A die-cutting gilding anti-scratching optimization method and system based on surface simulation

Abstract

This invention proposes an optimization method and system for anti-scratch in die-cutting and hot stamping based on surface simulation, belonging to the field of die-cutting simulation technology. The method includes: performing model replication analysis on a solid sample with scratch defects to extract defect model data; performing feature mapping on the defect model data to obtain a defect feature distribution; setting the common substructure in the defect feature distribution as a scratch defect mode; performing simulated geometric reconstruction on the original model data of the target object to obtain an initial feature distribution; performing global feature matching on the initial feature distribution based on the scratch defect mode to obtain the points to be protected; performing simulated local lifting on the original model data based on the points to be protected to obtain virtual protection data; and performing difference analysis on the virtual protection data and the original model data to obtain pressure compensation parameters. This invention can improve the efficiency of anti-scratch in die-cutting and hot stamping.

Description

Technical Field

[0001] This invention relates to the field of die-cutting simulation technology, and in particular to a method and system for optimizing die-cutting and hot stamping anti-scratch based on surface simulation. Background Technology

[0002] The current optimization of anti-scratch measures in die-cutting and hot stamping processes generally adopts a physical sampling and trial adjustment mode, relying on manual visual identification of scratch defects on the surface of finished products. It is impossible to perform digital modeling and analysis of scratch morphology, making it difficult to extract the unified feature structure of various scratch defects. The selection of protection points is affected by human judgment bias, resulting in insufficient accuracy in point positioning.

[0003] Traditional process optimization lacks surface simulation reconstruction and global feature matching methods. Local lifting parameters in the protected area lack quantitative calculation formulas, resulting in an inability to achieve a smooth transition between the lifted surface and the original product surface. Pressure compensation parameters rely on repeated measurements and tuning, leading to poor parameter matching. This limits the effectiveness of scratch defect control in die-cutting and hot-stamping products and makes process optimization time-consuming. Therefore, improving the efficiency of scratch prevention in die-cutting and hot-stamping has become an urgent problem to be solved. Summary of the Invention

[0004] This invention provides a surface simulation-based optimization method and system for preventing scratches during die-cutting and hot stamping, in order to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, this invention provides a surface simulation-based optimization method for anti-scratch during die-cutting and hot stamping, comprising:

[0006] S01. Perform model replication analysis on the entity sample with scratch defects to extract the defect model data of the entity sample;

[0007] S02. Perform feature mapping on the defect model data to obtain the defect feature distribution of the entity sample, and set the common substructure in the defect feature distribution as the scratch defect mode;

[0008] S03. Perform simulation geometric reconstruction on the original model data of the target object to obtain the initial feature distribution of the target object;

[0009] S04. Based on the scratch defect pattern, perform global feature matching on the initial feature distribution to obtain the points of the target object to be protected;

[0010] S05. Based on the points to be protected, simulate local lifting of the original model data to obtain virtual protection data of the target object;

[0011] S06. Perform difference analysis on the virtual protection data and the original model data to obtain the pressure compensation parameters of the target object.

[0012] In a preferred embodiment, the step of performing model replication analysis on the physical sample with scratch defects to extract the defect model data of the physical sample includes:

[0013] A sample model of the entity sample is obtained by digitally mapping and reconstructing the entity sample with scratch defects.

[0014] The surface morphology data of the sample model is used to extract the scratched defect area to obtain the defect model data of the entity sample.

[0015] In a preferred embodiment, the step of performing feature mapping on the defect model data to obtain the defect feature distribution of the entity sample includes:

[0016] The defect model data is decoupled by features to obtain the feature component sequence of the defect model data;

[0017] Spatial mapping and association are performed on the feature component sequences to obtain the defect arrangement structure of the entity sample;

[0018] By performing region connectivity aggregation on the defect arrangement structure, the defect feature distribution of the entity sample is obtained.

[0019] In a preferred embodiment, setting the common substructure in the defect feature distribution as a scratch defect pattern includes:

[0020] The defect feature distribution is compared and verified by layering to obtain the repeating regions of the defect feature distribution;

[0021] Contour analysis is performed on the repeated regions to extract the common substructure of the defect feature distribution;

[0022] The common substructure is feature-locked and stored to obtain the scratch defect pattern of the defect feature distribution.

[0023] In a preferred embodiment, the step of performing simulated geometric reconstruction on the original model data of the target object to obtain the initial feature distribution of the target object includes:

[0024] Perform a global geometric scan on the target object to obtain the original model data of the target object;

[0025] Parametric surface fitting is performed on the original model data to obtain a continuous geometric representation of the target object;

[0026] The continuous geometric representation is analyzed for morphological features to obtain the initial feature distribution of the target object.

[0027] In a preferred embodiment, the step of performing global feature matching on the initial feature distribution based on the scratch defect pattern to obtain the points to be protected on the target object includes:

[0028] Orientation feature extraction is performed on the initial feature distribution to obtain the orientation-sensitive feature map of the initial feature distribution;

[0029] Based on the scratch defect pattern, the direction-sensitive feature map is matched and compared to obtain the primary matching point of the target object;

[0030] Based on the surface damage extension path in the scratch defect pattern, the source of potential hazards is traced along the path of the primary matching points to obtain the derivative risk points of the target object.

[0031] The primary matching points and the derived risk points are collected and compiled across the entire domain to obtain the points of protection to be obtained for the target object.

