Rapid mold generation method based on large model data application

Abstract

The invention provides a rapid mold generation method based on large model data application. The method comprises the following steps: acquiring engineering data information of a to-be-generated mold, matching the engineering data information with mold product instances in a database, if matching succeeds, outputting the successfully matched mold product instances, and if matching fails, executing the following steps: extracting a DL line from DL data of the to-be-generated mold; identifying a reference line matched with the DL line from the reference line sample set; obtaining a reference coordinate position of a part associated with the matched reference line in the to-be-generated mold; and calling a virtual model of the matched part associated with the reference line from the database, and placing the virtual model according to the reference coordinate position. The production speed and efficiency of the mold are improved, and the mold design time is further saved.

Description

Technical Field

[0001] This invention relates to the field of intelligent manufacturing technology, and in particular to a method for rapid mold generation based on large model data applications. Background Technology

[0002] To address the issue of long design and debugging cycles in traditional molds, the applicant's prior patent application (publication number CN109376397A) provides an intelligent design method and system for three-dimensional molds. The main design concept of this intelligent three-dimensional mold design system is to use mathematical methods to establish virtual models (including target virtual model frames and virtual model components) for assembly in three-dimensional space. Each virtual model in this design system contains a set of mathematical formulas and a dynamic algorithm. When specific feature parameters are input into the virtual model, a static target entity is calculated. When the target frame feature parameters of the mold to be designed are input, a step-by-step reverse calculation method based on engineering experience is used to establish the corresponding mathematical association model. Then, the optimal target mold frame and components are sequentially retrieved through virtual model calculations and assembled in space, quickly and accurately achieving the desired target effect. Although this patent can reduce the design and debugging time of molds to 2 hours to 2 days, for each new mold design requirement, a new virtual model (including the target virtual model framework and virtual model parts) needs to be built. It cannot inherit historical mold product instances and component virtual models, resulting in a waste of time, manpower and resources, and making it impossible to generate molds faster.

[0003] In both traditional and intelligent mold design, reference lines (i.e., algorithmic design lines, or AD lines) are typically drawn manually or automatically on the mold body. Components (or virtual model components) within the mold determine their coordinates according to engineering rules based on these reference lines to avoid interference. The components (or virtual model components) are then positioned according to these coordinates to complete the mold design. However, regenerating reference lines and recalculating component coordinates for each new mold design requirement, without referencing historical data, leads to wasted time and manpower. The speed of mold generation and the efficiency of mold design need further improvement. Summary of the Invention

[0004] The present invention aims to at least solve the technical problems in the prior art that the historical resources are not fully utilized, resulting in a waste of time and manpower, and the efficiency of mold design still needs to be improved, and provides a method for rapid mold generation based on the application of large model data.

[0005] To achieve the above-mentioned objectives of the present invention, according to a first aspect of the present invention, the present invention provides a method for rapid mold generation based on large model data application, comprising: acquiring engineering data information of the mold to be generated; matching the engineering data information with mold product instances in a database; if the matching is successful, outputting the successfully matched mold product instance; if the matching fails, performing the following steps: extracting DL lines from the DL data of the mold to be generated; identifying reference lines matching the DL lines from a set of reference line samples; acquiring the reference coordinate positions of the components associated with the matched reference lines in the mold to be generated; retrieving virtual models of the components associated with the matched reference lines from the database, and placing the virtual models according to the reference coordinate positions.

[0006] To achieve the above-mentioned objectives of the present invention, according to a second aspect of the present invention, a computer program product is provided, comprising a computer program / instructions, characterized in that, when the computer program / instructions are executed by a processor, they implement the steps of the method described in the first aspect of the present invention.

[0007] To achieve the above-mentioned objectives of the present invention, according to a third aspect of the present invention, an electronic device is provided, the electronic device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform a rapid mold generation method based on large model data applications as described in the first aspect of the present invention.

[0008] This invention first matches the engineering data of the mold to be generated with mold product instances in the database. If the database already contains mold product instances with the same or similar engineering data, the 3D model of the mold in the database is directly output. If no matching mold product instance is found in the database, the DL line of the mold to be generated is matched with reference lines in the reference line sample set to obtain a matching reference line. The reference coordinate positions of the components associated with the matching reference line are obtained, and the virtual model retrieved from the database is placed according to the reference coordinate positions to complete the mold model generation. This invention makes full use of the previously constructed and pre-stored mold product instances and component virtual models in the database, and uses the association between DL lines and reference lines to match DL lines and reference lines, quickly arranging the virtual models of components, thereby achieving rapid mold generation and saving mold design time. Attached Figure Description

[0009] Figure 1 This is a flowchart illustrating a preferred embodiment of the rapid mold generation method based on large model data application of the present invention.

