Tab error correction method and device, equipment and storage medium

Abstract

The invention provides a tab error correction method, device and equipment and a storage medium, and the method comprises the steps: shooting a morphological image of a tab on a path for conveying the tab from a material coil to a lamination table; inputting the morphological image into a preset morphological analysis model, and when whether a preset first defect type or a preset second defect type exists in the morphological image is analyzed, determining a target tab and outputting a corresponding first correction instruction or a corresponding second correction instruction if the preset first defect type or the preset second defect type exists in the morphological image; according to the first correction instruction, driving preset laser equipment to execute a preset first correction scheme on the target tab; or according to the second correction instruction, driving preset laser equipment to execute a preset second correction scheme on the target tab; and at the correction confirmation station, whether the target tab has a preset first defect type or a preset second defect type is evaluated within preset evaluation times until the evaluation result is that the target tab does not exist, and the target tab is confirmed as a qualified product.

Description

Technical Field

[0001] This application relates to the field of electrode detection technology, and in particular to an electrode error correction method, apparatus, device and storage medium. Background Technology

[0002] In the automated production of lithium-ion battery cells, the tabs are often subject to plastic deformation defects such as folding or warping along the path from the material roll to the stacking station due to mechanical vibration, tension fluctuations, or guide deviations. Current technologies largely rely on visual inspection systems at the end of the production line for quality screening. However, this post-production inspection method can only identify and reject defective products, and cannot perform online repairs of identified defects, leading to material waste and increased production costs. Furthermore, for minor warping defects, traditional mechanical straightening methods suffer from drawbacks such as slow response, insufficient precision in straightening force control, and a tendency to cause secondary damage. Summary of the Invention

[0003] This application provides a method, apparatus, device, and storage medium for correcting errors in tabs, which is used to proactively intervene in the tab production process and correct specific types of defects online, thereby improving production first-pass yield and material utilization.

[0004] In a first aspect, embodiments of this application provide a method for correcting errors in a tab, the method comprising: The morphological image of the electrode is captured on the path of conveying the electrode from the roll to the stacking table; When the morphological image is input into a preset morphological analysis model, and the morphological image is analyzed to see if there is a preset first defect type or a preset second defect type, if there is, the target electrode is determined and the corresponding first correction command or second correction command is output. The preset laser device is driven to perform a preset first correction scheme on the target electrode according to the first correction instruction; or the preset laser device is driven to perform a preset second correction scheme on the target electrode according to the second correction instruction. At the correction and confirmation station, within a preset number of evaluations, the target electrode is evaluated to determine whether there is a preset first defect type or a preset second defect type, until the evaluation result is that there is no defect, at which point the target electrode is confirmed as a qualified product.

[0005] Secondly, embodiments of this application provide a tab error correction device, the tab error correction device comprising: An image capturing module is used to capture morphological images of the tabs along the path from the material roll to the stacking table; The defect analysis module is used to input the morphological image into a preset morphological analysis model, analyze whether there is a preset first defect type or a preset second defect type in the morphological image, and if so, determine the target electrode and output the corresponding first correction instruction or second correction instruction. The defect correction module is used to drive a preset laser device to perform a preset first correction scheme on the target electrode according to the first correction instruction; or to drive a preset laser device to perform a preset second correction scheme on the target electrode according to the second correction instruction. The result evaluation module is used at the correction confirmation station to evaluate whether the target electrode has a preset first defect type or a preset second defect type within a preset number of evaluations, until the evaluation result is that it does not exist, and then the target electrode is confirmed as a qualified product.

[0006] Thirdly, embodiments of this application provide a tab correction device, the tab correction device including a memory and a processor; The memory is used to store computer programs; The processor is configured to execute the computer program and, in executing the computer program, implement the tab error correction method as described in any of the embodiments of this application.

[0007] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to implement the tab error correction method as described in any of the embodiments of this application.

