Battery sample reliability analysis method and device, computer equipment and storage medium

Abstract

The invention relates to a battery sample reliability analysis method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring an electroluminescent image of an aged battery sample; dividing the electroluminescent image into a plurality of image blocks, and determining respective gray coefficients of the image blocks; for each image block, determining the reliability influence weight of the image block on the battery sample according to the position of the image block in the battery sample; and based on each reliability influence weight, carrying out statistical analysis on the respective gray coefficient of each image block, and determining a reliability analysis result of the battery sample. Wherein the electroluminescent image is a grayscale image; the gray coefficient is determined according to the gray value of the image block and is used for representing the damage degree of the position where the image block is located. By adopting the method, the accuracy of a reliability analysis result can be improved.

Description

Technical Field

[0001] This application relates to the field of reliability analysis technology, and in particular to a method, apparatus, computer equipment, computer-readable storage medium, and computer program product for reliable analysis of battery samples. Background Technology

[0002] During the manufacturing process of solar cells, defects are inevitably generated or introduced. These defects cause non-uniformity in the cell's structure, seriously threatening its conversion efficiency, lifespan, and reliability. Therefore, it is necessary to conduct reliability analysis on cell samples to provide constructive feedback for optimizing solar cell processes and improving reliability.

[0003] Electroluminescence (EL) imaging is a widely used reliability assessment technique in the field of solar cells. It enables non-destructive, visual inspection of cells and is suitable for characterizing and analyzing surface defects. Currently, reliability analysis of cell samples based on this technology mainly relies on visual observation by experts, which is time-consuming and labor-intensive, and the accuracy of the analysis results depends on personal experience, resulting in significant errors. Summary of the Invention

[0004] Therefore, it is necessary to provide a battery sample reliability analysis method, apparatus, computer equipment, computer-readable storage medium, and computer program product that can improve the accuracy of reliability analysis results, in order to address the above-mentioned technical problems.

[0005] Firstly, this application provides a method for reliability analysis of battery samples. The method includes:

[0006] Acquire electroluminescence images of battery samples after aging treatment; the electroluminescence images are grayscale images;

[0007] The electroluminescent image is divided into multiple image blocks, and the grayscale coefficient of each image block is determined. The grayscale coefficient is determined based on the grayscale value of the image block and is used to characterize the degree of damage at the location of the image block.

[0008] For each image block, the influence weight of the image block on the reliability of the battery sample is determined according to the position of the image block in the battery sample;

[0009] Based on the reliability impact weights of each image block, statistical analysis is performed on their respective grayscale coefficients to determine the reliability analysis results of the battery sample.

[0010] In one embodiment, determining the reliability impact weight of each image block on the battery sample based on its position within the battery sample includes:

[0011] Determine the weighted influence factor that matches the sample type of the battery sample;

[0012] For each image block, the weight coefficient corresponding to the weight influence factor is determined based on the position of the image block in the battery sample.

[0013] The weighting coefficients are used to determine the influence weight of the image patch on the reliability of the battery sample.

[0014] In one embodiment, the sample type of the battery sample includes a battery module; the weighting influence factor includes a structural factor and a distance factor; determining the weighting coefficient corresponding to the weighting influence factor based on the position of the image patch in the battery sample includes:

[0015] Based on the location of the image block in the battery sample, the local battery structure contained in the image block is determined;

[0016] Based on the importance of the local battery structure in the battery assembly, the structure weight coefficient corresponding to the structure factor of the image block is determined; the structure weight coefficient is positively correlated with the importance.

[0017] Based on the position of the image patch in the battery sample, a distance weight coefficient corresponding to the distance factor of the image patch is determined; the distance weight coefficient is positively correlated with the distance between the position and the edge of the battery assembly.

[0018] In one embodiment, determining the distance weight coefficient corresponding to the distance factor of the image patch based on the position of the image patch in the battery sample includes:

[0019] Determine the equivalent center and equivalent radius of the battery sample;

[0020] Based on the position of the image block in the battery sample, determine the equivalent distance between the image block and the equivalent center;

[0021] Based on the ratio between the equivalent distance and the equivalent radius, the distance weight coefficient corresponding to the distance factor of the image patch is determined; the distance weight coefficient is inversely correlated with the ratio.

[0022] In one embodiment, determining the grayscale coefficient of each of the image blocks includes:

[0023] Obtain the grayscale value of each pixel in the image block;

[0024] The statistical results of each gray value are determined as the image block gray value of the image block;

[0025] The grayscale value of the image block is matched with the image grayscale scale configured for the battery sample to determine the grayscale coefficient corresponding to the grayscale value of the image block.

[0026] In one embodiment, the step of statistically analyzing the grayscale coefficients of each image patch based on the respective reliability impact weights to determine the reliability analysis result of the battery sample includes:

[0027] Based on the reliability influence weight of each image block, the grayscale coefficients are weighted and summed to obtain the statistical value of the damage degree of the battery sample.

[0028] Based on the number of each image block, the above-mentioned damage statistics are averaged to obtain the reliability analysis results of the battery sample.

[0029] In one embodiment, the method further includes:

[0030] Obtain sample information for each of the multiple battery samples participating in the aging process; the sample information includes parameter information corresponding to each of the multiple sample parameters.

[0031] Based on the reliability analysis results of each battery sample, factorial analysis is performed on the reliability of each battery sample to determine the factorial analysis results for each sample parameter.