[0010] Figure 2 This is a schematic diagram of the feature points of the DL line on the inner front door panel in Example 1 of the present invention;

[0011] Figure 3 This is a schematic diagram of the feature points of the pressure plater DL line of the rear door outer panel in Example 2 of the present invention;

[0012] Figure 4 This is a schematic diagram of the feature points of the DL line of the backdoor in Example 3 of the present invention;

[0013] Figure 5 This is a schematic diagram of the feature points of the DL line of the top cover in Example 4 of the present invention;

[0014] Figure 6 This is a schematic diagram of the structure of an electronic device in a preferred embodiment of the present invention. Detailed Implementation

[0015] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0016] In the description of this invention, it should be understood that the terms "longitudinal", "lateral", "up", "down", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0017] In the description of this invention, unless otherwise specified and limited, it should be noted that the terms "installation", "connection" and "linking" should be interpreted broadly. For example, they can refer to mechanical or electrical connections, or internal connections between two components. They can be direct connections or indirect connections through an intermediate medium. Those skilled in the art can understand the specific meaning of the above terms according to the specific circumstances.

[0018] This invention discloses a method for rapid mold generation based on large model data applications. In a preferred embodiment, such as... Figure 1 As shown, the method includes:

[0019] Step S1: Obtain the engineering data information of the mold to be generated.

[0020] In this embodiment, the engineering data information of the mold to be generated preferably includes, but is not limited to, the mold name, mold model, standard system, purpose, and manufacturer.

[0021] Step S2: Match the engineering data information with the mold product instances in the database. If the match is successful, proceed to step S3; if the match fails, proceed to step S4.

[0022] In this embodiment, multiple historically constructed mold product instances can be stored in the database in advance. Each mold product instance is set with instance information, which preferably includes, but is not limited to, mold name, mold model, standard system, manufacturer, and purpose. The engineering data information and the instance information are compared dimension by dimension, and the number of identical or compatible dimensions is counted. If the ratio of this number to the total number of dimensions of the instance information is greater than or equal to a preset percentage, the engineering data information is considered to be successfully matched with the mold product instance corresponding to the instance information, and step S3 is executed. If the ratio is less than the preset percentage, the engineering data information is considered to be unmatched with the mold product instance corresponding to the instance information, and step S4 is executed. The preset percentage is preferably, but not limited to, 80%-100%.

[0023] Step S3: Output the successfully matched mold product examples.

[0024] Step S4 specifically includes:

[0025] Step S41: Extract DL lines from the DL data of the mold to be generated;

[0026] In this embodiment, DL is an abbreviation for Die Layout, representing the mold layout. DL lines represent the process design lines of the mold. DL data includes multiple DL lines, which preferably include, but are not limited to, the outer and inner contour lines of the molded product, as well as the outer and inner contour lines of the components on the mold. Taking a stamping mold as an example, its components include a mold base, upper and lower mold plates, guide pillars, side pins, mold end, blank holder, blank holder guide plate, and pressure plate, etc.

[0027] In Example 1, the mold to be generated is a stamping mold for the inner panel of the front door. One of the obtained DL lines is the outer contour line of the inner panel of the front door, such as... Figure 2 As shown. In Example 2, the mold to be generated is a stamping mold for the outer panel of a car rear door. One of the obtained DL lines is the outer contour line of the pressure plate of this mold, as shown. Figure 3 As shown. In Example 3, the mold to be generated is a stamping mold for the outer panel of a car rear door. One of the obtained DL lines is the rear door outline of the mold, as shown. Figure 4 As shown. In Example 2, the top cover stamping die to be generated is obtained by using one of the DL lines as the top cover outline of the die, as shown. Figure 5 As shown.

[0028] Step S42: Identify the reference line that matches the DL line from the reference line sample set.

[0029] In this embodiment, the reference line sample set can be derived from mold product instances in a database. The reference line sample set includes multiple reference lines, each with multiple feature points. The distribution characteristics of these feature points characterize the shape of the reference line and the associated component mounting point characteristics.

[0030] In mold design, to avoid interference between components, corresponding reference lines are typically generated based on the deep learning (DL) lines of the components. Specifically, this is achieved by simultaneously scaling down, scaling up, or translating the DL lines. The DL lines and their corresponding reference lines share similar or identical shapes. The components (or virtual model components) in the mold determine their position coordinates based on these reference lines and according to engineering rules. Therefore, in step S42, reference lines matching the DL lines can be identified from the reference line sample set based on their shape matching degree. In other words, the matching process involves searching the reference line sample set for reference lines with the same or similar shape characteristics as the DL lines.

