Intelligent quotation system and method for cutting die

Abstract

The invention discloses an intelligent quotation system and method for a cutting die. The intelligent quotation system comprises a data input module, a graph analysis calculation module, a pricing calculation module, a preview display module and a data storage module, the method comprises the following steps: step 1, inputting data; 2, analyzing and calculating the graph; step 3, pricing calculation; 4, outputting a result; the data input module supports multi-format drawing analysis and non-drawing parameterization modeling, and is matched with the functions of file analysis, layer identification, bridge location completion and makeup analysis, so that the manual operation threshold is greatly reduced, the quotation efficiency and precision are improved, and the layer identification accuracy can be gradually improved by recording user habits; a double-track evaluation mechanism of the graph analysis calculation module realizes objective quantification of processing difficulty, and the pricing calculation module is combined with a continuous marginal coefficient algorithm to ensure quotation accuracy and smooth price transition; and high efficiency, precision, standardization and convenience of cutting die quotation are integrally realized.

Description

Technical Field

[0001] This invention relates to the field of industrial automation pricing technology, specifically to an intelligent pricing system and method for die-cutting molds. Background Technology

[0002] Die-cutting mold quotation is a pricing service offered by the die-cutting mold industry for customized die-cutting mold products. Its core function is to calculate a reasonable selling price based on customer needs, combined with factors such as material consumption, processing difficulty, and production efficiency, providing a basis for transactions between supply and demand parties. Currently, die-cutting mold quotations typically employ a manual quotation model. Professional technicians receive drawings provided by customers, manually identify layer types in the drawings, measure graphic dimensions, calculate geometric parameters such as perimeter or area, count the number of pieces to be assembled, estimate material usage and processing difficulty, and finally provide a quotation based on material unit prices and processing costs.

[0003] The manual quotation model has the following drawbacks in practical applications: 1. Low quotation efficiency: The process of manually analyzing drawings and calculating parameters is cumbersome, time-consuming, and prone to errors; 2. Difficulty in drawing analysis: The drawings provided by customers are complex in format, containing multiple layers of information. Manually identifying layer types requires extensive experience, and some customers upload monochrome drawings, such as all entities in a DXF file being white or black. The lack of color differentiation information further increases the difficulty of automatic identification; 3. Omission of bridge position identification: When manually analyzing drawings, bridge positions in die-cutting drawings are often missed, leading to inaccurate perimeter calculations and affecting the accuracy of the quotation; 4. Cumbersome panel calculation: Drawings often contain multiple repeating graphics, and manually counting the number of panels and calculating the total perimeter is very cumbersome and time-consuming; 5. Lack of quotation capability without drawings: When customers need a quick estimate but do not provide drawings, the manual quotation model cannot meet the needs; 6. Subjective complexity assessment: Current technology relies on manual judgment of the difficulty of graphic processing, lacking a unified standard, resulting in poor quotation consistency. Summary of the Invention

[0004] The purpose of this invention is to provide an intelligent pricing system and method for die-cutting molds to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: an intelligent pricing system for die-cutting molds, comprising a data input module, a graphic analysis and calculation module, a pricing calculation module, a preview display module, and a data storage module. The data storage module establishes data connections with the data input module, the graphic analysis and calculation module, the pricing calculation module, and the preview display module, respectively. The graphic analysis and calculation module establishes data connections with the data input module and the pricing calculation module, respectively, and the pricing calculation module establishes a data connection with the preview display module.

[0006] The data input module serves as the data acquisition entry point, supporting both illustrated and unillustrated quotations across all scenarios. It acquires the basic data required for die-cutting pricing, providing raw input for subsequent analysis and calculation. The graphic analysis and calculation module assesses the complexity of die-cutting processing, generating complexity coefficients to provide a reference for processing difficulty in pricing calculations. The pricing calculation module calculates various costs and the total price of the die-cutting based on geometric parameters, complexity coefficients, material specifications, and continuous marginal coefficient algorithms. The preview and display module visualizes pricing-related information, supporting previewing, real-time updates, and file export of pricing results. The data storage module stores configuration data, historical data, user preferences, and template data required for system operation, supporting system adaptive learning and historical traceability.

[0007] The data input module includes a drawing input submodule and a manual input submodule. The drawing input submodule is used to handle quotations with drawings, and the manual input submodule is used to handle quotations without drawings.

[0008] The image parsing and calculation module includes a local algorithm evaluation submodule and an AI vision evaluation submodule. The local algorithm evaluation submodule is used to calculate the complexity score based on geometric features, manufacturing difficulty, and similarity indicators. The AI ​​vision evaluation submodule is used to call a visual model to assist in evaluating the complexity and integrate it with the local algorithm results.

[0009] The pricing calculation module includes a tool material pricing submodule, a sheet material pricing submodule, a punching pricing submodule, a cutting time calculation submodule, and a die weight calculation submodule. The tool material pricing submodule is used to calculate the tool material cost based on the total cutting circumference. The sheet material pricing submodule is used to calculate the sheet material cost based on the die boundary dimensions. The punching pricing submodule is used to calculate the punching cost based on the hole type, quantity, and size. The cutting time calculation submodule is used to estimate the die processing and cutting time. The die weight calculation submodule is used to calculate the weight of the finished die.

[0010] The preview display module includes a drawing information preview submodule and a real-time quotation submodule. The drawing information preview submodule is used for the visual display of graphics, and the real-time quotation submodule is used to update the details of various costs, complexity rating and total quotation when parameters change.