[0032] Secondly, this application also provides a battery sample reliability analysis device. The device includes:

[0033] The acquisition module is used to acquire electroluminescence images of battery samples after aging treatment; the electroluminescence images are grayscale images.

[0034] An image segmentation module is used to divide the electroluminescent image into multiple image blocks and determine the grayscale coefficient of each image block; the grayscale coefficient is determined based on the grayscale value of the image block and is used to characterize the degree of damage at the location of the image block;

[0035] The reliability impact weight determination module is used to determine the reliability impact weight of each image block on the battery sample based on the position of the image block in the battery sample.

[0036] The statistical analysis module is used to perform statistical analysis on the grayscale coefficients of each image block based on the reliability influence weights, and to determine the reliability analysis results of the battery sample.

[0037] Thirdly, this application also provides a computer device. This computer device includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described method.

[0038] Fourthly, this application also provides a computer-readable storage medium. This computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the steps of the above-described method.

[0039] Fifthly, this application also provides a computer program product. This computer program product includes a computer program that, when executed by a processor, implements the steps of the above-described method.

[0040] The aforementioned battery sample reliability analysis method, apparatus, computer equipment, computer-readable storage medium, and computer program product divide the electroluminescence image of the aged battery sample into multiple image blocks. For each image block, on the one hand, based on the grayscale value of the image block, a grayscale coefficient is determined to characterize the degree of damage at the location of the image block, enabling a finer-grained assessment of the degree of damage and improving the accuracy of the reliability analysis results. On the other hand, based on the location of the image block in the battery sample, the reliability influence weight of the image block on the battery sample is determined. Based on each reliability influence weight, statistical analysis is performed on the grayscale coefficients of each image block to determine the reliability analysis results of the battery sample. This allows different weights to be assigned to different image blocks according to actual conditions, which is beneficial to further improving the accuracy of the reliability analysis results. Attached Figure Description

[0041] Figure 1 This is a diagram illustrating the application environment of a battery sample reliability analysis method in one embodiment.

[0042] Figure 2 This is a flowchart illustrating a battery sample reliability analysis method in one embodiment;

[0043] Figure 3 This is a schematic diagram of the battery assembly configuration in one embodiment;

[0044] Figure 4 This is a schematic diagram of the electroluminescence image of a battery assembly in one embodiment;

[0045] Figure 5 This is a schematic diagram of an EL image grayscale ruler in one embodiment;

[0046] Figure 6 This is a schematic diagram illustrating the modeling of the severity index of EL image reliability in one embodiment;

[0047] Figure 7 This is a schematic diagram illustrating the process of determining the reliability severity index of an EL image in one embodiment;

[0048] Figure 8 This is a statistical table of the reliability severity index of battery samples with different materials and processes in one embodiment;

[0049] Figure 9 This is a schematic diagram illustrating the significance analysis results of the reliability severity index factor of EL images in one embodiment.

[0050] Figure 10 This is a main effect plot of the reliability severity index of EL images in one embodiment;

[0051] Figure 11 This is a schematic diagram illustrating the interaction of the severe index factor on the reliability of EL images in one embodiment;

[0052] Figure 12 This is a structural block diagram of a battery sample reliability analysis device in one embodiment;

[0053] Figure 13 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0054] 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.

[0055] In one embodiment, the battery sample reliability analysis method provided in this application can be applied to, for example... Figure 1In the application environment shown, the acquisition device 101 can acquire electroluminescence images of the battery sample; the computer device 102 can interact with the acquisition device 101 to obtain the electroluminescence images acquired by the acquisition device 101. The connection between the computer device 102 and the acquisition device 101 can be wired or wireless. The wireless connection can be, for example, Bluetooth, Wi-Fi, etc., and is not limited here. The computer device 102 can be a terminal or a server. The terminal can be, but is not limited to, various desktop computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can be smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, etc. Portable wearable devices can be smartwatches, smart bracelets, head-mounted devices, etc. A server can be a standalone physical server, a server cluster or distributed system consisting of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms.

[0056] In the process of executing the battery sample reliability analysis method, computer device 102: acquires an electroluminescence image of the battery sample after aging treatment; divides the electroluminescence image into multiple image blocks and determines the grayscale coefficient of each image block; for each image block, determines the reliability influence weight of the image block on the battery sample based on its position in the battery sample; based on each reliability influence weight, statistically analyzes the grayscale coefficient of each image block to determine the reliability analysis result of the battery sample. The electroluminescence image is a grayscale image; the grayscale coefficient is determined based on the grayscale value of the image block and is used to characterize the degree of damage at the location of the image block.

[0057] In one embodiment, such as Figure 2 As shown, a method for reliability analysis of battery samples is provided, which can be applied to... Figure 1 Taking computer device 102 as an example, the following steps are included:

[0058] Step S202: Obtain an electroluminescence image of the battery sample after aging treatment.

[0059] Electroluminescence, also known as electroluminescence (EL), is a physical phenomenon where an electric field is generated by applying a voltage to two electrodes. Electrons excited by this field collide with luminescent centers, causing energy level transitions, changes, and recombination, resulting in light emission. The battery sample can be a bare cell or a battery module obtained by encapsulating battery cells. Taking the case where the battery sample is a battery module as an example... Figure 3 As shown, the glass panel 1, the encapsulating film 2, and the backplate 3 form a sandwich structure, wherein the battery cell 4 is encapsulated in the encapsulating film.