[0031] Step S43: Obtain the reference coordinate positions of the parts associated with the matching reference lines in the mold to be generated.

[0032] In this embodiment, each reference line can be associated with at least one component. Based on the reference line, the reference coordinate position of the at least one component associated with the reference line in the mold to be generated can be calculated using engineering rules. In Example 1, if the extracted DL line is the outer contour line of the front door inner panel, all components positioned based on the outer contour line of the front door inner panel are considered as components associated with the matching reference line. In Example 2, if the extracted DL line is the outer contour line of the mold's pressure platen, then the component associated with the reference line matching the pressure platen's outer contour line is the pressure platen itself.

[0033] Step S44: Retrieve the virtual models of the components associated with the matching reference lines from the database, and place the virtual models of the components according to the reference coordinate positions.

[0034] In this embodiment, in step S4, multiple DL lines can be extracted simultaneously from the DL data of the mold to be generated. After each DL line is matched with a reference line, the reference coordinate position of its associated component is obtained according to the matched reference line, and the virtual model of the component is placed according to the reference coordinate position.

[0035] It should be noted that in step S42, if the DL line cannot find a matching reference line in the reference line sample set, then the next DL line is selected to execute step S42.

[0036] It should be noted that after all DL lines in the DL data of the mold to be generated have completed steps S42-S44, if all parts in the mold to be generated have obtained reference coordinate positions and the virtual model has been placed, then the model generation of the mold to be generated is completed. If some parts in the mold to be generated cannot obtain reference coordinate positions (i.e., the DL lines corresponding to these parts cannot be matched with reference lines in the reference line sample set) and / or cannot obtain virtual models from the database (i.e., virtual models have not yet been created for these parts), then it is necessary to calculate reference coordinate positions and / or generate virtual models for these parts. Even so, it greatly speeds up the mold generation process, saves manpower, and makes full use of historical data resources.

[0037] In this embodiment, the generated mold is associated with engineering data information and stored in the database to achieve database updates.

[0038] In a preferred embodiment, feature points are used to determine whether the DL line matches the reference line, thereby simplifying the matching process, improving matching speed, and fully utilizing the shape similarity between the DL line and the reference line to ensure matching accuracy. Therefore, step S42, identifying the reference line that matches the DL line from the reference line sample set, includes:

[0039] Step A: Extract multiple feature points on the DL line.

[0040] Feature points are preferably, but not limited to, inflection points on the DL line where the curvature exceeds the curvature threshold, or corner points on the DL line, or inflection points of convex parts of the DL line.

[0041] Step B involves identifying a reference line that matches the DL line based on the feature points of the DL line and the feature points of the reference lines in the reference line sample set.

[0042] Specifically, in step B, the match between the DL line and the reference line can be determined by comparing the number of feature points on the DL line and the number of feature points on the reference line. Alternatively, the coordinates of the feature points of the DL line can be sorted in a counter-clockwise / clockwise direction to form a first sequence, and the coordinates of the feature points of the reference line can be sorted in a counter-clockwise / clockwise direction to form a second sequence. The correlation coefficient between the first and second sequences can be calculated. If the correlation coefficient is greater than a preset correlation coefficient threshold, the DL line is considered to match the reference line; otherwise, the DL line is considered not to match the reference line.

[0043] In one example of this implementation, step A specifically includes:

[0044] Step A101: Select the corner section based on the curvature of the DL line, that is, the set of points in the DL line whose curvature is greater than the curvature threshold is taken as the corner section.

[0045] Step A102: Aggregate and divide the selected corner portion into blocks. The aggregation radius is the point selection step distance, and the minimum block includes at least 3 points.

[0046] Step A103: Select the straight line formed by the last data point of the two preceding and following protrusions (target protrusions) and the first data point. Calculate whether the distance from the highest point of the target protrusion to the straight line is less than a preset distance value. If it is less than or equal to the preset distance value, the target protrusion is removed; if it is greater than the preset distance value, the target protrusion is retained. This process can remove smaller protrusions and unwanted data points selected in step A101.

[0047] In another example of this implementation, step A specifically includes:

[0048] Step A201: Smooth the DL line. Disassemble the smoothed DL line into sequentially connected segments and obtain the endpoints of the segments. Delete duplicate endpoints.

[0049] Step A202: Sort counterclockwise starting from the extreme endpoints of the first quadrant.