[0032] In a preferred embodiment, the step of simulating local elevation of the original model data based on the points to be protected to obtain virtual protection data for the target object includes:

[0033] Density clustering expansion is performed on the points to be protected to obtain the target lifting area of ​​the target object;

[0034] Based on the scratch defect pattern, the original model data points in the target lifting area are lifted in reverse normal direction to obtain the reference compensation surface unit of the target object.

[0035] The overlapping areas of adjacent reference compensation surface units are weighted and fused to obtain a smooth raised surface of the target object;

[0036] The smooth, raised surface is discretized and sampled using a mesh to obtain virtual protection data for the target object.

[0037] In a preferred embodiment, the step of performing reverse normal lifting on the original model data points within the target lifting area based on the scratch defect pattern to obtain the reference compensation surface element of the target object includes:

[0038] The target uplift amount is calculated for the original model data points within the target uplift area to obtain the target uplift value of the data points. The calculation formula for the target uplift value is as follows:

[0039]

[0040] In the formula, For the first The target elevation value for each of the data points. The index number of the data point. The preset reference lift amount, The depth value of the scratch defect pattern at the data point. The maximum depth of the scratch defect pattern. The distance from the data point to the center of the corresponding point to be protected is denoted as . The standard deviation parameter of the target lifting region. The average curvature of the original model data at the data point. The preset feature length, Let be the arc length parameter extending radially outward from the center of the point to be protected along the curved surface. Let be the second derivative of the depth value of the scratch defect pattern along the direction of the arc length parameter. It is an exponential function;

[0041] Based on the target elevation value of the data points, the normal offset of the original model data points is adjusted to obtain the elevation set of the target object.

[0042] The raised point set is then subjected to shape-preserving smoothing trimming to obtain the reference compensated surface element of the target object.

[0043] In a preferred embodiment, the step of performing difference analysis on the virtual protection data and the original model data to obtain the pressure compensation parameters of the target object includes:

[0044] Extract the simulation space offset of the virtual protection data relative to the original model data, and generate the actual offset sequence of the virtual protection data;

[0045] The actual offset sequence is clustered by region distribution to obtain the continuous lifting blocks of the target object;

[0046] Boundary fitting and extraction are performed on the continuous lifting blocks to obtain the lifting profile of the continuous lifting blocks;

[0047] The pressure response quantization analysis of the raised profile is performed to obtain the pressure compensation parameters of the raised profile.

[0048] To address the aforementioned problems, this invention also provides a surface simulation-based anti-scratch optimization system for die-cutting and hot stamping, the system comprising:

[0049] The model replication and analysis module is used to perform model replication and analysis on entity samples with scratch defects, and extract the defect model data of the entity samples.

[0050] The feature mapping module is used to perform feature mapping on the defect model data to obtain the defect feature distribution of the entity sample, and to set the common substructure in the defect feature distribution as the scratch defect mode.

[0051] The simulation geometry reconstruction module is used to perform simulation geometry reconstruction on the original model data of the target object to obtain the initial feature distribution of the target object;

[0052] The global matching and point-finding module is used to perform global feature matching on the initial feature distribution based on the scratch defect pattern to obtain the points to be protected on the target object.

[0053] The simulation local lifting module is used to simulate local lifting of the original model data based on the points to be protected, so as to obtain virtual protection data of the target object;

[0054] The difference analysis and parameter calculation module is used to perform difference analysis on the virtual protection data and the original model data to obtain the pressure compensation parameters of the target object.

[0055] Compared with the prior art, the present invention has the following beneficial effects:

[0056] 1. This invention extracts defect data by digitally replicating and analyzing physical samples, refines a unified scratch defect pattern through feature mapping, performs full-domain feature matching on the product model based on the defect pattern, traces and collects explicit and derivative risk points, accurately delineates all areas to be protected, realizes refined identification of scratch hazard points, and optimizes the accuracy of protective area delineation.

[0057] 2. This invention calculates the elevation value of the points using a quantitative formula, generates a smooth virtual protection model through surface fusion and trimming, and accurately outputs pressure compensation parameters based on the difference analysis between the original model and the virtual protection data. The parameters are adapted to the curved surface structure of the product, and the process parameters are pre-optimized by simulation modeling, which effectively improves the scratch resistance of die-cut hot stamping products and enhances the overall process optimization efficiency. Attached Figure Description

[0058] Figure 1 This is a flowchart illustrating an optimization method for anti-scratch in die-cutting and hot stamping based on surface simulation, according to an embodiment of the present invention.

[0059] Figure 2 This is a functional block diagram of a surface simulation-based die-cutting and hot stamping anti-scratch optimization system according to an embodiment of the present invention.

[0060] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0061] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0062] This application provides a surface simulation-based method for optimizing the anti-scratch properties of die-cutting and hot stamping. The execution entity of this surface simulation-based method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the surface simulation-based method can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0063] Reference Figure 1 The diagram shown is a flowchart illustrating a surface simulation-based method for optimizing die-cutting and hot-stamping scratch prevention according to an embodiment of the present invention. In this embodiment, the surface simulation-based method for optimizing die-cutting and hot-stamping scratch prevention includes:

[0064] S01. Perform model replication analysis on the entity sample with scratch defects to extract the defect model data of the entity sample;

[0065] In this embodiment of the invention, the step of performing model replication analysis on a physical sample with scratch defects to extract defect model data of the physical sample includes:

[0066] A sample model of the entity sample is obtained by digitally mapping and reconstructing the entity sample with scratch defects.