[0060] Specifically, the battery sample can first undergo an aging process, and then be subjected to electroluminescence treatment to obtain an electroluminescence image of the battery sample. The aging process can involve subjecting the battery sample to charge-discharge cycles and load placement under certain conditions to simulate the aging process under actual usage conditions. Furthermore, the electroluminescence image is a grayscale image. A grayscale image is an image where each pixel has only one sampled color. These images are typically displayed as grayscale ranging from the darkest black to the brightest white; theoretically, this sampling can be any shade of color, even different shades of color at different brightness levels. Grayscale images differ from black-and-white images. In the field of computer graphics, black-and-white images only have black and white colors, while grayscale images have many levels of color depth between black and white. Figure 4 As shown, after electroluminescence processing, an electroluminescence image of the battery sample can be obtained. The location of the battery cell is presented as a grayscale image.

[0061] It should be noted that the battery sample has a three-dimensional structure, while the electroluminescence image of the battery sample is a two-dimensional image obtained by image acquisition along the thickness direction of the battery sample. For example... Figure 3 In this process, images are acquired along the Z-axis of the battery sample to obtain electroluminescence images of the battery sample in the XY plane. For ease of understanding, the following explanations will use the case where the grayscale image is a two-dimensional image as an example.

[0062] It is understandable that for a bare cell, since the entire cell sample possesses electroluminescence, the image size of the electroluminescence image of the cell sample is consistent with the size of the cell sample. For a cell assembly, since the components other than the cells do not possess electroluminescence properties, the obtained electroluminescence image is actually the electroluminescence image of the cells in the cell sample; that is, the size of the electroluminescence image is smaller than the size of the cell sample.

[0063] Step S204: Divide the electroluminescent image into multiple image blocks and determine the grayscale coefficient of each image block.

[0064] The grayscale coefficient, determined based on the grayscale values ​​of an image patch, characterizes the degree of damage at the location of that patch. Specifically, there is a mapping relationship between the grayscale values ​​and grayscale coefficients of an electroluminescent image, which can be recorded using an EL image grayscale ruler. This grayscale ruler can be either a discrete or continuous grayscale ruler. Figure 5In the discrete grayscale scale shown, 0% indicates the most severe damage to the solar cell, and 100% indicates no damage. It is understood that in other embodiments, 0% could represent no damage and 100% could represent the most severe damage; this is not a limitation. Specifically, the computer device can divide the electroluminescent image into multiple image blocks and determine the grayscale coefficient of each image block based on the grayscale value of each pixel contained within the image block.

[0065] The specific method for dividing the image into blocks is not unique. Optionally, the computer device can divide the electroluminescent image into multiple image blocks based on a set grid shape. This set grid shape can be square, triangular, or other shapes, and the grid shape and size of different image blocks can be the same or different. Optionally, the computer device can divide the image blocks based on the structural information of the battery cells in the battery sample to ensure that the structures at different locations within the same image block are similar, so that a more accurate reliability influence weight can be determined subsequently.

[0066] The specific method for determining the grayscale coefficient is not unique. For example, a computer device can match the grayscale value at the center of an image block with a grayscale ruler to determine the grayscale coefficient of the image block. This image block center can be, for example, the geometric center or centroid of the image block. In some embodiments, the computer device can also determine the grayscale values ​​of multiple pixels in the image block, and then determine the grayscale value at the center of the image block by averaging or performing linear interpolation (similar to the shape function in finite element theory). Alternatively, the computer device can acquire the grayscale values ​​of each pixel in the image block and determine the image block grayscale as the statistical result of each grayscale value; then match the image block grayscale with an image grayscale ruler configured for the battery sample to determine the grayscale coefficient corresponding to the image block grayscale.

[0067] It is understandable that the denser the grid division of an image patch, the more accurate the determined grayscale value. Based on this, the grid can be finer for areas with higher reliability weights, i.e., important areas in the battery sample, to improve accuracy.

[0068] Step S206: For each image block, determine the weight of the image block's influence on the reliability of the battery sample based on the image block's position in the battery sample.

[0069] The reliability impact weight characterizes the degree of influence of the image patch's location on the overall reliability of the battery sample, and is positively correlated with the importance of the image patch's location. Specifically, engineers can score and evaluate each location on the battery sample according to the specific needs of the aging experiment, determining the reliability impact weight of each location. The computer equipment can then obtain the reliability impact weight of each location and determine the reliability impact weight of the image patch based on the location correspondence. Alternatively, the computer equipment can directly use the RPN (risk priority number) from DFMEA (Design Failure Mode and Effects Analysis) as the reliability impact weight. The reliability impact weight ranges from 0 to 100%, with a larger number indicating a greater risk and requiring priority and active improvement. 0 represents the least important location; 100% represents the most important location, the area of ​​strong interest in the corresponding experiment.