[0011] The data storage module includes a configuration data repository, a historical quotation repository, a user preference repository, and a standard graphic template library. The configuration data repository stores the configuration data required for system operation. The historical quotation repository stores the user's historical quotation details and quotation status. The user preference repository stores the user's layer type confirmation history, confidence calculation data, and personalized operation preferences. The standard graphic template library stores the standard graphic parametric models required for quotation scenarios without graphics.

[0012] The drawing input submodule includes a file parsing unit, a layer recognition unit, a bridge detection unit, a panel analysis unit, and a geometric calculation unit. The file parsing unit is used to parse the vector file structure and extract the core data of entities, layers, and boundaries required for pricing. The layer recognition unit is used to accurately identify layer types through a multi-layer strategy. The bridge detection unit is used to automatically detect breakpoints in the die-cutting drawing, verify their validity, and complete the breakpoints. It detects bridges by calculating the gap distance between adjacent endpoints, verifying whether the tangential direction of the endpoints is aligned, verifying whether the normal offset of the connection line is within the threshold range, and detecting whether there are obstructing entities in the bridge corridor. The panel analysis unit is used to identify repetitive graphic patterns in the die-cutting drawing and count the number of panels. It identifies repetitive graphics by detecting closed graphics, calculating the dimensional fingerprint of each closed graphic, grouping similar graphics by dimensional fingerprint, and counting the number of graphics in each group. The geometric calculation unit is used to calculate the geometric parameters of the die-cutting, including the perimeter of straight lines, arcs, circles, and polylines, and the area of ​​rectangles, circles, ellipses, annexes, and rectangular spacers.

[0013] The manual input submodule includes an intelligent parameter input unit, a panel parameter configuration unit, and a geometric parameter derivation unit. The intelligent parameter input unit is used to dynamically generate corresponding parameter input controls based on the standard graphic type selected by the user. The panel parameter configuration unit is used to support multi-panel layout configuration and shared cutting mode. The geometric parameter derivation unit is used to automatically calculate the geometric parameters of the die-cutting mold based on the basic size parameters input by the user.

[0014] The file parsing unit includes an entity information extraction subunit, a layer information extraction subunit, and a boundary information extraction subunit. The entity information extraction subunit is used to extract entity information, which includes the start and end coordinates of a straight line, the center, radius, start angle, and end angle of an arc, the center and radius of a circle, the vertex sequence and closure marker of a lightweight polyline, and the control points and order of a spline curve. The layer information extraction subunit is used to extract layer information, which includes the layer name, layer color, and layer visibility. The boundary information extraction subunit is used to extract boundary information, which includes the width and height of the overall bounding box of the graphic.

[0015] The layer recognition unit includes a color rule recognition subunit, a user habit recognition subunit, a manual confirmation subunit, and a monochrome drawing recognition subunit. The color rule recognition subunit is used to automatically identify the layer type according to a preset color mapping rule. The user habit recognition subunit is used to record the user's layer type confirmation history and improve the recognition confidence based on the number of confirmations. The manual confirmation subunit is used to provide a human-computer interaction interface for users to confirm or modify the layer type when the overall confidence of automatic recognition is less than a preset threshold. The monochrome drawing recognition subunit identifies the layer type for drawings where all entities have the same color through a multi-dimensional strategy.

[0016] The multi-dimensional strategy includes layer name recognition, automatic classification based on geometric features, inference based on positional relationships, and recognition based on line type attributes.

[0017] A method for intelligent pricing of die-cutting molds includes: Step 1, data input; Step 2, graphic analysis and calculation; Step 3, pricing calculation; and Step 4, result output.

[0018] In step one above, the basic data of the die-cutting mold is input through the data input module (1);

[0019] In step two above, the complexity coefficient of die-cutting is generated by the graphic analysis and calculation module (2);

[0020] In step three above, the pricing calculation module (3) calls the geometric parameters, complexity coefficients and configuration data, and combines them with the continuous marginal coefficient algorithm to calculate various costs and the total price.

[0021] In step four above, the quotation results are visualized through the preview display module (4).

[0022] In step one, the basic data of the die-cutting mold is input through the data input module (1) as follows: For the scenario of quotation with drawings, the die-cutting mold drawings are uploaded through the drawing input submodule (11), the drawings are automatically parsed, and entity information, layer information and boundary information are extracted; For the scenario of quotation without drawings, the standard graphic template is selected through the manual input submodule (12), the basic size parameters are input, the panel configuration parameters are set, and the geometric parameters of the die-cutting mold are automatically calculated.

[0023] In step three, the continuous marginal coefficient algorithm is specifically as follows: A continuous smoothing function is used to interpolate the base unit price within a preset coefficient range to obtain the base marginal coefficient. A decreasing function is then used to calculate the gradient coefficient based on the pricing unit number. The base marginal coefficient is multiplied by the gradient coefficient to obtain the final marginal coefficient for each pricing unit. Here, the number of pricing units refers to the number of graphics on a single sheet of material. The total price can be expressed using the following formula:

[0024]

[0025] Among them, C m Total price; C munit N is the unit price; N is the number of pricing units. Let be the final marginal coefficient of the i-th pricing unit. The total price calculation mode includes the first-item full-price mode and the unified marginal mode. The first-item full-price mode sets the final marginal coefficient of the first pricing unit to 1, and applies the calculated marginal coefficient from the second unit onwards. The unified marginal mode applies the calculated marginal coefficient to all pricing units uniformly.

