A curved surface explosion-proof film fitting track generation method and system and a storage medium

Abstract

The application relates to the technical field of device manufacturing, and discloses a curved-surface explosion-proof film fitting track generation method, a system and a storage medium. The method comprises the following steps: acquiring curvature characteristic data, determining the bending degree of each region of a curved-surface screen, and constructing a curvature characteristic mapping table; the moving speed of a fitting tool is adaptively adjusted, the continuous angle change of the posture adjustment of the fitting tool is calculated, and a continuous basic fitting track is generated; the curvature characteristics of key control points are analyzed, the fitting pressure is dynamically adjusted, and a curved-surface explosion-proof film fitting track is generated; the running state of the fitting tool is monitored in real time, the moving speed is adjusted, and the continuous contact state of the fitting tool and the curved-surface screen is maintained; a film material deformation buffer index is constructed, a deformation buffer real-time state is extracted, and a risk region is identified; the curvature deviation of the real-time curvature and the reference value of the mapping table is calculated, the tool parameters are adjusted, and the curved-surface explosion-proof film fitting track is optimized to meet the preset fitting standard. The application improves the accuracy of the curved-surface explosion-proof film fitting track.

Description

Technical Field

[0001] This application relates to the field of device manufacturing technology, and in particular to a method, system and storage medium for generating bonding trajectory of curved explosion-proof film. Background Technology

[0002] Curved screens, with their wide viewing angle and superior appearance, have become the mainstream development direction for display devices. Explosion-proof film lamination is a key process in their production, and the rationality of the lamination trajectory directly determines the lamination quality and finished product yield. Currently, most methods for generating lamination trajectories for curved explosion-proof films rely on empirical parameters from flat screens, employing fixed speed and posture trajectory planning methods without considering the uneven Gaussian curvature distribution of curved screens for differentiated design. Existing technologies cannot accurately identify the characteristic differences in high curvature areas and curvature transition areas, resulting in lamination trajectories lacking fine-tuning. This easily leads to large tool vibrations, high positioning errors, and defects in the explosion-proof film such as wrinkles, bubbles, stress concentration, and even cracking. Furthermore, traditional trajectory generation does not consider the deformation buffering requirements of the explosion-proof film, resulting in a disconnect between parameter adjustment and trajectory optimization during the lamination process. Insufficient monitoring and intervention in curvature abrupt change areas lead to a high risk of lamination defects and difficulty in improving product yield. As the curvature design of curved screens becomes increasingly complex, existing methods can no longer meet the production requirements of high-precision, high-yield explosion-proof film lamination. There is an urgent need for a high-precision lamination trajectory generation method that adapts to the curvature characteristics of curved surfaces and meets the deformation requirements of explosion-proof films. Summary of the Invention

[0003] To address the aforementioned technical problems, this application provides a method, system, and storage medium for generating the bonding trajectory of a curved explosion-proof film, which improves the accuracy of the bonding trajectory of the curved explosion-proof film.

[0004] In a first aspect, this application provides a method for generating the bonding trajectory of a curved explosion-proof film, the method comprising:

[0005] Curvature feature data of the curved screen surface is obtained by laser scanning. Based on the curvature feature data, the degree of bending of each region of the curved screen is determined, and a curvature feature mapping table is constructed.

[0006] The movement speed of the bonding tool is adjusted according to the curvature region type by combining the curvature feature mapping table, the continuous angle change of the bonding tool posture adjustment is calculated, and the tool speed and posture parameters are fused to generate a continuous basic bonding trajectory.

[0007] Identify the key control points of the continuous basic bonding trajectory, analyze the curvature characteristics of the key control points and dynamically adjust the bonding pressure, and generate the curved explosion-proof film bonding trajectory based on the adjusted bonding pressure.

[0008] Based on the real-time monitoring of the bonding trajectory of the curved explosion-proof film, the operating status of the bonding tool is monitored, pressure data is obtained, and the moving speed is adjusted according to a preset threshold to maintain continuous contact between the bonding tool and the curved screen.

[0009] Based on the continuous contact state, a membrane deformation buffer index is constructed, the real-time deformation buffer state of the curved explosion-proof membrane is extracted, the real-time deformation buffer state is correlated with the Gaussian curvature value of the current region and time-series tracking is performed to identify risk areas.

[0010] For the risk area, calculate the curvature deviation between the real-time curvature and the baseline value of the mapping table, adjust the tool parameters based on the curvature deviation, and optimize the bonding trajectory of the curved explosion-proof film to meet the preset bonding standard.

[0011] Secondly, this application provides a curved explosion-proof film bonding trajectory generation system, the system comprising:

[0012] The basic trajectory generation unit is used to acquire curvature feature data of the curved screen surface through laser scanning, determine the degree of bending of each region of the curved screen based on the curvature feature data, and construct a curvature feature mapping table; combine the curvature feature mapping table to adapt and adjust the moving speed of the bonding tool according to the curvature region type, calculate the continuous angle change of the bonding tool posture adjustment, fuse the tool speed and posture parameters, and generate a continuous basic bonding trajectory.

[0013] The bonding adjustment unit is used to identify key control points of the continuous basic bonding trajectory, analyze the curvature characteristics of the key control points and dynamically adjust the bonding pressure, and generate a curved explosion-proof film bonding trajectory based on the adjusted bonding pressure; based on the curved explosion-proof film bonding trajectory, the unit monitors the operating status of the bonding tool in real time, acquires pressure data and adjusts the moving speed according to a preset threshold to maintain continuous contact between the bonding tool and the curved screen.

[0014] The trajectory optimization unit is used to construct a membrane deformation buffer index based on the continuous contact state, extract the real-time deformation buffer state of the curved explosion-proof membrane, associate the real-time deformation buffer state with the Gaussian curvature value of the current area and perform time-series tracking to identify risk areas; calculate the curvature deviation between the real-time curvature and the mapping table reference value for the risk areas, adjust the tool parameters based on the curvature deviation, and optimize the bonding trajectory of the curved explosion-proof membrane to meet the preset bonding standard.

[0015] A third aspect of this application provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the above-described method for generating a curved explosion-proof film bonding trajectory.

[0016] Compared with the prior art, the beneficial effects of the present invention are at least as follows:

[0017] First, surface curvature feature data is collected via laser scanning to construct a normalized two-dimensional curvature feature mapping table. This quantifies the degree of bending in each region and marks areas of uneven curvature, providing accurate surface foundation data for trajectory planning and achieving deep adaptation between the trajectory and the actual curvature of the surface. High, low, and transition zones are divided according to curvature characteristics, and the speed of the bonding tool is differentiated accordingly. The speed is linearly reduced based on the membrane deformation buffering requirements, and the attitude adjustment angle is corrected. A continuous basic trajectory is generated by fusing parameters through cubic spline interpolation and then optimized through secondary smoothing to control tool vibration and speed fluctuations, adapting to the differentiated deformation requirements of the membrane material. Next, key control points of the basic trajectory are identified, and bonding pressure is adjusted differentiated according to Gaussian curvature. Feedback loops are used for fine-tuning until stress uniformity is achieved, integrating pressure parameters into the basic trajectory to eliminate local stress concentration and resolve wrinkles and bubbles caused by uneven pressure. Real-time monitoring of bonding pressure data is performed, and a two-level adaptive threshold related to curvature is constructed after moving average filtering. This enables gradient linear control of tool speed and dynamic optimization of trajectory speed parameters, maintaining continuous contact between the tool and the screen and avoiding bonding defects caused by sudden pressure changes. Finally, the contact area of ​​the bonding tool is divided into sub-regions, and pressure and speed data are collected. A membrane deformation buffer index is constructed using normalized weighting. Combined with Gaussian curvature time-series tracking, bonding defect risk areas are accurately identified, enabling early defect prediction and target area localization, thus improving optimization targeting. Curvature deviations are calculated for risk areas, and tool parameters are adjusted when they exceed a threshold. Quadratic Bezier curve interpolation is used to plan local quadratic trajectories, with progressive pressure application and iterative optimization. Finally, the local and overall trajectories are merged to achieve precise repair of risk areas, avoiding full-domain trajectory reconstruction and ensuring overall trajectory continuity. Attached Figure Description

[0018] 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.

[0019] Figure 1 This is a flowchart of a method for generating the bonding trajectory of a curved explosion-proof film in an embodiment of this application;

[0020] Figure 2 This is a schematic representation of the curvature feature mapping in an embodiment of this application.

[0021] Figure 3 This is a schematic diagram illustrating the pressure adaptive control effect of an embodiment of this application;

[0022] Figure 4 This is a schematic diagram of a curved explosion-proof film bonding trajectory generation system according to an embodiment of this application. Detailed Implementation

[0023] This application provides a method, system, and storage medium for generating the bonding trajectory of a curved explosion-proof film. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0024] For ease of understanding, the specific process of the embodiments of this application is described below. Please refer to [link / reference]. Figure 1 One embodiment of the method for generating the bonding trajectory of a curved explosion-proof film in this application includes:

[0025] Step S1: Obtain curvature feature data of the curved screen surface through laser scanning, determine the degree of bending of each area of ​​the curved screen based on the curvature feature data, and construct a curvature feature mapping table.

[0026] Determining the degree of curvature in different areas of the curved screen includes:

[0027] A laser scanner is used to scan the entire curved screen, collecting Gaussian curvature values ​​and principal curvature data at each scan point, and summarizing them to generate curvature feature data. The curvature feature data is divided into local regions of a preset size, and the extreme values ​​and average values ​​of Gaussian curvature in each local region are calculated. The curvature difference between adjacent regions is compared and the rate of curvature change is calculated to identify regions with uneven Gaussian curvature distribution whose rate of curvature change exceeds a preset value. The actual Gaussian curvature value of each local region is divided by the maximum Gaussian curvature value of the entire curved screen, and normalized to generate the degree of bending within a preset interval. A curvature feature mapping table is constructed with the vertical position of the curved screen as the row and the horizontal position as the column. The curvature gradient between adjacent cells in the curvature feature mapping table is estimated using a linear interpolation method to determine the curvature change trend of each local region, and regions with uneven Gaussian curvature distribution and the degree of bending are marked.