[0008] This application provides a method for correcting errors in tabs. The method includes: taking a morphological image of the tab along the path from the coil to the stacking table; inputting the morphological image into a preset morphological analysis model to analyze whether a preset first defect type or a preset second defect type exists in the morphological image; if so, identifying the target tab and outputting a corresponding first correction instruction or second correction instruction; driving a preset laser device to perform a preset first correction scheme on the target tab according to the first correction instruction; or driving a preset laser device to perform a preset second correction scheme on the target tab according to the second correction instruction; and at a correction confirmation station, evaluating whether the target tab has the preset first defect type or the preset second defect type within a preset number of evaluations until the evaluation result is that it does not exist, and confirming the target tab as a qualified product. In the above method, by capturing images of the tab's shape and inputting them into a preset model, the defects of the tab are identified and classified in real time, and corresponding correction instructions are generated. Based on the correction instructions, the laser equipment is driven to execute a targeted solution. The non-contact and high-precision characteristics of laser processing are used to accurately repair folding and warping defects. A closed-loop processing mechanism is formed through the cyclic evaluation mechanism of the correction confirmation station to ensure that each tab meets the qualification standard, thereby significantly reducing material loss while improving the product qualification rate. Attached Figure Description

[0009] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 A schematic flowchart illustrating a method for correcting errors in a tab, as provided in an embodiment of this application; Figure 2 This is a schematic block diagram of a tab error correction device provided in an embodiment of this application. Detailed Implementation

[0011] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described below with reference to the accompanying drawings.

[0012] The terms "first" and "second," etc., used in the specification, claims, and drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0013] The term "embodiment" as used herein means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0014] It should be understood that in this application, "at least one (item)" means one or more, "more than one" means two or more, "at least two (items)" means two or three or more, and "and / or" is used to describe the relationship between related objects, indicating that there can be three relationships. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the related objects before and after are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0015] Please see Figure 1 , Figure 1 This is a schematic flowchart illustrating a method for correcting tab errors according to an embodiment of this application. Figure 1 As shown, the specific steps of the error correction method for this electrode include: S101-S104.

[0016] S101. Take a picture of the shape of the tab along the path from the material roll to the stacking table.

[0017] For example, in the middle of the path where the tab strip is stably unwound from its coil and transported to the stacking table at a constant speed via a tension control system and a series of guide rollers, an image acquisition area with uniform illumination is set up. Within this area, a high-speed industrial camera with a resolution of at least 5 megapixels is deployed, using a parallel backlight source at a specific angle perpendicular to the tab plane to capture images. The shooting process must ensure that the tab is in a naturally extended state, free from external mechanical stress interference, to obtain a morphological image that clearly reflects the true physical shape of the tab. The morphological image must completely include key information such as the tab's outline edges, surface flatness, and the presence of any abnormal protrusions or depressions. To adapt to the production line speed, the image exposure time is controlled to within one-thousandth of a second to eliminate motion blur and ensure the accuracy of subsequent analysis. The obtained morphological image will serve as the original data source for the entire correction process, and its quality directly determines the accuracy of subsequent analysis and correction.

[0018] S102. Input the morphological image into the preset morphological analysis model. If the preset first defect type or preset second defect type exists in the morphological image, and if it does, determine the target electrode and output the corresponding first correction command or second correction command.

[0019] For example, the acquired morphological image is transmitted to the image processing unit in real time and input into a pre-set morphological analysis model trained on a large number of samples. The pre-set morphological analysis model is an algorithm model based on a deep convolutional neural network architecture, capable of pixel-level segmentation and feature extraction of the input morphological image. The pre-set morphological analysis model compares the extracted contour features with an internal standard template library. If it identifies an irregular, continuous outward convex contour exceeding the tolerance range at the edge of the tab, it determines that a pre-set first defect type exists, namely tab folding; if it identifies an unexpected, continuous curvature change or permanent bending in a local area on the tab surface, it determines that a pre-set second defect type exists, namely tab warping. Once the existence of either defect type is confirmed, the model will accurately frame the defective tab in the image, mark it as the target tab, and generate a structured instruction document. If the instruction document targets a folding defect, it is a first correction instruction, which includes the precise pixel coordinates of the defect and the geometric dimensions of the folded area; if it targets a warping defect, it is a second correction instruction, which includes the curvature of the warping, its axis, and the affected area.