[0050] Step A203, aggregation. Specifically, aggregate points counterclockwise along the smoothed DL line at 15mm increments, with no more than 5 points in each aggregation group.

[0051] Step A204: Execute along the DL line, take the endpoint of the middle position in each aggregate class, and determine whether the endpoint of the middle position can be deleted: if the distance between the line connecting the two endpoints before and after the endpoint of the middle position and the endpoint of the middle position is no more than 15mm, then delete the endpoint of the middle position; otherwise, keep the endpoint of the middle position.

[0052] Step A205: Repeat the deletion process of step A204. After each deletion, compare the number of points with the number before deletion. If the number of points has not changed, end the loop and obtain the final valid points.

[0053] In a preferred embodiment, to improve the accuracy of identifying reference lines that accurately match the DL line from the reference line sample set, step B involves identifying the reference line that matches the DL line based on the feature points of the DL line and the feature points of the reference lines in the reference line sample set, including:

[0054] Step B101: Select one or more initial reference lines from the reference line sample set whose number of feature points matches the number of feature points of the DL line.

[0055] Specifically, if the number of feature points of the reference line is equal to the number of feature points of the DL line, or if the difference between the number of feature points of the DL line and the number of feature points of the reference line is less than the difference threshold, then the number of feature points of the DL line and the reference line are considered to match, and the reference line is used as an initial reference line. Otherwise, the reference line is discarded, and the next reference line in the reference line sample set is matched and identified.

[0056] Step B102 involves decomposing the DL line into multiple vectors using multiple feature points along the DL line. Specifically, the multiple feature points sequentially decompose the DL line into connected line segments, and each line segment is converted into a vector. The number of vectors is equal to the number of feature points on the DL line. Each vector has a scale (i.e., length) and a direction.

[0057] Step B103 involves decomposing the initial reference line into multiple vectors using multiple feature points on the initial reference line. Similarly, the reference line is sequentially decomposed into connected line segments by multiple feature points, and each line segment is converted into a vector. The number of vectors is equal to the number of feature points on the reference line. Each vector has a scale and a direction.

[0058] Step B104: Perform vector matching between the DL line and the initial reference line. When the number of DL line vectors that successfully match the reference line reaches a preset requirement, the initial reference line is used as the reference line for matching the DL line.

[0059] Specifically, step B104 involves iterating through the vectors of the DL line. If a DL line vector can be found among multiple vectors of the reference line whose ratio difference is within a threshold range and whose direction angle is within an angle threshold range, the number of DL line vectors successfully matched to the reference line is incremented by 1. If a DL line vector cannot be found among multiple vectors of the reference line whose ratio difference is within a threshold range and whose direction angle is within an angle threshold range, the number of DL line vectors successfully matched to the reference line remains unchanged. After the iteration is complete, it is determined whether the number of DL line vectors successfully matched to the reference line reaches a preset requirement. The preset requirement is that the proportion of the number of DL line vectors successfully matched to the reference line in the total number of DL line vectors is greater than or equal to a first proportion threshold, which ranges from 70% to 99%.

[0060] Preferably, if the proportion of vectors of a DL line that cannot successfully match each reference line in the reference line sample set is greater than or equal to a second proportion threshold in the total number of vectors of the DL line, the DL line is considered a new class, a corresponding reference line is generated, and the reference line is saved in the reference line sample set. The second proportion threshold is 60%-80%, preferably 70%.

[0061] Since vectors with larger proportions better reflect the shape characteristics of the DL line, to improve work efficiency, preferably, vectors with proportions greater than or equal to a length threshold are selected from all vectors of the DL line for the above-mentioned traversal matching process. Simultaneously, vectors with proportions greater than or equal to the length threshold are selected from all vectors of the reference line to participate in the above-mentioned traversal matching. The length threshold is preferably, but not limited to, a range of 30mm to 100mm, and is more preferably 50mm, 60mm, or 70mm.

[0062] In a preferred embodiment, to ensure that the extracted feature points more accurately reflect the shape or positioning features of the DL line, step A involves extracting multiple feature points on the DL line, including:

[0063] Step A301: Disassemble the DL line into sequentially connected segments and pick up the endpoints of the segments.