[0028] Specifically, a laser scanner is used to perform a full-area surface scan of the curved screen. As a non-contact surface feature acquisition device, the laser scanner can accurately scan various points on the surface of the curved screen. During the scanning process, the Gaussian curvature value and principal curvature data of each scanned point on the curved screen are collected and summarized to generate curvature feature data. The Gaussian curvature value refers to the product of the two principal curvatures at a certain point on the surface, which is the core quantitative indicator characterizing the degree of local spatial curvature of the surface. The principal curvatures are the maximum and minimum curvatures in different directions at a certain point on the surface. Together, they constitute curvature feature data that can comprehensively reflect the curvature characteristics of the curved screen surface.

[0029] A local region is the smallest analytical unit defined to achieve refined curvature analysis of curved screens, avoiding the loss of curvature features caused by indiscriminate processing of the entire curved screen. The input to this partitioning process is the summarized curvature feature data, and the output is the curvature dataset corresponding to each local region after being divided according to a preset size. Based on the curvature dataset of each local region, the extreme values ​​and average values ​​of Gaussian curvature in each local region are calculated. The extreme values ​​are the maximum and minimum values ​​of Gaussian curvature among all scan points in each local region, and the average value is the arithmetic mean of the Gaussian curvature values ​​of all scan points in that local region. The extreme values ​​and average values ​​can accurately reflect the Gaussian curvature distribution characteristics of a single local region. Subsequently, based on the Gaussian curvature of each local region... The average value is compared with the curvature difference of adjacent regions and the rate of curvature change is calculated. The curvature difference is the absolute difference between the average Gaussian curvature of two adjacent local regions. The rate of curvature change is the ratio of the curvature difference to the physical distance between adjacent local regions. It is a quantitative indicator that characterizes the degree of drastic change in surface curvature from one local region to another. The region with uneven Gaussian curvature distribution is the region on the curved screen with drastic curvature changes and irregular bending degree. It is also the core area where defects such as wrinkles and bubbles are prone to occur during the subsequent bonding of the curved explosion-proof film. The input of this recognition process is the rate of curvature change of each adjacent region and the preset change value. The output is the physical position, boundary range and corresponding curvature characteristics of the region with uneven Gaussian curvature distribution on the curved screen.

[0030] Normalization is a numerical processing method that converts Gaussian curvature values ​​of different magnitudes into dimensionless values ​​within a preset range. It can effectively eliminate the magnitude differences in curvature values ​​themselves, and achieve a unified quantitative comparison of the degree of curvature in different local areas of the curved screen. The degree of curvature is the core quantitative indicator that characterizes the degree of curvature in a local area of ​​the curved screen after normalization. The larger the value, the higher the degree of curvature in that area. The input to the normalization calculation process is the actual Gaussian curvature value of each local area and the maximum Gaussian curvature value of the entire curved screen. The output is the degree of curvature value of each local area within the preset range. After this step, the degree of curvature in local areas at different locations of the curved screen becomes comparable, providing standardized data for the subsequent construction of the curvature feature mapping table.

[0031] The curvature feature mapping table is a two-dimensional data table that associates the physical spatial position of a curved screen with the corresponding curvature feature data. It is the core basic data carrier for subsequent speed adjustment, posture calculation, pressure adjustment, and bonding trajectory generation of the bonding tool. During construction, the vertical physical position of the curved screen is used as the row and the horizontal physical position as the column. The physical spatial coordinates of the curved screen are precisely matched with the curvature feature data. The input to this construction process is the physical spatial coordinates of the curved screen, the bending degree value of each local area, and the curvature feature data. The output is the curvature feature mapping table. The linear interpolation method is a numerical calculation method that estimates the curvature value at any position between cells by fitting a linear function to the known curvature feature data of two adjacent cells. The curvature gradient is a quantitative indicator that characterizes the rate and direction of curvature change in space. It exists in vector form and contains two gradient components, vertical and horizontal, corresponding to the curvature change characteristics of the vertical row and horizontal column in the curvature feature mapping table, respectively. The input to this linear interpolation calculation process is the bending degree value of adjacent cells in the curvature feature mapping table, and the output is the vertical and horizontal curvature gradient between adjacent cells. The curvature change trend refers to the direction and rate of change of the degree of bending in various local areas of a curved screen with spatial position. Specifically, it manifests as the degree of bending increasing or decreasing with the longitudinal and transverse coordinates of the curved screen, or as abrupt or gradual changes in the degree of bending. This is directly related to the deformation buffering requirements during the application of the curved explosion-proof film and is a crucial basis for adjusting the speed and posture of the subsequent application tools. Based on this curvature gradient value, the curvature change trend in various local areas of the curved screen can be accurately determined. Specifically, this is achieved by comprehensively judging the magnitude and positive / negative attribute of the curvature gradient value. The magnitude of the curvature gradient value represents the rate of curvature change; a larger gradient value indicates a higher rate of change. The more drastic the change in curvature along a direction, the smaller the gradient value, indicating a more gradual change in curvature. The curvature gradient value in regions with uneven Gaussian curvature distribution will be significantly greater than in other regions, corresponding to abrupt changes in curvature. The positive or negative attribute of the curvature gradient value characterizes the direction of curvature change. A positive gradient value indicates an increasing trend in curvature along that coordinate direction, while a negative gradient value indicates a decreasing trend. By combining the longitudinal and transverse curvature gradient values ​​of each local region in the curvature feature mapping table, the overall curvature change trend of that local region can be comprehensively determined, clarifying the specific direction and rate of change of curvature in the longitudinal and transverse directions.Finally, the curvature feature mapping table is used to mark the uneven Gaussian curvature distribution areas and the degree of bending in each local area. The location, boundary features, and bending degree values ​​of each local area of ​​the uneven Gaussian curvature distribution areas are accurately associated with and labeled with the physical space coordinates of the mapping table. The curvature feature mapping table can intuitively and accurately reflect the degree of bending, curvature change trend, and uneven Gaussian curvature distribution areas with high risk of defects at each physical location of the curved screen. This provides comprehensive and accurate basic data support for each step of subsequent bonding trajectory generation, ensuring that the parameter adjustment of the bonding tool and the generation of the bonding trajectory are highly adapted to the actual curvature characteristics of the curved screen.

[0032] For example, Figure 2 This diagram illustrates the intended representation of curvature feature mapping, showcasing the mesh division, Gaussian curvature distribution, and normalized bending degree of a local area on a curved screen. The curved screen is divided into a 5×5 regular mesh, with each mesh corresponding to a local analysis region. The shades of gray represent the levels of Gaussian curvature values, and the center of each mesh is labeled with coordinate numbers (row, column). The horizontal and vertical axes are labeled with row and column identifiers, visually demonstrating the spatial distribution and meshed management structure of the curvature feature data.

[0033] Step S2: Combine the curvature feature mapping table to adapt and adjust the moving speed of the bonding tool according to the curvature region type, calculate the continuous angle change of the bonding tool's posture adjustment, and use the cubic spline interpolation method to fuse the tool speed and posture parameters to generate a continuous basic bonding trajectory.

[0034] The generation of continuous basic fitting trajectories includes:

[0035] Based on the curvature feature mapping table, the degree of curvature and the curvature change trend are used to calculate the difference in curvature in each region. Based on preset difference intervals and preset curvature thresholds, the curvature region type of each region is determined. Curvature region types include high curvature regions, low curvature regions, and curvature transition regions. For high curvature regions, the deformation buffering time required for the explosion-proof film is calculated based on the degree of curvature. Based on this deformation buffering time, the moving speed of the bonding tool is linearly reduced from the standard moving speed to a low speed suitable for the deformation of the explosion-proof film. For low curvature regions, the standard moving speed of the bonding tool is maintained to ensure the bonding efficiency of the curved explosion-proof film. From the curvature... Curvature feature data of the curvature transition region is extracted from the feature mapping table. The continuous angle change is calculated by multiplying the curvature difference between adjacent points by a preset coefficient to obtain the attitude adjustment angle within a preset range. The attitude adjustment angle is corrected by the arctangent function to generate attitude parameters. The moving speed and attitude parameters of the bonding tool in each region are extracted to form a key point sequence. The cubic spline interpolation algorithm is applied to interpolate the key point sequence. Densified interpolation points are inserted in the curvature transition region to control the tool speed fluctuation within a preset range. The speed and attitude parameters are fused to generate a continuous basic bonding trajectory that adapts to the deformation buffering requirements of the curved explosion-proof film.

[0036] After generating the continuous basic fitting trajectory, the process also includes:

[0037] Based on the curvature feature mapping table, the moving speed distribution of the bonding tool in each curvature region type is dynamically planned and a speed distribution curve is formed; the curvature gradient data of the curvature transition region is extracted, and the change value of the tilt angle of the tool relative to the surface normal is calculated based on the speed distribution curve geometric analysis. A continuous posture adjustment scheme is generated according to the preset adjustment step size, and the adjustment step size is reduced for small curved screens; the basic bonding trajectory is optimized by secondary smoothing interpolation, and the density of interpolation points is increased at the locations of abrupt changes in trajectory curvature to control the vibration amplitude of the bonding tool within the preset vibration range.