[0020] S103. Drive the preset laser device to perform a preset first correction scheme on the target electrode according to the first correction command. Or drive the preset laser device to perform a preset second correction scheme on the target electrode according to the second correction command.

[0021] For example, after the instruction document is parsed, the control system will drive a preset laser device into working mode. If the received instruction is the first correction instruction, the preset first correction scheme is executed: the galvanometer system of the laser device will guide the laser focus to be precisely positioned at the root of the folded area of ​​the target electrode tab according to the coordinate and size information in the instruction. Using laser parameters with a peak power of 500W and a pulse width in the nanosecond range, a high-speed scan is performed along the connection line between the folded part and the main electrode tab. The folded part is vaporized and separated through laser ablation, while ensuring a smooth transition at the edge of the main electrode tab. If the received instruction is the second correction instruction, the preset second correction scheme is executed: the laser is set to a low-power continuous wave mode, with a power range between 50 and 100 watts. The laser spot is guided to scan at a uniform speed along the warped ridge of the target electrode tab. Local thermal stress is generated in the warped area through controllable heat input, causing the electrode tab material to undergo plastic deformation to offset the initial internal stress, thereby restoring its flat shape. The entire laser action process is completed synchronously during the electrode tab delivery process without interruption.

[0022] S104. At the correction and confirmation station, within the preset number of assessments, assess whether the target electrode tab has a preset first defect type or a preset second defect type, until the assessment result is that it does not exist, and then confirm the target electrode tab as a qualified product.

[0023] For example, the target tab after laser correction continues to move along the production line to an independent correction confirmation station. At this station, an imaging system with the same configuration as in step S101 is used to capture a high-resolution morphological image of the target tab again. The image is then fed into a preset morphological analysis model for defect analysis to assess whether the previously existing preset first defect type or preset second defect type has been successfully eliminated after the correction intervention. The evaluation process is set with a preset number of evaluations, for example, three. If the first evaluation shows that neither type of defect exists, the target tab is immediately confirmed as a qualified product and allowed to flow into the subsequent stacking process. If the evaluation finds that the defect still exists but its severity has been reduced, and the current cumulative number of evaluations has not reached the upper limit, the process returns to the correction instruction generation step for laser treatment after parameter optimization; this cycle continues until the defect is eliminated. If the preset number of evaluations has been exhausted and the defect is still not eliminated, the target tab is determined to be an unrepairable product, and a rejection procedure is initiated to isolate it.

[0024] This application provides a method for correcting errors in tabs. The method includes: taking a morphological image of the tab along the path from the coil to the stacking table; inputting the morphological image into a preset morphological analysis model to analyze whether a preset first defect type or a preset second defect type exists in the morphological image; if so, identifying the target tab and outputting a corresponding first correction instruction or second correction instruction; driving a preset laser device to perform a preset first correction scheme on the target tab according to the first correction instruction; or driving a preset laser device to perform a preset second correction scheme on the target tab according to the second correction instruction; and at a correction confirmation station, evaluating whether the target tab has the preset first defect type or the preset second defect type within a preset number of evaluations until the evaluation result is that it does not exist, and confirming the target tab as a qualified product. In the above method, by capturing images of the tab's shape and inputting them into a preset model, the defects of the tab are identified and classified in real time, and corresponding correction instructions are generated. Based on the correction instructions, the laser equipment is driven to execute a targeted solution. The non-contact and high-precision characteristics of laser processing are used to accurately repair folding and warping defects. A closed-loop processing mechanism is formed through the cyclic evaluation mechanism of the correction confirmation station to ensure that each tab meets the qualification standard, thereby significantly reducing material loss while improving the product qualification rate.