[0067] The surface morphology data of the sample model is used to extract the scratched defect area to obtain the defect model data of the entity sample.

[0068] The process involves collecting spatial location information from all angles of the physical sample's surface. The data collection is conducted progressively along all contour lines of the sample's outer surface, covering the entire surface, including areas with scratches and undamaged areas. The spatial orientation of each point on the sample's surface is recorded continuously, ensuring the collection of all points across the entire surface. The data collection proceeds along the direction of the curved surface, without skipping any small areas. After all point information is collected, the scattered points are matched to their corresponding spatial orientations and systematically pieced together, adhering strictly to the actual contours of the sample's surface. Scattered points are arranged along the outline, and adjacent points are connected and arranged according to the actual adjacency relationship on the entity surface to ensure that the information arrangement after splicing is consistent with the actual structure of the entity sample. After all the scattered points are spliced ​​in an orderly manner, a complete information collection is formed. This collected content is determined as the sample model. The sample model fully carries the point information of the entire surface of the entity sample collected in the early stage. The point details corresponding to the scratched area are completely preserved in the information content of the sample model. The point content corresponding to the intact area is also completely included in the sample model. No point-related content is deleted throughout the process. The information content corresponding to all the morphological features of the entity sample surface is completely replicated. The sample model is generated based on the information content after the entire splicing and collection.

[0069] Retrieve all surface morphology-related information stored within the sample model, distinguish the attributes of the corresponding areas of the entity sample surface for each piece of information, and identify whether scratches remain at the location corresponding to each individual morphology information. Complete the attribute differentiation of all morphology information, isolate and remove all morphology information corresponding to intact, unscratched areas, and only collect the surface morphology information corresponding to the areas with scratches. Maintain the original arrangement order of the collected scratch-related morphology information within the sample model without changing the order. Based on the original arrangement structure, select and retain various types of information. The morphological information is uniformly integrated and stored. The integration process does not change the details of any morphological information. All surface concave and convex details contained in each morphological information are preserved as is. After integration, a unique dataset is generated. This dataset is determined as defect model data. The defect model data only contains the scratch-specific morphological information extracted from the sample model. It does not contain any morphological content corresponding to the intact areas of the physical sample. The original morphological arrangement details of the scratched area are completely preserved. After the entire process of differentiation, screening, removal of redundant content, and regularization and integration, the final defect model data is produced.

[0070] The beneficial effect is that the sample model obtained by digital mapping and reconstruction based on full-surface point acquisition and orderly splicing can completely contain all surface morphology information of the physical sample, and completely retain all kinds of details corresponding to scratched and intact parts, avoiding the situation of missing local morphology content when manually summarizing sample information. Targeted area interception of the sample model can accurately screen out the morphology content corresponding to the intact areas of the product, and only collect and organize the morphology information related to scratches to form defect model data. This ensures that the final retained data is all focused on scratch defects, eliminating interference from irrelevant data at the source. This provides accurate and pure data support for subsequent defect feature analysis and scratch pattern summarization, and solidifies the data foundation for subsequent die-cutting, hot stamping, and anti-scratch optimization operations.

[0071] S02. Perform feature mapping on the defect model data to obtain the defect feature distribution of the entity sample, and set the common substructure in the defect feature distribution as the scratch defect mode;

[0072] In this embodiment of the invention, the step of performing feature mapping on the defect model data to obtain the defect feature distribution of the entity sample includes:

[0073] The defect model data is decoupled by features to obtain the feature component sequence of the defect model data;

[0074] Spatial mapping and association are performed on the feature component sequences to obtain the defect arrangement structure of the entity sample;

[0075] By performing region connectivity aggregation on the defect arrangement structure, the defect feature distribution of the entity sample is obtained.

[0076] The step of setting the common substructure in the defect feature distribution as a scratch defect pattern includes:

[0077] The defect feature distribution is compared and verified by layering to obtain the repeating regions of the defect feature distribution;

[0078] Contour analysis is performed on the repeated regions to extract the common substructure of the defect feature distribution;

[0079] The common substructure is feature-locked and stored to obtain the scratch defect pattern of the defect feature distribution.

[0080] The various scratch-related contents contained within the defect model data are broken down one by one. The breakdown is carried out item by item according to the inherent attributes of the corresponding shape extension, indentation direction and edge contour. The breakdown process completely sorts out all the information items stored in the data, without omitting any sub-content. After the breakdown is completed, the sub-content with the same attributes are collected together. The collected sub-content is then arranged in an orderly manner according to the original fixed arrangement order in the defect model data. After all the contents are broken down and arranged in an orderly manner, a continuous set of contents sequence is generated. This set of contents sequence is determined as the feature component sequence.

[0081] By comparing the original spatial location of each sub-content in the feature component sequence with its corresponding location in the defect model data, the directional reference information of each individual content is locked. Multiple component contents with consistent directional reference information are then linked together. The linking operation strictly follows the spatial distribution pattern of the original data and connects the scattered contents along the direction of the scratch extension. The directional association and end-to-end connection of all component contents are completed one by one. After all the contents are linked and integrated, a complete and coherent content layout is formed. This formed layout is determined as the defect layout structure.