[0038] Specifically, the difference in bending degree refers to the quantitative difference between the normalized bending degree value of each local area and the average bending degree value of the entire curved screen. It can intuitively reflect the relative difference in bending degree of each area. Combined with the preset difference range and preset curvature threshold, the curvature region type of each area is determined. The preset difference range refers to the numerical range marked to distinguish the characteristics of bending degree change. The preset curvature threshold refers to the critical bending degree value that divides high curvature and low curvature. The curvature region type is a high curvature region, low curvature region and curvature transition region divided according to the bending degree and curvature change trend. It is the core basis for realizing the differential adjustment of the fitting tool parameters. In the process of classifying the curvature regions of various local areas on a curved screen, the division of curvature transition regions is determined based on the differences in bending degree and curvature change trend between adjacent local areas in the curvature feature mapping table. Specifically, firstly, regions with bending degree greater than a preset curvature threshold are classified as high curvature regions, and regions with bending degree less than or equal to the threshold are classified as low curvature regions. Then, the curvature change trend of continuous local areas between high and low curvature regions is judged. When the bending degree of a local area is between that of the high and low curvature regions, and the curvature gradient of this area along the screen extension direction is within a preset gradient range and the curvature change is continuous and gradual, this local area is determined to be a curvature transition region. The preset gradient range is a numerical range pre-defined according to the curved screen structure to distinguish between gentle transitions and abrupt changes. The curvature gradient in the curvature transition area will not have a large value like in the high curvature area, nor will it be as stable as in the low curvature area. Instead, it will show a continuous gradual change characteristic from high to low or from low to high. This division method can accurately identify the connection area between the high curvature area and the low curvature area, providing a clear regional basis for the subsequent continuous adjustment of the posture of the fitting tool and the smooth transition of speed. The input of this division process is the bending degree and curvature change trend data in the curvature feature mapping table, and the output is the curvature area type labeling result corresponding to each local area of ​​the curved screen. After the labeling is completed, the fitting tool movement speed can be adapted and adjusted based on different curvature area types.

[0039] For the marked high-curvature areas, the deformation buffering time requirement of the explosion-proof film is first calculated based on the degree of bending in these areas. The deformation buffering time refers to the minimum adaptation time required for the curved explosion-proof film to complete the bending deformation along the curved contour of the high-curvature area. It is positively correlated with the degree of bending; the greater the degree of bending, the longer the deformation buffering time required for the explosion-proof film. The specific calculation is achieved through a quantitative calculation formula calibrated by the lamination process, using the normalized bending degree value ω of the high-curvature area as the core variable, combined with the basic deformation buffering time calibrated based on the physical properties of the explosion-proof film. The calculation is performed using the following formula: Where t is the final calculated deformation buffer time of the explosion-proof film, k is the correlation coefficient between the degree of curvature bending and the deformation buffer time, which is calibrated according to the actual bonding process requirements of the curved explosion-proof film, and ε is the process correction factor, which is used to compensate for the time deviation caused by external factors such as the processing error of different curved screens, the bonding force of the explosion-proof film, and the bonding environment.

[0040] Based on the deformation buffer time requirement, the moving speed of the bonding tool is linearly reduced from the standard moving speed to a low speed adapted to the deformation of the explosion-proof film. The standard moving speed refers to the preset constant moving speed of the bonding tool when bonding the explosion-proof film in the low curvature area, taking into account both bonding efficiency and bonding quality. The linear speed reduction means that the moving speed of the bonding tool is gradually reduced from the standard moving speed through a linear function fitting method to avoid defects such as wrinkles and cracks in the explosion-proof film due to insufficient deformation time caused by sudden speed changes. The input of this linear speed reduction process is the deformation buffer time requirement and the standard moving speed, and the output is the low speed value of the bonding tool adapted to the high curvature area and the corresponding speed adjustment command.

[0041] The bending degree in the low curvature area is small, and the deformation buffering requirement of the explosion-proof film is low. There is no need to adjust the moving speed. Maintaining the standard moving speed can maximize the overall bonding efficiency of the curved explosion-proof film while ensuring the bonding quality. The input of this process is the judgment result of the low curvature area, and the output is the control command to maintain the standard moving speed of the bonding tool.

[0042] After completing the differentiated adjustment of the bonding tool's moving speed, the curvature feature data of the curvature transition region is extracted from the curvature feature mapping table. The curvature transition region refers to the local curved surface area connecting the high curvature region and the low curvature region. Its bending degree has a continuous and gradual characteristic, which is the key area for adjusting the bonding tool's posture and also one of the high-incidence areas for explosion-proof film bonding defects. Then, the continuous angular change of the bonding tool's posture adjustment is calculated by multiplying the curvature difference between adjacent points by a preset coefficient. The curvature difference between adjacent points refers to the quantified difference in the bending degree of adjacent physical positions within the curvature transition region. The preset coefficient is a conversion coefficient between the curvature difference and the angle increment, calibrated according to the explosion-proof film material, bonding tool structure, and bonding process requirements. This realizes the quantification conversion of the numerical difference of curvature features into the angular quantity of the tool posture. The continuous angular change is not a single angle value, but refers to the continuous angular change of the bonding tool in the curvature transition region. When the tool moves continuously within the curvature transition region along the curved surface, the sequence of continuous and gradual angle increments generated by the gradual change in curvature of adjacent points is the core data for the continuous posture adjustment of the bonding tool. The calculation process first obtains the posture adjustment angle increment between each group of adjacent points within the curvature transition region by multiplying the curvature difference between adjacent points with a preset coefficient. Then, these angle increments are accumulated sequentially along the spatial position of the curvature transition region to finally form the continuous angle change of the bonding tool within this region. Based on this continuous angle change, the posture adjustment angle within a preset range can be obtained. The posture adjustment angle refers to the tilt angle of the bonding tool relative to the tangent plane at a certain position on the curved screen. Its core function is to ensure effective bonding between the bonding tool and the surface of the curved screen and avoid the problem of loose bonding of the explosion-proof film due to mismatch between the tool posture and the curved surface contour.

[0043] The arctangent function, as a geometric correction function, can eliminate the calculation error of the attitude adjustment angle caused by the gradual change of curvature, correct the deviation between the angle and the actual contour of the curved surface caused by simple numerical calculation, and make the calculated attitude adjustment angle match the actual contour height of the curved screen. The input of the arctangent function correction process is the attitude adjustment angle within a preset range and the curvature gradient data of the curvature transition region. The output is the corrected attitude adjustment angle sequence. Then, the corrected continuous angle change values ​​are standardized and integrated to generate the attitude parameters of the fitting tool. The attitude parameters refer to the set of attitude adjustment angles corresponding to each physical position point of the fitting tool on the curved screen, which is the core attitude data for subsequent trajectory generation.

[0044] After completing the differential adaptation of the moving speed and attitude parameters of the bonding tool, the moving speed and attitude parameters of the bonding tool in the high curvature region, low curvature region and curvature transition region are extracted to form a key point sequence. The key point sequence refers to the discrete data sequence composed of the feature position points of each curvature region and the corresponding speed and attitude parameters, which is the basic data carrier for realizing continuous trajectory generation.

[0045] Cubic spline interpolation is a numerical interpolation algorithm that constructs piecewise cubic polynomial functions, ensuring the continuity of the first and second derivatives of each polynomial at key points. This allows for the fitting of discrete keypoint sequences to continuous curves. The algorithm converts discrete velocity and attitude parameters into continuous parameter variation curves, preventing velocity jitter and abrupt attitude changes during tool application. Its inputs are the velocity-attitude corresponding keypoint sequence and the locations of encrypted interpolation points in the curvature transition region. The output is a continuous velocity-attitude parameter fitting curve. Encrypted interpolation points are inserted in the curvature transition region; these points are additional interpolation calculation points added to the feature region of gradual curvature change. Their function is to improve the precision of trajectory generation in this region. Simultaneously, the algorithm's fitting control keeps tool velocity fluctuations within a preset range. This preset range refers to the critical velocity fluctuation value specified to ensure stable deformation of the curved explosion-proof film. Velocity fluctuations exceeding this range can easily lead to defects such as uneven stretching and wrinkles in the explosion-proof film. Finally, based on the continuous velocity-attitude parameter fitting curve obtained by the cubic spline interpolation algorithm, the velocity and attitude parameters are fused to generate a continuous basic bonding trajectory that adapts to the deformation buffering requirements of the curved explosion-proof film. The continuous basic bonding trajectory refers to the initial bonding trajectory that only completes the adaptation of velocity and attitude parameters, without incorporating bonding pressure parameters. It is the basis for the subsequent generation of pressure-adaptive curved explosion-proof film bonding trajectory. The input of this fusion generation process is the continuous velocity-attitude parameter fitting curve, and the output is the continuous basic bonding trajectory that adapts to the deformation buffering requirements of the curved explosion-proof film.

[0046] Specifically, after generating the above-mentioned continuous basic fitting trajectory, the basic trajectory needs to be refined and optimized in multiple dimensions. First, based on the curvature feature mapping table, the moving speed distribution of the fitting tool in each curvature region type is dynamically planned and a speed distribution curve is formed. The speed distribution curve is a continuous curve of the moving speed of the fitting tool in the entire curved screen as the spatial physical position changes. It can intuitively reflect the global distribution characteristics of the tool speed. The input of this planning process is the curvature region type labeling result of the curvature feature mapping table and the adaptation speed value of each region. The output is the fitting tool speed distribution curve in the entire curved screen. Next, the curvature gradient data of the curvature transition region is extracted. The curvature gradient data refers to vector data that characterizes the rate and direction of change of the curvature transition region's bending degree. Combined with the generated velocity distribution curve, the change value of the tilt angle of the fitting tool relative to the surface normal is calculated through geometric analysis. The surface normal is a straight line perpendicular to the tangent plane at a certain position on the curved screen, which is the spatial baseline for adjusting the posture of the fitting tool. Geometric analysis refers to deriving the correlation between the tilt angle of the fitting tool and the curvature gradient and the movement speed through spatial geometric relationships, so that the calculated change value of the tilt angle is adapted to the tool's movement speed. The input of this calculation process is the curvature gradient data and velocity distribution curve of the curvature transition region, and the output is the change value of the tilt angle of the fitting tool relative to the surface normal.

[0047] Subsequently, a continuous posture adjustment scheme is generated according to the preset adjustment step size. The preset adjustment step size refers to the minimum angle increment for adjusting the posture of the bonding tool, which is the basis for ensuring the continuous and smooth change of the tool posture. For small curved screens, the adjustment step size is reduced. Small curved screens refer to curved screens whose physical size is smaller than the preset size threshold. Their curvature feature areas are more densely distributed. Reducing the adjustment step size can improve the posture adjustment precision of this type of screen and meet the high-precision explosion-proof film bonding requirements. The input of the scheme generation process is the tilt angle change value and the preset adjustment step size, and the output is the continuous posture adjustment scheme of the bonding tool.