[0025] To more clearly illustrate the technical solution of this application, the technical solution of this application will be described below through specific embodiments. It should be noted that the specific embodiments are used to expand the description of the technical solution of this application, and are not intended to limit this application.

[0026] In some embodiments, capturing a morphological image of the electrode tab along the path from the stock roll to the stacking table includes: segmenting the morphological image to extract the edge contour line and surface topography data of the electrode tab; comparing the edge contour line with a standard contour template, and if there is contour overflow or abnormal protrusion, determining that there is a first defect type, the first defect type being electrode tab folding; comparing the surface topography data with a standard flatness, and if there is continuous curvature or bending, determining that there is a second defect type, the second defect type being electrode tab warping; based on the first defect type or the second defect type, locking the defective electrode tab in the morphological image as the target electrode tab, and generating identification information, the identification information including: defect type, location, and geometric dimensions; and according to the identification information, calling a preset laser parameter library to generate a first correction command or a second correction command.

[0027] For example, after acquiring the morphological image, it is sent to a digital image processing flow for analysis. Using image segmentation technology based on grayscale thresholding and edge detection algorithms, the visual representation of the tab is accurately separated from the complex background, and key data defining the tab's shape boundary, namely the edge contour line, is extracted. Simultaneously, by calculating the gradient and height information of pixels in the image, the three-dimensional surface features of the tab are reconstructed, forming surface morphology data. Next, the extracted edge contour line is compared pixel-level with a standard contour template stored in the database as a reference. This comparison process calculates the overlap and deviation of the contours. When a significant overflow exceeding a preset tolerance threshold (e.g., exceeding the standard contour by 0.1 mm) or a sharp, abnormally protruding shape is detected in the edge contour line, the tab is determined to have a first defect type according to preset logical rules. This first defect type is defined as tab folding in the business logic. In parallel, the surface morphology data is compared and analyzed with a standard flatness dataset characterizing ideal flatness. Surface curvature and continuity are calculated. When a continuous arc spanning a certain length (e.g., exceeding 2 mm) or a bend with obvious creases is detected on the tab surface, a second defect type is identified, defined as tab warping. After the defect identification, the processing logic backtracks to the original morphological image, precisely locating the pixel region where the defect occurs in the image coordinate space, thus pinpointing the specific tab that caused the defect and identifying it as the target tab. Subsequently, the system generates structured identification information, which must fully include the identified defect type, the precise location coordinates of the defect in the tab coordinate system, and the geometric dimensions of the defect (e.g., the width and height of the fold, or the arc length and sagitta of the warp). Finally, based on this identification information, the system will access a preset laser parameter library, which stores optimized laser processing parameters corresponding to different defect characteristics. Through querying and matching, a first correction instruction (for folding) or a second correction instruction (for warping) that can drive the laser equipment to perform specific actions will be generated.

[0028] In some embodiments, a preset first correction scheme is executed on the target tab by driving a preset laser device according to a first correction instruction, including: locating the fold root position of the tab according to the identification information in the first correction instruction; controlling the laser device to emit a laser beam and adjusting the spot focus to the fold root position; controlling the laser beam to scan along the connection line between the fold root and the main tab, and removing the fold portion from the main tab by laser ablation; during the removal process, coaxial vision is used to monitor the laser cutting path to ensure that the edge of the tab remains smooth after the cutting is completed.

[0029] For example, after the first correction instruction is generated, the correction process enters the physical execution phase. Based on the identification information contained in the first correction instruction, especially the detailed defect location coordinates, the scanning galvanometer and focus positioning system of the laser equipment are controlled to accurately position the laser action point at the fold root of the target electrode, that is, the physical starting point where the folded part connects to the electrode body.