[0026] Compared with existing technologies, the beneficial effects of this invention are as follows: The data input module of this invention supports multi-format drawing parsing and parametric modeling without drawings. Combined with file parsing, layer recognition, bridge completion, and imposition analysis functions, it significantly reduces the threshold for manual operation, improves the efficiency and accuracy of quotation, and can gradually improve the accuracy of layer recognition by recording user habits. The dual-track evaluation mechanism of the graphic parsing and calculation module realizes the objective quantification of processing difficulty, and the pricing calculation module combines the continuous marginal coefficient algorithm to ensure the accuracy of quotation and the smooth transition of price. Overall, it realizes the high efficiency, accuracy, standardization, and convenience of die-cutting quotation. Attached Figure Description

[0027] Figure 1 This is a system structure block diagram of the present invention;

[0028] Figure 2 This is a structural block diagram of the drawing input submodule of the present invention;

[0029] Figure 3 This is a structural block diagram of the manual input submodule of the present invention;

[0030] Figure 4 This is a block diagram of the pricing calculation module of the present invention;

[0031] Figure 5 This is a preview diagram of the module structure of the present invention;

[0032] Figure 6 This is a block diagram of the data storage module structure of the present invention;

[0033] Figure 7 This is a flowchart of the method of the present invention;

[0034] Figure 8 This is a flowchart illustrating the multi-graph integrated quotation process of the present invention;

[0035] Figure 9 This is a flowchart illustrating the drawing analysis process of the present invention;

[0036] Figure 10 This is a flowchart of the layer recognition process of the present invention;

[0037] Figure 11 This is a flowchart of the bridge site detection algorithm of the present invention;

[0038] Figure 12 This is a flowchart of the layout recognition process of the present invention;

[0039] Figure 13 This is a diagram illustrating the complexity evaluation dimensions of the present invention.

[0040] Figure 14 This is a schematic diagram of the drawing-free quotation interface of the present invention;

[0041] Figure 15 This is a flowchart of the pricing calculation module of the present invention;

[0042] Figure 16 This is a flowchart of the intelligent recognition process for monochrome drawings according to the present invention.

[0043] In the diagram: 1. Data Input Module; 11. Drawing Input Submodule; 111. File Parsing Unit; 1111. Entity Information Extraction Submodule; 1112. Layer Information Extraction Submodule; 1113. Boundary Information Extraction Submodule; 112. Layer Recognition Unit; 1121. Color Rule Recognition Submodule; 1122. User Habit Recognition Submodule; 1123. Manual Confirmation Submodule; 1124. Monochrome Drawing Recognition Submodule; 113. Bridge Location Detection Unit; 114. Imposition Analysis Unit; 115. Geometric Calculation Unit; 12. Manual Input Submodule; 121. Intelligent Parameter Input Unit; 122. Imposition Parameter configuration unit; 1, 2, 3; Geometric parameter derivation unit; 2; Graphic analysis and calculation module; 21; Local algorithm evaluation submodule; 22; AI visual evaluation submodule; 3; Pricing calculation module; 31; Tool material pricing submodule; 32; Sheet metal pricing submodule; 33; Punching pricing submodule; 34; Cutting time calculation submodule; 35; Die weight calculation submodule; 4; Preview display module; 41; Drawing information preview submodule; 42; Real-time quotation submodule; 5; Data storage module; 51; Configuration data storage repository; 52; Historical quotation storage repository; 53; User preference storage repository; 54; Standard graphic template library. Detailed Implementation

