Battery piece defect detection method and device, computer device, and readable storage medium

Abstract

The application relates to a battery piece defect detection method and device, computer equipment and a readable storage medium. The method comprises the following steps: obtaining a to-be-detected battery piece image; inputting the to-be-detected battery piece image into a first defect detection model to obtain a main defect, wherein the main defect comprises a continuous defect area; inputting the to-be-detected battery piece image into a second defect detection model to obtain a local defect, wherein the local defect comprises at least one defect source point, and the defect source point comprises a defect endpoint or a defect intersection point; and determining a target defect based on the main defect and the local defect. The battery piece defect detection method provided in the application can complement or correct the omissions or errors of the main defect detection result through the local defect, the local defect can more accurately express the position of the induced defect, important reference bases for analyzing the reasons for causing the battery piece defect are provided, and the accuracy and detection efficiency of the battery piece defect detection are effectively improved.

Description

Technical Field

[0001] This application relates to the field of solar cell testing technology, and in particular to a method, apparatus, computer equipment, and readable storage medium for detecting defects in solar cells. Background Technology

[0002] In recent years, with the excessive use of non-renewable energy sources such as oil and coal and the resulting increasingly severe environmental problems, new energy technologies have received attention and are developing rapidly. Solar photovoltaic power generation technology is a typical representative of new energy technologies, and its widespread application can effectively alleviate the current energy problems. However, the manufacturing process of photovoltaic cells is quite complex, and each production stage can potentially damage the cells, leading to various cell defects.

[0003] In traditional solar cell defect detection techniques, the datasets used to train defect detection models often fail to cover all defects due to the varying sizes and shapes of defects within the cells. Furthermore, localized areas of defects are themselves defects, potentially leading to detection results that pinpoint specific defect regions. This results in poor regression performance of the detection algorithm and consequently, low accuracy in defect detection. Therefore, there is a pressing need for a detection technique that can improve the accuracy of solar cell defect detection. Summary of the Invention

[0004] Therefore, it is necessary to provide a method, apparatus, computer equipment, and readable storage medium for detecting defects in solar cells that can improve the accuracy of defect detection in solar cells, in order to address the aforementioned technical problems.

[0005] In a first aspect, this application provides a method for detecting defects in solar cells. The method includes:

[0006] Acquire images of the battery cells to be inspected;

[0007] The image of the battery cell to be detected is input into the first defect detection model to obtain the main defect, which includes a continuous defect region;

[0008] The image of the battery cell to be detected is input into the second defect detection model to obtain local defects. The local defects include at least one defect source point, and the defect source point includes a defect endpoint or a defect intersection point.

[0009] The target defect is determined based on the main defect and the local defects.

[0010] In one embodiment, determining the target defect based on the main defect and the local defects includes:

[0011] Determine the main defect region of the main defect and the local defect region of the local defect;

[0012] The target defect is determined based on the degree of overlap between the main defect region and the local defect region.

[0013] In one embodiment, determining the target defect based on the overlap between the main defect region and the local defect region includes:

[0014] If the overlap is greater than zero, the target defect is determined based on the main defect region and the local defects;

[0015] If the overlap is not greater than zero, then the target defect is determined based on the main defect.

[0016] In one embodiment, determining the target defect based on the local defect includes:

[0017] If the local defect is a hidden oblique crack, and the local defect includes at least two hidden oblique crack initiation points, then determine the position of the hidden oblique crack initiation points in the image of the battery cell to be inspected;

[0018] If all the oblique hidden crack source points are internal endpoints, then the main defect region, the oblique hidden crack, and all internal endpoints are determined as the target defect;

[0019] If the oblique hidden crack source point includes at least one boundary endpoint, then the main defect region, the oblique hidden crack, and all boundary endpoints are determined as the target defect.

[0020] In one embodiment, determining the target defect based on the local defect includes:

[0021] If the local defect includes intersecting hidden cracks and oblique hidden cracks, the main defect region, the intersecting hidden cracks, and all the source points of the intersecting hidden cracks are determined as the target defect.