[0082] Following the boundaries of each sub-content within the defect layout structure, the connection boundaries of adjacent contents are checked one by one. Adjacent contents that can be seamlessly connected at the boundary are grouped together into contiguous content units. All contents contained in the layout structure are continuously traversed to complete the contiguous collection of scattered contents until all contents with connectivity conditions within the layout structure are integrated. The overall set of contiguous collected contents is then determined as the defect feature distribution.

[0083] Multiple independently generated defect feature distribution contents are retrieved, and the content information corresponding to various defect feature distributions is superimposed layer by layer based on the same reference benchmark. After superposition, the outline direction and content layout style of each layer are checked layer by layer. The range blocks where the layout of different levels of content completely overlap are marked. After the comparison and verification work of all levels is completed, all marked overlapping blocks are summarized, and all the summarized and gathered block contents are uniformly designated as duplicate areas.

[0084] By gradually tracing the boundary extension along the complete boundary line of the repeated area, the entire internal content range enclosed by the boundary line is completely delineated. Without altering the original layout details of the internal content, all content within the boundary range is separated from the original content system. The separated complete set of content is then organized and determined as a common substructure.

[0085] The complete outline shape and internal content layout rules of the fixed common substructure are completely fixed, and no detail of the common substructure is changed. The complete set of common substructure content with fixed shape is archived for a long time. The standardized reference content formed after shape locking and content recording is determined as the scratch defect mode.

[0086] The beneficial effects are that by breaking down and integrating defect-related data step by step, the inherent patterns of scratch-related information can be analyzed layer by layer. Based on the defect feature distribution obtained from the breakdown and aggregation, the arrangement details of various scratches can be fully gathered, accurately distinguishing between differentiated and common defects. By using a layered comparison method to filter duplicate areas and then extract common substructures, unique scratch details of various samples can be eliminated, and the unified morphological characteristics of all scratches can be condensed. Locking in the archived scratch defect patterns can form a standardized reference benchmark, providing a unified reference for subsequent identification of potential product hazards, reducing interference from invalid information, and ensuring that the judgment criteria for subsequent scratch-resistant optimization work remain consistent.

[0087] S03. Perform simulation geometric reconstruction on the original model data of the target object to obtain the initial feature distribution of the target object;

[0088] In this embodiment of the invention, the step of performing simulated geometric reconstruction on the original model data of the target object to obtain the initial feature distribution of the target object includes:

[0089] Perform a global geometric scan on the target object to obtain the original model data of the target object;

[0090] Parametric surface fitting is performed on the original model data to obtain a continuous geometric representation of the target object;

[0091] The continuous geometric representation is analyzed for morphological features to obtain the initial feature distribution of the target object.

[0092] A comprehensive information collection operation is carried out along the entire contour of the target object's surface. The collection range includes all exposed surfaces of the target object, including curved transition points, edge corners, and flat surfaces. The shape details of each point are continuously collected along the surface extension direction without skipping any minute area of ​​the surface. After the shape information corresponding to all points is completely collected, it is systematically integrated according to the inherent positional arrangement logic of various information on the target object's surface. The integration process strictly follows the original positional association of scattered information on the surface, without altering the shape details inherent in any information. After all the scattered information is neatly gathered, a complete set of integrated content is formed. This integrated content is the original model data.

[0093] The entire original model data is retrieved, containing all shape-related content. Based on the natural extension patterns of the target object's surface, the scattered shape content that is adjacent to each other in space is sequentially connected end to end. During the connection process, gaps between adjacent content are filled in according to the natural trend of surface changes. The entire set of scattered content is continuously connected, gradually eliminating the state of independent separation of individual information. This allows the originally fragmented shape content to be interconnected and form an uninterrupted whole, completely restoring the uninterrupted shape of the target object's surface. The complete content carrier formed after all content connection work is completed is the continuous geometric representation.

[0094] The entire surface-related content of the continuous geometric representation is dissected one by one. Based on the inherent differences in the concave and convex shape, the extension direction of the outline, and the bending changes of the surface, the content is classified and sorted. The shape attributes corresponding to each segment of content are identified one by one. Sub-content with the same attribute is gathered and grouped into the same content group. The classification and collection of all content within the continuous geometric representation is completed in sequence. The collection process retains the original shape features of various details without content deletion. After all classification groups are summarized and integrated, a set of feature collection content is formed. This set of collected content is determined as the initial feature distribution.

[0095] The beneficial effects include: full-domain geometric scanning can completely capture all surface morphology-related content of the target object, thus preserving the complete surface details of the original model data and avoiding the problem of missing local information. By relying on surface fitting to process scattered data and form a continuous geometric representation, the fragmentation of the original data is overcome, and the coherent surface appearance of the target object is fully restored. The initial feature distribution obtained through morphological feature analysis systematically organizes all surface features, fully presenting the morphological regularity of the target object. This allows for adaptation and comparison with previously compiled scratch defect patterns, providing a complete and reliable reference for subsequent full-domain matching to find potential hazard points, and solidifying the data foundation for screening anti-scratch points in die-cut products.

[0096] S04. Based on the scratch defect pattern, perform global feature matching on the initial feature distribution to obtain the points of the target object to be protected;

[0097] In this embodiment of the invention, the step of performing global feature matching on the initial feature distribution based on the scratch defect pattern to obtain the points to be protected on the target object includes:

[0098] Orientation feature extraction is performed on the initial feature distribution to obtain the orientation-sensitive feature map of the initial feature distribution;

[0099] Based on the scratch defect pattern, the direction-sensitive feature map is matched and compared to obtain the primary matching point of the target object;

[0100] Based on the surface damage extension path in the scratch defect pattern, the source of potential hazards is traced along the path of the primary matching points to obtain the derivative risk points of the target object.