[0048] Finally, a secondary smoothing interpolation optimization is performed on the continuous basic bonding trajectory. This secondary smoothing interpolation optimization involves further interpolating the trajectory curve based on the cubic spline interpolation to improve its smoothness. Specifically, the density of interpolation points is increased at locations where the curvature of the trajectory changes abruptly. These locations are also high-incidence vibration points during the bonding tool's operation. Increasing the density of interpolation points at these locations effectively reduces the vibration amplitude of the tool and keeps it within a preset vibration range. This preset vibration range is a critical value for tool vibration amplitude calibrated to ensure the bonding quality of the curved explosion-proof film. Vibration amplitude exceeding this range can easily lead to bonding defects such as local detachment and wrinkles in the explosion-proof film. The input to this secondary smoothing interpolation optimization process is the continuous basic bonding trajectory and the data of the trajectory curvature abrupt change locations. The output is the secondary optimized continuous basic bonding trajectory. This optimized trajectory effectively reduces vibration and positioning errors during the bonding tool's operation, providing more accurate and stable basic trajectory data for the subsequent generation of pressure-adaptive curved explosion-proof film bonding trajectories.

[0049] Step S3: Identify the key control points of the continuous basic bonding trajectory, analyze the curvature characteristics of the key control points and dynamically adjust the bonding pressure, and generate the bonding trajectory of the curved explosion-proof film based on the adjusted bonding pressure.

[0050] The process of generating the curved explosion-proof film bonding trajectory includes:

[0051] Key control points are identified at trajectory inflection points and locations where the curvature gradient exceeds a preset threshold on the continuous basic bonding trajectory. Gaussian curvature values ​​of these key control points are obtained based on a curvature feature mapping table, and each key control point is classified into high-curvature feature points and low-curvature feature points based on these values. For areas containing high-curvature feature points, bonding pressure is increased by a first preset ratio to eliminate local stress concentration. For areas containing low-curvature feature points, bonding pressure is decreased by a second preset ratio to maintain uniform contact between the explosion-proof film and the screen. The adjusted bonding pressure is applied to the bonding tool, and the stress distribution is monitored through feedback loops, with the bonding pressure fine-tuned according to preset increments until the stress distribution in each area is uniform. The pressure distribution parameters corresponding to the final stress homogenization are integrated into the tool speed and attitude parameter settings of the continuous basic bonding trajectory to generate a pressure-adapted curved explosion-proof film bonding trajectory.

[0052] Specifically, key control points refer to the core feature locations on the continuous basic bonding trajectory where curvature characteristics change abruptly, easily causing defects in the bonding of explosion-proof films. They are precise targets for differentiated adjustment of bonding pressure. The identification method is to identify the trajectory turning points and locations where the curvature gradient is greater than a preset gradient threshold on the continuous basic bonding trajectory as key control points. The trajectory turning point refers to the location where the movement direction of the bonding tool changes in the continuous basic bonding trajectory. This location is prone to local pressure changes due to the change in trajectory direction. The curvature gradient is a quantitative indicator that characterizes the rate of change of the curvature of the continuous basic bonding trajectory curve in space. The larger the value, the more drastic the change in trajectory curvature. The preset gradient threshold is a critical value that is pre-calibrated according to the requirements of the curved explosion-proof film bonding process to distinguish between normal and abrupt curvature gradients. The location where the curvature gradient exceeds this threshold is the location of curvature feature abrupt change. After identification, curvature feature analysis and classification can be carried out on each key control point based on the curvature feature mapping table.

[0053] The Gaussian curvature values ​​of each key control point are obtained based on the curvature feature mapping table. These Gaussian curvature values ​​are the actual Gaussian curvature values ​​of the curved screen obtained by laser scanning in step S1, which precisely match the spatial coordinates of the key control points. They are not the curvature values ​​of the trajectory curve. Their core function is to reflect the curvature degree of the actual area of ​​the curved screen corresponding to the key control point, and they are the core basis for the classification of key control points. Based on these Gaussian curvature values, each key control point is classified into high curvature feature points and low curvature feature points. The curvature feature threshold used for classification is based on the curvature distribution of the curved screen. The pre-defined threshold values ​​for features define high curvature feature points as key control points with Gaussian curvature values ​​greater than the threshold, corresponding to high curvature regions of the curved screen. Low curvature feature points are key control points with Gaussian curvature values ​​less than or equal to the threshold, corresponding to low curvature regions of the curved screen. This classification process takes the spatial coordinates of the key control points, the curvature feature mapping table, and the preset curvature feature threshold as inputs, and outputs a dataset of key control points labeled with high and low curvature feature point types, providing a clear basis for region division for differentiated adjustment of fitting pressure.

[0054] After classifying the key control points, the bonding pressure is dynamically adjusted dynamically and differentially for the regions containing different types of feature points. The first preset ratio is a pressure increase ratio pre-calibrated based on the material characteristics of the explosion-proof film and the stress distribution characteristics of the high curvature area of ​​the surface. Due to the large curvature of the surface in this area, local stress concentration is prone to occur during the bonding of the explosion-proof film. Stress concentration refers to the phenomenon that the stress generated during the bonding deformation of the explosion-proof film accumulates in a local area. If it is not dispersed in time, it can easily lead to defects such as cracking and wrinkling of the explosion-proof film. Targeted increase of bonding pressure can effectively disperse the locally accumulated stress and achieve uniform stress distribution; the second preset ratio For example, the pressure reduction ratio is pre-calibrated based on the bonding requirements of low curvature areas. In these areas, the curvature is small and the deformation buffering requirement of the explosion-proof film is low. If high pressure is used for bonding, it is easy to cause excessive compression, which can lead to problems such as displacement of the explosion-proof film and the formation of air bubbles. Targeted reduction of bonding pressure can effectively avoid excessive compression and maintain uniform contact between the explosion-proof film and the curved screen. This differential adjustment process of bonding pressure takes the regional marking results of high and low curvature feature points, the first preset ratio, the second preset ratio, and the basic bonding pressure value of the bonding tool as input, and outputs the initial adjusted bonding pressure value of the area where each key control point is located.

[0055] The initially adjusted bonding pressure is applied to the bonding tool. This application process takes the initially adjusted bonding pressure value as input and outputs a pressure execution command for the bonding tool, causing the tool to complete the bonding action according to the adjusted pressure value. Subsequently, a feedback loop monitors stress distribution and fine-tunes the bonding pressure according to a preset increment. The feedback loop refers to a closed-loop optimization method where, after applying the adjusted bonding pressure to the bonding tool, the stress distribution during the bonding process is monitored in real time, and the bonding pressure is iteratively fine-tuned based on the monitoring results until the stress distribution in all areas of the curved screen reaches a uniform state. The core function of this method is to eliminate pressure adjustment errors caused by process errors and parameter calibration deviations during the initial pressure adjustment process, achieving refined optimization of the bonding pressure. Stress distribution refers to the magnitude and spatial distribution of stress on different areas of the explosion-proof film during the bonding process between the explosion-proof film and the curved screen. Monitoring of this stress is achieved through a pressure sensor and stress detection module mounted on the bonding tool. It can acquire stress data of the entire area of ​​the explosion-proof film in real time. The stress distribution monitoring process takes the real-time pressure data of the bonding process and the bonding status data of the explosion-proof film as input, and outputs the stress distribution monitoring results of the entire curved screen. The preset increment is the smallest unit for fine-tuning the bonding pressure, which is pre-calibrated according to the precision requirements of the bonding process. Its core function is to ensure the smoothness of the pressure fine-tuning process and avoid bonding defects of the explosion-proof film due to sudden pressure changes. If the stress distribution monitoring results show that the stress distribution in a certain area is uneven, the bonding pressure of the corresponding area is increased or decreased in units of the preset increment. The pressure fine-tuning process takes the stress distribution monitoring results and the preset increment as input and outputs the fine-tuned bonding pressure value. The above feedback loop will continue to be executed until the stress distribution in each area of ​​the curved screen reaches a uniform state. The stress uniform state means that the stress on each area of ​​the explosion-proof film is within the preset stress uniform range, without obvious stress accumulation or insufficient pressure.

[0056] After the stress distribution in each region reaches a uniform state, the pressure distribution parameters corresponding to the final stress homogenization are extracted. The pressure distribution parameters refer to the correlation dataset of the bonding pressure value and the spatial coordinates of each region when the stress in each region of the curved screen reaches a uniform state. This dataset can intuitively reflect the optimal bonding pressure of the bonding tool at different positions on the curved screen. Subsequently, the pressure distribution parameters are integrated into the tool speed and attitude parameter settings of the continuous basic bonding trajectory. This fusion process involves accurately matching and integrating the pressure distribution parameters with the original speed and attitude parameters in the continuous basic bonding trajectory across the entire spatial coordinate domain. This ensures that each spatial position point of the continuous basic bonding trajectory corresponds to a unique and suitable speed, attitude, and pressure parameter, achieving deep coupling of the three parameters. This fusion process takes the stress homogenization pressure distribution parameters and the speed-attitude parameter set of the continuous basic bonding trajectory as inputs, and outputs a pressure-adapted curved explosion-proof film bonding trajectory. This trajectory integrates the three core bonding parameters of speed, attitude, and pressure, and can fully adapt to the curvature characteristics of the curved screen and the bonding deformation requirements of the explosion-proof film, providing a precise execution trajectory for real-time monitoring and optimization of the subsequent bonding process.

[0057] Step S4: Based on the real-time monitoring of the bonding tool's operating status according to the bonding trajectory of the curved explosion-proof film, pressure data is obtained and the moving speed is adjusted according to the preset threshold to maintain continuous contact between the bonding tool and the curved screen.