[0030] The laser is controlled to emit a laser beam of a specific wavelength, and the optical system is dynamically adjusted so that the focal spot of the laser beam falls precisely on the pre-positioned fold root position, ensuring energy concentration.

[0031] The laser beam is controlled to perform high-speed, precise scanning along a preset path, connecting the root of the folded portion to the main electrode tab. During this process, the laser beam interacts with the electrode tab material, and through laser ablation, the material at the folded portion is instantly vaporized or peeled off, thereby completely removing the folded portion from the main electrode tab.

[0032] Throughout the entire excision process, a coaxial vision monitoring unit integrated into the laser head works continuously to capture images of the laser cutting path area in real time. Through real-time analysis of the cutting trajectory, it ensures that the laser cutting path accurately follows the predetermined route. After the cutting is completed, it verifies that the cutting edge of the main electrode can meet the preset smoothness and flatness requirements, avoiding burrs, molten metal, or secondary damage.

[0033] In some embodiments, driving a preset laser device to perform a preset second correction scheme on the target tab according to a second correction instruction includes: locating the raised ridge line of the tab warping region according to the identification information in the second correction instruction; controlling the laser device to emit a low-power-density laser beam to linearly scan and heat the raised ridge line at a non-ablative energy level; and during the linear scanning and heating process, monitoring the deformation feedback of the tab in real time and dynamically adjusting the scanning speed or power of the laser according to the deformation feedback.

[0034] For example, the specific implementation path differs when the execution process is based on the second correction instruction. Based on the identification information contained in the second correction instruction, particularly the descriptive data of the warp morphology, the most prominent ridge line within the warped area of ​​the target tab is located, i.e., the area of ​​highest stress in the warped deformation. A laser device is controlled to emit a strictly controlled low-power-density laser beam. The energy level of this laser beam is precisely calibrated to ensure that it acts below a non-ablative energy threshold when applied to the tab material, preventing material removal and primarily generating a thermal effect. This laser beam is controlled to perform a uniform or variable-speed linear scanning heating along the located ridge line, generating a controllable temperature field in the ridge region through laser energy injection. During this continuous linear scanning heating process, a set of highly sensitive deformation sensors (such as macro cameras or laser displacement meters) monitors the deformation feedback of the tab under thermal stress in real time, i.e., the recovery of its flatness. The monitored deformation feedback data is fed into the control algorithm in real time. Based on the preset deformation-energy relationship model, the algorithm dynamically adjusts the scanning speed of the laser beam or the real-time output power on the scanning path. For example, when the deformation feedback shows that the recovery rate is slowing down, the laser power can be appropriately increased or the scanning speed can be reduced to achieve adaptive closed-loop correction of warping defects.

[0035] In some embodiments, at the correction confirmation station, within a preset number of evaluations, the presence of a preset first defect type or a preset second defect type of the target electrode is evaluated until the evaluation result indicates that the target electrode does not exist, and the target electrode is confirmed as a qualified product. This includes: after the laser equipment completes the correction operation, the target electrode is transported to the correction confirmation station; at the correction confirmation station, a post-correction morphological image of the target electrode is captured; the post-correction morphological image is input again into a preset morphological analysis model for analysis to obtain a correction analysis result; if neither the first defect type nor the second defect type exists in the correction analysis result, the target electrode is marked as a qualified product and allowed to flow into the stacking process; if the first defect type or the second defect type still exists in the correction analysis result, it is determined whether the current number of evaluations has reached the preset number of evaluations; if not, a correction instruction is generated again and a new round of correction is started; if the target electrode has reached the preset number of evaluations, the target electrode is marked as a non-qualified product and rejected.

