Two-dimensional code adhesive tape multi-layer microstructure AI visual guidance composite manufacturing method

Abstract

The application relates to a two-dimensional code adhesive tape multilayer microstructure AI visual guidance composite manufacturing method in the field of information technology, which comprises the following steps: forming a micron-level pit array on the upper surface of an adhesive tape base material layer through a micro-imprinting process to construct a microstructure foundation; performing selective ablation treatment on the inside of the pit array by focusing a laser beam to obtain preliminary two-dimensional code information unit distribution data; setting a position mark layer according to the adhesion force distribution parameters, generating position deviation data during the compounding of multilayer materials through multi-angle image acquisition; adjusting the postures of each functional layer by adopting closed-loop regulation instructions to obtain adjusted interlayer position matching parameters; analyzing the two-dimensional code pattern boundary definition in combination with the interlayer position matching parameters to output stable contour data; and performing digital simulation verification on information traceability records through the stable contour data to determine final two-dimensional code adhesive tape manufacturing parameter configurations.

Description

Technical Field

[0001] This invention relates to the field of information technology, and in particular to a method for manufacturing a multi-layer microstructure AI visual-guided composite of QR code tape. Background Technology

[0002] In the manufacturing process of power battery cells, QR code tape serves as a key information traceability carrier, directly affecting the efficiency of production quality traceability and fault location. However, existing processes face a series of technical challenges.

[0003] First, when traditional laser marking is applied directly to the surface of the substrate, severe laser energy scattering causes interference between adjacent information units, resulting in blurred QR code pattern boundaries and an inability to form a high-resolution microstructure array. Second, this boundary blurring further weakens the anchoring effect of the subsequent transparent protective layer, leading to insufficient bonding strength between the protective layer and the substrate. Under the high temperature and pressure conditions of cell assembly, this can easily peel off, exacerbating pattern distortion and introducing additional interface reflection interference, amplifying the loss of clarity. Third, the lack of a precise alignment mechanism in the multi-layer composite process results in relative positional deviations of tens of micrometers between functional layers, causing localized occlusion or stretching deformation of the QR code pattern. This, combined with the aforementioned boundary blurring and adhesion problems, creates a vicious cycle, ultimately reducing the QR code recognition rate after composite processing to below 70%.

[0004] In battery cell service scenarios, the digital verification of such tape traceability records is inefficient, with an average response time of more than 5 seconds, which cannot meet the millisecond-level real-time monitoring requirements and seriously restricts the closed-loop feedback of intelligent manufacturing and batch defect traceability. Summary of the Invention

[0005] This invention provides a method for manufacturing multilayer microstructure AI vision-guided composites of QR code tape, mainly including:

[0006] A micron-scale pit array is formed on the surface of the adhesive tape substrate layer using a micro-imprinting process to construct the microstructure foundation. A laser beam is focused on the pit array for selective ablation to obtain preliminary QR code information unit distribution data. Based on this distribution data, the constraint effect of the pit sidewalls on the laser energy is analyzed, and a transparent protective layer is filled to determine the adhesion distribution parameters between the protective layer and the substrate layer. An alignment mark layer is set according to these adhesion distribution parameters, and positional deviation data for multi-layer material composite is generated through multi-angle image acquisition. Closed-loop control commands are used to adjust the posture of each functional layer, obtaining the adjusted interlayer position matching parameters. The QR code pattern boundary clarity is analyzed based on these interlayer position matching parameters, and stable contour data is output. Digital simulation verification of information traceability recording is performed using this stable contour data to determine the final QR code tape manufacturing parameter configuration. Furthermore, the step of forming a micron-scale pit array on the upper surface of the adhesive tape substrate layer using a microimprinting process to construct a microstructure foundation includes: acquiring initial state data of the upper surface of the adhesive tape substrate layer; forming a micron-scale pit array with a predetermined spatial arrangement using a microimprinting process; determining the array distribution pattern for the pit array and acquiring the geometric parameters of each pit; adjusting the pressure distribution of the microimprinting process according to the geometric parameters to obtain a uniform microstructure foundation; extracting array distribution features from the microstructure foundation and determining whether the distribution uniformity meets a preset threshold; if the distribution uniformity meets the preset threshold, outputting the microstructure foundation data; if not, adjusting the microimprinting process parameters and reconstructing the microstructure foundation; and generating a basic distribution model for subsequent laser processing using the microstructure foundation data to guide the determination of the laser beam focusing position. Furthermore, the selective ablation process using a laser beam focused on the interior of the pit array to obtain preliminary QR code information unit distribution data includes: acquiring laser beam parameters and focusing the laser beam on the interior of each pit in the pit array; determining the preliminary ablation region based on the focusing process and extracting energy distribution data of the ablation region; constraining the laser energy using the pit sidewalls to determine the level of scattering interference and obtain interference reduction results; if the interference reduction results meet a preset threshold, verifying the integrity of the QR code unit through surface flatness detection; extracting feature data from the ablation region to determine the unit position coordinates; constructing an information distribution matrix based on the unit position coordinates to generate preliminary QR code information unit distribution data; and verifying the accuracy of the ablation process using the information distribution matrix to provide data support for subsequent boundary clarity analysis.Furthermore, the step of analyzing the constraint effect of the pit sidewall on laser energy and filling a transparent protective layer based on the distribution data of the QR code information units includes: acquiring the distribution data of the QR code information units, extracting the constraint features of the pit sidewall, and determining the laser energy limiting effect; calculating the boundary sharpness index of each information unit based on the limiting effect, and generating a boundary sharpness distribution map; filling a transparent protective layer on the pit structure according to the boundary sharpness distribution map, using the pit to provide an anchoring structure to enhance the bonding strength; acquiring the interface data between the protective layer and the substrate layer from the anchoring structure, and judging the uniformity of the adhesion distribution parameters; if the adhesion distribution parameters are uniform, adjusting the thickness gradient of the protective layer to determine the final adhesion distribution parameters; verifying the bonding stability between the protective layer and the substrate layer through the final adhesion distribution parameters, providing a basis for the subsequent setting of the alignment mark layer. Furthermore, the step of setting an alignment mark layer based on the adhesion distribution parameters and generating position deviation data during multi-layer material composite through multi-angle image acquisition includes: acquiring the adhesion distribution parameters and determining the alignment mark positions on the upper surface of the transparent protective layer; setting an alignment mark layer for the alignment mark positions to generate a mark layer structure; acquiring image data of the mark layer structure from different angles using multiple industrial cameras to obtain acquired image data; comparing the acquired image data with a preset theoretical template and determining the preliminary position deviation value using pixel coordinate matching; adjusting the composite process based on the preliminary position deviation value and the real-time correction parameters of the composite equipment sensors to generate final position deviation data; and verifying the accuracy of multi-layer material composite through the final position deviation data to provide a basis for subsequent functional layer attitude adjustment. Furthermore, the step of adjusting the attitude of each functional layer using closed-loop control commands to obtain adjusted inter-layer position matching parameters includes: acquiring the position deviation data and determining the closed-loop control commands; driving the precision alignment actuator through the closed-loop control commands to adjust the attitude of each functional layer; acquiring data from the real-time detection feedback mechanism from the attitude adjustment to determine the alignment status of the multi-layer structure; if there is a deviation in the alignment status, generating an error compensation calculation based on the feedback mechanism data; updating the inter-layer position matching data according to the error compensation calculation; obtaining the adjusted matching parameters from the updated position matching data; performing an alignment verification loop through the matching parameters to determine the final high-precision alignment result, and outputting the adjusted inter-layer position matching parameters to support subsequent boundary sharpness analysis.Furthermore, the step of analyzing the QR code pattern boundary clarity by combining the interlayer position matching parameters and outputting stable contour data includes: acquiring the interlayer position matching parameters and extracting QR code pattern composite data from the information unit independence characteristics; determining the multi-layer material stacking accuracy for the composite data and analyzing the boundary clarity improvement effect of the QR code pattern after multi-layer composite; extracting pattern boundary contour data from the analysis results and determining whether the boundary clarity index reaches a preset threshold; if the boundary clarity index is lower than the preset threshold, then backtracking to the laser beam focusing ablation process to adjust the energy distribution inside the pit; regenerating the pattern boundary contour data based on the adjusted energy distribution; and outputting stable contour data through the regenerated contour data to provide a data foundation for subsequent information traceability and verification. Furthermore, the step of digitally simulating and verifying information traceability records using the stable contour data to determine the final QR code tape manufacturing parameter configuration includes: acquiring the stable contour data, verifying information traceability records through digital simulation, and generating a correlation result between the traceability records and the microstructure array distribution; for the correlation result, using a traceability efficiency index calculation method to determine whether the traceability efficiency index has reached a preset standard threshold; if the traceability efficiency index has not reached the preset standard threshold, then backtracking to the alignment marker layer setting stage and adjusting the image acquisition angle to the array distribution alignment position; acquiring new image data based on the adjusted image acquisition angle and re-performing digital simulation verification; obtaining the improved value of the traceability efficiency index from the re-verification; if the improved value reaches the preset standard threshold, then outputting the final QR code tape manufacturing parameter configuration to improve the microstructure array distribution. Furthermore, the step of analyzing the constraint effect of the pit sidewall on laser energy based on the distribution data of the QR code information units includes: acquiring the distribution data of the QR code information units and extracting the constraint features of the pit sidewall corresponding to each information unit; analyzing the distribution state of laser energy inside the pit based on the constraint features to determine the energy limitation effect; calculating the boundary sharpness index of each information unit based on the energy limitation effect and generating a boundary sharpness distribution map; extracting the sharpness data of key areas from the boundary sharpness distribution map and determining whether it meets the preset sharpness standard; if the sharpness data meets the preset standard, outputting the constraint effect verification result; if it does not meet the standard, adjusting the laser beam parameters and re-performing the ablation process; and determining the optimization scheme for the constraint of laser energy by the pit sidewall based on the verification result, providing technical support for subsequent protective layer filling.Furthermore, the digital simulation verification of information traceability records using the stable contour data includes: acquiring the stable contour data, combining it with information traceability records from the cell manufacturing process, and constructing a digital simulation verification model; analyzing the matching degree between the traceability records and the microstructure array distribution for the verification model, and generating a correlation result; calculating the traceability efficiency index based on the correlation result, and determining whether a preset standard threshold has been reached; if the traceability efficiency index has not reached the preset standard threshold, then backtracking to the alignment marker layer setting stage and adjusting the image acquisition angle; acquiring updated image data through the adjusted image acquisition angle, and re-performing the simulation verification; extracting the updated value of the traceability efficiency index from the re-verification; if the updated value reaches the preset standard threshold, then outputting the final verification result to determine the stability of the QR code tape manufacturing parameter configuration.