[0044] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0045] Please see the appendix Figure 1 -Appendix Figure 16This invention provides an embodiment of an intelligent pricing system for die-cutting molds, comprising a data input module 1, a graphic analysis and calculation module 2, a pricing calculation module 3, a preview and display module 4, and a data storage module 5. The data storage module 5 establishes data connections with the data input module 1, the graphic analysis and calculation module 2, the pricing calculation module 3, and the preview and display module 4, respectively. The graphic analysis and calculation module 2 establishes data connections with the data input module 1 and the pricing calculation module 3, and the pricing calculation module 3 establishes a data connection with the preview and display module 4. The data input module 1 serves as the data acquisition entry point, supporting both graphic-based and non-graphic-based pricing scenarios, and obtaining the basic data required for die-cutting mold pricing. According to the data, the system provides raw input for subsequent analysis and calculation. The graphic analysis and calculation module 2 is used to evaluate the complexity of die-cutting machining and generate a complexity coefficient, providing a reference for machining difficulty in pricing calculations. The pricing calculation module 3 is used to calculate various costs and the total price of the die-cutting based on geometric parameters, complexity coefficients, material specifications, and continuous marginal coefficient algorithms. The preview and display module 4 is used to visually display pricing-related information, supporting previewing pricing results, real-time updates, and file export. The data storage module 5 stores the configuration data, historical data, user preferences, and template data required for system operation, supporting system adaptive learning and historical traceability. The data input module 1 includes a drawing input submodule 11 and... The manual input submodule 12 and drawing input submodule 11 are used to handle quotations with drawings, while the manual input submodule 12 is used to handle quotations without drawings. The graphic analysis and calculation module 2 includes a local algorithm evaluation submodule 21 and an AI visual evaluation submodule 22. The local algorithm evaluation submodule 21 is used to calculate the complexity score based on geometric features, manufacturing difficulty, and similarity indicators. The AI ​​visual evaluation submodule 22 is used to call a visual model to assist in evaluating complexity and integrate it with the local algorithm results. The pricing calculation module 3 includes a cutting material pricing submodule 31, a sheet metal pricing submodule 32, a punching pricing submodule 33, a cutting time calculation submodule 34, and a cutting die weight calculation submodule. Block 35 includes the following sub-modules: Tool Material Pricing Sub-module 31 calculates tool material costs based on the total cutting circumference; Sheet Material Pricing Sub-module 32 calculates sheet material costs based on the die boundary dimensions; Punching Pricing Sub-module 33 calculates punching costs based on hole type, quantity, and size; Cutting Time Calculation Sub-module 34 estimates the die processing and cutting time; and Die Weight Calculation Sub-module 35 calculates the weight of the finished die. The Preview Display Module 4 includes the Drawing Information Preview Sub-module 41 and the Real-time Quotation Sub-module 42. The Drawing Information Preview Sub-module 41 provides a visual representation of the graphics, while the Real-time Quotation Sub-module 42 updates the detailed cost breakdown, complexity rating, and total quotation when parameters change.Data storage module 5 includes a configuration data repository 51, a historical quotation repository 52, a user preference repository 53, and a standard graphic template library 54. The configuration data repository 51 stores the configuration data required for system operation; the historical quotation repository 52 stores the user's historical quotation details and quotation status; the user preference repository 53 stores the user's layer type confirmation history, confidence calculation data, and personalized operation preferences; and the standard graphic template library 54 stores the standard graphic parametric model required for quotation scenarios without drawings. The drawing input submodule 11 includes a file parsing unit 111, a layer recognition unit 112, a bridge location detection unit 113, a panel layout analysis unit 114, and a geometric calculation unit 115. The file parsing unit 111 is used to parse the vector file structure and extract the core data of entities, layers, and boundaries required for pricing. The layer recognition unit 112 is used to accurately identify the layer type through a multi-layer strategy. The bridge detection unit 113 is used to automatically detect breakpoints in the die-cutting drawing, verify their validity, and complete the breakpoints. It detects bridge positions by calculating the gap distance between adjacent endpoints, verifying whether the endpoint tangents are aligned, verifying whether the connection normal offset is within the threshold range, and detecting whether there are obstructing entities in the bridge position corridor. The panel analysis unit 114 is used to identify repeating graphic patterns in the die-cutting drawing and count the number of panels. It detects closed graphics, calculates the size fingerprint of each closed graphic, and groups similar graphics according to the size fingerprint. The system identifies duplicate graphics by grouping and counting the number of graphics in each group. The geometric calculation unit 115 calculates the geometric parameters of the die-cutting mold, including the perimeter of straight lines, arcs, circles, and polylines, and the area of ​​rectangles, circles, ellipses, annexes, and rectangular spacers. The manual input submodule 12 includes an intelligent parameter input unit 121, a panel parameter configuration unit 122, and a geometric parameter derivation unit 123. The intelligent parameter input unit 121 dynamically generates corresponding parameter input controls based on the standard graphic type selected by the user. The panel parameter configuration unit 122 supports multi-panel layout configuration and shared-die mode. The geometric parameter derivation unit 123 automatically calculates the geometric parameters of the die-cutting mold based on the basic dimension parameters input by the user. The component parsing unit 111 includes an entity information extraction subunit 1111, a layer information extraction subunit 1112, and a boundary information extraction subunit 1113. The entity information extraction subunit 1111 is used to extract entity information, which includes the starting and ending coordinates of a straight line, the center, radius, starting angle, and ending angle of an arc, the center and radius of a circle, the vertex sequence and closure mark of a lightweight polyline, and the control points and order of a spline curve. The layer information extraction subunit 1112 is used to extract layer information, which includes the layer name, layer color, and layer visibility. The boundary information extraction subunit 1113 is used to extract boundary information, which includes the width and height of the overall bounding box of the graphic.The layer recognition unit 112 includes a color rule recognition subunit 1121, a user habit recognition subunit 1122, a manual confirmation subunit 1123, and a monochrome drawing recognition subunit 1124. The color rule recognition subunit 1121 automatically identifies layer types according to preset color mapping rules. The user habit recognition subunit 1122 records the user's layer type confirmation history and improves the recognition confidence based on the number of confirmations. The manual confirmation subunit 1123 provides a human-computer interaction interface for users to confirm or modify layer types when the overall confidence of automatic recognition is less than a preset threshold. The monochrome drawing recognition subunit 1124 identifies layer types for drawings where all entities have the same color through a multi-dimensional strategy. This multi-dimensional strategy includes layer name-based recognition, automatic classification based on geometric features, inference based on positional relationships, and recognition based on line type attributes.

[0046] A method for intelligent pricing of die-cutting molds includes: Step 1, data input; Step 2, graphic analysis and calculation; Step 3, pricing calculation; and Step 4, result output.

[0047] In step one above, the basic data of the die-cutting mold is input through the data input module 1. Specifically, for the scenario with drawings, the die-cutting mold drawings are uploaded through the drawing input submodule 11, and the drawings are automatically parsed to extract entity information, layer information and boundary information. For the scenario without drawings, the standard graphic template is selected through the manual input submodule 12, the basic size parameters are input, the panel configuration parameters are set, and the geometric parameters of the die-cutting mold are automatically calculated.