[0036] For example, after the laser equipment completes the correction operation on the target tab, the target tab is transported by the production line conveyor to a specially established correction confirmation station, independent of the correction station. Upon arrival at the correction confirmation station, the image acquisition device located at the station is immediately activated to capture an image of the target tab's corrected morphology after laser processing, using the same imaging standard and resolution as in the initial inspection stage. Subsequently, this corrected morphology image is used as input and re-input into the previously used preset morphology analysis model to perform the same analysis process, thereby obtaining a correction analysis result regarding the current tab state. The obtained correction analysis result is logically judged: if the preset first defect type (tab folding) and second defect type (tab warping) judgment flags in the correction analysis result are both "absent", then the target tab is immediately marked as a qualified product in the information system, and a control command is triggered, allowing the target tab to smoothly flow into the subsequent stacking process. Conversely, if the correction analysis results still show either the first defect type or the second defect type as "existing," an evaluation loop control logic needs to be initiated: It must be determined whether the cumulative number of evaluations performed on the target electrode has reached a preset number of evaluations (e.g., set to 3). If the current number of evaluations has not reached this preset limit, the control system will automatically trigger an instruction to generate a new or optimized correction instruction and immediately initiate a new round of laser correction operations. If the current number of evaluations has reached the preset limit, it means that the target electrode is considered difficult to repair or the correction is ineffective. It must be marked as a non-conforming product in the information system, and a rejection device will be triggered to physically remove it from the production flow.

[0037] In some embodiments, generating a correction instruction again and starting a new round of correction further includes: acquiring a morphological optimization image of the target tab from the previous correction, inputting the morphological optimization image into a preset morphological analysis model to obtain a correction evaluation result; if the correction evaluation result includes: the defect size has decreased but not been completely eliminated, then in the current correction cycle, the same laser action mechanism is maintained, and the laser energy density or number of scans is reduced proportionally according to a preset rule based on the reduction ratio of the defect size; if the evaluation result indicates that the defect morphology has changed but has not been eliminated, then in the current correction cycle, the laser action mechanism is switched, and the laser parameters are recalculated based on the correction evaluation result.

[0038] For example, the decision loop for generating a correction instruction again and initiating a new round of correction includes a feedback-based parameter optimization sub-process. First, a morphological optimization image taken after laser treatment of the target electrode in the previous correction cycle is needed. This image reflects the actual effect of the previous correction operation. This morphological optimization image is then input into a pre-set morphological analysis model for analysis, resulting in a quantitative correction evaluation result. This result should include quantitative characteristics such as the current size and shape of the defect. This correction evaluation result is then analyzed: if the result clearly shows that the defect size (such as fold area or warpage height) has decreased compared to the previous cycle but has not been completely eliminated, then in the current correction cycle, the control system will decide to maintain the same laser action mechanism as the previous cycle (i.e., continue using ablation or heating modes). However, based on the calculated reduction ratio of the defect size, according to a pre-set rule in the algorithm (e.g., if the defect size decreases by 50%, the laser energy density or number of scans will also be reduced by 20-30%), the laser energy density or the number of laser beam scans for this operation will be reduced proportionally. In another scenario, if the correction assessment results indicate that the morphology of the defect has changed significantly but has not been eliminated (e.g., folding becomes warping, or the warping axis shifts), then in the current correction cycle, the control system will decide to switch the laser action mechanism (e.g., switch from ablation mode to heating mode, or vice versa), and based on the new defect morphology reflected by the new correction assessment results, recalculate a set of applicable laser parameters (such as wavelength, pulse frequency, and power density) to cope with the dynamic changes in the defect morphology.

[0039] In some embodiments, the preset morphological analysis model is a convolutional neural network model obtained by deep learning training on sample images, including: normal electrode images, folded electrode images, and warped electrode images.

[0040] For example, the construction of the pre-built morphological analysis model is an offline machine learning process. Essentially, this pre-built morphological analysis model is a high-precision classification and recognition model obtained by training a deep convolutional neural network on a massive number of precisely labeled sample images. The sample images used for training need to be representative and diverse, including a large number of normal electrode images acquired under different lighting and angle conditions, images of folded electrodes covering various degrees and shapes, and images of warped electrodes with various typical warping patterns. These sample images need to be preprocessed and data augmented before being used in training to improve the robustness of the model. The training process iteratively optimizes the network weights, enabling the model to automatically extract deep features from the input electrode morphology images and accurately distinguish between normal electrodes, folded electrodes, and warped electrodes.