[0007] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0008] This invention discloses an AI-guided composite manufacturing method for multi-layer microstructures of QR code tape. It proposes an integrated solution to address the challenges of microstructure array construction, laser energy confinement, and multi-layer material alignment accuracy in the manufacturing of QR code tape for power battery cells. By forming a micron-level pit array on the substrate layer and constructing QR code information units using laser ablation, the boundary clarity is improved by combining the confinement effect of the pit sidewalls on the laser energy. Simultaneously, a transparent protective layer is filled to enhance adhesion. High-precision alignment of the multi-layer structure is achieved through multi-angle image acquisition and closed-loop control, ultimately optimizing the QR code pattern contour data and information traceability efficiency. This invention ensures that boundary clarity and traceability efficiency meet standards through a backtracking adjustment mechanism, outputting stable manufacturing parameter configurations. This significantly improves the digital accuracy and process integration of power battery cell information traceability, providing innovative technical support for high-precision information recording in industrial manufacturing. Attached Figure Description

[0009] Figure 1 This is a flowchart illustrating an AI vision-guided composite manufacturing method for a multilayer microstructure of QR code tape according to the present invention. Detailed embodiments.

[0010] To further understand the content of this invention, a detailed description of the invention is provided in conjunction with the accompanying drawings and embodiments. The specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention. It should also be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0011] like Figure 1 This embodiment of a multi-layer microstructure AI vision-guided composite manufacturing method for QR code tape may specifically include:

[0012] Step S101: A micron-level pit array is formed on the upper surface of the QR code tape substrate layer of the power battery cell using a micro-imprinting process. The microstructure base is constructed according to a predetermined spatial arrangement. Then, a laser beam is focused on the interior of each pit for selective ablation. The laser energy is constrained by the pit sidewalls to reduce scattering interference, and preliminary QR code information unit distribution data is obtained.

[0013] A micron-scale array of pits is formed on the surface of the battery cell tape substrate layer using a microimprinting process. A predetermined spatial arrangement is used to construct the microstructure foundation, resulting in an array distribution pattern. For this array distribution pattern, laser beam parameters are acquired, and selective ablation is performed by focusing the laser beam within each pit to determine the initial ablation area. Based on these initial ablation areas, the laser energy is constrained using the pit sidewalls, and energy distribution data is acquired to determine the level of scattering interference and obtain interference reduction results. If the interference reduction results meet a preset threshold, the integrity of the QR code unit is verified through surface flatness detection. Features are extracted from the ablation marks to determine the unit's position coordinates. Based on these unit position coordinates, an information distribution matrix is ​​constructed to obtain the QR code information unit distribution data.