[0048] In step two above, the complexity coefficient of die-cutting is generated through the graphic analysis and calculation module 2;

[0049] In step three above, the pricing calculation module 3 calls the geometric parameters, complexity coefficients, and configuration data, and combines them with the continuous marginal coefficient algorithm to calculate various fees and the total price. Specifically, the continuous marginal coefficient algorithm involves: using a continuous smoothing function to interpolate the base unit price within a preset coefficient range to obtain the base marginal coefficient; using a decreasing function to calculate the gradient coefficient based on the pricing unit number; and multiplying the base marginal coefficient by the gradient coefficient to obtain the final marginal coefficient for each pricing unit. The number of pricing units refers to the number of graphics on a single sheet of material. The total price can be expressed using the following formula:

[0050]

[0051] Among them, C m Total price; C munit N is the unit price; N is the number of pricing units. Let be the final marginal coefficient of the i-th pricing unit. The total price calculation modes include the first-item full-price mode and the unified marginal coefficient mode. The first-item full-price mode sets the final marginal coefficient of the first pricing unit to 1, and applies the calculated marginal coefficient from the second unit onwards. The unified marginal coefficient mode applies the calculated marginal coefficient to all pricing units.

[0052] In step four above, the quotation results are visualized through the preview display module 4.

[0053] Based on the above, the advantages of this invention are that, when in use, the data input module 1 serves as the data acquisition entry point. Templates from the standard graphic template library 54 can be selected via the manual input submodule 12, basic dimension parameters can be input via the intelligent parameter input unit 121, and layout configuration can be performed via the layout parameter configuration unit 122. Then, the geometric parameter derivation unit 123 calculates the core parameters in real time, thereby achieving the function of providing a quotation without drawings. Alternatively, single or multiple DXF / AI / PDF / CDR format vector drawings can be uploaded via the drawing input submodule 11. The file parsing unit 111 parses the structure of DXF / AI / PDF / CDR and other vector files, extracts entity information via the entity information extraction submodule 1111, extracts layer information via the layer information extraction submodule 1112, extracts boundary information via the boundary information extraction unit 1113, the layer recognition unit 112 uses a multi-layer recognition strategy for layer recognition, and the color rule recognition submodule 1121 automatically maps layer types based on layer colors. For example, white maps to cutting material (cutting line), red to indentation (fold line), blue to serrated edge (serrated line), cyan to sheet metal (material boundary), magenta to punching hole (feeding hole), and green to text / annotation. The user habit recognition subunit 1122 records the user's confirmation history of layer types. Confidence increases with the number of confirmations. The confidence calculation formula can be expressed as: Confidence = Base value + Number of confirmations × 0.02, with a maximum confidence level of 0.98. For example, if a user uploads a DXF file, the system recognizes a yellow layer, classifying it as unknown by default, and displays a layer confirmation interface. The user manually changes the yellow layer to a serrated edge, and the system records the confirmation history. When the user uploads a file containing a yellow layer again, the system checks the history and finds that the yellow layer was previously confirmed as a serrated edge, so it automatically identifies it as such with a confidence level of 0.82. After the user confirms again, the number of confirmations accumulates to 2, increasing the confidence level to 0.84. After multiple confirmations, the confidence level approaches the maximum of 0.If the system detects a yellow layer as a toothed knife, the system will always identify it as such. When the confidence level of the automatic identification is insufficient, the manual confirmation subunit 1123 provides a human-computer interaction interface for the user to confirm or modify the layer type. When all entities in the drawing are detected to have the same color, the monochrome drawing identification subunit 1124 enables a multi-dimensional intelligent identification strategy. This strategy can infer the type by parsing keywords in the layer name. For example, if the layer name keywords include "knife," "cut," or "cutting," it is mapped to a knife material; if they include "press," "crease," "fold," or "score," it is mapped to an indentation; if they include "tooth," "tooth," or "perf," it is mapped to a toothed knife; if they include "hole," "hole," or "punch," it is mapped to a punch; and if they include "board," it is mapped to a hole. If the name is "frame", it is mapped to the edge of the sheet metal. For entities that cannot be identified by name, they can be intelligently classified according to their geometric attributes. For example, if the geometric attribute is "circle" and the diameter is less than 15mm, it can be classified as a punch; if the geometric attribute is "circle" and the diameter is less than 15mm, it can be classified as a circular blade outline; if the geometric attribute is "dashed line" or "dotted line", it can be classified as an indentation line; if the geometric attribute is the outermost closed outline, it can be classified as the blade outer frame; if the geometric attribute is a regularly arranged array of small circles, it can be classified as a punch group. For entities that are still undetermined, they can be inferred by analyzing the spatial relationship between entities. For example, by calculating the bounding boxes of all closed shapes, the closed outline corresponding to the outermost bounding box can be inferred as the blade outer frame, which is completely contained within the outer frame. Small circles can be inferred as punching holes, straight lines connecting two areas can be inferred as indentation lines, and line segments tangent to or intersecting the outer frame can be inferred as crease lines. Line type attributes can also be used for identification; for example, solid lines represent cutting materials, dashed lines represent indentation lines, dotted lines represent indentation lines or annotation lines, center lines represent auxiliary lines or annotation lines, and hidden lines represent indentation lines. AI vision-assisted recognition can also be used. After rendering the drawing as an image, the AI ​​vision model is called to identify the overall structure and layout of the drawing, infer the functional type of each area, return classification suggestions for user confirmation, and integrate with local algorithm results to improve accuracy. The formula for calculating the confidence score of the monochrome drawing recognition subunit 1124 is: Overall Confidence Score = max(Name Confidence, Geometric Confidence, Location Confidence, Line Type Confidence). For example, when all entities... If the overall confidence level is ≥0.6, there is no need to enable the user visual assignment mode. If the overall confidence level is <0.6, the user visual assignment mode is enabled, highlighting the entities to be classified. Users can select and specify types in batches. After user confirmation, the system records the user's classification preferences for subsequent learning. The bridge detection unit 113 can realize intelligent detection and completion of bridge positions. Bridge positions are intentionally left breakpoints in the die-cutting drawing to maintain the connection between the product and the waste. Large bridge positions are breakpoints with a gap of about 5mm, and small bridge positions are breakpoints with a gap of 1-1.6mm. The detection algorithm is as follows: First, group the entities by layer, then traverse each layer, perform endpoint analysis on line segment entities, and calculate the gap distance for each pair of close endpoints. The formula is: gap = sqrt((endpoint A).x-endpoint Bx). 