[0022] In one embodiment, after determining the target defect based on the main defect and the local defects, the method further includes:

[0023] Determine the size ratio between the image of the battery cell to be tested and the size of the battery cell to be tested;

[0024] The actual distribution of the target defect on the battery cell under test is determined based on the location information of the target defect in the image of the battery cell under test and the size ratio.

[0025] In one embodiment, after determining the target defect based on the main defect and the local defects, the method further includes:

[0026] Determine the comparison result between the target defect and the preset defect;

[0027] Based on the comparison results, unusable defective battery cells are identified and removed.

[0028] Secondly, this application also provides a battery cell defect detection device. The device includes:

[0029] Image acquisition module, used to acquire images of the battery cell to be inspected;

[0030] The first defect detection module is used to input the image of the battery cell to be detected into the first defect detection model to obtain the main defect, wherein the main defect includes a continuous defect region;

[0031] The second defect detection module is used to input the image of the battery cell to be detected into the second defect detection model to obtain local defects. The local defects include at least one defect point, and the defect point includes a defect endpoint or a defect intersection.

[0032] The defect determination module is used to determine the target defect based on the main defect and the local defects.

[0033] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method described in any one of the first aspects above.

[0034] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any one of the first aspects above.

[0035] The aforementioned cell defect detection method, apparatus, computer equipment, and readable storage medium detect main defects and local defects using first and second detection models, and determine target defects based on the main defects and local defects. On the one hand, this fully considers the inherent relationship between the characteristics of main defects and local defects, using local defects to supplement or correct omissions or errors in the main defect detection results. On the other hand, local defects can more accurately express the location of the defect, providing important reference for analyzing the causes of cell defects, enabling faster location of defect-causing links in the production process, and reducing potential further losses during cell production. The cell defect detection method provided in this application effectively improves the accuracy and efficiency of cell defect detection by fully considering the correlation between main defect characteristics and local defects.

[0036] Details of one or more embodiments of this application are set forth in the following drawings and description to make other features, objects and advantages of this application more readily apparent. Attached Figure Description

[0037] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0038] Figure 1 This is a flowchart illustrating a cell defect detection method in one embodiment;

[0039] Figure 2 This is a schematic diagram of one embodiment where a whole battery cell is divided into 6 smaller battery cell pieces;

[0040] Figure 3 This is a schematic diagram of the battery chip to be tested in one embodiment;

[0041] Figure 4 This is a schematic diagram showing overall and local defects of the battery cell in one embodiment;

[0042] Figure 5 This is a schematic diagram of a microcrack defect in a battery cell in one embodiment;

[0043] Figure 6 This is a schematic diagram of a microcrack defect in a battery cell in another embodiment;

[0044] Figure 7 This is a schematic diagram of cross-crack defects in a solar cell in one embodiment;

[0045] Figure 8 This is a schematic diagram of the defect distribution of the battery cells in one embodiment;

[0046] Figure 9 This is a schematic diagram of the cell testing process in one embodiment;

[0047] Figure 10 This is a structural block diagram of a cell defect detection device in one embodiment. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0049] Unless otherwise defined, the technical or scientific terms used in this application shall have the general meaning as understood by one of ordinary skill in the art to which this application pertains. Words such as “a,” “an,” “an,” “the,” “the,” and “these,” used in this application, do not indicate quantitative limitation and may be singular or plural. The terms “comprising,” “including,” “having,” and any variations thereof used in this application are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that comprises a series of steps or modules (units) is not limited to the listed steps or modules (units) but may include steps or modules (units) not listed, or may include other steps or modules (units) inherent to such processes, methods, products, or devices. The terms “connected,” “linked,” and “coupled,” used in this application, are not limited to physical or mechanical connections but may include electrical connections, whether direct or indirect. The term “multiple” used in this application refers to two or more. The "and / or" operator describes the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: A alone, A and B simultaneously, and B alone. Typically, the character " / " indicates that the objects before and after it are in an "or" relationship. The terms "first," "second," and "third," etc., used in this application are merely for distinguishing similar objects and do not represent a specific ordering of the objects.