[0014] In one embodiment, a micron-scale array of pits is formed on the surface of a QR code tape substrate layer for a power battery cell using a microimprinting process. This process first prepares the substrate layer, typically using a flexible polymer material such as polyester film to accommodate the curved surfaces of the battery cell. Microimprinting involves using a pre-fabricated mold with a raised pattern on its surface corresponding to a predetermined spatial arrangement. When the mold is pressed into the substrate layer, uniform pressure and appropriate temperature are applied to plastically deform the substrate surface, forming an array of pits with a depth of 1 to 10 micrometers. These pits are arranged in a rectangular grid or honeycomb pattern to construct the microstructure foundation, ensuring uniform spacing between each pit to support stable encoding of subsequent QR code information. This array design helps improve the tape's abrasion resistance and maintains the readability of the QR code in the power battery production environment. Furthermore, the implementation of the microimprinting process includes mold design and imprinting parameter control. The mold can be fabricated using photolithography, with surface protrusions corresponding to the potential positions of QR code units. During imprinting, the pressure is controlled within the range of 0.5 to 2 MPa, and the temperature is maintained between 50 and 100 degrees Celsius to avoid excessive deformation of the substrate.

[0015] For example, on a production line for QR code tape for power battery cells, the substrate surface is first cleaned to remove impurities, then a positioning mold is used for imprinting, followed by cooling and curing of the array structure. This method ensures that the accuracy of the pit array reaches the micrometer level, providing a reliable foundation for laser processing. Through this step, the construction of the microstructure foundation can adapt to tapes of different battery specifications, such as the bonding requirements of cylindrical or square cells.

[0016] Preferably, a laser beam is then focused into each pit for selective ablation. The laser beam uses an Nd:YAG laser with a wavelength of 1064 nm and a power adjustable between 1 and 5 watts. The focusing process reduces the laser beam diameter to 80% of the pit size using an optical lens system, ensuring energy is concentrated at the bottom of the pit. The ablation process involves laser pulses irradiating the material surface, causing localized vaporization or melting removal, thus creating depth variations or material removal zones within the pit. These zones correspond to information units in a QR code, such as a black and white dot matrix. Utilizing the pit sidewalls to confine the laser energy is crucial; these sidewalls act as reflective boundaries, reducing scattering interference from the laser beam.

[0017] Specifically, the inclination angle of the pit sidewalls is designed to be between 45 and 60 degrees. When the laser is incident, the energy reflected by the sidewalls returns to the bottom, enhancing the ablation efficiency and reducing thermal damage to the surrounding materials. This constraint mechanism, verified through optical simulation, can reduce scattering loss to less than 20% of the original, thereby obtaining clear QR code information unit distribution data.

[0018] In one possible implementation, the detailed steps of laser ablation include a scanning system controlling a laser beam to locate each pit individually. The scanning employs a galvanometer system, combined with computer-aided design software to pre-set paths, ensuring selective ablation of only designated pits.

[0019] For example, for a 5mm square QR code area, the array contains hundreds of pits, and the laser processes it at a rate of 10 to 50 pulses per second, with each pulse lasting on the order of nanoseconds. Through this processing, preliminary QR code information unit distribution data is formed, including the binary representation of the unit, used to store battery serial numbers or production information. This data distribution is applied in power battery quality traceability systems to ensure information durability.

[0020] It should be noted that the principle of confining laser energy on the sidewalls of the recesses is based on optical reflection and energy localization. The sidewall material is chosen to have a high reflectivity coating, such as an aluminum film, to further enhance the confinement effect. When the laser beam enters the recess, the unabsorbed energy is reflected multiple times on the sidewalls, concentrating in the target area and reducing interference from scattering to adjacent recesses. In practical operation, this mechanism is optimized by adjusting the laser incident angle to improve energy utilization. In this way, the resulting QR code information unit distribution data exhibits high contrast and a low error rate, making it suitable for automated identification scenarios of power battery cells.

[0021] Specifically, in another embodiment, for QR code tapes used with high-capacity power battery cells, the pit depth of the micro-imprint array can be adjusted to 5 micrometers to match the depth requirements of laser ablation. During laser processing, pulse width modulation is introduced to control the ablation degree, forming multi-level grayscale units and further expanding the information capacity. This variant demonstrates the flexibility of the technical solution, adapting to the encoding needs of different battery types while maintaining domain consistency.

[0022] Understandably, the overall effect of this process is to improve the durability and readability of QR codes. On the power battery assembly line, the QR code tapes formed retain their information integrity even after being subjected to vibration and temperature changes, enabling efficient traceability through this technology. Furthermore, in one implementation, the acquisition of initial QR code information unit distribution data involves optical scanning verification. An array image is captured using a CCD camera, and software analyzes the unit distribution to ensure compliance with predetermined encoding standards. This step confirms the accuracy of the ablation process, supporting subsequent tape applications.

[0023] Step S102: Based on the preliminary QR code information unit distribution data, analyze the limiting effect of the pit sidewall constraint on laser energy, extract the boundary clarity index of each information unit, fill the pit structure with a transparent protective layer, use the pit to provide an anchoring structure for the protective layer to enhance the bonding strength, and determine the adhesion distribution parameters between the protective layer and the substrate layer.

[0024] Preliminary QR code information unit distribution data is obtained, and the concave sidewall constraint features are extracted from the data to determine the laser energy limiting effect. Based on the laser energy limiting effect, the boundary sharpness index of each information unit is calculated to obtain a boundary sharpness distribution map. According to the boundary sharpness distribution map, a transparent protective layer is filled into the concave structure, utilizing the concave to provide an anchoring structure and enhance bonding strength. From the filled anchoring structure, the interface data between the protective layer and the substrate layer is obtained to determine the uniformity of the adhesion distribution parameters. If the adhesion distribution parameters are uniform, the thickness gradient of the protective layer is adjusted based on the interface data to determine the final adhesion distribution parameters.

[0025] In one implementation, based on the preliminary QR code information unit distribution data, the influence of pit sidewall constraints on laser energy is first analyzed.

[0026] Specifically, the analysis is based on the micro-pit structure on the QR code substrate, which is formed by laser etching. The geometry of the sidewalls, such as slope and curvature, limits the diffusion of laser energy.

[0027] For example, in the QR code manufacturing process, when a laser beam irradiates a substrate, sidewall confinement prevents excessive energy scattering, thereby maintaining the precise distribution of information units. The energy confinement effect can be determined by measuring the relationship between the sidewall angle and the laser incident angle.

[0028] For example, when the sidewall angle is 45 degrees, the energy reflectivity increases by 20%, which helps optimize the etching depth and reduce thermal damage. Furthermore, the boundary sharpness index for each information unit is extracted. This index is quantified using image processing techniques involving edge detection of the QR code distribution data.

[0029] It should be noted that the boundary sharpness index is defined as the average gradient value of the edge pixels of the information unit. The calculation process includes applying Gaussian filtering to the initial data to remove noise, and then using the Sobel operator to extract the gradient.