[0070] Furthermore, by combining factors such as the experimental objective or the sample type of the battery sample, a weighted influence factor matching the battery sample can be determined, thereby determining the reliability influence weight of each image block in the battery sample. In some possible implementations, for the battery assembly, the computer equipment can also determine the reliability influence weight based on the distance between the image block's location and the edge of the battery assembly. Specifically, during the aging process of the battery assembly, moisture may penetrate the encapsulation film between the glass panel and the backsheet and enter the assembly's interior. The farther the image block is from the edge of the battery assembly, the lower the probability of moisture intrusion and damage during the aging process. Therefore, if the image block is damaged, the overall reliability of the battery assembly is likely to be poor. However, for image blocks closer to the edge of the battery assembly, the probability of moisture intrusion and damage is higher. Even if the image block is damaged, it is impossible to judge the overall reliability of the entire battery assembly based on this. Based on this, image blocks farther from the edge of the battery assembly can be assigned a relatively larger reliability influence weight; that is, the reliability influence weight is positively correlated with the distance between the image block's location and the edge of the battery assembly.

[0071] Step S208: Based on the reliability impact weights, perform statistical analysis on the grayscale coefficients of each image block to determine the reliability analysis results of the battery sample.

[0072] Specifically, the computer equipment can perform statistical analysis on the grayscale coefficients of each image patch based on its respective reliability influence weight to determine the reliability analysis result of the battery sample. The specific algorithm for statistical analysis can be to multiply each grayscale coefficient by its corresponding reliability influence weight to obtain a weighted value for the grayscale coefficients, and then use the statistical result of these weighted values ​​as the reliability analysis result of the battery sample. This statistical result can be, for example, any one of the following: sum, mean, median, or cluster center. Furthermore, after obtaining the weighted values ​​of the grayscale coefficients, outliers can be removed before subsequent statistical analysis. In summary, this embodiment does not limit the specific algorithm for statistical analysis.

[0073] It should be noted that electroluminescence images can also be obtained for battery samples before aging treatment, and by executing steps S204 to S208, the reliability analysis results of the battery samples before aging treatment can be obtained. Then, the computer equipment can perform further analysis by analyzing the reliability differences of battery samples before and after aging treatment.

[0074] The aforementioned battery sample reliability analysis method divides the electroluminescence image of the aged battery sample into multiple image blocks. For each image block, on the one hand, based on the grayscale value of the image block, a grayscale coefficient is determined to characterize the degree of damage at the location of the image block, enabling a finer-grained assessment of the degree of damage and improving the accuracy of the reliability analysis results. On the other hand, based on the position of the image block in the battery sample, the reliability influence weight of the image block on the battery sample is determined. Based on each reliability influence weight, the grayscale coefficients of each image block are statistically analyzed to determine the reliability analysis results of the battery sample. This method allows different weights to be assigned to different image blocks according to the actual situation, which is beneficial to further improving the accuracy of the reliability analysis results.

[0075] In one embodiment, step S206 includes: determining a weighted influence factor that matches the sample type of the battery sample; for each image block, determining a weight coefficient corresponding to the weighted influence factor based on the position of the image block in the battery sample; and determining the influence weight of the image block on the reliability of the battery sample based on the weight coefficient.

[0076] The battery sample can be either a bare cell or a battery module. Specifically, when the sample type is a battery module, the computer equipment can determine the risk factors related to the local structure of the battery and the distance factors related to the distance between the image patch and the module edge as weighted influence factors. When the sample type is a bare cell, since the probability of moisture intrusion is the same at all locations, the computer equipment can determine the risk factors related to the local structure of the battery as weighted influence factors. The risk factors characterize the importance of the local structure of the battery in the battery sample; the distance factors characterize the consistency between the reliability of the image patch's location and the reliability of the entire battery sample. After determining the weighted influence factors, the computer equipment can determine the weight coefficient corresponding to the weighted influence factor for each image patch based on its location in the battery sample, and then determine the influence weight of the image patch on the reliability of the battery sample based on the weight coefficient.

[0077] It can be understood that when there is only one weighting factor, the computer device can determine the weight coefficient corresponding to that weighting factor as the reliability impact weight of the image patch on the battery sample. When there are multiple weighting factors, the computer device can determine the reliability impact weight of the image patch on the battery sample by combining the weight coefficients corresponding to each weighting factor. That is, the reliability impact weight can be expressed as a function of each weighting factor. The specific form of this function is not unique; for example, it can include logarithmic functions, exponential functions, etc. For example, the computer device can also determine the reliability impact weight of the image patch on the battery sample by the sum, product, or quotient of each weight coefficient.

[0078] In the above embodiments, different weighting factors are matched for different sample types, which can meet the needs of different application scenarios and improve the flexibility of battery sample reliability analysis methods.

[0079] In one embodiment, the sample type of the battery sample includes a battery module; the weighting influence factors include a structure factor and a distance factor. In this embodiment, determining the weighting coefficients corresponding to the weighting influence factors based on the position of the image patch in the battery sample includes: determining the local battery structure contained in the image patch based on its position in the battery sample; determining the structure weighting coefficient corresponding to the structure factor of the image patch based on the importance of the local battery structure in the battery module; and determining the distance weighting coefficient corresponding to the distance factor of the image patch based on its position in the battery sample.

[0080] Among them, the structure weight coefficient is positively correlated with the importance of the local structure of the battery in the battery module; the distance weight coefficient is positively correlated with the distance between the position of the image patch in the battery sample and the edge of the battery module. Specifically, the more important the local structure of the battery, the greater the impact on the overall reliability of the battery module if the local structure is damaged. Based on this, relatively large structure weight coefficients can be assigned to important local structures of the battery.