[0058] Adjusting the movement speed according to a preset threshold includes:

[0059] The pressure data includes instantaneous pressure values, pressure change rate over a preset time period, and the difference between the peak pressure and the baseline pressure. The pressure data is processed using a moving average filter. A first-level pressure threshold and a second-level pressure threshold are preset, with the second-level threshold being greater than the first. The first-level pressure threshold is adaptively adjusted based on the curvature characteristics of the current region displayed in the curvature feature mapping table, constructing a two-level adaptive pressure threshold associated with the curvature characteristics. If the instantaneous pressure value exceeds the first-level pressure threshold and the pressure change rate is greater than a preset change rate threshold, the current moving speed of the bonding tool is reduced to a preset percentage range of its original speed. If the instantaneous pressure value exceeds the second-level pressure threshold, the moving speed is reduced to a lower preset percentage of its original speed. The adjustment of the bonding tool's moving speed uses a linear decay method with a preset time constant. If the instantaneous pressure value falls below the first-level pressure threshold, the moving speed of the bonding tool is gradually restored according to a preset speed increment, synchronously and dynamically optimizing the speed parameters of the curved explosion-proof film bonding trajectory while maintaining the tool's posture and trajectory direction unchanged.

[0060] Specifically, based on the curved explosion-proof film bonding trajectory generated in step S3, the actual operating status of the bonding tool is monitored in real time across the entire domain through pressure sensors and motion detection modules mounted on the bonding tool. The focus is on collecting pressure data during the bonding process. This pressure data includes instantaneous pressure values, pressure change rate within a preset time period, and the difference between the peak pressure and the baseline pressure. The instantaneous pressure value refers to the instantaneous quantified value of the bonding pressure detected by the bonding tool at each sampling moment during the bonding process, directly reflecting the actual bonding pressure status at the current moment and serving as the core basic data for judging pressure anomalies. The pressure change rate refers to the ratio of the change in instantaneous pressure value within a preset time period to that preset time period, characterizing the speed of change in bonding pressure. An excessively large value indicates a sudden change in bonding pressure, which can easily lead to bonding defects such as uneven stretching and wrinkles in the explosion-proof film. The baseline pressure refers to the pre-calibrated bonding pressure benchmark value for the corresponding area based on the curvature characteristics of each area in the curvature feature mapping table, serving as the basic reference for judging whether the pressure is abnormal. The difference between the peak pressure and the baseline pressure refers to the quantified difference between the peak pressure detected within a preset time period and the baseline pressure of that area, used to intuitively determine the specific extent of pressure exceeding the standard.

[0061] Moving average filtering is a classic time-domain digital filtering algorithm. Its core principle is to calculate the arithmetic average of several continuously collected pressure data samples to obtain the filtered pressure data at the corresponding sampling time. The core function of this algorithm is to effectively eliminate random errors caused by sensor noise, slight disturbances in the contact environment, and slight vibrations of the tool during the pressure data acquisition process, ensuring the authenticity and stability of the pressure data and avoiding misjudgments and improper adjustments caused by false data. The input of this algorithm is a continuously collected unfiltered pressure data sequence, specifically including the instantaneous pressure value sequence, the pressure change rate sequence, and the difference sequence between the pressure peak and the baseline pressure. The output is a filtered and smoothed pressure data sequence. By averaging adjacent collected data, high-frequency noise in the data is filtered out, and the true trend of pressure data change is preserved.

[0062] The first-level pressure threshold serves as an early warning threshold for pressure anomalies, triggering a slight adjustment to the movement speed of the bonding tool. The second-level pressure threshold is an emergency threshold for pressure anomalies, triggering a deeper adjustment to the movement speed of the bonding tool. This two-tiered setting of pressure thresholds achieves gradient control of pressure anomalies, avoiding the problems of overly rapid or slow control caused by a single threshold. Then, based on the curvature characteristics of the current bonding area displayed in the curvature feature mapping table, the first-level pressure threshold is adaptively adjusted. Finally, a two-level adaptive pressure threshold associated with curvature characteristics is constructed. This adaptive adjustment process dynamically adjusts based on curvature characteristics such as the degree of bending, curvature gradient, and curvature region type of the current area. The system adjusts the pressure threshold accordingly. For example, in high curvature areas and curvature transition areas, where the explosion-proof film requires high deformation buffering, sudden pressure changes are likely to occur during the bonding process. The first-level pressure threshold is appropriately lowered to provide early warning of pressure anomalies. In low curvature areas, the first-level pressure threshold remains unchanged to balance control accuracy and bonding efficiency. The input to the construction process of this two-level adaptive pressure threshold is the preset initial two-level pressure threshold and the curvature feature data of the current region in the curvature feature mapping table. The output is a two-level adaptive pressure threshold that adapts to the curvature features of the current region. This makes the pressure threshold no longer a fixed value across the entire region, but rather deeply adapted to the curvature features of the surface, greatly improving the accuracy of pressure anomaly detection.

[0063] Based on the established two-level adaptive pressure threshold, and combined with filtered pressure data, the moving speed of the bonding tool is dynamically adjusted. First, the instantaneous pressure value and pressure change rate after filtering are judged. If the instantaneous pressure value exceeds the first-level pressure threshold, and the pressure change rate is greater than a preset change rate threshold, the current moving speed of the bonding tool is reduced to a preset percentage range of the original speed. The preset change rate threshold is a critical value pre-calibrated according to the bonding process requirements to determine whether the pressure change is an abnormal sudden change. This threshold setting effectively distinguishes between normal small pressure fluctuations and abnormal sudden changes, avoiding unnecessary speed adjustments caused by normal fluctuations. If the instantaneous pressure value directly exceeds the second-level pressure threshold, it indicates a serious pressure anomaly, and the moving speed of the bonding tool is directly reduced to a lower percentage of the original speed. The system features a preset ratio for emergency speed reduction, and the adjustment of the bonding tool's movement speed employs a linear decay method with a preset time constant. The preset time constant is a quantitative parameter characterizing the rate of linear speed decay, pre-calibrated based on the motion characteristics of the bonding tool and the deformation and buffering characteristics of the explosion-proof film. The linear decay method means that the bonding tool's movement speed starts from the current speed and decreases continuously and smoothly to the target speed at a preset decay rate, rather than a step-like decrease. The inputs to this adjustment method are the current movement speed of the bonding tool, the target speed, and the preset time constant, and the output is a control curve for the linear speed decay. In short, its working principle is to fit the speed change process through a linear function to ensure the smoothness of the speed adjustment and avoid secondary bonding defects such as uneven stretching of the explosion-proof film and vibration of the bonding tool caused by sudden speed changes.

[0064] After adjusting the speed of the bonding tool, the bonding pressure is continuously monitored. If the filtered instantaneous pressure value falls below the first-level pressure threshold, it indicates that the current bonding pressure has returned to the normal range. The speed of the bonding tool is then gradually restored according to the preset speed increment. The preset speed increment is the smallest unit of speed restoration, pre-calibrated according to the bonding process accuracy requirements. The gradual restoration method ensures the smoothness of the speed recovery process and avoids sudden speed increases that could cause pressure anomalies again. Simultaneously, during the speed restoration process, the speed parameters of the curved explosion-proof film bonding trajectory are dynamically optimized while keeping the tool posture and trajectory direction unchanged. This dynamic trajectory optimization is not a complete reconstruction of the bonding trajectory, but rather an update of the real-time adjusted speed parameters to the parameter set of the curved explosion-proof film bonding trajectory without changing the tool posture and movement direction of the original trajectory. This allows the speed parameters of the bonding trajectory to adapt to the actual pressure state of the bonding process in real time, achieving dynamic updates of the bonding trajectory. The inputs to this dynamic optimization process are the adjusted speed parameters and the complete parameter set of the original curved explosion-proof film bonding trajectory, and the output is the optimized curved explosion-proof film bonding trajectory after updating the speed parameters. By real-time acquisition, filtering, gradient threshold judgment, smooth adjustment and gradual recovery of the pressure data, and synchronous dynamic optimization of the bonding trajectory speed parameters, the real-time linkage control of the bonding tool's moving speed and bonding pressure can be achieved. The bonding pressure is always controlled within a range that adapts to the curvature characteristics of the current area, thereby maintaining continuous contact between the bonding tool and the curved screen. This continuous contact refers to the bonding tool maintaining stable contact with the curved screen surface during the bonding process, without gaps or excessive compression. This is the core condition for ensuring a tight bond between the explosion-proof film and the curved screen, avoiding bonding defects such as bubbles, wrinkles, and local detachment. It also lays a stable foundation for the subsequent identification of bonding defect risk areas and the final optimization of the bonding trajectory.

[0065] For example, Figure 3 This diagram illustrates the effect of pressure adaptive control, demonstrating the linkage mechanism between pressure data and movement speed. Figure 3 In the diagram, subplot (a) shows instantaneous pressure data and two levels of adaptive thresholds, where the first level threshold is dynamically adjusted based on curvature characteristics, and the second level threshold is a fixed value; subplot (b) shows pressure change rate analysis, setting a change rate threshold to identify pressure abrupt changes; subplot (c) shows that the movement speed is adaptively adjusted according to the pressure state, including two modes: emergency deceleration and gradient deceleration; subplot (d) visually displays the pressure-speed linkage control mechanism in the form of dual Y-axis, using arrows to indicate the linkage relationship, showing the correspondence between pressure exceeding the standard area and speed adjustment response.

[0066] Step S5: Construct the membrane deformation buffer index based on the continuous contact state, extract the real-time deformation buffer state of the curved explosion-proof membrane, associate the real-time deformation buffer state with the Gaussian curvature value of the current area and perform time-series tracking to identify the risk area of ​​the bonding defect.

[0067] Among them, the risk areas for identifying fitting defects include:

[0068] Based on continuous contact, the contact surface of the bonding tool is divided into multiple sub-regions. Pressure distribution sampling values ​​are obtained for each sub-region, and the set value and actual execution value of the bonding tool's moving speed within a preset control cycle are recorded. Based on the pressure distribution sampling values, the local gradient of pressure distribution between sub-regions and the dynamic rate of change of speed adjustment in adjacent control cycles are calculated. The local gradient and dynamic rate of change are normalized and weighted to calculate the membrane deformation buffer index. The membrane deformation buffer index is compared with a preset index threshold, classifying it into multiple buffer state levels and setting them as the real-time deformation buffer state. The real-time deformation buffer state of the curved explosion-proof membrane is correlated and compared with the Gaussian curvature value of the current sub-region. The real-time deformation buffer state of any sub-region is time-tracked for a preset control cycle. If the Gaussian curvature value reaches a preset first threshold, and the real-time deformation buffer state is continuously at a high-risk buffer state level, any sub-region is determined to be a risk area.