[0050] The terms “module”, “unit”, etc., used below refer to a combination of software and / or hardware that can perform a predetermined function. Although the apparatus described in the following embodiments is preferably implemented in hardware, implementation in software, or a combination of software and hardware, is also possible and contemplated.

[0051] In one embodiment, such as Figure 1 As shown, a method for detecting defects in battery cells is provided. This embodiment illustrates the method applied to a terminal; however, it is understood that the method can also be applied to a server, or to a system including both a terminal and a server, and implemented through interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0052] S201: Obtain the image of the battery cell to be tested.

[0053] The solar cell mentioned in this application refers to a monocrystalline silicon or polycrystalline silicon solar cell used in photovoltaic power generation. In the embodiments of this application, the image of the solar cell to be tested may include an image of the entire solar cell, or images of smaller solar cell pieces cut from the entire solar cell. Specifically, in some embodiments, the solar cell to be tested may include a single 210*210mm image of the solar cell. If the entire solar cell is divided into six 210*35mm solar cell pieces, the solar cell to be tested may also include images of at least one of these smaller solar cell pieces. Figure 2 As shown, the entire battery cell can be divided into 1-6 smaller battery pieces, and the image of each smaller battery piece can be obtained as the image of the battery cell to be detected. Figure 3 This is an image of the battery cell to be inspected obtained in one embodiment; it is an image of a segmented battery cell. It can be understood that after determining the defects in each segment, the defect distribution of the entire battery cell can be obtained.

[0054] In this embodiment, based on the characteristic that solar cells emit light when illuminated by certain light sources, and because the light emission characteristics of defective and normal locations differ significantly under illumination, the defect status of the solar cell can be determined by acquiring an image of the solar cell under illumination. Acquiring the image of the solar cell to be inspected includes illuminating the solar cell with a light source and then scanning it with a camera to obtain the image. The light source includes an LED light source or a laser light source, and the camera can include a TDI industrial line scan camera, etc. Illuminating the solar cell with an external light source does not affect the solar cell itself, and the TDI industrial line scan camera has high sensitivity and good image quality.

[0055] In some embodiments of this application, an LED light source, a TDI industrial line scan camera, and an industrial control computer can be installed on the battery cell production line with a reserved space of approximately 0.5 meters. An external encoder is installed on the motor of the conveyor belt. The encoder generates a trigger signal based on a preset absolute distance the conveyor belt moves and sends it to the TDI camera. Upon receiving the trigger signal, the TDI camera acquires an image of the battery cell to be inspected. The reserved installation space of approximately 0.5 meters allows the defect detection model in this embodiment to have a buffer time of about 1 second, enhancing the reliability of defect detection. Even occasional detection delays will not affect the normal operation of the production line. Furthermore, installing an external encoder on the conveyor belt motor allows the generation of a trigger signal based on a preset absolute distance the conveyor belt moves, ensuring consistent battery cell images are obtained at different conveyor speeds. This avoids image stretching or compression caused by fixed-frequency triggering methods when the conveyor belt speed changes.

[0056] S203: Input the image of the battery cell to be detected into the first defect detection model to obtain the main defect, wherein the main defect includes a continuous defect region.

[0057] In this embodiment, the first defect detection model can be used to identify and match defects in the input battery cell to be inspected, and output the main defect based on the defect features. Depending on the training method of the first defect detection model, obtaining the main defect can include marking the main defect in the image of the battery cell to be inspected, outputting the information of the main defect, or a combination of both methods. In this embodiment, the main defect includes a continuous defect region, which includes a region with continuous defect features formed by the same defect. The defect features include defect lines or defect surfaces formed by the defect. Specifically, the continuous defect region can include a continuous hidden crack defect line or a continuous black spot defect surface. In some embodiments, the continuous defect region can also include the bounding rectangle of the continuous defect region. In other embodiments, the main defect can also include a main defect type and / or a main defect source point. The main defect type can include oblique hidden cracks, intersecting hidden cracks, etc., and the main defect source point can include defect endpoints or defect intersections.