[0081] For battery modules, the farther an image patch is from the module's edge, the lower the probability of moisture intrusion and damage during the aging process. Therefore, if this image patch is damaged, the overall reliability of the battery module is likely to be poor. However, for image patches closer to the module's edge, the probability of moisture intrusion and damage during the aging process is higher. Even if this image patch is damaged, it cannot be used to determine the overall reliability of the battery module. Therefore, image patches farther from the module's edge can be assigned a relatively larger distance weighting coefficient.

[0082] In the above embodiments, when the battery sample is a battery assembly, the reliability influence weight is determined by comprehensively considering the importance of the local battery structure contained in the image block, as well as the position of the image block in the battery assembly and the distance between the edge of the battery assembly. This can ensure the accuracy of the reliability influence weight, and thus ensure the accuracy of the reliability analysis results determined based on the reliability influence weight.

[0083] In one embodiment, determining the distance weight coefficient corresponding to the distance factor of the image patch based on the position of the image patch in the battery sample includes: determining the equivalent center and equivalent radius of the battery sample; determining the equivalent distance between the image patch and the equivalent center based on the position of the image patch in the battery sample; and determining the distance weight coefficient corresponding to the distance factor of the image patch based on the ratio between the equivalent distance and the equivalent radius.

[0084] The distance weighting coefficient is inversely correlated with the ratio. Specifically, the computer device can determine the equivalent center of the battery sample as its geometric center or centroid, and determine its equivalent radius based on its area. In a specific implementation, the equivalent radius of the battery sample can be expressed as... Where S1 is the projected area of ​​the battery sample on the image sampling plane. It can be understood that when the projection of the battery sample onto the XY plane is circular, the equivalent radius is the radius.

[0085] After determining the equivalent center, the computer device can determine the equivalent distance between the image block and the equivalent center based on the image block's location within the battery sample. For example, the computer device can determine the equivalent distance as the distance between the geometric center or centroid of the image block and the equivalent center. Taking the equivalent center as the origin and the image block's center coordinates as (x, y) as an example, the distance between the image block and the equivalent center is: Finally, the computer device determines the distance weight coefficient corresponding to the distance factor of the image patch based on the ratio between the equivalent distance and the equivalent radius.

[0086] In a specific implementation, the reliability impact weight of an image patch can be expressed as W / d, where W is the structure weight coefficient corresponding to the structure factor, 1 / d is the distance weight coefficient corresponding to the distance factor, and d can be expressed as:

[0087]

[0088] It should be noted that, in other embodiments, the distance weight coefficient corresponding to the distance factor of the image patch can also be determined based on the difference between the equivalent radius and the equivalent distance. The distance weight coefficient is positively correlated with this difference.

[0089] In the above embodiments, by calculating the equivalent radius and equivalent distance, and then determining the distance weighting coefficient, it can be applied to battery samples of different shapes, which is beneficial to improving the flexibility of battery sample reliability analysis methods.

[0090] In one embodiment, determining the grayscale coefficient of each image block includes: obtaining the grayscale value of each pixel in the image block; determining the image block grayscale of the image block by statistically analyzing the grayscale values; and matching the image block grayscale with an image grayscale scale configured for the battery sample to determine the grayscale coefficient corresponding to the image block grayscale.

[0091] The image grayscale ruler is used to record the mapping relationship between the grayscale values ​​and grayscale coefficients of the electroluminescent image. This image grayscale ruler can be a discrete grayscale ruler or a continuous grayscale ruler. As mentioned earlier, an electroluminescent image is a light-emitting image. Based on this, after dividing the image into blocks, the computer device can determine the multiple pixels contained in the image block and further obtain the grayscale value of each pixel. Then, statistical calculations are performed on each grayscale value, and the statistical result is determined as the image block grayscale. The specific method of statistical calculation can be, for example, averaging, taking the median, etc. Finally, the computer device matches the image block grayscale with the image grayscale ruler configured for the battery sample to determine the grayscale coefficient corresponding to the image block grayscale.

[0092] It should be noted that when the image grayscale scale is a continuous grayscale scale, the computer device can determine the grayscale coefficient of the image block based on the grayscale of the image block and the mapping relationship between the grayscale values ​​and grayscale coefficients recorded in the image grayscale scale; when the image grayscale scale is a discrete grayscale scale, the computer device can determine the grayscale coefficient corresponding to the image grayscale block by interpolating the grayscale values ​​in the image grayscale scale.

[0093] In the above embodiments, by combining the gray values ​​of each pixel in the image block to determine the gray coefficient of the image block, the accuracy of the gray coefficient can be ensured, thereby ensuring the accuracy of the reliability analysis results determined based on the gray coefficient.

[0094] In one embodiment, step S208 includes: weighting and summing each grayscale coefficient based on the reliability influence weight of each image block to obtain a statistical value of the damage degree of the battery sample; and averaging the statistical value of the damage degree according to the number of each image block to obtain the reliability analysis result of the battery sample.

[0095] Specifically, the computer equipment can perform a weighted summation of each grayscale coefficient based on the reliability impact weight of each image patch to obtain a statistical value of the damage level of the battery sample. This statistical value of damage level can characterize the overall damage level of the battery sample. Then, the computer equipment can average the statistical value of damage level according to the number of each image patch to obtain the reliability analysis result of the battery sample.