2 +(Endpoint Ay - Endpoint By) 2Next, the validity of the bridge position is verified using tangential alignment: the angle between the tangent directions at both ends is less than a threshold; normal offset: the normal offset of the connecting line relative to the tool line is less than a threshold; and corridor collision detection: checking whether there are other entities blocking the corridor of the bridge position. Finally, fill line segments are generated, and straight line entities connecting the breakpoints are created, maintaining the same color attributes as the original layer. Specific layers such as marking lines, board borders, blanking holes, and text are excluded, or specific color layers are excluded. The panel analysis unit 114 automatically identifies the repeating graphic patterns in the drawing. First, closed graphic detection is performed, then the boundary frame size of each closed graphic is calculated, a dimension fingerprint is generated, and the graphics are grouped according to the dimension fingerprint. Graphics in the same group are considered as identical. In the same layout unit, the final layout statistics are performed, outputting the number of repetitions for each graphic, the sample coordinates of each graphic, and the total number of layouts. The geometric calculation unit 115 performs length and area calculations for different entity types, such as calculating the length of a straight line and the circumference of a circle, and the area of ​​annular rings and rectangular spacers. When multiple vector drawings in DXF / AI / PDF / CDR format are input through the drawing input submodule 11, the system parses each file independently, generates independent analysis results, and then uses either an independent calculation mode or a combined calculation mode. The independent calculation mode calculates the price for each drawing independently, displays the details of each drawing, and summarizes the total price. The combined calculation mode merges multiple drawings into a single sheet. The system automatically optimizes board utilization, shares board costs, and reduces overall costs. It ultimately applies batch discounts based on the total quantity, uses a shared price adjustment coefficient across multiple images, and outputs an integrated quote. The graphics analysis and calculation module 2 employs a dual-track parallel complexity evaluation mechanism. The local algorithm evaluation submodule 21 calculates complexity scores based on the analytical data. The scoring dimensions include geometric complexity: for example, scoring curve complexity based on the arc ratio: +0.7 points for an arc ratio ≥ 85% or > 200 arcs, +0.6 points for an arc ratio ≥ 70%, +0.5 points for an arc ratio ≥ 50%, and +0.4 points for an arc ratio ≥ 30%. For example, scoring based on the estimated number of vertices: +0.4 points for 20 vertices. (17-2) Zero vertices: +0.3 points; 13-16 vertices: +0.2 points; For example, scoring based on the number of irregular holes: 15 holes: +0.3 points; 10-15 holes: +0.25 points; For example, scoring based on manufacturing difficulty: minimum feature size < 2mm: +0.2 points; 2mm < minimum feature size < 3mm: +0.2 points; 15 small bridge positions: +0.3 points; 10-15 small bridge positions: +0.2 points; For example, scoring based on similarity index: regular arrangement reduces complexity; the presence of repeated graphics reduces complexity; This can be expressed by the following formula: Complexity score = base score + geometric score + manufacturing score - similarity deduction, Complexity score = min(3.0. Complexity score), which can be divided into four levels: simple, normal, complex, and extremely complex. The AI ​​visual evaluation submodule 22 can be used as an auxiliary evaluation, rendering the parsed drawings into images, calling the visual model to analyze the images, extracting AI recognition results, such as the number of panels, graphic type judgment, and complexity suggestions, and finally merging them with the local algorithm results. In the pricing calculation module 3, the cutting material pricing submodule 31 calculates the cutting material cost based on the cutting perimeter and cutting material specification parameters. First, it obtains the total perimeter of the cutting material layer, then queries the unit price according to the cutting material brand and specifications, then applies the price adjustment coefficient, and finally calculates the cost according to the following formula. Calculating cutting tool cost: Cutting tool cost = Total perimeter × Cutting tool unit price × Price adjustment coefficient; Sheet metal pricing submodule 32 calculates sheet metal cost based on the die boundary dimensions and sheet metal specifications. First, obtain the overall boundary frame dimensions of the die, then add the outer edge distance, calculate the sheet metal area, look up the unit price based on the sheet metal thickness, and apply the price adjustment coefficient to calculate the sheet metal cost. This can be expressed by the following formula: Sheet metal cost = Sheet metal area × Sheet metal unit price × Price adjustment coefficient; Punching pricing submodule 33 calculates punching cost based on the type, size, and quantity of holes. First, count the quantity of each type of hole, look up the unit price based on the hole type, and match the hole diameter... The pricing tiers are assigned, and the punching cost is calculated using a price adjustment coefficient. This can be expressed by the following formula: Single-type punching cost = Number of holes × Unit price of hole × Price adjustment coefficient; the total punching cost is the sum of the costs for each type of punching. The cutting time calculation submodule 34 estimates the processing and cutting time based on the die complexity and dimensions, using the following formula: Basic cutting time = Total circumference × Cutting time per unit circumference; Complexity correction time = Basic cutting time × (Complexity coefficient - 1); Tool change time = Number of layers × Single tool change time; Clamping time = Basic clamping time + Dimension correction; Total cutting time = Basic cutting time + Complexity correction. The time includes cutting time, tool change time, and clamping time. The cutting time per unit circumference is set according to the material and tool type, such as 1 minute / meter. The complexity coefficient is provided by the graphic analysis and calculation module 2. The single tool change time can be set to 5 minutes, and the basic clamping time is set to 10 minutes. The die weight calculation submodule 35 calculates the weight of the finished die based on the die size and material specifications. The formula is as follows: Die weight = Die circumference × Die height × Die thickness × Steel density; Plate weight = Plate area × Plate thickness × Wood density; Total die weight = Die weight + Plate weight. For example, the material density parameter is 7.85 g / cm³ for steel. 3 The wood density is 0.6-0.8 g / cm³. 3The die weight can be used for logistics cost estimation, packaging material selection, and equipment load-bearing verification. In the preview display module 4, the drawing information preview submodule 41 can realize real-time rendering of SVG graphics, and instantly re-render the graphic preview when parameters change. It supports multi-panel layout display, can display dimension lines and values, and uses different colors to distinguish different layers. The real-time quotation submodule 42 can automatically update the calculated price when parameters change, and can display various cost details and price change trends. In the data storage module 5, the configuration data storage repository 51 is used to store configuration data, such as pricing rules, material unit price, price adjustment coefficient, continuous marginal coefficient, panel default parameters, batch discount rules, etc. The historical quotation storage repository 52 is used to store users' historical quotation details and quotation status. The user preference storage repository 53 is used to store users' layer type confirmation history, confidence calculation data, personalized operation preferences, etc. The standard graphic template library 54 is used to store the standard graphic parametric models required for quotation scenarios without drawings.