[0058] In this embodiment of the application, after acquiring the image of the battery cell to be inspected, a training dataset for establishing a first defect detection model is further included, and the first defect detection model is trained based on the training dataset. In some embodiments, after acquiring the image of the battery cell to be inspected, qualified sample images and defective sample images can be further acquired, and data augmentation operations such as rotation, mirroring, flipping, and grayscale transformation can be performed on the sample images. Different defects in the battery cell need to be distinguished and the defect type needs to be determined. Different defect types also correspond to different defect source points. For example, for cross-crack defects, the cross point is used as the defect source point rather than the endpoint.

[0059] In some embodiments, continuous defect regions are labeled when processing defect sample images. In other embodiments, defect types and defect source points may also be labeled in the defect sample images.

[0060] In this embodiment, the first defect detection model may include model components trained using machine learning methods, or it may be a combination of multiple models or computing engines; this application does not impose any limitations on this. The machine learning methods may include K-nearest neighbors, perceptron, support vector machine, logistic regression, maximum entropy, etc., and the corresponding generated model components may include Naive Bayes, Hidden Markov Models, etc. Of course, in other embodiments, the machine learning methods may also include deep learning methods, reinforcement learning methods, etc., and the generated model components may include Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), LeNet, ResNet, Long Short-Term Memory (LSTM), Bi-LSTM, etc., and this application does not impose any limitations on this.

[0061] S205: Input the image of the battery cell to be detected into the second defect detection model to obtain local defects. The local defects include at least one defect source point, and the defect source point includes a defect endpoint or a defect intersection point.

[0062] In traditional solar cell defect detection techniques, different defects in solar cells vary in size and shape. For example, the length and curve shape of microcracks differ, and even a localized area of ​​a defect is still considered a defect. Because the dataset used to train the defect detection model cannot cover all defects, the final detection result may be localized to a specific area of ​​the defect, leading to poor regression performance of the detection algorithm and consequently, low accuracy in defect detection. Figure 4 As shown, the actual defect in the solar cell is region 11, and region 13 is the defect detection result of the detection model. Region 13 corresponds to the local defect region 12 of the actual defect, and the detection model did not detect the actual defect region 11. On the other hand, a local region of one type of defect may be another type of defect. For example, the local region of a cross-shaped hidden crack may be an oblique hidden crack. When the target region detected by the detection algorithm before regression is located in the local region, the cross-shaped hidden crack will be misclassified as an oblique hidden crack, resulting in a decrease in the accuracy of defect detection. Furthermore, traditional solar cell defect detection models have difficulty determining the location of defects in the solar cell. Neither target detection algorithms nor semantic segmentation algorithms can achieve high regression accuracy. As mentioned above, the training set cannot cover all defects, which also leads to a decrease in regression accuracy. Therefore, the defect detection model also has difficulty detecting defects located at the edge or inside the solar cell.

[0063] Based on this, in this embodiment of the application, the image of the solar cell to be inspected is input into a second defect detection model to obtain local defects. The second defect detection model can be used to identify and match local defects in the input solar cell to be inspected, and output local defects based on the features of the local defects. The local defect includes at least one defect source point, which includes a defect endpoint or a defect intersection. The defect endpoint or defect intersection can include an interior point located inside the solar cell, or an edge point located at the edge of the solar cell. In some embodiments, the local defect can also include a locally continuous defect region containing the defect source point, or the bounding rectangle of the locally continuous defect region. In other embodiments, the local defect can also include a local defect type corresponding to the defect source point. The local defect type can include oblique hidden cracks, intersecting hidden cracks, etc.

[0064] In this embodiment, after acquiring the image of the battery cell to be detected, the method further includes establishing a training dataset for a second defect detection model, and training the second defect detection model based on the training dataset. In some embodiments, after acquiring the image of the battery cell to be detected, qualified sample images and local defect sample images can be further acquired, and data enhancement operations such as rotation, mirroring, flipping, and grayscale transformation can be performed on the sample images. In some embodiments, local defect sample images containing at least one defect source point are selected, and the defect source point is labeled. Specifically, labeling the defect source point includes labeling defect endpoints or defect intersections. In some embodiments, labeling the defect source point also includes labeling defect interior points or defect edge points. In other embodiments, the corresponding defect type can also be labeled based on the defect source point. In still other embodiments, local continuous defect regions containing the defect source point can also be labeled.