[0030] In one possible implementation, a standard QR code module is considered acceptable if its boundary clarity index is above a threshold of 0.8, which ensures the accuracy of recognition during scanning.

[0031] For example, in high-density QR code scenarios, this extraction step can be applied to data under different lighting conditions, demonstrating its versatility in the field of QR code printing. Simultaneously, a transparent protective layer is filled into the recessed structure. This protective layer uses a polymer material, such as polycarbonate, and is applied to the recesses via spraying or deposition.

[0032] Preferably, the filling process controls the thickness to be 5 to 10 micrometers to cover the pit surface without excessively flattening it.

[0033] In one embodiment, on the QR code storage medium production line, the substrate is first cleaned, and then a protective layer material is injected to ensure uniform distribution. This step improves the abrasion resistance of the QR code. Dimples are used to provide an anchoring structure for the protective layer to enhance bonding strength.

[0034] Specifically, the rough sidewalls and bottom of the pit form anchoring points, which increase the contact area between the protective layer and the substrate, thereby improving mechanical interlocking.

[0035] For example, during the analysis, the anchoring structure was observed using a scanning electron microscope, showing that a pit depth of 2 micrometers improved the bonding strength by 30%, preventing peeling in QR code label applications. In one embodiment, this enhancement is suitable for outdoor QR codes, such as logistics labels, where environmental factors like humidity would test adhesion. The adhesion distribution parameters between the protective layer and the substrate layer are determined. These parameters are derived through tensile testing and surface energy calculations, and include the average adhesion value and the coefficient of spatial variation. Further, the testing process involves applying a uniform force to the QR code sample and measuring the peel threshold.

[0036] For example, an average adhesion of 10 Newtons per square centimeter and a coefficient of variation of less than 0.1 indicate a uniform distribution.

[0037] It should be noted that this determination step is based on the aforementioned pit constraints and boundary indicators to ensure that the parameters reflect the actual limiting effect. In QR code verification scenarios, this parameter is used for quality control, resulting in technical benefits such as extended service life. In another embodiment, the above steps are combined and applied to flexible QR code substrates, such as plastic films. First, the pit sidewall constraints are analyzed, and the laser energy is adjusted to 50 joules per square centimeter to match the thermal sensitivity of the flexible material. Then, the boundary sharpness index is extracted to ensure that the index is stable under bending conditions. When filling the protective layer, the pit anchoring is used to enhance the bonding, and the adhesion distribution parameters are found to be higher in the edge area. This is suitable for QR codes on wearable devices, demonstrating the flexibility of the solution.

[0038] Understandably, through these steps, the technical solution achieves durability and accuracy optimization in the QR code field without introducing subjective evaluation, relying solely on objective measurement and process description.

[0039] Step S103: Based on the adhesion distribution parameters, an alignment mark layer is set on the upper surface of the transparent protective layer. Multiple industrial cameras are used to collect alignment mark images from different angles. The collected image data is compared with a preset theoretical template to generate position deviation data when the multilayer material is composited.

[0040] Based on adhesion distribution parameters, alignment mark positions are obtained from the upper surface of the transparent protective layer, and an alignment mark layer is established to obtain a mark layer structure. Multiple industrial cameras are used to capture alignment mark images from different angles to obtain image data. The acquired image data is compared with a preset theoretical template, and the initial positional deviation during multilayer material lamination is determined by pixel coordinate matching to obtain an initial deviation value. Based on this initial deviation value, the lamination process is adjusted using real-time correction parameters obtained from sensors in the lamination equipment to generate final positional deviation data.

[0041] In one embodiment, an alignment mark layer is provided on the upper surface of the transparent protective layer according to the adhesion distribution parameters.

[0042] Specifically, adhesion distribution parameters refer to the adhesion strength indices of different regions on a material surface. These parameters can be obtained through surface tension testing or simulation software. For example, in the field of multilayer material composites, the transparent protective layer is first divided into central and edge regions. Then, based on the adhesion distribution map obtained from testing, the region with higher adhesion is selected to set the marking layer. The purpose of this is to ensure that the marking is not easily detached during the lamination process, thereby improving alignment stability.

[0043] For example, in the production of multilayer composites for flexible displays, the adhesion distribution parameters of the transparent protective layer, such as polyester film, show an adhesion of 50 Newtons per square meter in the central area and 30 Newtons per square meter in the edge area. Therefore, a cross-shaped marking layer is preferentially printed in the central area to support subsequent image acquisition. Furthermore, alignment mark images are acquired from different angles using multiple industrial cameras.

[0044] In one possible implementation, multiple industrial cameras are positioned above and to the side of the production line; for example, three cameras could capture images from directly above, at a 45-degree left angle, and at a 45-degree right angle, respectively, ensuring a stereoscopic view covering the marking layer. This multi-angle acquisition reduces light and shadow interference and perspective distortion.

[0045] It should be noted that camera parameters, such as resolution set to 1920 x 1080 pixels and acquisition frequency set to 10 frames per second, are used. In scenarios involving multi-layered materials, such as composite optical films, the marker image acquisition process first fixes the material position, then synchronously triggers the camera to capture images, generating multiple image data to capture the precise position of the marker in three-dimensional space, thus providing a comprehensive data foundation for deviation calculation.

[0046] Preferably, the acquired image data is compared with a preset theoretical template.

[0047] Specifically, the theoretical template is an ideal marked image generated based on design drawings, such as a digital template containing standard cross marks. The comparison process includes image preprocessing and feature matching. First, the acquired image is subjected to grayscale conversion and edge enhancement, and then a template matching algorithm is used to calculate similarity, for example, by calculating pixel differences to quantize the offset.

[0048] In one embodiment, for the production of multilayer composite glass, if the similarity is less than 90% during comparison, it is identified as a deviation area. This method can objectively identify positional shifts and support the deviation generation of the claims.

[0049] For example, when generating positional deviation data for multilayer material composites, the deviation value is calculated based on the comparison results.

[0050] Specifically, the deviation data includes the offsets of the X-axis, Y-axis, and rotation angle. For example, by comparing the coordinate difference of the center point of the mark, the X-axis deviation is 0.5 mm and the Y-axis deviation is 0.3 mm.

[0051] In one embodiment, during the composite solar panel process, adhesion parameters are further integrated, and the protective layer position is adjusted if the deviation exceeds a threshold, such as 1 mm. This generation method ensures the accuracy of the composite process, reducing material waste and improving product yield in business operations. In another embodiment, the adhesion distribution parameters can be obtained in conjunction with environmental factors.

[0052] For example, in multilayer composites in humid environments, parameter testing includes humidity effect simulation, first measuring dry adhesion and then adjusting to wet value distribution to optimize the placement of the marking layer. This versatility enhances the flexibility of the technical solution within the same field. Furthermore, the angle selection of multiple industrial cameras can be adjusted according to material thickness.