[0096] In a specific embodiment, taking the case where the battery sample is divided into m image blocks in the X direction and n image blocks in the Y direction as an example, the statistical value of the damage degree can be expressed as follows: ,in Indicates the position as Image patch (hereinafter referred to as image patch) The weighted value of the grayscale coefficients. For a given image patch Specifically, the weighted value of the grayscale coefficients of this image patch can be expressed as:

[0097]

[0098] Where W is the structural weight coefficient corresponding to the structural factor, 1 / d is the distance weight coefficient corresponding to the distance factor, and H is the gray coefficient.

[0099] The reliability severity index ρ of the battery sample can then be expressed as:

[0100]

[0101] The reliability severity index ρ can be used to represent the reliability analysis results of battery samples. In this embodiment, the larger the value of ρ, the more severe the reliability failure of the corresponding experimental group. It can be understood that in other embodiments, with different statistical algorithms for ρ, a smaller reliability severity index may indicate a more severe reliability failure of the corresponding experimental group.

[0102] In a specific embodiment, using a virtual EL image as an example, the method for determining the reliability severity index ρ of an EL image is explained in detail. Figure 6 The diagram shows a modeling schematic for the reliability severity index ρ of an EL image. A Cartesian coordinate system can be established with the center of the battery cell as the origin and the Z-direction (the thickness direction of the battery cell) as the coordinate axis. The EL image is then a two-dimensional image on the XY plane. The point (x, y) is the center point of the image patch, and dx and dy are the dimensions of the image patch in the X and Y directions, respectively. A computer can calculate the reliability severity index of the battery sample based on the following formula.

[0103]

[0104]

[0105]

[0106] Where (-a,+a) represents the coordinate region of the evaluated battery sample EL image in the X-axis direction, and (-b,+b) represents the coordinate region of the evaluated battery sample EL image in the Y-axis direction. S1 is the projected area of ​​the battery sample in the XY plane. S2 is the sum of the projected areas of the battery cell EL images in the XY plane, m is the number of image blocks contained in the EL image in the X-axis direction, and n is the number of image blocks contained in the EL image in the Y-axis direction. In this embodiment, each image block has the same size. For point The structural weight coefficient at this point For image blocks The center coordinates. Point in the XY plane The ratio of the distance to the origin to the equivalent radius of the battery sample. This applies when the battery sample is a bare cell. Take the constant 1. For point The grayscale coefficient corresponding to the grayscale value can be determined using a grayscale ruler. Represents image blocks The weighted value of the grayscale coefficient.

[0107] like Figure 7As shown, taking a 300mm*300mm battery module as an example, the area where the battery cells are located is divided into four 2*2 grid units, resulting in four image blocks. The structure of the battery module in the Z direction is as follows... Figure 3 As shown, from top to bottom, the components are: top cover glass, front film, battery cells, rear film, and rear cover. Furthermore, the center of the battery cell matrix (S1) coincides with the center of the glass (S2). For ease of calculation, in this case, S1 and S2 are two squares with coincident centers, and S1 = 4 * S2.

[0108] like Figure 7 As shown in (a), the grayscale coefficients (H) of the four image blocks are 80%, 25%, 25%, and 80% respectively (in order: starting from the top left cell, clockwise); Figure 7 As shown in (b), the engineer assigned importance scores to different regions of the specific process battery based on the experimental purpose and principle. In this embodiment, the importance scores (i.e., structural weight coefficients W) of the four image blocks are 80%, 30%, 30%, and 80%, respectively; Figure 7 As shown in (c), the ratio (d) of the distance from the four image patches to the origin to the equivalent radius of the battery sample is 31.33%. Therefore, according to the above formula, the reliability severity index ρ of the battery sample in this embodiment can be determined as:

[0109] =3.511

[0110] In the above embodiments, by weighted summation and averaging of the grayscale coefficients, reliability analysis results characterizing the average reliability of the battery sample can be obtained, thereby ensuring the accuracy of the reliability analysis results.

[0111] In one embodiment, the battery sample reliability analysis method further includes: obtaining sample information of each of the multiple battery samples participating in the aging process; and performing factorial analysis on the reliability of each battery sample based on the reliability analysis results of each battery sample to determine the factorial analysis results for each sample parameter.

[0112] The sample information includes parameter information corresponding to each of the multiple sample parameters. These sample parameters may include, for example, materials and processes. The battery structures of the various battery samples may be the same or different. Factorial analysis is a process of determining whether the effect of multiple factors acting together deviates from the sum of their individual effects, and how this deviation occurs. In classic factorial designs, in addition to considering the main effects of each research factor, the interaction between factors is also considered. That is, factorial analysis can include main effect analysis, interaction analysis, etc. Specifically, computer equipment can acquire the sample information of multiple battery samples participating in the aging process, and based on the reliability analysis results of each battery sample, perform factorial analysis on the reliability of the battery samples to determine the factorial analysis results for each sample parameter.

[0113] For example, consider a sample where the sample information includes material information and process information, and the material information includes three materials: MAT1, MAT2, and MAT3, and the process information includes three processes: TLS, TLP, and TLR. Figure 8 As shown, AI represents the EL images of nine groups of battery samples after undergoing the same reliability treatment. The material and process categories of each group of samples are shown in the figure. By analyzing each EL image, the reliability severity index ρ of each group of samples can be determined. It should be noted that in some embodiments, the change in the reliability severity index Δρ can also be used as an evaluation index. That is, for battery samples before and after aging treatment, their electroluminescence images are obtained respectively, and reliability analysis is performed based on the electroluminescence images to obtain the reliability severity index of the battery samples before and after aging treatment. The change in the reliability severity index Δρ is obtained by calculating the difference, and further factorial analysis is performed based on Δρ. Figure 8 As shown, when performing EL image reliability analysis, engineers generally obtain qualitative analysis results based on visual observation and theoretical experience: the brightness of the EL images of processes A, B, and C is similar, indicating that there is no significant difference between the three processes TLS, TLP, and TLR under material MAT1; the EL performance of processes D to I gradually deteriorates, indicating that MAT2 is superior to MAT3. Under the conditions of MAT2 and MAT3 materials, the superiority of the processes is TLS>TLP>TLR.