[0065] In this embodiment of the application, the training method of the second defect detection model can refer to the training method of the first defect detection model in step S203 above, and will not be repeated here.

[0066] S207: Determine the target defect based on the main defect and the local defect.

[0067] In this embodiment, a target defect can be determined based on the main defect and the local defects. The target defect includes the main defect and / or the local defects. Since the second detection model determines the local defect region or defect type based on the defect source point, its accuracy in detecting defect source points and defect types is higher than that of the first detection model. The local defects are compared with the main defect output by the first detection model to obtain a comparison result, and the target defect can be determined based on the comparison result. In some embodiments, the target defect includes at least one of a target defect region, a target defect source point, and a target defect type, wherein the target defect region includes a continuous defect region.

[0068] In some embodiments, the target defect can be determined based on the degree of overlap between the main defect region and the local defect region. The local defects output by the second defect model include at least one defect source point, which may be located in the same or different defects. Therefore, the degree of overlap can be used to determine whether the local defect is consistent with the main defect. If they are consistent, the local defect is determined to be the target defect, and the main defect region is also determined to be the target defect region; if they are inconsistent, the main defect is determined to be the target defect.

[0069] The cell defect detection method provided in this application detects main defects and local defects using first and second detection models, and determines the target defect based on the main and local defects. On the one hand, it fully considers the inherent relationship between the characteristics of main defects and local defects, using local defects to supplement or correct omissions or errors in the main defect detection results. On the other hand, local defects can more accurately express the location of the defect, providing important reference for analyzing the causes of cell defects, enabling faster location of the defect-causing link in the production process, and reducing potential further losses during cell production. By fully considering the correlation between main defect characteristics and local defects, the cell defect detection method provided in this application effectively improves the accuracy and efficiency of cell defect detection.

[0070] In this embodiment of the application, to determine whether both the main defect and the local defect belong to the target defect, a method for determining the target defect is also provided. In step S207, determining the target defect based on the main defect and the local defect includes:

[0071] S301: Determine the main defect region of the main defect and the local defect region of the local defect.

[0072] S303: Determine the target defect based on the overlap between the main defect region and the local defect region.

[0073] In this embodiment, multiple defects may exist in the same battery cell to be tested. When the first defect detection model outputs the main defect, it may mark multiple defect regions. When the second defect detection model outputs the local defect, it may also mark multiple defect source points. Therefore, the target defect can be determined by identifying the main defect region of the main defect and the local defect region of the local defect, and then determining the target defect based on the overlap between the main defect region and the local defect region, that is, determining the main defect and the local defect that belong to the target defect. In some embodiments, if the overlap between the main defect region and the local defect region is greater than a preset threshold, the defect region of the main defect and the defect source point of the local defect are determined as the target defect; if the overlap is not greater than the preset threshold, the main defect is determined as the target defect. In other embodiments, determining the target defect based on the overlap between the main defect region and the local defect region in step S303 includes:

[0074] S3031: If the overlap is greater than zero, then the target defect is determined based on the main defect region and the local defects.

[0075] S3033: If the overlap is not greater than zero, then the target defect is determined based on the main defect.

[0076] In this embodiment, the overlap ratio can be determined by calculating IOU (Intersection over Union) or one or more of its derivative concepts, such as DIOU (Distance-Intersection over Union). In some embodiments of this application, determining the main defect region of the main defect includes determining the bounding rectangle of the main defect region as A, and determining the local defect region of the local defect includes determining the bounding rectangle of the local defect region as B. Then, the overlap ratio is determined as IOU, which can be determined by equation (1):

[0077]

[0078] In equation (1), |A∩B| is the area of ​​the rectangle where rectangle A and rectangle B intersect, and |A∪B| is the area of ​​the circumscribed rectangles of rectangle A and rectangle B.