[0053] For example, for thicker composite panels, an additional bottom camera can be added to capture data from below, supplementing the view's completeness and ensuring the accuracy of the comparison data.

[0054] It should be noted that the theoretical template can be generated using CAD software and updated regularly to match different composite batches. During comparison, if curved materials are involved, the template needs to be projected to accommodate non-planar markings.

[0055] In one embodiment, the output format of the position deviation data is a digital report, including a deviation vector. For example, in the case of a composite electronic circuit board, the generated data is directly imported into the control system to adjust the position of the robotic arm and achieve automated correction.

[0056] For example, in actual operation, the entire process, from adhesion parameter analysis to deviation generation, forms a closed-loop control, which can achieve millimeter-level precision alignment in the field of multilayer material composites.

[0057] Preferably, in scenarios requiring high precision, such as optical lens composites, an image enhancement step is added, such as applying a filtering algorithm, to further refine the comparison accuracy, thereby supporting a wider range of applications.

[0058] Step S104: For the generated position deviation data, a closed-loop control command is used to drive the precision alignment actuator to adjust the attitude of each functional layer. Through a real-time detection and feedback mechanism, high-precision alignment of the multi-layer structure is achieved, and the adjusted inter-layer position matching parameters are obtained.

[0059] The generated position deviation data is acquired, and a closed-loop control command is determined based on this data. This command drives a precision alignment actuator. The precision alignment actuator adjusts the attitude of each functional layer, and data from the attitude adjustment is obtained from a real-time detection feedback mechanism. The real-time detection feedback mechanism is used to determine the alignment status of the multi-layer structure. If there is an alignment deviation, error compensation calculation is generated based on the feedback mechanism data, and the inter-layer position matching is updated based on the error compensation calculation. Adjusted matching parameters are obtained from the inter-layer position matching, and an alignment verification loop is performed using these parameters. The final high-precision alignment is determined through this alignment verification loop, yielding the adjusted inter-layer position matching parameters.

[0060] In one implementation, to obtain the generated position deviation data, it is first necessary to understand the source and generation process of the position deviation data.

[0061] Specifically, positional deviation data is obtained by scanning the multi-layer structure using optical inspection equipment. For example, in display panel manufacturing, a high-resolution camera captures edge markers of each functional layer, and then the relative offsets between these points are calculated. This offset includes horizontal displacement, vertical displacement, and rotation angle, forming a deviation vector for subsequent adjustment. This process ensures data accuracy because optical inspection can capture micron-level differences in real time, avoiding errors introduced by manual measurement. In this way, the deviation data becomes the basic input for closed-loop control. Furthermore, closed-loop control commands drive a precision alignment actuator to adjust the posture of each functional layer. Here, the closed-loop control command refers to the control signal generated based on the deviation data, and the command includes parameters for adjustment amplitude and direction.

[0062] In one possible implementation, the control command is calculated by a control algorithm that considers each component of the deviation vector and generates corresponding drive signals to be sent to the actuator. The precision alignment actuator typically consists of a stepper motor and a fine-tuning platform, enabling precise movement of the functional layer in three-dimensional space.

[0063] It should be noted that this attitude adjustment process involves multi-axis linkage, such as simultaneously correcting deviations in the x-axis, y-axis, and theta axis (Z rotation axis) to achieve initial alignment between layers. The key to this step lies in the closed-loop characteristic of the command, that is, the command is not issued all at once, but is optimized and adjusted through continuous iteration.

[0064] For example, in a multi-layer alignment scenario for a display panel, the specific operational process of adjusting the actuator's posture can be implemented as follows: Assuming deviation data shows a 2-micrometer horizontal offset between the upper and lower layers, the control command drives the motor to gradually move the upper layer in 0.1-micrometer steps, while monitoring the offset change. If the offset decreases, the adjustment continues; if it increases, the adjustment is reversed. This iterative adjustment demonstrates the advantages of a closed loop, enabling real-time response to variability factors on the manufacturing line, such as thermal expansion caused by temperature, thereby improving alignment stability. The operational goal of this process is to reduce the defect rate caused by inter-layer misalignment, which can significantly improve the yield rate in actual production.

[0065] Preferably, high-precision alignment of the multi-layered structure is achieved through a real-time detection feedback mechanism. This mechanism involves the deployment of a sensor network, such as using a laser rangefinder or CCD camera to continuously monitor positional changes after adjustment. The feedback mechanism works as follows: after the actuator adjusts, the sensor immediately acquires new data, compares it with the target matching parameters, and generates an error signal. This signal is fed back to the control system to generate the next set of instructions. This mechanism forms a closed loop, ensuring alignment accuracy at the sub-micron level.

[0066] In one embodiment, the frequency of the feedback loop is set to 10 times per second to meet the needs of high-speed production lines and avoid cumulative errors.

[0067] It is understandable that obtaining the adjusted interlayer position matching parameters is the output of the entire process.

[0068] Specifically, the matching parameters include the final relative position coordinates and alignment error values. For example, after adjustment, a parameter set is calculated based on the interlayer offset being less than 0.5 micrometers through final detection. These parameters can be stored in the system for quality control or subsequent lamination processes. In the display panel field, this parameter matching ensures accurate pixel layer stacking and improves the uniformity of the display effect. In another implementation, the sensitivity of the feedback mechanism can be adjusted for multilayer structures of different thicknesses.

[0069] For example, higher-precision optical sensors are used for thin-film layers, while mechanical tactile feedback is combined for thicker layers. This variation demonstrates the flexibility of the technical solution, but remains confined to the field of precision manufacturing, ensuring the universal applicability of the alignment process.

[0070] In one embodiment, the entire alignment process is integrated into an automated production line: from the generation of deviation data to the end of attitude adjustment and feedback loops, the system coordinates the modules through a central controller. This integrated approach includes low-latency data transmission design, such as using high-speed Ethernet to connect sensors and actuators to achieve millisecond-level response times. The operational benefits of this design are improved production efficiency and reduced inconsistencies caused by human intervention.

[0071] For example, when aligning multiple layers of flexible display panels, deviation data may include additional distortion caused by bending, requiring additional curvature compensation calculations in the control commands. Through real-time feedback, the system gradually straightens the interlayer posture, and the final matching parameters reflect the positional accuracy after bending correction. This scenario highlights the potential of this technology in flexible materials.

[0072] Step S105: Based on the adjusted interlayer position matching parameters and the independence characteristics of information units, analyze the edge clarity improvement effect of the QR code pattern after multi-layer composite, extract the final pattern boundary contour data, and determine if the boundary clarity index is lower than the preset threshold. Then, backtrack to the laser beam focusing ablation process to adjust the energy distribution inside the pit; otherwise, output stable contour data.