[0054] 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 invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A die-cutting mold intelligent pricing system, comprising a data input module (1), a graphic analysis and calculation module (2), a pricing calculation module (3), a preview display module (4), and a data storage module (5), characterized in that: The data storage module (5) establishes data connections with the data input module (1), the graphics analysis and calculation module (2), the pricing calculation module (3), and the preview display module (4), respectively. The graphics analysis and calculation module (2) establishes data connections with the data input module (1) and the pricing calculation module (3), respectively, and the pricing calculation module (3) establishes data connections with the preview display module (4).

2. The intelligent quotation system for die-cutting molds according to claim 1, characterized in that: The data input module (1) serves as the data acquisition entry point, supporting all scenarios of quotations with and without drawings, acquiring the basic data required for die pricing, and providing the original input for subsequent analysis and calculation. The graphic analysis and calculation module (2) is used to evaluate the processing complexity of the die and generate a complexity coefficient, providing a processing difficulty reference for pricing calculation. The pricing calculation module (3) is used to calculate various costs and the total price of the die based on geometric parameters, complexity coefficients, material specifications, and continuous marginal coefficient algorithms. The preview and display module (4) is used to visualize and display relevant pricing information, supporting the preview of pricing results, real-time updates, and file export. The data storage module (5) stores the configuration data, historical data, user preferences, and template data required for system operation, supporting the system's adaptive learning and historical traceability.

3. The intelligent quotation system for die-cutting molds according to claim 2, characterized in that: The data input module (1) includes a drawing input submodule (11) and a manual input submodule (12). The drawing input submodule (11) is used to process quotations with drawings, and the manual input submodule (12) is used to process quotations without drawings.

4. The intelligent quotation system for die-cutting molds according to claim 2, characterized in that: The graphics analysis and calculation module (2) includes a local algorithm evaluation submodule (21) and an AI visual evaluation submodule (22). The local algorithm evaluation submodule (21) is used to calculate the complexity score based on geometric features, manufacturing difficulty, and similarity indicators. The AI ​​visual evaluation submodule (22) is used to call a visual model to assist in evaluating the complexity and integrate it with the local algorithm results.

5. The intelligent quotation system for die-cutting molds according to claim 2, characterized in that: The pricing calculation module (3) includes a tool material pricing submodule (31), a sheet material pricing submodule (32), a punching pricing submodule (33), a cutting time calculation submodule (34), and a die weight calculation submodule (35). The tool material pricing submodule (31) is used to calculate the tool material cost based on the total circumference of the cut. The sheet material pricing submodule (32) is used to calculate the sheet material cost based on the die boundary dimensions. The punching pricing submodule (33) is used to calculate the punching cost based on the hole type, quantity, and size. The cutting time calculation submodule (34) is used to estimate the die processing and cutting time. The die weight calculation submodule (35) is used to calculate the weight of the finished die.

6. The intelligent quotation system for die-cutting molds according to claim 2, characterized in that: The preview display module (4) includes a drawing information preview submodule (41) and a real-time quotation submodule (42). The drawing information preview submodule (41) is used for the visual display of graphics, and the real-time quotation submodule (42) is used to update the details of various costs, complexity rating and total quotation when parameters change.

7. The intelligent quotation system for die-cutting molds according to claim 2, characterized in that: The data storage module (5) includes a configuration data repository (51), a historical quotation repository (52), a user preference repository (53), and a standard graphic template library (54). The configuration data repository (51) is used to store the configuration data required for system operation. The historical quotation repository (52) is used to store the user's historical quotation details and quotation status. The user preference repository (53) is used to store the user's layer type confirmation history, confidence calculation data, and personalized operation preferences. The standard graphic template library (54) is used to store the standard graphic parameterized model required for the quotation scenario without a graphic.