[0069] Specifically, based on the continuous contact state maintained in step S4, the contact surface of the bonding tool is first divided into multiple sub-regions. Each sub-region is the smallest contact analysis unit defined to achieve refined monitoring of the bonding state. This division avoids the problem of insufficient accuracy caused by indiscriminate monitoring of the entire contact surface of the bonding tool, and can accurately capture changes in the bonding state of local areas. Subsequently, the pressure distribution sampling value of each sub-region is obtained, and the set value and actual execution value of the bonding tool's moving speed within the preset control cycle are recorded. The pressure distribution sampling value refers to the quantitative value of the spatial distribution of bonding pressure collected in real time by the pressure sensing module in each sub-region under continuous contact state, which directly reflects the actual bonding pressure state of each local area. The preset control cycle is a continuous data acquisition time interval pre-calibrated according to the requirements of bonding process monitoring accuracy and efficiency. The speed set value refers to the theoretical moving speed value pre-planned for each region in the bonding trajectory of the curved explosion-proof film. The actual speed execution value refers to the actual moving speed value of the bonding tool within the preset control cycle during actual operation. The numerical deviation between the two can intuitively reflect the actual execution effect of the bonding tool speed control.

[0070] The local gradient of pressure distribution is a quantitative indicator that characterizes the rate and direction of change of the bonding pressure distribution between adjacent sub-regions. Its value directly reflects the non-uniformity of the pressure distribution on the contact surface of the bonding tool. The larger the gradient value, the more drastic the pressure change between adjacent sub-regions and the more uneven the pressure distribution. The dynamic change rate of speed adjustment in adjacent control cycles refers to the ratio of the change in the speed adjustment of the bonding tool within two adjacent preset control cycles to the duration of a single preset control cycle. This indicator can characterize the fluctuation of the speed control of the bonding tool. The larger the dynamic change rate, the more unstable the speed control is, which is prone to sudden abnormalities in the deformation buffer of the explosion-proof film. The above two indicators are key characteristic quantities that can reflect the bonding state of the explosion-proof film from the two core dimensions of pressure and speed. They are also the core basic data for constructing the film deformation buffer index. The calculation process of this characteristic quantity takes the sampled value of the pressure distribution of each sub-region and the speed adjustment data of adjacent preset control cycles as input, and outputs the local gradient value of the pressure distribution between each sub-region and the dynamic change rate value of the speed adjustment in adjacent control cycles.

[0071] First, the local gradient of the pressure distribution and the dynamic rate of change of the velocity adjustment are normalized. Then, the normalized eigenvalues ​​are weighted to calculate the membrane deformation buffer index. Normalization is a classic method for standardizing characteristic quantities. Its core principle is to convert original characteristic values ​​of different magnitudes and dimensions into dimensionless values ​​within a preset range through linear transformation. The input of this method is the original value of the local gradient of the pressure distribution and the original value of the dynamic rate of change of the velocity adjustment. The output is the normalized standardized eigenvalues, eliminating the differences in magnitude and dimension between different characteristic quantities, enabling characteristic quantities that were not originally additive to be comprehensively calculated. The weighting process is based on the requirements of the curved explosion-proof membrane lamination process, resulting in two normalized standardized eigenvalues. The numerical processing method assigns preset weight coefficients, which are calibrated based on the degree of influence of each feature quantity on the deformation buffering capacity of the explosion-proof membrane. Among them, the local gradient of pressure distribution has a higher influence on the membrane deformation buffering capacity, and is assigned a relatively higher weight coefficient. The input of this processing is two normalized standardized feature values ​​and preset weight coefficients, and the output is a weighted comprehensive feature value, which is the membrane deformation buffering index. This index is the core indicator for quantitatively characterizing the real-time deformation adaptability of the curved explosion-proof membrane in each sub-region. The larger the index value, the weaker the real-time deformation buffering capacity of the explosion-proof membrane in that sub-region, and the higher the risk of bonding defects such as wrinkles, bubbles, and stress concentration. The smaller the index value, the better the membrane deformation buffering capacity and the lower the risk of bonding defects.

[0072] The calculated membrane deformation buffer index is compared with a preset index threshold. Based on the comparison results, the membrane deformation buffer capacity is divided into multiple buffer state levels, and each buffer state level is set as the real-time deformation buffer state. The preset index threshold is a critical value pre-calibrated according to the characteristics of the explosion-proof membrane material and the requirements of the curved surface bonding process, used to distinguish different levels of deformation buffer capacity. For example, based on the relationship between the membrane deformation buffer index and the preset index threshold, multiple buffer state levels such as low risk, medium risk, and high risk can be divided. The real-time deformation buffer state is a direct representation of the deformation buffer capacity of the curved explosion-proof membrane in each sub-region at the current moment. This state classification process takes the membrane deformation buffer index value and the preset index threshold as inputs and outputs the real-time deformation buffer state corresponding to each sub-region.

[0073] Subsequently, the Gaussian curvature values ​​are correlated and compared. This correlation comparison involves a precise dual matching of spatial coordinates and numerical characteristics between the real-time deformation buffer state, which characterizes the film material's own bonding state, and the Gaussian curvature value, which characterizes the physical features of the curved screen. Since the magnitude of the Gaussian curvature value directly reflects the degree of curvature in a local area of ​​the curved screen, areas with larger Gaussian curvature values ​​have higher deformation requirements for the explosion-proof film and are naturally high-risk areas for bonding defects. Combining this with the real-time deformation buffer state of the film material enables a multi-dimensional and accurate assessment of bonding defect risks. Time-series tracking refers to the continuous monitoring, recording, and trend analysis of the real-time deformation buffer state of any sub-region for multiple preset control cycles. This method can effectively avoid misjudgments of risk caused by accidental data from a single monitoring session, ensuring the accuracy of risk assessment. This correlation tracking process takes the real-time deformation buffer state of each sub-region, the corresponding Gaussian curvature value of the sub-region, and the preset control cycle as inputs, and outputs the correlation tracking data of the sub-region deformation buffer state-Gaussian curvature value for multiple preset control cycles.

[0074] The preset first threshold is a Gaussian curvature critical value, pre-calibrated based on the curvature characteristics of the curved screen, representing a local area of ​​the curved screen with a high degree of curvature. When the Gaussian curvature value reaches this threshold, it indicates that the sub-region belongs to the high curvature zone of the curved screen. The deformation difficulty of the explosion-proof film in this area is significantly higher than in other areas. Furthermore, if it remains in a high-risk buffer state for multiple consecutive cycles, it means that the explosion-proof film in this sub-region is continuously in a state of insufficient deformation buffering capacity, inevitably leading to bonding defects. The risk area for bonding defects is where, during the bonding process of the curved explosion-proof film, the film material's deformation buffering capacity is continuously insufficient, and the corresponding curvature characteristics of the curved screen area are in a high-risk state, easily resulting in wrinkles, bubbles, stress concentration, and explosion-proof film bursting. For sub-regions with bonding defects such as cracks, this risk assessment process takes continuous state-curvature correlation tracking data, a preset first threshold, and a high-risk buffer state level as inputs. The output is the spatial coordinates and range labeling results of the bonding defect risk area in the parametric coordinate system of the curved screen. The risk area assessment achieved in this way combines the objectivity of the curved surface physical characteristics with the real-time nature of the film bonding state. It can accurately locate the potential high-incidence areas of bonding defects, providing accurate target area positioning for subsequent steps to optimize the bonding trajectory of the risk area and perform local secondary bonding. This effectively avoids the problems of full-area inspection and blind optimization in traditional processes, and greatly improves the efficiency and accuracy of bonding trajectory optimization of curved explosion-proof films.

[0075] Step S6: Calculate the curvature deviation between the real-time curvature and the baseline value of the mapping table for the risk area, adjust the tool parameters based on the curvature deviation, and optimize the bonding trajectory of the curved explosion-proof film to meet the preset bonding standard.

[0076] Among these, optimizing the bonding trajectory of the curved explosion-proof film to meet the preset bonding standards includes:

[0077] The risk area is spatially matched with the curvature feature mapping table to extract the curvature quantification data of the risk area, including real-time curvature and the mapping table baseline value. A standardized difference calculation is performed between the real-time curvature and the mapping table baseline value to obtain the basic curvature deviation. The basic curvature deviation is then weighted and corrected based on the membrane deformation buffer index of the risk area to generate the curvature deviation. If the curvature deviation exceeds a preset deviation threshold, the posture and pressure parameters of the bonding tool are adjusted. A quadratic Bezier curve interpolation method is used to re-plan the local secondary bonding trajectory of the risk area, and local secondary bonding is performed using a progressive pressure method. The bonding effect is then re-monitored. If the preset standard is not met, the tool parameters are iteratively adjusted and the local secondary bonding trajectory is reconstructed after a preset number of iterations. If the preset standard is met, the local secondary bonding trajectory is merged with the overall pressure-adapted curved explosion-proof membrane bonding trajectory to ensure that the optimized curved explosion-proof membrane bonding trajectory meets the preset bonding standard.

[0078] Specifically, spatial matching refers to precisely aligning the spatial coordinates of the risk area in the parametric coordinate system of the curved screen with the coordinate system of the curvature feature mapping table. This achieves a one-to-one correspondence between the physical location of the risk area and the data location in the mapping table, avoiding curvature data extraction errors caused by coordinate misalignment. The inputs to this matching process are the spatial boundary coordinate data of the risk area, the coordinate system of the curvature feature mapping table, and the curvature data. The output is the precise positioning result of the risk area in the curvature feature mapping table. The curvature quantification data specifically includes real-time curvature and mapping table reference value. Real-time curvature refers to the Gaussian curvature value of the risk area collected in real time by the curvature sensing module during the bonding process. This value is dynamically changing due to factors such as bonding tool operation, explosion-proof film deformation, and stress distribution. It is the core data reflecting the current actual curvature characteristics of the risk area. The mapping table reference value refers to the Gaussian curvature reference value of the local area that perfectly matches the spatial location of the risk area in the curvature feature mapping table constructed in step S1. This value is the original static curvature value of the curved screen before bonding and is not affected by various factors during the bonding process. It is the core reference for curvature deviation calculation.