[0073] The adjusted interlayer position matching parameters are obtained, and the composite data of the QR code pattern is extracted from the independence characteristics of the information units to determine the multi-layer material stacking accuracy. Regarding the multi-layer material stacking accuracy, the boundary clarity improvement effect of the QR code pattern after multi-layer composite is analyzed to obtain pattern boundary contour data. The final pattern boundary contour is extracted from the pattern boundary contour data, and the boundary clarity index is judged. If the boundary clarity index is lower than a preset threshold, the process is backtracked to the laser beam focusing ablation process to adjust the energy distribution inside the pit. Stable contour data is output based on the adjusted energy distribution inside the pit.

[0074] In one implementation, the multi-layer QR code patterns are first composited according to the adjusted interlayer position matching parameters. The interlayer position matching parameters refer to the coordinate adjustment values ​​used to ensure precise alignment of each layer of the QR code pattern when multiple layers of materials are stacked. For example, interlayer offsets are measured using an optical scanning device, and geometric transformation algorithms are applied to correct positional deviations. This parameter adjustment helps maintain the overall consistency of the pattern. Combined with the independence of information units—that is, each black and white module in the QR code acts as an independent information carrier, unaffected by interference from adjacent modules—the robustness of the pattern is enhanced during the composite process.

[0075] Specifically, when composite multi-layer QR codes, the bottom layer pattern is first located, and then the top layer is superimposed, ensuring that the boundaries of each information unit do not overlap or become blurred. Furthermore, the effect of improved boundary sharpness after multi-layer composite is analyzed. This analysis involves evaluating the sharpness of the composite pattern edges, for example, by scanning the pattern using edge detection operators and calculating the gradient values ​​of boundary pixels to quantify the sharpness improvement. The independence of information units comes into play here, as the independent encoding of each unit allows its boundaries to be optimized independently during composite without affecting the overall information integrity. In this way, the improvement in boundary sharpness after composite can be observed; for example, originally blurred edges become sharper through inter-layer matching, thereby improving the scanning and recognition rate of the QR code.

[0076] Preferably, the final pattern boundary contour data is extracted. This step is achieved through image processing techniques, such as contour tracking methods, which trace the boundary lines pixel by pixel from the composite pattern to form closed contour curve data. The specific process includes first binarizing the pattern to separate the black and white modules, and then applying a chain code algorithm to record the sequence of boundary points, ensuring that the extracted data accurately reflects the geometry of the pattern. This extraction method not only covers standard QR code scenarios but is also applicable to deformed multi-layered composite patterns, demonstrating the versatility of the technology.

[0077] In one possible implementation, the boundary sharpness index is determined to be below a preset threshold. The boundary sharpness index can be defined as the average sharpness value of the boundary contour data, for example, by calculating the curvature and contrast of the contour line segments. If the index is below the threshold, such as a set value of 0.8 (out of 1), it indicates that the boundary is not sharp enough and requires further optimization.

[0078] It should be noted that if the boundary sharpness index falls below a preset threshold, the process reverts to the laser beam focusing ablation stage to adjust the energy distribution within the pits. Laser beam focusing ablation refers to the process of using a laser device to create tiny pits on a material surface to etch a QR code pattern. The energy distribution within the pits involves the power density distribution of the laser beam; for example, the energy is higher in the central region to create a deep pit, while the energy gradually decreases at the edges to avoid over-ablation. During adjustment, the proportion of energy at the center can be increased or the focusing depth can be modified to improve the smoothness of the pit edges. This backtracking mechanism ensures iterative optimization of the pattern boundary. For example, in actual QR code production, a preliminary pattern is first ablated, then composited and inspected. If the sharpness is insufficient, the process returns to adjust the laser parameters and redo the ablation.

[0079] For example, in the composite scenario of QR codes on multi-layer plastic films, initial ablation may lead to uneven energy distribution in the pits, resulting in blurred boundaries. By backtracking and adjusting the energy distribution to a gradient mode, that is, decreasing the energy by 20% from the center of the pit to the edge, the clarity of the composite boundary is improved to above the threshold. This adjustment process explains in detail the principle of energy distribution: uneven energy distribution causes thermal deformation of the material, affecting the sharpness of the boundary, while uniform adjustment can stabilize the material response and achieve a clear pattern. Furthermore, in another embodiment, for multi-layer paper QR codes, the energy distribution of laser ablation can be dynamically adjusted according to the material thickness. For example, low energy is used for thin layers to prevent penetration, while energy is increased for thick layers to ensure the pit depth. This diversity of scenarios demonstrates the flexibility of the technology without exceeding the scope of QR code processing.

[0080] Understandably, if the boundary sharpness index reaches or exceeds a preset threshold, stable contour data is output. This output includes formatting the extracted boundary contour data into a vector file for subsequent printing or scanning applications. Through this judgment, the system avoids unnecessary backtracking and improves production efficiency.

[0081] Specifically, in the production of QR code security labels, the entire process from parameter adjustment to output forms a closed loop, ensuring the stability of the final pattern's boundary clarity. For example, when composite three-layer patterns, data is directly output after the clarity index meets the standard for anti-counterfeiting verification.

[0082] In one embodiment, the method improves the durability and recognition accuracy of QR codes. For example, after multi-layer composite, the stability of the boundary contour data allows the pattern to still be effectively scanned in curved or abrasive environments.

[0083] Step S106: Using the obtained stable contour data, digital simulation verification is performed on the information traceability records in the manufacturing of power battery cells. Combining the characteristics of strong process integration, the traceability records are associated with the distribution of microstructure arrays. If the traceability efficiency index does not meet the preset standard, the process is traced back to the alignment mark layer setting stage to optimize the image acquisition angle; otherwise, the final QR code tape manufacturing parameter configuration is output.

[0084] Stable contour data is obtained from the cell manufacturing process, and traceability records are verified through digital simulation to obtain the correlation results between the traceability records and the microstructure array distribution. For these correlation results, a traceability efficiency index is calculated, which is determined by the ratio of array distribution matching degree to record completeness in the correlation results. It is then determined whether the traceability efficiency index reaches a preset standard threshold. If not, the process backtracks to the alignment marker layer setting stage. In this stage, the current image acquisition angle is obtained, and the angle is adjusted to align with the array distribution, resulting in adjusted image data. Based on this adjusted image data, digital simulation verification is performed again to determine the improved traceability efficiency index value. If the improved traceability efficiency index value reaches the preset standard threshold, the final QR code tape manufacturing parameter configuration is output, and defect pattern recognition is used to correlate the traceability records to improve the microstructure array distribution.

[0085] In one implementation, stable contour data is obtained through image processing technology, which is derived from surface scanning during the manufacturing process of power battery cells.

[0086] Specifically, stable contour data refers to the edge contour information of the cell casing or internal components. This information is captured by optical sensors and then filtered to remove noise, ensuring data stability. This data is used for subsequent traceability record verification, providing a reliable geometric basis. In the field of power battery cell manufacturing, this acquisition method is applicable to different types of cells, such as production lines for cylindrical or prismatic cells, ensuring the accuracy of the traceability system. Furthermore, digital simulation verification of information traceability records involves building virtual models to simulate the actual production environment. Digital simulation verification refers to using software tools to simulate and test traceability records.