[0079] If IOU > 0, then the defect region of the main defect and the defect source point of the local defect are determined as the target defect, and the target defect may also include the defect type corresponding to the local defect source point; if IOU ≤ 0, then the main defect is determined as the target defect.

[0080] In this embodiment, the target defect is determined by the main defect region and the local defect region. The main defect and the local defects included in the target defect can be determined. The overall distribution of the target defect can be determined by the main defect. The defect source point of the target defect can be accurately determined by the target defect. The type of the target defect can also be determined. Therefore, the accuracy of cell defect detection can be further improved.

[0081] To further improve the detection accuracy of battery cell defect sources, in this embodiment of the application, the determination of the target defect in step S207 includes:

[0082] S401: If the local defect is a hidden oblique crack, and the local defect includes at least two hidden oblique crack source points, then determine the position of the hidden oblique crack source points in the image of the battery cell to be tested.

[0083] S403: If all the oblique hidden crack source points are internal endpoints, then the main defect region, the oblique hidden crack, and all internal endpoints are determined as the target defect.

[0084] S405: If the oblique hidden crack source point includes at least one boundary endpoint, then the main defect region, the oblique hidden crack, and all boundary endpoints are determined as the target defect.

[0085] In this embodiment of the application, if the local defect output by the second defect detection model is a hidden oblique crack and includes at least two hidden oblique crack source points, the defect source points causing the hidden oblique crack can be further determined by judging the position of the hidden oblique crack source points in the image of the battery cell to be detected. In a specific embodiment, such as Figure 5 As shown, the local defect source points include interior points as shown in 22 and edge points as shown in 23. Therefore, edge point 23 is preferentially determined as the defect source point of the target defect. Furthermore, the local defect corresponding to edge point 23 is a hidden oblique crack, and the main defect region is 21. Therefore, 21, 23, and the hidden oblique crack are determined as the target defect. In another embodiment, as... Figure 6 As shown, the identified target defect includes the oblique hidden crack initiation point 17 output by the second detection model, where 17 is the endpoint of the cell boundary. The target defect also includes the main defect region 18 output by the first detection model.

[0086] If a microcrack in a solar cell contains both a boundary microcrack origin and an internal microcrack origin, the location that causes the microcrack is typically the boundary microcrack origin. In this embodiment, when the local defects detected by the second detection model include both boundary and internal microcrack origins, the boundary endpoint is preferentially identified as the origin of the target defect. This allows for accurate location of the defect-inducing origin among multiple microcrack origins, further improving the accuracy of solar cell defect detection results.

[0087] To further improve the detection accuracy of battery cell defect types, in this embodiment of the application, the determination of the target defect by the local defect in step S207 includes:

[0088] S501: If the local defect includes intersecting hidden cracks and oblique hidden cracks, the main defect region, the intersecting hidden cracks, and all the source points of the intersecting hidden cracks are determined as the target defect.

[0089] In this embodiment, if the local defects output by the second defect detection model include multiple defect types, the actual defect type of the target defect can be further determined. Since the local manifestation of cross-crack defects in solar cells is oblique cracks, if the local defects output by the second defect detection model include both cross-crack origin points and oblique crack origin points, the defect type of the target defect is determined to be a cross-crack. In a specific embodiment, such as... Figure 7 As shown, the local defects output by the second defect detection model include oblique hidden crack source point edge point 31, oblique hidden crack source point edge point 32, oblique hidden crack source point interior point 33, oblique hidden crack source point interior point 34, oblique hidden crack source point interior point 35, and intersecting hidden crack source point intersection point 36. The main defects output by the first defect detection model include intersecting hidden cracks 37 in the main defect region. It is understandable that this induces... Figure 5 The source point of the defect should be the intersection point 36. Therefore, other detected defect source points, edge points, and inner points are excluded, and the main defect region 37, the cross-crack, and all cross-crack source points 36 are identified as the target defect.