[0079] Standardized difference calculation is a quantitative calculation method that performs a unified numerical transformation on the original differences between real-time curvature values ​​of different magnitudes and the baseline values ​​of the mapping table. Its core function is to eliminate the magnitude differences in the curvature values ​​themselves, obtain a dimensionless basic curvature deviation, and improve the consistency and accuracy of curvature deviation comparisons in different risk areas. In this application, the input of this calculation method is the real-time curvature value of the risk area and the corresponding baseline value of the mapping table, and the output is the standardized basic curvature deviation value. Through a preset unified numerical transformation rule, the original difference between the real-time curvature and the baseline value of the mapping table is converted into a standardized deviation value within the same numerical range, ensuring the objectivity of deviation judgment.

[0080] The weighted correction process assigns weighting coefficients based on the magnitude of the membrane deformation buffer index. The membrane deformation buffer index is a core indicator that quantifies the real-time deformation buffering capability of the explosion-proof film in risk areas. The larger the value, the weaker the membrane deformation buffering capability in that area, and the higher the sensitivity to curvature deviation. Therefore, a weighting coefficient greater than 1 is assigned to such areas to amplify the curvature deviation value and achieve accurate judgment of the curvature deviation degree in highly sensitive risk areas. Areas with better membrane deformation buffering capability are assigned a weighting coefficient equal to 1, and the basic curvature deviation is directly used as the final curvature deviation. The inputs of this weighted correction process are the basic curvature deviation value, the membrane deformation buffer index of the risk area, and the preset weighting coefficient allocation rules. The output is the weighted corrected curvature deviation. This curvature deviation is a core indicator that quantifies the degree of deviation of the real-time curvature of the risk area from the original reference curvature of the curved screen. The larger the value, the more serious the deviation of the real-time curvature, the worse the adaptability of the explosion-proof film to the contour of the curved screen, and the greater the range of adjustment required for the bonding tool parameters.

[0081] The preset deviation threshold is a critical value pre-calibrated based on the precision requirements of the curved explosion-proof film bonding process. It distinguishes whether the curvature deviation is within the acceptable range of the process. If the curvature deviation exceeds the preset deviation threshold, it indicates that the curvature deviation of the risk area has exceeded the allowable range of the process. It is necessary to make targeted adjustments to the posture and pressure parameters of the bonding tool. The posture parameter adjustment here is to combine the real-time curvature characteristics of the risk area and fine-tune the tilt angle of the bonding tool relative to the surface normal so that the bonding angle of the bonding tool matches the height of the curved contour of the risk area. The pressure parameter adjustment is to make fine-tuning based on the original pressure distribution parameters and according to the specific magnitude of the curvature deviation according to a preset ratio to avoid new bonding defects caused by excessive parameter adjustment. The input of the tool parameter adjustment process is the curvature deviation value, the preset tool parameter adjustment rules, and the curvature characteristic data of the risk area. The output is the adjusted posture and pressure parameters of the bonding tool.

[0082] After adjusting the posture and pressure parameters of the bonding tool, the local secondary bonding trajectory of the risk area is replanned using the quadratic Bezier curve interpolation method. The quadratic Bezier curve interpolation method is a classic parametric curve interpolation algorithm. Its core principle is to construct a quadratic polynomial curve by setting three feature points: the start point, the end point, and a control point of the bonding trajectory in the risk area, so as to achieve smooth interpolation of the bonding trajectory in the risk area. The core role of this algorithm in this application is to ensure that the planned local secondary bonding trajectory has good continuity and smoothness, and to avoid secondary bonding defects such as bonding tool vibration and uneven stretching of explosion-proof film caused by abrupt trajectory changes. Its input is the spatial boundary coordinates of the risk area, the start and end point parameters of the bonding trajectory, and the coordinates of the preset control point. The output is the local secondary bonding trajectory curve of the risk area. A continuous and smooth quadratic curve is fitted by the spatial coordinates of the three feature points, which serves as the bonding trajectory specific to the risk area.

[0083] The progressive pressure method refers to the bonding tool starting with a lower initial pressure during the local secondary bonding process, gradually and smoothly increasing to the adjusted target pressure as the local secondary bonding trajectory progresses, rather than directly applying the target pressure for bonding. The core function of this method is to gradually release the local stress accumulated in the risk area, avoiding bonding defects such as wrinkles and bubbles in the explosion-proof film caused by sudden pressure. Its inputs are the adjusted target pressure value and the local secondary bonding trajectory curve of the risk area, and the output is the pressure control curve of progressive pressure. After the local secondary bonding process is completed, the bonding effect of the area is re-monitored through pressure sensing and stress detection modules. The core content of bonding effect monitoring is the pressure distribution state, stress distribution state, and deformation bonding state of the explosion-proof film in the risk area, in order to determine whether the preset bonding standard has been met. The preset bonding standard refers to the judgment standard pre-established according to the bonding process requirements of curved explosion-proof film, specifically that the explosion-proof film is tightly bonded to the curved screen, without wrinkles, bubbles, or local stress concentration, and the pressure and stress are evenly distributed throughout the entire area.

[0084] If the bonding effect after re-monitoring does not meet the preset bonding standard, the tool parameters are adjusted iteratively and the local secondary bonding trajectory is reconstructed according to the preset number of iterations. The preset number of iterations is the maximum number of iterations pre-calibrated based on the bonding process efficiency and accuracy requirements to avoid infinite iterations affecting production efficiency. Each iteration will finely adjust the posture and pressure parameters of the bonding tool based on the previous bonding effect monitoring data. Then, based on the finely adjusted parameters, the local secondary bonding trajectory is re-planned using the quadratic Bezier curve interpolation method until the bonding effect meets the preset bonding standard or the maximum number of iterations is reached. The input of this iterative optimization process is the previous bonding effect monitoring data, the finely adjusted tool parameters, and the spatial coordinate data of the risk area. The output is the reconstructed local secondary bonding trajectory curve. If the bonding effect of the partial secondary bonding meets the preset bonding standard, the local secondary bonding trajectory is then fused with the overall pressure-adapted curved explosion-proof film bonding trajectory. Trajectory fusion refers to the precise spatial coordinate alignment of the speed, attitude, and pressure parameters of the local secondary bonding trajectory with the corresponding parameters of the overall trajectory, so that the local secondary bonding trajectory is seamlessly integrated into the overall trajectory, ensuring the continuity and consistency of the overall bonding trajectory after fusion, and avoiding problems such as trajectory breakpoints and parameter abrupt changes. The input of this fusion process is the local secondary bonding trajectory of the risk area and the overall pressure-adapted curved explosion-proof film bonding trajectory, and the output is the fused optimized curved explosion-proof film bonding trajectory.

[0085] Through the above process, the optimized bonding trajectory of the curved explosion-proof film retains the overall trajectory's adaptability to the curvature characteristics of the curved screen, while also achieving precise repair of bonding defect risk areas. It fully meets the preset bonding standards, ensuring that the curved explosion-proof film can achieve a high-precision, defect-free bonding effect throughout the entire curved screen area.

[0086] The above describes a method for generating a curved explosion-proof film bonding trajectory in the embodiments of this application. The following describes a system for generating a curved explosion-proof film bonding trajectory in the embodiments of this application. Please refer to [link / reference]. Figure 4 One embodiment of the curved explosion-proof film bonding trajectory generation system in this application includes:

[0087] The basic trajectory generation unit is used to acquire curvature feature data of the curved screen surface through laser scanning, determine the degree of bending of each area of ​​the curved screen based on the curvature feature data, and construct a curvature feature mapping table; combine the curvature feature mapping table to adapt and adjust the moving speed of the bonding tool according to the curvature area type, calculate the continuous angular change of the bonding tool's posture adjustment, fuse the tool speed and posture parameters, and generate a continuous basic bonding trajectory.

[0088] The bonding adjustment unit is used to identify key control points of the continuous basic bonding trajectory, analyze the curvature characteristics of the key control points and dynamically adjust the bonding pressure, and generate the bonding trajectory of the curved explosion-proof film based on the adjusted bonding pressure; based on the bonding trajectory of the curved explosion-proof film, the unit monitors the running status of the bonding tool in real time, acquires pressure data and adjusts the moving speed according to a preset threshold to maintain the continuous contact between the bonding tool and the curved screen.

[0089] The trajectory optimization unit is used to construct the membrane deformation buffer index based on the continuous contact state, extract the real-time deformation buffer state of the curved explosion-proof membrane, associate the real-time deformation buffer state with the Gaussian curvature value of the current area and perform time-series tracking to identify risk areas; calculate the curvature deviation between the real-time curvature and the mapping table reference value for the risk areas, adjust the tool parameters based on the curvature deviation, and optimize the bonding trajectory of the curved explosion-proof membrane to meet the preset bonding standard.

[0090] This application also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when the instructions are executed on a computer, cause the computer to perform the steps of the method for generating a curved explosion-proof film bonding trajectory.