[0087] For example, information such as the batch number and production date of the battery cells can be input into the model to simulate the completeness of the traceability path. This verification method, combined with the strong integration of processes, means that the traceability system is seamlessly integrated with the manufacturing process, such as embedding traceability data in real time during the battery cell assembly stage.

[0088] It should be noted that the strong process integration is reflected in the system's ability to directly interface with existing equipment such as laser marking machines, avoiding additional hardware requirements and thus improving overall efficiency.

[0089] For example, associating traceability records with microstructure array distribution is achieved through a mapping algorithm. Microstructure array distribution refers to the microstructures within the battery cell, such as the array arrangement of electrode materials; these distributions affect battery performance and traceability accuracy. The specific process involves first extracting key parameters from the traceability records, such as material batches, and then matching and associating them with the coordinate data of the microstructure array.

[0090] For example, if the traceability record shows a certain batch of electrode materials, it is correlated with the density distribution at the corresponding location in the array. This correlation ensures the accuracy of information traceability and can be applied to quality control scenarios in the manufacturing of power battery cells, such as quickly locating the source of the problem when detecting defects.

[0091] In one possible implementation, if the traceability efficiency index fails to meet the preset standard, the process reverts to the alignment marker layer setting stage. The traceability efficiency index refers to the response time and accuracy of the traceability process.

[0092] For example, the preset standard might be a response time of less than 5 seconds and an accuracy rate of over 95%. If these standards are not met, backtracking means returning to the alignment mark layer settings, i.e., adjusting the position level of the marks on the cell surface. Specifically, optimizing the image acquisition angle involves analyzing image distortion at the current angle and then adjusting to the optimal angle by rotating the sensor, such as from 45 degrees to 30 degrees, to improve the clarity of the contour data. This backtracking mechanism enhances the robustness of the system in battery manufacturing.

[0093] Preferably, after the traceability efficiency index reaches the preset standard, the final manufacturing parameter configuration for the QR code tape is output. The QR code tape refers to the traceability label used on the battery cell, and the manufacturing parameter configuration includes tape size, QR code density, and printing resolution.

[0094] In one embodiment, the parameters are configured as a tape width of 20 mm and a QR code pixel density of 300 dpi to ensure easy scanning on the battery assembly line. This output is applicable to various battery production scenarios, such as the manufacturing of new energy vehicle batteries or energy storage batteries, providing a standardized traceability solution.

[0095] Understandably, in another implementation scenario of power battery cell manufacturing, the entire process can be integrated into an automated production line. First, stable profile data is acquired, then simulation verification and correlation are performed. If efficiency is not up to standard, the angle is optimized; otherwise, the parameters are directly output. This integration demonstrates the versatility of the technical solution.

[0096] For example, in high-capacity production lines, simulation verification can handle batch data, while in small-scale customized production, the focus is on optimizing image angles to accommodate different cell shapes. In this way, the system achieves a balance between the reliability of information traceability and process efficiency.

[0097] Specifically, the correlation process of the microstructure array distribution is further detailed as follows: First, microscopic images of the battery cells are acquired, and array features such as spacing and arrangement patterns are extracted; then, information in the traceability records, such as supplier codes, is mapped to these features to form a correlation database.

[0098] For example, if the array distribution is uneven, the traceability record will mark potential risks, thus guiding subsequent optimization. This detailed correlation helps prevent quality problems in battery manufacturing and ensures the integrity of the traceability record. Furthermore, the principle behind backtracking to optimize the image acquisition angle lies in improving data quality. The specific process includes evaluating the contour stability at the current angle; if blurring exists, adjustments are calculated, such as adjusting to a vertical angle based on geometric transformation principles. In actual battery production lines, this optimization can reduce human intervention and increase automation levels.

[0099] In one embodiment, the effectiveness can be verified after the QR code tape manufacturing parameters are configured.

[0100] For example, scanning tests confirm improved traceability efficiency. This configuration's flexibility allows parameters to be adjusted for different battery types, such as adding a encryption layer for high-energy-density cells.

[0101] It should be noted that the entire technical solution provides an efficient traceability mechanism in the field of power battery cell manufacturing. Through simulation verification and correlation distribution, it ensures information traceability while optimizing process parameters to meet production needs.

[0102] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to any specific implementation. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for manufacturing a multi-layer microstructure AI visual-guided composite of QR code tape, characterized in that, include: A micron-scale pit array is formed on the surface of the adhesive tape substrate layer using a micro-imprinting process to construct the microstructure foundation. A laser beam is focused on the pit array for selective ablation to obtain preliminary QR code information unit distribution data. Based on this distribution data, the constraint effect of the pit sidewalls on the laser energy is analyzed, and a transparent protective layer is filled to determine the adhesion distribution parameters between the protective layer and the substrate layer. An alignment mark layer is set according to these adhesion distribution parameters, and positional deviation data for multi-layer material composite is generated through multi-angle image acquisition. Closed-loop control commands are used to adjust the posture of each functional layer, obtaining the adjusted interlayer position matching parameters. The QR code pattern boundary clarity is analyzed based on these interlayer position matching parameters, and stable contour data is output. Digital simulation verification of information traceability recording is performed using this stable contour data to determine the final QR code tape manufacturing parameter configuration.

2. The method for manufacturing a multi-layer microstructure AI visual-guided composite of QR code tape as described in claim 1, characterized in that, The process of forming a micron-scale pit array on the upper surface of an adhesive tape substrate layer using a microimprinting process to construct a microstructure foundation includes: acquiring initial state data of the upper surface of the adhesive tape substrate layer; forming a micron-scale pit array with a predetermined spatial arrangement using a microimprinting process; determining the array distribution pattern for the pit array and acquiring the geometric parameters of each pit; adjusting the pressure distribution of the microimprinting process according to the geometric parameters to obtain a uniform microstructure foundation; extracting array distribution features from the microstructure foundation and determining whether the distribution uniformity meets a preset threshold; if the distribution uniformity meets the preset threshold, outputting the microstructure foundation data; if not, adjusting the microimprinting process parameters and reconstructing the microstructure foundation; and generating a basic distribution model for subsequent laser processing using the microstructure foundation data to guide the determination of the laser beam focusing position.

3. The method for manufacturing a multi-layer microstructure AI visual-guided composite of QR code tape as described in claim 1, characterized in that, The process of selectively ablating the interior of the pit array using a laser beam to obtain preliminary QR code information unit distribution data includes: acquiring laser beam parameters and focusing the laser beam on the interior of each pit in the pit array; determining the preliminary ablation region based on the focusing process and extracting energy distribution data of the ablation region; constraining the laser energy using the pit sidewalls to determine the level of scattering interference and obtain interference reduction results; if the interference reduction results meet a preset threshold, verifying the integrity of the QR code units through surface flatness detection; extracting feature data from the ablation region to determine the unit position coordinates; constructing an information distribution matrix based on the unit position coordinates to generate preliminary QR code information unit distribution data; and verifying the accuracy of the ablation process using the information distribution matrix to provide data support for subsequent boundary clarity analysis.