[0101] The primary matching points and the derived risk points are collected and compiled across the entire domain to obtain the points of protection to be obtained for the target object.

[0102] The initial feature distribution is systematically analyzed, including all morphological extensions. Following the inherent extension directions of each morphology, the corresponding orientation details are broken down and refined. The orientation category of each detail is distinguished, and all sub-categories belonging to the same orientation are collected and organized. The original arrangement order of each detail in the initial feature distribution is strictly followed, ensuring complete recording of all directional information within the initial feature distribution, without omitting any orientation detail. After all content is categorized, collected, and arranged in an orderly manner, the resulting set of content constitutes the orientation-sensitive feature atlas.

[0103] All morphological details carried by the finalized scratch defect pattern are retrieved and used as a unified reference standard. Various orientation features within the scratch defect pattern are extracted sequentially. Each extracted feature is compared and verified with the orientation features of each region in the orientation-sensitive feature map. During the verification process, the location where each detail can be completely matched is marked. After the full-range item-by-item comparison is completed, all marked locations are collected and stored. All the collected and summarized location contents are determined as the primary matching points.

[0104] The complete record of the damage extension path within the scratch defect pattern is extracted. Based on this fixed damage extension trajectory, starting from the location of each primary matching point, the surrounding area is gradually investigated outward along the trajectory. The corresponding locations that can generate new damage hazards following the damage development pattern are continuously searched along the extension path. All the hazard locations identified along the path are collected and saved one by one. After the entire source tracing and investigation work is completed, all the collected location information is summarized, and the resulting set of information is determined as the derivative risk points.

[0105] All marked primary matching points and derivative risk points are collected centrally. The distribution information of the two types of points is sorted out in a unified manner. The same locations marked repeatedly in the two types of points are found and retained separately. After removing duplicate content, all the remaining valid points are integrated and collected according to the original distribution pattern. The point compilation work of the whole area is completed. The complete set of points obtained after collection and compilation is determined as the points to be protected.

[0106] The beneficial effects include extracting orientation information from the initial feature distribution to generate a direction-sensitive feature map, which can comprehensively analyze the extension patterns of various morphologies and accurately extract and isolate the directional information related to scratches. Based on the established scratch defect patterns, content comparison and screening are conducted to screen initial matching points, accurately locating visible hidden dangers consistent with known scratch features. Tracing the damage extension path yields derivative risk points, filling in the gaps in hidden damage locations surrounding the potential hazards and addressing omissions caused by single comparisons. Integrating these two types of points completes the compilation of points to be protected, achieving full coverage of both visible and hidden risk locations, accurately delineating the protection range, and providing a complete and reliable positioning basis for subsequent targeted local protection optimization.

[0107] S05. Based on the points to be protected, simulate local lifting of the original model data to obtain virtual protection data of the target object;

[0108] In this embodiment of the invention, the step of simulating local elevation of the original model data based on the points to be protected to obtain virtual protection data for the target object includes:

[0109] Density clustering expansion is performed on the points to be protected to obtain the target lifting area of ​​the target object;

[0110] Based on the scratch defect pattern, the original model data points in the target lifting area are lifted in reverse normal direction to obtain the reference compensation surface unit of the target object.

[0111] The overlapping areas of adjacent reference compensation surface units are weighted and fused to obtain a smooth raised surface of the target object;

[0112] The smooth, raised surface is discretized and sampled using a mesh to obtain virtual protection data for the target object.

[0113] The step of performing reverse normal lifting on the original model data points within the target lifting area based on the scratch defect pattern to obtain the reference compensation surface element of the target object includes:

[0114] The target uplift amount is calculated for the original model data points within the target uplift area to obtain the target uplift value of the data points. The calculation formula for the target uplift value is as follows:

[0115]

[0116] In the formula, For the first The target elevation value for each of the data points. The index number of the data point. The preset reference lift amount, The depth value of the scratch defect pattern at the data point. The maximum depth of the scratch defect pattern. The distance from the data point to the center of the corresponding point to be protected is denoted as . The standard deviation parameter of the target lifting region. The average curvature of the original model data at the data point. The preset feature length, Let be the arc length parameter extending radially outward from the center of the point to be protected along the curved surface. Let be the second derivative of the depth value of the scratch defect pattern along the direction of the arc length parameter. It is an exponential function;

[0117] Based on the target elevation value of the data points, the normal offset of the original model data points is adjusted to obtain the elevation set of the target object.

[0118] The raised point set is then subjected to shape-preserving smoothing trimming to obtain the reference compensated surface element of the target object.

[0119] The distribution information of all points to be protected was sorted out one by one. Points that are close to each other in space were grouped into point groups. The outer perimeter of each point group was defined by extending outward from the center of each point group. The boundary range generated by the outward expansion of each group was continuously spliced ​​together. The expansion range that could be seamlessly connected was continuously merged and connected. After all points have completed the range expansion and connection integration, a complete and contiguous range was finally formed. This range was determined as the target elevation area.

[0120] Information on the changes in the depth of damage, the positional relationship between the points and the protection center, and the bending changes of the corresponding curved surfaces are retrieved from the scratch defect pattern record. Based on the comprehensive information of multiple types, the lifting scale corresponding to each original model data point within the target lifting area is determined. The lifting scale information of each data point after finalization is sorted out one by one, and the lifting scale corresponding to a single data point is uniformly recorded as the target lifting value.