[0064] In this embodiment, the DL line is preferably, but not limited to, decomposed into sequentially connected line segments using polygon splitting, curve approximation, or arc segmentation methods. To improve processing speed, relevant software platforms, such as AutoCAD or CAT IA, can be used to decompose the DL line and pick the endpoints of the line segments, deleting duplicate endpoints during the picking process. To better extract feature points and avoid introducing noise, it is further preferred that the DL line be smoothed before performing step A301, and then the smoothed DL line is decomposed into sequentially connected line segments, picking the endpoints of the line segments. The smoothing process can use existing curve smoothing algorithms, such as B-spline curve smoothing algorithms and BMP smoothing algorithms, or it can utilize the smoothing function module of relevant software platforms.

[0065] Step A302: Perform the first loop process to obtain the first endpoint set. The first loop process includes:

[0066] In each iteration, starting from the extreme endpoint of the first quadrant, the system moves counterclockwise along the DL line with a first preset step size until returning to the starting point. After each step, the endpoints are aggregated to obtain a first aggregate class. Intermediate endpoints of the first aggregate class that satisfy the first preset deletion condition are deleted to obtain a first endpoint set. If the difference between the number of endpoints in the first endpoint set obtained in this iteration and the number of endpoints in the first endpoint set obtained in the previous iteration is less than a preset difference threshold, the first iteration stops; otherwise, the next iteration begins. The number of endpoints in the first aggregate class does not exceed N1, where N1 is a positive integer ranging from 3 to 7, preferably 5. Figure 2 , Figure 3 , Figure 4 and Figure 5As shown, the extreme endpoints in the first quadrant represent the endpoints located in the first quadrant that are farthest from the coordinate center. The first preset step size can be set by the user, ranging from 10mm to 30mm, preferably 15mm. The first preset deletion condition is preferably, but not limited to, that the difference between the curvature of the middle endpoint of the first aggregation class and the curvature of its adjacent endpoints is less than a preset curvature difference threshold. The preset difference threshold ranges from 0 to 3, preferably 0.

[0067] And / or, perform a second loop process to obtain a second endpoint set, the second loop process including:

[0068] In each iteration, starting from the extreme endpoint of the fourth quadrant, the algorithm steps clockwise along the DL line with a second preset step size until returning to the starting point. After each step, the endpoints are aggregated to obtain a second aggregate class. Intermediate endpoints of the second aggregate class that satisfy the second preset deletion condition are deleted to obtain a second endpoint set. If the difference between the number of endpoints in the second endpoint set obtained in this iteration and the number of endpoints in the second endpoint set obtained in the previous iteration is less than a preset difference threshold, the second iteration stops; otherwise, the next iteration begins. The number of endpoints in the second aggregate class does not exceed N2, where N2 is a positive integer ranging from 3 to 7, preferably 5. Figure 2 , Figure 3 , Figure 4 and Figure 5 As shown, the extreme endpoints in the fourth quadrant represent the endpoints located in the fourth quadrant that are farthest from the coordinate center. The second preset step size can be set by the user, with a value ranging from 10mm to 30mm, preferably 15mm. The second preset deletion condition is preferably, but not limited to, that the difference between the curvature of the middle endpoint of the second aggregation class and the curvature of its adjacent endpoints is less than a preset curvature difference threshold.

[0069] Step A303: Determine multiple feature points of the DL line using the first endpoint set and / or the second endpoint set.

[0070] In this embodiment, when step A302 only includes the first loop processing, determining multiple feature points of the DL line using the first endpoint set in step A303 can be achieved by directly using the first endpoint set as the multiple feature points of the DL line. When step A302 only includes the second loop processing, determining multiple feature points of the DL line using the second endpoint set in step A303 can also be achieved by directly using the second endpoint set as the multiple feature points of the DL line. When step A302 includes both the first and second loop processing, the two loop processing can be executed simultaneously or sequentially. In step A303, determining multiple feature points of the DL line using both the first and second endpoint sets can specifically involve calculating the union of the first and second endpoint sets and using the union as the multiple feature points of the DL line.

[0071] In this embodiment, to simplify calculations and effectively filter out endpoints that cannot well represent the shape characteristics of the DL line, a more preferred first preset deletion condition is that the distance between the middle endpoint of the first aggregation class and the line connecting its preceding and following endpoints is less than or equal to a first distance threshold; a second preset deletion condition is that the distance between the middle endpoint of the second aggregation class and the line connecting its preceding and following endpoints is less than or equal to a second distance threshold. The first distance threshold ranges from 10mm to 20mm, preferably 15mm. The second distance threshold ranges from 10mm to 20mm, preferably 15mm. In one example, let point A be the middle endpoint of a certain first aggregation class / a certain second aggregation class, and points B and C be the preceding and following endpoints of point A, respectively. Connect points B and C to obtain line BC, and obtain the distance d from point A to line BC. If d is less than the first distance threshold, then point A is deleted; otherwise, point A is retained.