[0114] However, as Figure 8As shown, when using the scheme of this application for EL image reliability analysis, because the reliability severity index includes the importance weight W(x,y) of the image region, the eccentricity ratio d(x,y) of the region points, and the grayscale coefficient of the region image, the scheme of this application transforms the traditional qualitative description into a continuous quantitative index characterizing the reliability severity of EL images. Compared with qualitative description, it provides rich quantifiable experimental data. It can fully utilize statistical knowledge to perform factor significance analysis, factor correlation analysis, factor main effect analysis, variance analysis, etc. At the same time, it can reduce the number of experiments, uncover more main effects and interaction effects between factors, and greatly accelerate the new product development process.

[0115] like Figure 9-11 The results show that, when performing reliability analysis on EL images, based on the severity index of the EL images combined with statistical analysis, the following conclusions can be drawn:

[0116] 1. Factor significance analysis showed that, at a 95% confidence level, factor B (different processes) had no significant impact on the reliability of EL images of the experimental samples.

[0117] 2. Factor influence analysis: Based on the EL reliability severity index of the experimental samples, the influence of materials is greater than that of the process, with a ratio of 4 / 1.3 = 3.1 times.

[0118] 3. Under material MAT1, the effects of the three processes are not significant;

[0119] 4. Factor interaction analysis: There is no interaction between materials MAT2 and MAT3 and process TLP and TLR factors (parallel), while there is an interaction between materials MAT2 and MAT3 and process TLR and TLS factors (non-parallel).

[0120] In the above embodiments, factorial analysis based on reliability analysis results can achieve multi-factor screening and analysis, overcoming the limitations and inaccuracies of visual judgment under multi-factor conditions.

[0121] It should be understood that although the steps in the flowcharts of the above embodiments 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 above embodiments 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 in other steps.

[0122] Based on the same inventive concept, this application also provides a battery sample reliability analysis apparatus for implementing the battery sample reliability analysis method described above. The solution provided by this apparatus is similar to the solution described in the above method; therefore, the specific limitations in one or more embodiments of the battery sample reliability analysis apparatus provided below can be found in the limitations of the battery sample reliability analysis method described above, and will not be repeated here.

[0123] In one embodiment, such as Figure 12 As shown, a battery sample reliability analysis device is provided, including: an acquisition module 1201, an image segmentation module 1202, a reliability influence weight determination module 1203, and a statistical analysis module 1204, wherein:

[0124] The acquisition module 1201 is used to acquire electroluminescence images of the battery sample after aging treatment; the electroluminescence images are grayscale images.

[0125] The image segmentation module 1202 is used to divide the electroluminescent image into multiple image blocks and determine the grayscale coefficient of each image block; the grayscale coefficient is determined according to the grayscale value of the image block and is used to characterize the degree of damage at the location of the image block;

[0126] The reliability impact weight determination module 1203 is used to determine the reliability impact weight of each image block on the battery sample based on the position of the image block in the battery sample.

[0127] The statistical analysis module 1204 is used to perform statistical analysis on the grayscale coefficients of each image block based on the reliability influence weights, and to determine the reliability analysis results of the battery sample.

[0128] In one embodiment, the reliability impact weight determination module 1203 includes: a weight impact factor determination unit, used to determine a weight impact factor that matches the sample type of the battery sample; a weight coefficient determination unit, used to determine the weight coefficient corresponding to the weight impact factor for each image block based on the position of the image block in the battery sample; and a reliability impact weight determination unit, used to determine the reliability impact weight of the image block on the battery sample based on the weight coefficient.

[0129] In one embodiment, the sample type of the battery sample includes a battery module; the weighting factors include a structure factor and a distance factor. In this embodiment, the weighting coefficient determination unit includes: a local structure determination component, used to determine the local battery structure contained in the image patch based on the position of the image patch in the battery sample; a structure weighting coefficient determination component, used to determine the structure weighting coefficient corresponding to the structure factor of the image patch based on the importance of the local battery structure in the battery module; the structure weighting coefficient is positively correlated with the importance; and a distance weighting coefficient determination component, used to determine the distance weighting coefficient corresponding to the distance factor of the image patch based on the position of the image patch in the battery sample; the distance weighting coefficient is positively correlated with the distance between the position and the edge of the battery module.

[0130] In one embodiment, the distance weight coefficient determination component is specifically used to: determine the equivalent center and equivalent radius of the battery sample; determine the equivalent distance between the image patch and the equivalent center based on the position of the image patch in the battery sample; determine the distance weight coefficient corresponding to the distance factor of the image patch based on the ratio between the equivalent distance and the equivalent radius; the distance weight coefficient is inversely correlated with the ratio.