[0121] Based on the determined target elevation value and the corresponding elevation scale, move the position of each original model data point in the target elevation area along the fixed vertical direction of the original surface of the data points. After all data points have completed the orientation offset adjustment, gather all the changed position content. The complete set of point content formed by gathering and summarizing is the raised point set.

[0122] The original shape and layout details of the raised points are fully preserved. Only the abrupt dividing points at the connection points are smoothed. All the connecting dividing points are polished one by one along the layout of the points. The block content after the entire point content is fully repaired is the reference compensation surface unit.

[0123] The overlapping content of each adjacent reference compensation surface element is identified one by one. The overlapping area is then processed by combining the inherent morphological details of the two elements. The arrangement details of the transition content are allocated according to the original content proportion of the two elements. The cutting boundaries at the unit connection position are eliminated. After all overlapping areas are fused, the overall content is determined as a smooth raised surface.

[0124] The overall content of the smooth raised surface is divided into equal grids according to fixed division rules. Within each grid, detailed content in a fixed position is extracted and stored. The entire grid is traversed one by one to complete the content sampling. All the sampled detailed content is organized and summarized. The complete set of content after summarization is determined as virtual protection data.

[0125] Each component is extracted from the finished product data of the previous process. The reference lifting amount is fixed and retained according to the preset process standards. The content related to the depth of single-point scratches is extracted point by point from the morphological record corresponding to the scratch defect pattern after forming. After summarizing the depth content of all points, the limit depth content in the total depth is selected. The interval information between the point and the center of the point to be protected is determined by sorting out the point layout record of the target lifting area. The content related to the discrete features of the region is generated by the regularization and collection of the full point distribution information of the target lifting area. The single-point curved surface bending information is taken from the surface morphological record of the original model data at the corresponding position. The feature length is fixed in advance according to the established process specifications. The second-order change content of the depth corresponding to the radial arc length is collected by sorting out the depth fluctuation law layer by layer along the curved surface trajectory extending outward from the center of the point to be protected.

[0126] The entire set of content is used to determine the target lift value corresponding to each original model data point within the target lift area by combining multiple measured data such as scratch depth, spatial location of points, and surface deformation characteristics. It integrates the damage depth pattern inherent in the scratch defect pattern, the point distribution characteristics of the target lift area, and the surface bending information recorded in the original model data into the determination of the single-point lift scale. This ensures that the final target lift value perfectly matches the scratch hazard characteristics and surface morphology around the point, providing a precise dimensional benchmark for subsequent adjustment of the normal offset of the original model data points. This guarantees that the resulting set of lift points can adapt to the process optimization requirements of local anti-scratching.

[0127] As the area extends from the center of the protected point outwards, the change in the distance between the point and the center is continuously controlled by the exponential constraint rules to manage the range of change of the lifting benchmark. The second-order change of the surface depth along the radial arc length is responsible for correcting the additional lifting scale brought about by the surface shape of a single point. As the distance between the point and the protection center continues to extend, the corresponding lifting benchmark continues to shrink. The second-order change of the depth of the local surface simultaneously completes the fine-tuning of the lifting scale of the corresponding point. Finally, a fixed change pattern is formed in which the lifting scale near the protection center matches the damage depth, and the lifting scale far away from the center gradually decreases.

[0128] The beneficial effect is that by focusing on key points and expanding the defined target elevation area, the protection can completely cover the surrounding potential hazards of the points to be protected, avoiding any gaps in the protection range. The elevation value corresponding to each point is determined based on the inherent characteristics of the scratch defects. After adjusting the point positions according to the corresponding scale, conformal repair is performed, and the generated benchmark compensation surface unit fits the original surface morphology. Overlapping areas of adjacent units are merged to eliminate abrupt changes in morphology at the connection points, resulting in a smooth and uniform transition of the formed elevation surface. The virtual protection data obtained through standardized sampling completely retains the optimized surface information, accurately restoring the overall morphology after the protection modification, providing a complete reference for subsequent pressure compensation calculations.

[0129] S06. Perform difference analysis on the virtual protection data and the original model data to obtain the pressure compensation parameters of the target object.

[0130] In this embodiment of the invention, the step of performing difference analysis on the virtual protection data and the original model data to obtain the pressure compensation parameters of the target object includes:

[0131] Extract the simulation space offset of the virtual protection data relative to the original model data, and generate the actual offset sequence of the virtual protection data;

[0132] The actual offset sequence is clustered by region distribution to obtain the continuous lifting blocks of the target object;

[0133] Boundary fitting and extraction are performed on the continuous lifting blocks to obtain the lifting profile of the continuous lifting blocks;

[0134] The pressure response quantization analysis of the raised profile is performed to obtain the pressure compensation parameters of the raised profile.

[0135] The spatial orientation records of the same location in the virtual protection data and the original model data are compared point by point. The spatial misalignment details of the same point in the two sets of data are marked one by one. The misalignment information of all points in the entire area of ​​the target object is fully collected. The inherent arrangement order of each point on the surface of the target object is strictly followed. All misalignment details are listed and collected in an orderly manner, without omitting any offset information generated by any point. After all the offset content is organized according to the original arrangement rules, a complete set of orderly content is formed. The content of this set is determined as the actual offset sequence.