[0072] In this embodiment, when the first loop processing is performed in step A302 to obtain the first endpoint set, and the second loop processing is performed to obtain the second endpoint set, in order to remove redundant endpoints, more preferably, in step A303, multiple feature points of the DL line are determined through the first endpoint set and the second endpoint set, including:

[0073] Step A3031: Find the union of the first endpoint set and the second endpoint set;

[0074] Step A3032: Extract multiple selection endpoints from the union set, where the distance between the selection endpoints is greater than a third distance threshold. The distance between the selection endpoints extracted in step A3032 is greater than the third distance threshold, which has a value range of 20mm-30mm, preferably 25mm.

[0075] Step A3033: Determine multiple feature points of the DL line by selecting multiple endpoints. Multiple endpoints can be directly used as the multiple feature points of the DL line.

[0076] To remove local noise, more preferably, in step A3033, multiple feature points of the DL line are determined by selecting multiple endpoints, including:

[0077] Each selected endpoint is connected to its adjacent selected endpoints before and after it, forming two vectors. The common endpoint of the two vectors is the selected endpoint. The angle between the two vectors is calculated. If the angle is greater than a preset angle threshold, the selected endpoint is retained and used as a feature point of the DL line. If the angle is less than or equal to the preset angle threshold, the selected endpoint is deleted. The angle threshold ranges from 5° to 15°, preferably 10°. The final obtained multiple feature points of the DL line are as follows: Figure 2 , Figure 3 , Figure 4 and Figure 5As shown by the white square mark in the middle.

[0078] Preferably, the method for obtaining multiple feature points of the reference line in the reference line sample set is as follows: for the DL line used to generate the reference line, the feature point extraction method described above in this invention is used to obtain multiple feature points. After generating the corresponding reference line using the DL line, the multiple feature points extracted from the DL line are mapped onto the reference line respectively, and the mapped points are used as feature points of the reference line.

[0079] In a preferred embodiment, to achieve intelligent mold design and improve automation and mold design efficiency, after the virtual models of all parts of the mold to be generated are placed, the method further includes:

[0080] Step C101: Determine the matching and positioning rules based on the engineering data information of the mold to be generated.

[0081] The matching and positioning rules are determined based on the manufacturer, model, application (e.g., drawing, stamping), and standard system (e.g., European standard system, Japanese standard system) in the engineering data. The matching and positioning rules include engineering rules, component distribution logic, proportional relationships, and feature variation relationships.

[0082] Step C102: Construct a neural network. In the neural network, a neuron is a component, and each component has a coordinate position. There are relationships between components (such as coordinate position relationships and distance relationships). Correspondingly, there are coordinate distance relationships between the neurons corresponding to the components. The position coordinates of each component are used as the attributes of the corresponding neuron.

[0083] Two interconnected components will have their position coordinates changed if the position coordinates of one component change, thus altering the distance between them and the coordinate distance between the corresponding two neurons. The position coordinates of the components can be the x, y, and z-axis coordinates of their centroid or center point. Preferably, the components use six spatial coordinates, representing the x, y, and z-axis coordinates of their centroid or center point, and the tilt angles of the components around the x, y, and z axes, respectively.

[0084] Repeat the following steps C1021-C1022 until the deep learning stopping condition is met:

[0085] Step C1021: Filter out neurons that do not conform to the matching localization rules and mark them as abnormal neurons. The target position coordinate range of each component can be inferred according to the matching localization rules. If the current position coordinate of the component corresponding to a certain neuron is not within its corresponding target position coordinate range, it is considered an abnormal neuron.

[0086] Step C1022: Generate multiple adjustment strategies. Predict the number of abnormal neurons generated after implementing each adjustment strategy through backpropagation. Select the adjustment strategy with the smallest predicted number of abnormal neurons to adjust the attributes of the abnormal neurons. Each adjustment strategy adjusts the attribute value of one abnormal neuron to meet the target position coordinate range of the corresponding component. Through backpropagation, predict the number of abnormal neurons generated after executing each adjustment strategy based on the matching localization rules.

[0087] The stopping condition for deep learning is that the coordinates of the component positions corresponding to all neurons satisfy the matching localization rule (specifically, the coordinates of the component positions corresponding to each neuron are within the range of its target position coordinates).

[0088] The present invention also discloses a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the above-mentioned rapid mold generation method based on large model data application provided by the present invention. The computer program product should be understood as a software product that mainly implements its solution through a computer program, such as a program product integrated in the cloud or software library.

