearance defect image features in the defect area. The second evaluation module is used to evaluate the degree of appearance defects in the defect area based on the type of appearance defect and the features of the appearance defect image, and to obtain the evaluation value of the degree of appearance defects in the defect area. The calculation module is used to determine the defect level assessment value of the photovoltaic panel based on the internal defect level assessment value and the appearance defect level assessment value of each defect area, so as to assess the defect level of the photovoltaic panel.

Images