[0136] The spatial distribution of all offset information within the actual offset sequence is analyzed item by item. Offset information that is adjacent to each other in space is gathered and grouped into separate content groups. The outer boundaries of each content group are continuously searched, and multiple content groups that can be seamlessly connected by boundary lines are continuously integrated into contiguous content. All information in the actual offset sequence is traversed and all connectable groups are integrated into contiguous content blocks. The contiguous content blocks formed after all the scattered offset information is integrated are determined as continuous lifting blocks.

[0137] By gradually tracing the edge trajectory of the outermost content of a single continuously raised block, the extension direction of the entire boundary is systematically analyzed. The extension lines of all boundaries around the block are fully tracked, and the block range enclosed by the entire boundary line is completely defined. Without altering the original offset information details inside the continuously raised block, the relevant content of the outer boundary obtained from the systematic analysis is extracted separately from the original block content. The extracted and summarized complete set of boundary content is determined as the raised outline.

[0138] The details of shape changes and spatial offsets at each edge of the lifting contour are dissected one by one. The stress adaptation content that needs to be supplemented for each contour detail during contact and pressure is matched in turn. The exclusive stress supplement details corresponding to each position of the entire contour are sorted out. All the stress supplement details that have been matched one by one are collected and organized in accordance with the original spatial arrangement of the lifting contour. The complete set of matching details formed after collection and integration is determined as the pressure compensation parameters.

[0139] The beneficial effects include extracting offset content and forming an actual offset sequence based on the point comparison of two sets of data, which can completely restore the spatial deformation details brought about by protection optimization and retain the deformation change information of each location. By relying on distributed aggregation to obtain continuous lifting blocks, and integrating scattered offset information to achieve contiguous and regular optimized areas, the overall scope corresponding to the protection modification is accurately located. The lifting contour obtained through boundary extraction completely delineates the outer boundary of the optimized area. Based on this, pressure compensation parameters are obtained through pressure adaptation analysis, allowing for precise adaptation between pressure control content and local lifting shape. This eliminates the scratch-inducing factors in the die-cutting contact stage from the pressure control level, stabilizing the surface forming quality of the product.

[0140] like Figure 2 The diagram shown is a functional block diagram of a surface simulation-based die-cutting and hot stamping anti-scratch optimization system provided in an embodiment of the present invention.

[0141] The surface simulation-based die-cutting and hot stamping anti-scratch optimization system 10 described in this invention can be installed in electronic devices. Depending on the functions implemented, the surface simulation-based die-cutting and hot stamping anti-scratch optimization system 10 may include a model replication and analysis module 11, a feature mapping and model fixing module 12, a simulation geometry reconstruction module 13, a global matching and point fixing module 14, a simulation local lifting module 15, and a difference analysis and parameter calculation module 16. The modules described in this invention can also be called units, referring to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, stored in the memory of the electronic device.

[0142] In this embodiment, the functions of each module / unit are as follows:

[0143] The model replication and analysis module 11 is used to perform model replication and analysis on entity samples with scratch defects, and extract the defect model data of the entity samples.

[0144] The feature mapping module 12 is used to perform feature mapping on the defect model data to obtain the defect feature distribution of the entity sample, and to set the common substructure in the defect feature distribution as the scratch defect mode.

[0145] The simulation geometry reconstruction module 13 is used to perform simulation geometry reconstruction on the original model data of the target object to obtain the initial feature distribution of the target object;

[0146] The global matching and positioning module 14 is used to perform global feature matching on the initial feature distribution based on the scratch defect pattern to obtain the points to be protected on the target object.

[0147] The simulation local lifting module 15 is used to simulate local lifting of the original model data based on the point to be protected, so as to obtain virtual protection data of the target object.

[0148] The difference analysis parameter calculation module 16 is used to perform difference analysis on the virtual protection data and the original model data to obtain the pressure compensation parameters of the target object.

[0149] In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0150] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0151] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0152] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0153] This application embodiment can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0154] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A surface simulation-based optimization method for anti-scratch in die-cutting and hot stamping, characterized in that, The method includes: S01. Perform model replication analysis on the entity sample with scratch defects to extract the defect model data of the entity sample; S02. Perform feature mapping on the defect model data to obtain the defect feature distribution of the entity sample, and set the common substructure in the defect feature distribution as the scratch defect mode; S03. Perform simulation geometric reconstruction on the original model data of the target object to obtain the initial feature distribution of the target object; S04. Based on the scratch defect pattern, perform global feature matching on the initial feature distribution to obtain the points of the target object to be protected; S05. Based on the points to be protected, simulate local lifting of the original model data to obtain virtual protection data of the target object; S06. Perform difference analysis on the virtual protection data and the original model data to obtain the pressure compensation parameters of the target object.

2. The surface simulation-based anti-scratch optimization method for die-cutting and hot stamping as described in claim 1, characterized in that, The process of performing model replication analysis on physical samples with scratch defects to extract defect model data of the physical samples includes: A sample model of the entity sample is obtained by digitally mapping and reconstructing the entity sample with scratch defects. The surface morphology data of the sample model is used to extract the scratched defect area to obtain the defect model data of the entity sample.

3. The surface simulation-based anti-scratch optimization method for die-cutting and hot stamping as described in claim 1, characterized in that, The step of performing feature mapping on the defect model data to obtain the defect feature distribution of the entity sample includes: The defect model data is decoupled by features to obtain the feature component sequence of the defect model data; Spatial mapping and association are performed on the feature component sequences to obtain the defect arrangement structure of the entity sample; By performing region connectivity aggregation on the defect arrangement structure, the defect feature distribution of the entity sample is obtained.