[0090] If a microcrack in a solar cell contains both a boundary microcrack origin and an internal microcrack origin, the location that causes the microcrack is typically the boundary microcrack origin. In this embodiment, when the local defects detected by the second detection model include both boundary and internal microcrack origins, the boundary endpoint is preferentially identified as the origin of the target defect. This allows for accurate location of the defect-inducing origin among multiple microcrack origins, further improving the accuracy of solar cell defect detection results.

[0091] In this embodiment of the application, after determining the target defect based on the main defect and the local defect in step S207, the method further includes:

[0092] S601: Determine the size ratio of the image of the battery cell to be tested and the size ratio of the battery cell to be tested.

[0093] S603: Determine the actual distribution of the target defect on the battery cell to be inspected based on the location information of the target defect in the image of the battery cell to be inspected and the size ratio.

[0094] In this embodiment, by determining the size ratio between the image of the battery cell to be inspected and the battery cell itself, the mapping from pixel scale to physical scale can be determined. Then, based on the location information of the target defect in the image of the battery cell to be inspected and the size ratio, the actual distribution of the target defect on the battery cell to be inspected can be determined, thereby obtaining the distribution of the target defect on the corresponding battery cell to be inspected. In a specific embodiment of this application, the distribution of the target defect on the battery cell to be inspected is as follows: Figure 8 As shown, from Figure 8 The defect distribution map can identify where defects are concentrated, and these defect-prone locations are typically where a particular mechanism operates during cell production. Therefore, the defect-generating mechanism can be quickly located based on the cell defect distribution map, allowing for timely reduction of potential losses in subsequent cell production.

[0095] In this embodiment of the application, after determining the target defect based on the main defect and the local defect in step S207, the method further includes:

[0096] S701: Determine the comparison result between the target defect and the preset defect.

[0097] S703: Based on the comparison results, determine the unusable defective battery cells and remove the defective battery cells.

[0098] In this embodiment, the preset defect may include defects that render the solar cell unusable. In some embodiments, the preset defect may include oblique microcracks with a length greater than a preset threshold, intersecting microcracks with a total length greater than a preset threshold, and black spot defects with a total area greater than a preset threshold. If the target defect is determined to be an unusable defect after comparison with the preset defect, the solar cell to be tested containing the unusable defect can be rejected, ensuring that all solar cells output from the production line are usable. This saves on the subsequent manual rejection process for unusable solar cells, reduces labor costs, and improves the efficiency of selecting usable solar cells.

[0099] The following specific embodiment illustrates the overall process of battery cell testing. Figure 9As shown, after a solar cell on the production line receives a photoelectric trigger and induces photoluminescence, the controller generates a trigger signal and sends it to the PLC device. Simultaneously, an external encoder controls a linear scan camera to acquire an image of the solar cell to be inspected. After the solar cell to be inspected is imaged, a training dataset is built based on qualified sample images and defective sample images to train a target defect detection model. After training, the image of the solar cell to be inspected is input into the target defect detection model, and a detection result image containing the target defect is obtained. Simultaneously, after receiving the trigger signal, the PLC device synchronously generates an identifier ID for the solar cell to be inspected and binds the identifier ID with the detection result image to form an ID-result pair, which is then sent to the PLC device. Based on the ID-result pair and a preset defect, the PLC device rejects solar cells on the production line that contain unusable defects.

[0100] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0101] Based on the same inventive concept, this application also provides a battery cell defect detection device 900 for implementing the aforementioned battery cell defect detection method. The solution provided by this device is similar to the implementation described in the above method; therefore, the specific limitations of one or more embodiments of the battery cell defect detection device 900 provided below can be found in the limitations of the battery cell defect detection method described above, and will not be repeated here.

[0102] In one embodiment, such as Figure 10 As shown, a battery cell defect detection device 900 is provided, comprising:

[0103] Image acquisition module 901 is used to acquire images of the battery cell to be inspected;

[0104] The first defect detection module 902 is used to input the image of the battery cell to be detected into the first defect detection model to obtain the main defect, wherein the main defect includes a continuous defect region;

[0105] The second defect detection module 903 is used to input the image of the battery cell to be detected into the second defect detection model to obtain local defects. The local defects include at least one defect point, and the defect point includes a defect endpoint or a defect intersection.