[0089] The present invention also discloses an electronic device, in one embodiment of which the electronic device includes at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to execute the rapid mold generation method based on large model data application provided by the present invention.

[0090] like Figure 6 The diagram shown is a schematic representation of an electronic device for a rapid mold generation method based on large model data applications, according to an embodiment of the present invention. The electronic device may include a processor 10, a memory 11, a communication bus 12, and a communication interface 13. It may also include a computer program stored in the memory 11 and executable on the processor 10, such as a program for a rapid mold generation method based on large model data applications.

[0091] In some embodiments, the processor 10 may be composed of integrated circuits, such as a single packaged integrated circuit or multiple integrated circuits with the same or different functions, including combinations of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor 10 is the control unit of the electronic device, connecting various components of the entire electronic device through various interfaces and lines. It executes programs or modules stored in the memory 11 (e.g., executing a rapid mold generation method based on large model data applications) and calls data stored in the memory 11 to perform various functions of the electronic device and process data.

[0092] The memory 11 includes at least one type of readable storage medium, including flash memory, portable hard drive, multimedia card, card-type memory (e.g., SD or DX memory), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 11 can be an internal storage unit of an electronic device, such as a portable hard drive. In other embodiments, the memory 11 can be an external storage device of the electronic device, such as a plug-in portable hard drive, Smart Media Card (SMC), Secure Digital (SD) card, Flash Card, etc. Furthermore, the memory 11 can include both internal and external storage units of the electronic device. The memory 11 can be used not only to store application software and various types of data installed on the electronic device, such as the code of a rapid mold generation method program based on large model data applications, but also to temporarily store data that has been output or will be output.

[0093] The communication bus 12 can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This bus can be divided into an address bus, a data bus, a control bus, etc. The bus is configured to enable communication between the memory 11 and at least one processor 10, etc.

[0094] Communication interface 13 is used for communication between the aforementioned electronic device and other devices, including a network interface and a user interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a Wi-Fi interface, Bluetooth interface, etc.), typically used to establish communication connections between the electronic device and other electronic devices. The user interface may be a display, an input unit (such as a keyboard), or, optionally, a standard wired or wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen, etc. The display may also be appropriately referred to as a screen or display unit, used to display information processed in the electronic device and to display a visual user interface.

[0095] Figure 6 Only electronic devices with components are shown; it will be understood by those skilled in the art that... Figure 6 The structure shown does not constitute a limitation on the electronic device and may include fewer or more components than shown, or combine certain components, or have different component arrangements. For example, although not shown, the electronic device may also include a power supply (such as a battery) to power the various components. Preferably, the power supply can be logically connected to at least one processor 10 through a power management device, thereby enabling functions such as charging management, discharging management, and power consumption management. The power supply may also include one or more DC or AC power supplies, recharging devices, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components. The electronic device may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be described in detail here.

[0096] It should be understood that the embodiments are for illustrative purposes only and are not limited to this structure in the scope of the patent application.

[0097] Furthermore, if the modules / units integrated in electronic device 1 are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, a computer-readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, or a read-only memory (ROM).