8. The intelligent quotation system for die-cutting molds according to claim 3, characterized in that: The drawing input submodule (11) includes a file parsing unit (111), a layer recognition unit (112), a bridge detection unit (113), a layout analysis unit (114), and a geometric calculation unit (115). The file parsing unit (111) is used to parse the vector file structure and extract the core data of entities, layers, and boundaries required for pricing. The layer recognition unit (112) is used to accurately identify the layer type through a multi-layer strategy. The bridge detection unit (113) is used to automatically detect breakpoints in the die-cutting drawing, verify the validity, and complete the breakpoints by calculating the gap distance between adjacent endpoints. The bridge position is detected by verifying whether the endpoint tangent is aligned, whether the connection normal offset is within the threshold range, and whether there are obstructing entities in the bridge position corridor. The panel analysis unit (114) is used to identify the repeating graphic patterns in the die-cutting drawing and count the number of panels. Repeating graphics are identified by detecting closed graphics, calculating the size fingerprint of each closed graphic, grouping similar graphics by size fingerprint, and counting the number of graphics in each group. The geometric calculation unit (115) is used to calculate the geometric parameters of the die-cutting, including the perimeter of straight lines, arcs, circles, polylines, and the area of ​​rectangles, circles, ellipses, annulus and rectangular pads.

9. The intelligent quotation system for die-cutting molds according to claim 3, characterized in that: The manual input submodule (12) includes an intelligent parameter input unit (121), a panel parameter configuration unit (122), and a geometric parameter derivation unit (123). The intelligent parameter input unit (121) is used to dynamically generate corresponding parameter input controls according to the standard graphic type selected by the user. The panel parameter configuration unit (122) is used to support multi-panel layout configuration and shared cutting mode. The geometric parameter derivation unit (123) is used to automatically calculate the geometric parameters of the cutting die according to the basic size parameters input by the user.

10. The intelligent quotation system for die-cutting molds according to claim 8, characterized in that: The file parsing unit (111) includes an entity information extraction subunit (1111), a layer information extraction subunit (1112), and a boundary information extraction subunit (1113). The entity information extraction subunit (1111) is used to extract entity information, which includes the starting coordinates and ending coordinates of a straight line, the center, radius, starting angle and ending angle of an arc, the center and radius of a circle, the vertex sequence and closure mark of a lightweight polyline, and the control points and order of a spline curve. The layer information extraction subunit (1112) is used to extract layer information, which includes the layer name, layer color and layer visibility. The boundary information extraction subunit (1113) is used to extract boundary information, which includes the width and height of the overall bounding box of the graphic.

11. The intelligent quotation system for die-cutting molds according to claim 8, characterized in that: The layer recognition unit (112) includes a color rule recognition subunit (1121), a user habit recognition subunit (1122), a manual confirmation subunit (1123), and a monochrome drawing recognition subunit (1124). The color rule recognition subunit (1121) is used to automatically recognize the layer type according to a preset color mapping rule. The user habit recognition subunit (1122) is used to record the user's layer type confirmation history and improve the recognition confidence based on the number of confirmations. The manual confirmation subunit (1123) is used to provide a human-computer interaction interface for the user to confirm or modify the layer type when the overall confidence of the automatic recognition is less than a preset threshold. The monochrome drawing recognition subunit (1124) identifies the layer type for all drawings with the same entity color through a multi-dimensional strategy.

12. The intelligent quotation system for die-cutting molds according to claim 11, characterized in that: The multi-dimensional strategy includes layer name recognition, automatic classification based on geometric features, inference based on positional relationships, and recognition based on line type attributes.

13. A method for intelligent pricing of die-cutting molds, comprising: step one, data input; step two, graphic analysis and calculation; step three, pricing calculation; and step four, result output; characterized in that: In step one above, the basic data of the die-cutting mold is input through the data input module (1); In step two above, the complexity coefficient of die-cutting is generated by the graphic analysis and calculation module (2); In step three above, the pricing calculation module (3) calls the geometric parameters, complexity coefficients and configuration data, and combines them with the continuous marginal coefficient algorithm to calculate various costs and the total price. In step four above, the quotation results are visualized through the preview display module (4).

14. The intelligent pricing method for die-cutting molds according to claim 13, characterized in that: In step one, the basic data of the die-cutting mold is input through the data input module (1) as follows: For the scenario of quotation with drawings, the die-cutting mold drawings are uploaded through the drawing input submodule (11), the drawings are automatically parsed, and entity information, layer information and boundary information are extracted; For the scenario of quotation without drawings, the standard graphic template is selected through the manual input submodule (12), the basic size parameters are input, the panel configuration parameters are set, and the geometric parameters of the die-cutting mold are automatically calculated.

15. The intelligent pricing method for die-cutting molds according to claim 13, characterized in that: In step three, the continuous marginal coefficient algorithm is specifically as follows: A continuous smoothing function is used to interpolate the base unit price within a preset coefficient range to obtain the base marginal coefficient. A decreasing function is then used to calculate the gradient coefficient based on the pricing unit number. The base marginal coefficient is multiplied by the gradient coefficient to obtain the final marginal coefficient for each pricing unit. Here, the number of pricing units refers to the number of graphics on a single sheet of material. The total price can be expressed using the following formula: Among them, C m Total price; C munit N is the unit price; N is the number of pricing units. Let be the final marginal coefficient of the i-th pricing unit. The total price calculation mode includes the first-item full-price mode and the unified marginal mode. The first-item full-price mode sets the final marginal coefficient of the first pricing unit to 1, and applies the calculated marginal coefficient from the second unit onwards. The unified marginal mode applies the calculated marginal coefficient to all pricing units uniformly.

Images