[0106] The defect determination module 904 is used to determine the target defect based on the main defect and the local defects.

[0107] Each module in the aforementioned battery cell defect detection device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0108] In one embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of the battery cell defect detection method described in any of the preceding embodiments.

[0109] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the cell defect detection method described in any of the preceding embodiments.

[0110] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0111] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0112] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0113] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for detecting defects in solar cells, characterized in that, The method includes: Acquire images of the battery cells to be inspected; The image of the battery cell to be detected is input into the first defect detection model to obtain the main defect, which includes a continuous defect region; The image of the battery cell to be detected is input into the second defect detection model to obtain local defects. The local defects include at least one defect point, and the defect point includes a defect endpoint or a defect intersection. The target defect is determined based on the main defect and the local defects; The step of determining the target defect based on the main defect and the local defects includes: determining the main defect region of the main defect and the local defect region of the local defects; and determining the target defect based on the overlap between the main defect region and the local defect region. Determining the target defect based on the overlap between the main defect region and the local defect region includes: if the overlap is greater than zero, then determining the target defect based on the main defect region and the local defect. The step of determining the target defect based on the main defect region and the local defects includes: if the local defect is a hidden oblique crack, and the local defect includes at least two hidden oblique crack source points, then determining the position of the hidden oblique crack source points in the image of the battery cell to be inspected; if the hidden oblique crack source points are all internal endpoints, then determining the main defect region, the hidden oblique crack, and all internal endpoints as the target defect; if the hidden oblique crack source points include at least one boundary endpoint, then determining the main defect region, the hidden oblique crack, and all boundary endpoints as the target defect. If the overlap is not greater than zero, then the target defect is determined based on the main defect.

2. The method according to claim 1, characterized in that, Determining the target defect based on the local defect includes: If the local defect includes intersecting hidden cracks and oblique hidden cracks, the main defect region, the intersecting hidden cracks, and all the source points of the intersecting hidden cracks are determined as the target defect.

3. The method according to claim 1, characterized in that, After determining the target defect based on the main defect and the local defects, the method further includes: Determine the size ratio between the image of the battery cell to be tested and the size of the battery cell to be tested; The actual distribution of the target defect on the battery cell under test is determined based on the location information of the target defect in the image of the battery cell under test and the size ratio.

4. The method according to claim 3, characterized in that, After determining the target defect based on the main defect and the local defects, the method further includes: Determine the comparison result between the target defect and the preset defect; Based on the comparison results, unusable defective battery cells are identified and removed.

5. A battery cell defect detection device, characterized in that, The device includes: Image acquisition module, used to acquire images of the battery cell to be inspected; The first defect detection module is used to input the image of the battery cell to be detected into the first defect detection model to obtain the main defect, wherein the main defect includes a continuous defect region; The second defect detection module is used to input the image of the battery cell to be detected into the second defect detection model to obtain local defects. The local defects include at least one defect point, and the defect point includes a defect endpoint or a defect intersection. The defect determination module is used to determine the target defect based on the main defect and the local defects; The step of determining the target defect based on the main defect and the local defects includes: determining the main defect region of the main defect and the local defect region of the local defects; and determining the target defect based on the overlap between the main defect region and the local defect region. Determining the target defect based on the overlap between the main defect region and the local defect region includes: if the overlap is greater than zero, then determining the target defect based on the main defect region and the local defect. The step of determining the target defect based on the main defect region and the local defects includes: if the local defect is a hidden oblique crack, and the local defect includes at least two hidden oblique crack source points, then determining the position of the hidden oblique crack source points in the image of the battery cell to be inspected; if the hidden oblique crack source points are all internal endpoints, then determining the main defect region, the hidden oblique crack, and all internal endpoints as the target defect; if the hidden oblique crack source points include at least one boundary endpoint, then determining the main defect region, the hidden oblique crack, and all boundary endpoints as the target defect. If the overlap is not greater than zero, then the target defect is determined based on the main defect.

6. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 4.

Images