[0091] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0092] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0093] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for generating the bonding trajectory of a curved explosion-proof film, characterized in that, The method includes: Curvature feature data of the curved screen surface is obtained by laser scanning. Based on the curvature feature data, the degree of bending of each region of the curved screen is determined, and a curvature feature mapping table is constructed. The movement speed of the bonding tool is adjusted according to the curvature region type by combining the curvature feature mapping table, the continuous angle change of the bonding tool posture adjustment is calculated, and the tool speed and posture parameters are fused to generate a continuous basic bonding trajectory. The generation of continuous basic fitting trajectories includes: Based on the bending degree and curvature change trend of the curvature feature mapping table, the bending degree difference of each region is calculated. Based on a preset difference interval and a preset curvature threshold, the curvature region type of each region is determined. The curvature region types include high curvature regions, low curvature regions, and curvature transition regions. For the high curvature region, the deformation buffering time requirement for the explosion-proof film is calculated based on the bending degree. Based on the deformation buffering time requirement, the moving speed of the bonding tool is linearly reduced from the standard moving speed to a low speed adapted to the deformation of the explosion-proof film. For the low curvature region, the standard moving speed of the bonding tool is maintained. From the curvature feature mapping... Curvature feature data of the curvature transition region is extracted from the table. The continuous angle change is calculated by multiplying the curvature difference between adjacent points by a preset coefficient to obtain the attitude adjustment angle within a preset range. The attitude adjustment angle is corrected by the arctangent function to generate the attitude parameters. The moving speed and attitude parameters of the bonding tool in each region are extracted to form a key point sequence. The cubic spline interpolation algorithm is applied to the key point sequence for interpolation calculation. Densified interpolation points are inserted in the curvature transition region to control the tool speed fluctuation within a preset range. The speed and attitude parameters are fused to generate a continuous basic bonding trajectory that adapts to the deformation buffering requirements of the curved explosion-proof film. Identify the key control points of the continuous basic bonding trajectory, analyze the curvature characteristics of the key control points and dynamically adjust the bonding pressure, and generate the curved explosion-proof film bonding trajectory based on the adjusted bonding pressure. Based on the real-time monitoring of the bonding trajectory of the curved explosion-proof film, the operating status of the bonding tool is monitored, pressure data is obtained, and the moving speed is adjusted according to a preset threshold to maintain continuous contact between the bonding tool and the curved screen. Based on the continuous contact state, a membrane deformation buffer index is constructed, the real-time deformation buffer state of the curved explosion-proof membrane is extracted, the real-time deformation buffer state is correlated with the Gaussian curvature value of the current region and time-series tracking is performed to identify risk areas. For the risk area, calculate the curvature deviation between the real-time curvature and the baseline value of the mapping table, adjust the tool parameters based on the curvature deviation, and optimize the bonding trajectory of the curved explosion-proof film to meet the preset bonding standard.

2. The method for generating the bonding trajectory of the curved explosion-proof film according to claim 1, characterized in that, Determine the degree of curvature in different areas of the curved screen, including: A laser scanner is used to perform a full-area scan of the curved screen, collecting Gaussian curvature values ​​and principal curvature data at each scan point, and summarizing them to generate the curvature feature data. The curvature feature data is divided into local regions of a preset size. The extreme values ​​and average values ​​of Gaussian curvature in each local region are calculated. The curvature difference between adjacent regions is compared and the rate of curvature change is calculated. Regions with uneven Gaussian curvature distribution whose rate of curvature change exceeds a preset value are identified. Divide the actual Gaussian curvature value of each local area by the maximum Gaussian curvature value of the entire curved screen, and normalize to generate the degree of bending within the preset range; A curvature feature mapping table is constructed with the vertical position of the curved screen as the row and the horizontal position as the column. The curvature gradient between adjacent cells is estimated by linear interpolation.

3. The method for generating the bonding trajectory of the curved explosion-proof film according to claim 1, characterized in that, After generating the continuous basic bonding trajectory, the following is also included: Based on the curvature feature mapping table, the moving speed distribution of the fitting tool in each curvature region type is dynamically planned and a speed distribution curve is formed. Extract curvature gradient data of the curvature transition region, calculate the change value of the tilt angle relative to the surface normal based on the velocity distribution curve geometric analysis tool, generate a continuous attitude adjustment scheme according to the preset adjustment step size, and reduce the adjustment step size for small curved screens. The basic bonding trajectory is optimized by secondary smoothing interpolation. The density of interpolation points is increased at locations where the trajectory curvature changes abruptly, and the vibration amplitude of the bonding tool is controlled within the preset vibration range.

4. The method for generating the bonding trajectory of the curved explosion-proof film according to claim 1, characterized in that, Generate the bonding trajectory for the curved explosion-proof film, including: The trajectory turning points and positions where the curvature gradient is greater than a preset gradient threshold on the continuous basic fitting trajectory are identified as the key control points; Based on the curvature feature mapping table, the Gaussian curvature values ​​of the key control points are obtained, and each key control point is classified into high curvature feature points and low curvature feature points based on the Gaussian curvature values. For areas with high curvature feature points, the bonding pressure is increased based on a first preset ratio to eliminate local stress concentration. For areas with low curvature feature points, the bonding pressure is reduced based on a second preset ratio to maintain uniform contact between the explosion-proof film and the screen. The adjusted bonding pressure is applied to the bonding tool, and the stress distribution is monitored through feedback loops. The bonding pressure is then finely adjusted according to the preset increment until the stress distribution in each area is uniform. The pressure distribution parameters corresponding to the final stress homogenization are incorporated into the tool speed and attitude parameter settings of the continuous base bonding trajectory to generate the pressure-adapted curved explosion-proof film bonding trajectory.

5. The method for generating the bonding trajectory of the curved explosion-proof film according to claim 1, characterized in that, Adjust the movement speed according to a preset threshold, including: The pressure data includes instantaneous pressure values, pressure change rate over a preset time period, and the difference between the peak pressure and the baseline pressure. The pressure data is then processed by moving average filtering. A first-level pressure threshold and a second-level pressure threshold are preset, with the second-level pressure threshold being greater than the first-level pressure threshold. The first-level pressure threshold is adaptively adjusted based on the curvature characteristics of the current region as displayed in the curvature feature mapping table, thus constructing a two-level adaptive pressure threshold associated with the curvature characteristics. If the instantaneous pressure value exceeds the first-level pressure threshold and the pressure change rate is greater than the preset change rate threshold, the current moving speed of the bonding tool will be reduced to a preset ratio range of the original speed. If the instantaneous pressure value exceeds the second-level pressure threshold, the moving speed will be reduced to a lower preset ratio of the original speed. The moving speed adjustment of the bonding tool adopts a linear decay method with a preset time constant. If the instantaneous pressure value drops below the first-level pressure threshold, the moving speed of the bonding tool is gradually restored according to the preset speed increment, and the speed parameters of the bonding trajectory of the curved explosion-proof film are dynamically optimized simultaneously to keep the tool posture and trajectory direction unchanged.

6. The method for generating the bonding trajectory of the curved explosion-proof film according to claim 1, characterized in that, Identify risk areas for fitting defects, including: Based on the continuous contact state, the contact surface of the bonding tool is divided into multiple sub-regions, the pressure distribution sampling value of each sub-region is obtained, and the set value and actual execution value of the bonding tool's moving speed within the preset control cycle are recorded. Based on the pressure distribution sampling values, calculate the local gradient of the pressure distribution between each sub-region and the dynamic change rate of the speed adjustment in adjacent control cycles. After normalizing the local gradient and the dynamic change rate, perform weighted processing to calculate the membrane deformation buffer index. The deformation buffer index of the membrane material is compared with a preset index threshold, and divided into multiple gradient buffer state levels, which are then set as the real-time deformation buffer state. The real-time deformation buffer status of the curved explosion-proof film is correlated and compared with the Gaussian curvature value of the current sub-region. The real-time deformation buffer status of any sub-region is time-tracked for a preset control period. If the Gaussian curvature value reaches a preset first threshold and the real-time deformation buffer status is continuously at the high-risk buffer status level, any sub-region is determined to be a risk area.

7. The method for generating the bonding trajectory of the curved explosion-proof film according to claim 1, characterized in that, Optimize the bonding trajectory of the curved explosion-proof film to meet the preset bonding standards, including: The risk region is spatially matched with the curvature feature mapping table to extract the curvature quantification data of the risk region. The curvature quantification data includes the real-time curvature and the baseline value of the mapping table. The basic curvature deviation is obtained by performing a standardized difference operation between the real-time curvature and the baseline value of the mapping table. The basic curvature deviation is then weighted and corrected based on the membrane deformation buffer index of the risk area to generate the curvature deviation. If the curvature deviation exceeds the preset deviation threshold, adjust the posture and pressure parameters of the bonding tool, replan the local secondary bonding trajectory of the risk area, perform local secondary bonding processing in a progressive pressure manner, and re-monitor the bonding effect. If the preset standard is not met, the tool parameters will be adjusted iteratively and the local secondary bonding trajectory will be reconstructed according to the preset number of iterations. If the preset standard is met, the local secondary bonding trajectory will be merged with the overall pressure-adapted curved explosion-proof film bonding trajectory, so that the optimized curved explosion-proof film bonding trajectory meets the preset bonding standard.

8. A system for generating a bonding trajectory for a curved explosion-proof film, used to implement the method for generating a bonding trajectory for a curved explosion-proof film as described in any one of claims 1-7, characterized in that, The system includes: The basic trajectory generation unit is used to acquire curvature feature data of the curved screen surface through laser scanning, determine the degree of bending of each region of the curved screen based on the curvature feature data, and construct a curvature feature mapping table; combine the curvature feature mapping table to adapt and adjust the moving speed of the bonding tool according to the curvature region type, calculate the continuous angle change of the bonding tool posture adjustment, fuse the tool speed and posture parameters, and generate a continuous basic bonding trajectory. The bonding adjustment unit is used to identify key control points of the continuous basic bonding trajectory, analyze the curvature characteristics of the key control points and dynamically adjust the bonding pressure, and generate a curved explosion-proof film bonding trajectory based on the adjusted bonding pressure; based on the curved explosion-proof film bonding trajectory, the unit monitors the operating status of the bonding tool in real time, acquires pressure data and adjusts the moving speed according to a preset threshold to maintain continuous contact between the bonding tool and the curved screen. The trajectory optimization unit is used to construct a membrane deformation buffer index based on the continuous contact state, extract the real-time deformation buffer state of the curved explosion-proof membrane, associate the real-time deformation buffer state with the Gaussian curvature value of the current area and perform time-series tracking to identify risk areas; calculate the curvature deviation between the real-time curvature and the mapping table reference value for the risk areas, adjust the tool parameters based on the curvature deviation, and optimize the bonding trajectory of the curved explosion-proof membrane to meet the preset bonding standard.

9. A computer-readable storage medium storing instructions thereon, characterized in that, When the instruction is executed by the processor, it implements a method for generating the bonding trajectory of a curved explosion-proof film as described in any one of claims 1-7.

Images