[0041] Please see Figure 2 , Figure 2 This is a schematic block diagram of a tab error correction device 200 provided in an embodiment of this application. The tab error correction device 200 is used to perform the aforementioned tab error correction method. The tab error correction device 200 can be configured in a server.

[0042] The server can be a standalone server, a server cluster, 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, content delivery networks (CDN), and big data and artificial intelligence platforms.

[0043] like Figure 2 As shown, the error correction device 200 for the electrode includes: an image capturing module 201, a defect analysis module 202, a defect correction module 203, and a result evaluation module 204.

[0044] Image capturing module 201 is used to capture images of the shape of the tabs on the path of conveying the tabs from the stock roll to the stacking table.

[0045] The defect analysis module 202 is used to input the morphological image into a preset morphological analysis model and analyze whether there is a preset first defect type or a preset second defect type in the morphological image. If there is, the target electrode is determined and the corresponding first correction instruction or second correction instruction is output.

[0046] The defect correction module 203 is used to drive a preset laser device to perform a preset first correction scheme on the target electrode according to a first correction instruction, or to drive a preset laser device to perform a preset second correction scheme on the target electrode according to a second correction instruction.

[0047] The result evaluation module 204 is used at the correction confirmation station to evaluate whether the target electrode has a preset first defect type or a preset second defect type within a preset number of evaluations, until the evaluation result is that it does not exist, and then the target electrode is confirmed as a qualified product.

[0048] This application provides a tab correction device, which includes a memory and a processor; the memory is used to store a computer program; the processor is used to execute the computer program and, when executing the computer program, implements a tab error correction method as described in any of the embodiments of this application.

[0049] This application provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it causes the processor to implement a tab error correction method as described in any of the embodiments of this application.

[0050] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for correcting errors in a tab, characterized in that, The method includes: The morphological image of the electrode is captured on the path of conveying the electrode from the roll to the stacking table; When the morphological image is input into a preset morphological analysis model, and the morphological image is analyzed to see if there is a preset first defect type or a preset second defect type, if there is, the target electrode is determined and the corresponding first correction command or second correction command is output. The preset laser device is driven to perform a preset first correction scheme on the target electrode according to the first correction instruction; or the preset laser device is driven to perform a preset second correction scheme on the target electrode according to the second correction instruction. At the correction and confirmation station, within a preset number of evaluations, the target electrode is evaluated to determine whether there is a preset first defect type or a preset second defect type, until the evaluation result is that there is no defect, at which point the target electrode is confirmed as a qualified product.

2. The error correction method for the electrode ear as described in claim 1, characterized in that, The step of capturing morphological images of the electrode along the path from the coil to the stacking table includes: The morphological image is segmented to extract the edge contour lines and surface topography data of the electrode. The edge contour line is compared with the standard contour template. If there is contour overflow or abnormal protrusion, it is determined that the first defect type exists. The first defect type is tab folding. The surface topography data is compared with the standard flatness. If there is a continuous arc or bend, it is determined that the second defect type exists. The second defect type is tab warping. Based on the first defect type or the second defect type, the electrode with the defect is identified in the morphological image as the target electrode, and identification information is generated, which includes: defect type, location and geometric dimensions. Based on the identification information, a preset laser parameter library is invoked to generate a first correction command or a second correction command.

3. The error correction method for the electrode ear as described in claim 2, characterized in that, The step of driving a preset laser device to perform a preset first correction scheme on the target electrode according to the first correction command includes: Based on the identification information in the first correction instruction, locate the folded root position of the electrode tab; Control the laser device to emit a laser beam and adjust the focus of the laser spot to the position of the fold root; The laser beam is controlled to scan along the connection line between the root of the folded portion and the main electrode tab, and the folded portion is removed from the main electrode tab by laser ablation; During the excision process, the laser cutting path is monitored by coaxial vision to ensure that the edge of the electrode remains smooth after the cutting is completed.