4. The method for manufacturing a multi-layer microstructure AI visual-guided composite of QR code tape as described in claim 1, characterized in that, The step of analyzing the constraint effect of the pit sidewalls on laser energy based on the distribution data of the QR code information units and filling a transparent protective layer includes: acquiring the distribution data of the QR code information units, extracting the constraint features of the pit sidewalls, and determining the laser energy limiting effect; calculating the boundary sharpness index of each information unit based on the limiting effect, and generating a boundary sharpness distribution map; filling a transparent protective layer on the pit structure according to the boundary sharpness distribution map, using the pits to provide an anchoring structure to enhance the bonding strength; acquiring the interface data between the protective layer and the substrate layer from the anchoring structure, and judging the uniformity of the adhesion distribution parameters; if the adhesion distribution parameters are uniform, adjusting the thickness gradient of the protective layer to determine the final adhesion distribution parameters; verifying the bonding stability between the protective layer and the substrate layer through the final adhesion distribution parameters, providing a basis for the subsequent setting of the alignment mark layer.

5. The method for manufacturing a multi-layer microstructure AI visual-guided composite of QR code tape as described in claim 1, characterized in that, The step of setting an alignment mark layer based on the adhesion distribution parameters and generating position deviation data for multi-layer material composite through multi-angle image acquisition includes: acquiring the adhesion distribution parameters and determining the alignment mark positions on the upper surface of the transparent protective layer; setting an alignment mark layer for the alignment mark positions to generate a mark layer structure; acquiring image data of the mark layer structure from different angles using multiple industrial cameras to obtain acquired image data; comparing the acquired image data with a preset theoretical template and determining the preliminary position deviation value using pixel coordinate matching; adjusting the composite process based on the preliminary position deviation value and the real-time correction parameters of the composite equipment sensors to generate final position deviation data; and verifying the accuracy of multi-layer material composite through the final position deviation data to provide a basis for subsequent functional layer attitude adjustment.

6. The method for manufacturing a multi-layer microstructure AI visual-guided composite of QR code tape as described in claim 1, characterized in that, The method of adjusting the attitude of each functional layer using closed-loop control commands to obtain adjusted inter-layer position matching parameters includes: acquiring the position deviation data and determining the closed-loop control commands; driving the precision alignment actuator through the closed-loop control commands to adjust the attitude of each functional layer; acquiring data from the real-time detection feedback mechanism from the attitude adjustment to determine the alignment status of the multi-layer structure; if there is a deviation in the alignment status, generating an error compensation calculation based on the feedback mechanism data; updating the inter-layer position matching data according to the error compensation calculation; obtaining the adjusted matching parameters from the updated position matching data; performing an alignment verification loop through the matching parameters to determine the final high-precision alignment result, and outputting the adjusted inter-layer position matching parameters to support subsequent boundary sharpness analysis.

7. The method for manufacturing a multi-layer microstructure AI visual-guided composite of QR code tape as described in claim 1, characterized in that, The step of analyzing the QR code pattern boundary clarity by combining the interlayer position matching parameters and outputting stable contour data includes: acquiring the interlayer position matching parameters and extracting composite data of the QR code pattern from the independence characteristics of information units; determining the multi-layer material stacking accuracy for the composite data and analyzing the boundary clarity improvement effect of the QR code pattern after multi-layer composite; extracting pattern boundary contour data from the analysis results and determining whether the boundary clarity index reaches a preset threshold; if the boundary clarity index is lower than the preset threshold, then backtracking to the laser beam focusing ablation process and adjusting the energy distribution inside the pit; regenerating the pattern boundary contour data based on the adjusted energy distribution; and outputting stable contour data through the regenerated contour data to provide a data foundation for subsequent information traceability and verification.

8. The method for manufacturing a multi-layer microstructure AI visual-guided composite of QR code tape as described in claim 1, characterized in that, The process of digitally simulating and verifying information traceability records using the stable contour data to determine the final QR code tape manufacturing parameter configuration includes: acquiring the stable contour data; verifying information traceability records through digital simulation; generating a correlation result between the traceability records and the microstructure array distribution; determining whether the traceability efficiency index reaches a preset standard threshold based on the correlation result; if the traceability efficiency index does not reach the preset standard threshold, then backtracking to the alignment marker layer setting stage and adjusting the image acquisition angle to the array distribution alignment position; acquiring new image data based on the adjusted image acquisition angle and re-performing the digital simulation verification; obtaining the improved value of the traceability efficiency index from the re-verification; and if the improved value reaches the preset standard threshold, outputting the final QR code tape manufacturing parameter configuration to improve the microstructure array distribution.

9. The method for manufacturing a multi-layer microstructure AI visual-guided composite of QR code tape as described in claim 1, characterized in that, The step of analyzing the constraint effect of the pit sidewall on laser energy based on the distribution data of the QR code information units includes: acquiring the distribution data of the QR code information units and extracting the constraint features of the pit sidewall corresponding to each information unit; analyzing the distribution state of laser energy inside the pit based on the constraint features to determine the energy limitation effect; calculating the boundary sharpness index of each information unit based on the energy limitation effect and generating a boundary sharpness distribution map; extracting the sharpness data of key areas from the boundary sharpness distribution map and determining whether it meets the preset sharpness standard; if the sharpness data meets the preset standard, outputting the constraint effect verification result; if it does not meet the standard, adjusting the laser beam parameters and re-performing the ablation process; and determining the optimization scheme for the constraint of laser energy by the pit sidewall based on the verification result, providing technical support for subsequent protective layer filling.

10. The method for manufacturing a multi-layer microstructure AI visual-guided composite of QR code tape as described in claim 1, characterized in that, The digital simulation verification of information traceability records using the stable contour data includes: acquiring the stable contour data, combining it with information traceability records from the cell manufacturing process, and constructing a digital simulation verification model; analyzing the matching degree between the traceability records and the microstructure array distribution for the verification model, and generating correlation results; calculating the traceability efficiency index based on the correlation results, and determining whether a preset standard threshold has been reached; if the traceability efficiency index has not reached the preset standard threshold, then backtracking to the alignment marker layer setting stage and adjusting the image acquisition angle; acquiring updated image data through the adjusted image acquisition angle, and re-performing the simulation verification; extracting the updated value of the traceability efficiency index from the re-verification; if the updated value reaches the preset standard threshold, then outputting the final verification result to determine the stability of the QR code tape manufacturing parameter configuration.

Images