4. The surface simulation-based anti-scratch optimization method for die-cutting and hot stamping as described in claim 1, characterized in that, The step of setting the common substructure in the defect feature distribution as a scratch defect pattern includes: The defect feature distribution is compared and verified by layering to obtain the repeating regions of the defect feature distribution; Contour analysis is performed on the repeated regions to extract the common substructure of the defect feature distribution; The common substructure is feature-locked and stored to obtain the scratch defect pattern of the defect feature distribution.

5. The surface simulation-based anti-scratch optimization method for die-cutting and hot stamping as described in claim 1, characterized in that, The process of performing simulated geometric reconstruction on the original model data of the target object to obtain the initial feature distribution of the target object includes: Perform a global geometric scan on the target object to obtain the original model data of the target object; Parametric surface fitting is performed on the original model data to obtain a continuous geometric representation of the target object; The continuous geometric representation is analyzed for morphological features to obtain the initial feature distribution of the target object.

6. The surface simulation-based anti-scratch optimization method for die-cutting and hot stamping as described in claim 1, characterized in that, The step of performing global feature matching on the initial feature distribution based on the scratch defect pattern to obtain the points of the target object to be protected includes: Orientation feature extraction is performed on the initial feature distribution to obtain the orientation-sensitive feature map of the initial feature distribution; Based on the scratch defect pattern, the direction-sensitive feature map is matched and compared to obtain the primary matching point of the target object; Based on the surface damage extension path in the scratch defect pattern, the source of potential hazards is traced along the path of the primary matching points to obtain the derivative risk points of the target object. The primary matching points and the derived risk points are collected and compiled across the entire domain to obtain the points of protection to be obtained for the target object.

7. The surface simulation-based anti-scratch optimization method for die-cutting and hot stamping as described in claim 1, characterized in that, The step of simulating local elevation of the original model data based on the points to be protected to obtain virtual protection data for the target object includes: Density clustering expansion is performed on the points to be protected to obtain the target lifting area of ​​the target object; Based on the scratch defect pattern, the original model data points in the target lifting area are lifted in reverse normal direction to obtain the reference compensation surface unit of the target object. The overlapping areas of adjacent reference compensation surface units are weighted and fused to obtain a smooth raised surface of the target object; The smooth, raised surface is discretized and sampled using a mesh to obtain virtual protection data for the target object.

8. The surface simulation-based anti-scratch optimization method for die-cutting and hot stamping as described in claim 7, characterized in that, The step of performing reverse normal lifting on the original model data points within the target lifting area based on the scratch defect pattern to obtain the reference compensation surface element of the target object includes: The target uplift amount is calculated for the original model data points within the target uplift area to obtain the target uplift value of the data points. The calculation formula for the target uplift value is as follows: In the formula, For the first The target elevation value for each of the data points. The index number of the data point. The preset reference lift amount, The depth value of the scratch defect pattern at the data point. The maximum depth of the scratch defect pattern. The distance from the data point to the center of the corresponding point to be protected is denoted as . The standard deviation parameter of the target lifting region. The average curvature of the original model data at the data point. The preset feature length, Let be the arc length parameter extending radially outward from the center of the point to be protected along the curved surface. Let be the second derivative of the depth value of the scratch defect pattern along the direction of the arc length parameter. It is an exponential function; Based on the target elevation value of the data points, the normal offset of the original model data points is adjusted to obtain the elevation set of the target object. The raised point set is then subjected to shape-preserving smoothing trimming to obtain the reference compensated surface element of the target object.

9. The surface simulation-based anti-scratch optimization method for die-cutting and hot stamping as described in claim 1, characterized in that, The step of performing difference analysis between the virtual protection data and the original model data to obtain the pressure compensation parameters of the target object includes: Extract the simulation space offset of the virtual protection data relative to the original model data, and generate the actual offset sequence of the virtual protection data; The actual offset sequence is clustered by region distribution to obtain the continuous lifting blocks of the target object; Boundary fitting and extraction are performed on the continuous lifting blocks to obtain the lifting profile of the continuous lifting blocks; The pressure response quantization analysis of the raised profile is performed to obtain the pressure compensation parameters of the raised profile.

10. A surface simulation-based die-cutting and hot stamping anti-scratch optimization system, characterized in that, The system for implementing the surface simulation-based die-cutting and hot stamping anti-scratch optimization method as described in claim 1 includes: The model replication and analysis module is used to perform model replication and analysis on entity samples with scratch defects, and extract the defect model data of the entity samples. The feature mapping module is used to perform feature mapping on the defect model data to obtain the defect feature distribution of the entity sample, and to set the common substructure in the defect feature distribution as the scratch defect mode. The simulation geometry reconstruction module is used to perform simulation geometry reconstruction on the original model data of the target object to obtain the initial feature distribution of the target object; The global matching and point-finding module is used to perform global feature matching on the initial feature distribution based on the scratch defect pattern to obtain the points to be protected on the target object. The simulation local lifting module is used to simulate local lifting of the original model data based on the points to be protected, so as to obtain virtual protection data of the target object; The difference analysis and parameter calculation module is used to perform difference analysis on the virtual protection data and the original model data to obtain the pressure compensation parameters of the target object.

Images