4. The error correction method for the electrode ear as described in claim 2, characterized in that, The step of driving a preset laser device to perform a preset second correction scheme on the target electrode according to the second correction command includes: Based on the identification information in the second correction instruction, locate the raised ridge line of the electrode warping region; The laser device is controlled to emit a low-power-density laser beam to linearly scan and heat the raised ridge at a non-ablative energy level. During the linear scanning heating process, the deformation feedback of the electrode tab is monitored in real time, and the scanning speed or power of the laser is dynamically adjusted according to the deformation feedback.

5. The error correction method for the electrode ear as described in claim 1, characterized in that, At the correction and confirmation station, within a preset number of evaluations, the presence of a preset first defect type or a preset second defect type of the target electrode is evaluated until the evaluation result indicates that the target electrode is not present, and the target electrode is then confirmed as a qualified product. This includes: After the laser device completes the correction operation, the target electrode is transported to the correction confirmation station; At the correction confirmation station, a corrected morphological image of the target electrode is captured; The corrected morphological image is then input into the preset morphological analysis model for analysis to obtain the correction analysis result; If neither the first defect type nor the second defect type exists in the correction analysis results, the target electrode tab is marked as a qualified product and is allowed to flow into the stacking process. If the first defect type or the second defect type still exists in the correction analysis results, it is determined whether the current number of evaluations has reached the preset number of evaluations; if not, a correction instruction is generated again and a new round of correction is started; if the target tab has been reached, it is marked as a non-conforming product and removed.

6. The error correction method for the electrode ear as described in claim 5, characterized in that, The process of regenerating the correction instruction and initiating a new round of correction also includes: Obtain the morphological optimization image of the target electrode from the previous correction, input the morphological optimization image into a preset morphological analysis model, and obtain the correction evaluation result. If the correction assessment result includes: the defect size is reduced but not completely eliminated, then in the current correction cycle, the same laser action mechanism is maintained, and the laser energy density or number of scans is reduced proportionally according to a preset rule based on the reduction ratio of the defect size. If the evaluation result indicates that the defect morphology has changed but has not been eliminated, then in the current correction cycle, the laser action mechanism is switched, and the laser parameters are recalculated based on the correction evaluation result.

7. The error correction method for the electrode ear as described in claim 1, characterized in that, The preset morphological analysis model is a convolutional neural network model obtained by deep learning training on sample images, including: normal electrode images, folded electrode images, and warped electrode images.

8. A device for correcting errors in a tab, characterized in that, The electrode error correction device is used to perform the electrode error correction method as described in any one of claims 1-7, and the electrode error correction device comprises: An image capturing module is used to capture morphological images of the tabs along the path from the material roll to the stacking table; The defect analysis module is used to input the morphological image into a preset morphological analysis model, analyze whether there is a preset first defect type or a preset second defect type in the morphological image, and if so, determine the target electrode and output the corresponding first correction instruction or second correction instruction. The defect correction module is used to drive a preset laser device to perform a preset first correction scheme on the target electrode according to the first correction instruction; or to drive a preset laser device to perform a preset second correction scheme on the target electrode according to the second correction instruction. The result evaluation module is used at the correction confirmation station to evaluate whether the target electrode has a preset first defect type or a preset second defect type within a preset number of evaluations, until the evaluation result is that it does not exist, and then the target electrode is confirmed as a qualified product.

9. A tab correction device, characterized in that, The electrode correction device includes a memory and a processor; The memory is used to store computer programs; The processor is configured to execute the computer program and, in executing the computer program, implement the electrode error correction method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, causes the processor to implement the electrode error correction method as described in any one of claims 1 to 7.

Images