[0131] In one embodiment, the image segmentation module 1202 is specifically used to: obtain the gray value of each pixel in the image block; determine the statistical result of each gray value as the image block gray value of the image block; and match the image block gray value with the image gray scale configured for the battery sample to determine the gray coefficient corresponding to the image block gray value.

[0132] In one embodiment, the statistical analysis module 1204 is specifically used to: perform weighted summation of each grayscale coefficient based on the reliability influence weight of each image block to obtain the damage degree statistics of the battery sample; and perform average processing on the above damage degree statistics according to the number of each image block to obtain the reliability analysis result of the battery sample.

[0133] In one embodiment, the battery sample reliability analysis device further includes a factorial analysis module, used to: acquire sample information for each of the multiple battery samples participating in the aging process; and perform factorial analysis on the reliability of each battery sample based on the reliability analysis results of each battery sample, to determine the factorial analysis results for each sample parameter. The sample information includes parameter information corresponding to each of the multiple sample parameters.

[0134] Each module in the aforementioned battery sample reliability analysis 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.

[0135] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 13 As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computational and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage medium. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a method for reliable analysis of battery samples. The display unit of the computer device is used to form a visually visible image. It can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0136] Those skilled in the art will understand that Figure 13 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0137] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0138] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0139] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0140] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. This computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. 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.

[0141] 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.

[0142] 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 patent 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 reliability analysis of battery samples, characterized in that, The method includes: Acquire electroluminescence images of battery samples after aging treatment; the electroluminescence images are grayscale images; The electroluminescent image is divided into multiple image blocks, and the grayscale coefficient of each image block is determined. The grayscale coefficient is determined based on the grayscale value of the image block and is used to characterize the degree of damage at the location of the image block. For each image block, the influence weight of the image block on the reliability of the battery sample is determined according to the position of the image block in the battery sample; Based on the reliability impact weights of each image block, statistical analysis is performed on their respective grayscale coefficients to determine the reliability analysis results of the battery sample.

2. The method according to claim 1, characterized in that, For each image block, determining the reliability impact weight of the image block on the battery sample based on its position within the battery sample includes: Determine the weighted influence factor that matches the sample type of the battery sample; For each image block, the weight coefficient corresponding to the weight influence factor is determined based on the position of the image block in the battery sample. The weighting coefficients are used to determine the influence weight of the image patch on the reliability of the battery sample.

3. The method according to claim 2, characterized in that, The sample type of the battery sample includes battery components; the weighting influence factors include structural factors and distance factors; determining the weighting coefficients corresponding to the weighting influence factors based on the position of the image patch in the battery sample includes: Based on the location of the image block in the battery sample, the local battery structure contained in the image block is determined; Based on the importance of the local battery structure in the battery assembly, the structure weight coefficient corresponding to the structure factor of the image block is determined; the structure weight coefficient is positively correlated with the importance. Based on the position of the image patch in the battery sample, a distance weight coefficient corresponding to the distance factor of the image patch is determined; the distance weight coefficient is positively correlated with the distance between the position and the edge of the battery assembly.

4. The method according to claim 3, characterized in that, The step of determining the distance weight coefficient corresponding to the distance factor of the image patch based on the position of the image patch in the battery sample includes: Determine the equivalent center and equivalent radius of the battery sample; Based on the position of the image block in the battery sample, determine the equivalent distance between the image block and the equivalent center; Based on the ratio between the equivalent distance and the equivalent radius, the distance weight coefficient corresponding to the distance factor of the image patch is determined; the distance weight coefficient is inversely correlated with the ratio.

5. The method according to claim 1, characterized in that, Determining the grayscale coefficient of each of the image blocks includes: Obtain the grayscale value of each pixel in the image block; The statistical results of each gray value are determined as the image block gray value of the image block; The grayscale value of the image block is matched with the image grayscale scale configured for the battery sample to determine the grayscale coefficient corresponding to the grayscale value of the image block.

6. The method according to claim 1, characterized in that, The step of statistically analyzing the grayscale coefficients of each image patch based on the reliability impact weights to determine the reliability analysis results of the battery sample includes: Based on the reliability influence weight of each image block, the grayscale coefficients are weighted and summed to obtain the statistical value of the damage degree of the battery sample. The reliability analysis results of the battery sample are obtained by averaging the statistical values ​​of the damage degree based on the number of each image block.

7. The method according to any one of claims 1 to 6, characterized in that, The method further includes: Obtain sample information for each of the multiple battery samples participating in the aging process; the sample information includes parameter information corresponding to each of the multiple sample parameters. Based on the reliability analysis results of each battery sample, factorial analysis is performed on the reliability of each battery sample to determine the factorial analysis results for each sample parameter.

8. A battery sample reliability analysis device, characterized in that, The device includes: The acquisition module is used to acquire electroluminescence images of battery samples after aging treatment; the electroluminescence images are grayscale images. An image segmentation module is used to divide the electroluminescent image into multiple image blocks and determine the grayscale coefficient of each image block; the grayscale coefficient is determined based on the grayscale value of the image block and is used to characterize the degree of damage at the location of the image block; The reliability impact weight determination module is used to determine the reliability impact weight of each image block on the battery sample based on the position of the image block in the battery sample. The statistical analysis module is used to perform statistical analysis on the grayscale coefficients of each image block based on the reliability influence weights, and to determine the reliability analysis results of the battery sample.

9. 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 described in any one of claims 1 to 7.

10. 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 described in any one of claims 1 to 7.

Images