[0098] In the description of this specification, the references to terms such as "an embodiment," "some embodiments," "example," "specific example," "a implementation," "a preferred implementation," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0099] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A mold rapid generation method based on a large model data application, characterized by, The method comprises the following steps: Obtaining engineering data information of a mold to be generated, matching the engineering data information with mold product instances in a database, outputting a matched mold product instance if the matching is successful, and executing the following steps if the matching fails: Extracting a DL line from DL data of the mold to be generated; Identifying a reference line matching the DL line from a reference line sample set; Obtaining a reference coordinate position of a component associated with the matched reference line in the mold to be generated; Retrieving a virtual model of the component associated with the matched reference line from the database and placing the virtual model according to the reference coordinate position. 2.The mold rapid generation method based on large model data application of claim 1, wherein, The method comprises the following steps: Extracting a plurality of feature points on the DL line; Identifying a reference line matching the DL line based on feature points of the DL line and feature points of the reference line in the reference line sample set. 3.The mold quick generation method based on large model data application of claim 2, wherein, The method comprises the following steps: Selecting one or more initial reference lines from the reference line sample set, the number of feature points of which matches the number of feature points of the DL line; Decomposing the DL line into a plurality of vectors through the plurality of feature points on the DL line; Decomposing the initial reference line into a plurality of vectors through the plurality of feature points on the initial reference line; Performing vector matching between the DL line and the initial reference line, and taking the initial reference line as the reference line matching the DL line when the number of vectors of the DL line successfully matched with the reference line reaches a preset requirement. 4.The mold rapid generation method based on large model data application of claim 2 or 3, wherein, The method comprises the following steps: Decomposing the DL line into sequentially connected line segments and picking up endpoints of the line segments; Performing a first loop process to obtain a first endpoint set, the first loop process comprising the following steps: In each loop, taking an extreme endpoint in a first quadrant as a starting point, stepping along the DL line counterclockwise at a first preset step distance until returning to the starting point, aggregating endpoints after each step to obtain a first aggregation class, deleting intermediate endpoints of the first aggregation class that meet a first preset deletion condition, obtaining the first endpoint set, and stopping the first loop process if the difference between the number of endpoints of the first endpoint set obtained in the current loop and the number of endpoints of the first endpoint set obtained in the previous loop is less than a preset difference threshold, otherwise, entering the next loop; wherein the number of endpoints of the first aggregation class does not exceed N1, and N1 is a positive integer; And / or, performing a second loop process to obtain a second endpoint set, the second loop process comprising the following steps: In each loop, taking an extreme endpoint in a fourth quadrant as a starting point, stepping along the DL line clockwise at a second preset step distance until returning to the starting point, aggregating endpoints after each step to obtain a second aggregation class, deleting intermediate endpoints of the second aggregation class that meet a second preset deletion condition, obtaining the second endpoint set, and stopping the second loop process if the difference between the number of endpoints of the second endpoint set obtained in the current loop and the number of endpoints of the second endpoint set obtained in the previous loop is less than a preset difference threshold, otherwise, entering the next loop; wherein the number of endpoints of the second aggregation class does not exceed N2, and N2 is a positive integer; Determining the plurality of feature points of the DL line through the first endpoint set and / or the second endpoint set. 5.The mold quick generation method based on large model data application of claim 4, wherein, The first preset deletion condition is that the distance between the intermediate endpoint of the first aggregation class and the line connecting the front and rear adjacent endpoints of the intermediate endpoint is less than or equal to a first distance threshold value. The second preset deletion condition is that the distance between the intermediate endpoint of the second aggregation class and the line connecting the front and rear adjacent endpoints of the intermediate endpoint is less than or equal to a second distance threshold value. 6.The mold quick generation method based on large model data application of claim 4 or 5, wherein, The first loop processing is performed to obtain a first endpoint set, and the second loop processing is performed to obtain a second endpoint set, and a plurality of feature points of the DL line are determined through the first endpoint set and the second endpoint set, including: The union set of the first endpoint set and the second endpoint set is obtained; A plurality of selected endpoints are extracted from the union set, and the distance between the selected endpoints is greater than a third distance threshold value; A plurality of feature points of the DL line are determined through the plurality of selected endpoints. 7.The mold quick generation method based on large model data application of claim 6, wherein, The plurality of feature points of the DL line are determined through the plurality of selected endpoints, including: Each selected endpoint is connected with the front and rear adjacent selected endpoints to form two vectors, the included angle of the two vectors is calculated, if the included angle is greater than a preset included angle threshold value, the selected endpoint is retained and the selected endpoint is taken as a feature point of the DL line, if the included angle is less than or equal to the preset included angle threshold value, the selected endpoint is deleted. 8.The mold quick generation method based on large model data application of claim 6, wherein, When the virtual models of all the parts of the mold to be generated are placed, further comprising: determining a matching positioning rule according to the engineering data information of the mold to be generated; constructing a neuron network, one neuron in the neuron network as one part, each part having a coordinate position, there being an association relationship between the parts, there being a coordinate distance relationship between the neurons corresponding to the parts, taking the position coordinate of each part as the attribute of the corresponding neuron, and the following steps are executed in a loop until a deep learning stop condition is reached: filtering out neurons that do not meet the matching positioning rule, denoted as abnormal neurons; generating a plurality of adjustment strategies, predicting the number of abnormal neurons generated after the implementation of each adjustment strategy through back propagation, and selecting the adjustment strategy with the smallest number of abnormal neurons after implementation to adjust the attributes of the abnormal neurons; The deep learning stop condition is that the position coordinates of all the parts corresponding to the neurons meet the matching positioning rule.

9. A computer program product comprising computer programs / instructions, characterized in that, The computer program / instructions are executed by the processor to implement the steps of the method of any one of claims 1-8.

10. An electronic device, comprising: The electronic device includes: at least one processor; and a memory connected in communication with the at least one processor; wherein the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the mold rapid generation method based on the large model data application of any one of claims 1 to 8.

Images