Intelligent industrial mold manufacturing method and system and medium

Abstract

The invention relates to an industrial mold intelligent manufacturing method and system and a medium, and the method comprises the steps: building a mold 3D initial model and a mold model performance model of the initial model through obtaining a product design parameter set and a product performance parameter set in advance; optimizing the mold 3D initial model according to a product design parameter actual value set and a product performance parameter actual value set which are obtained in advance to obtain a mold 3D optimized model, and setting an industrial mold manufacturing process parameter set of an actual mold corresponding to the mold 3D optimized model; according to the method, a simulation industrial product and a plurality of modal application performance simulation test value sets corresponding to the simulation industrial product are manufactured, and when the modal application performance simulation test value sets meet preset application performance simulation test conditions, the modal application performance simulation test value sets are obtained. And if not, correspondingly taking a mold design parameter set corresponding to the model after 3D optimization and a current industrial mold manufacturing process parameter set as an industrial mold optimal design parameter set and an industrial mold optimal manufacturing process parameter set respectively, and intelligently manufacturing the mold.

Description

Technical Field

[0001] The present invention relates to the field of intelligent manufacturing, and in particular to an industrial mold intelligent manufacturing method, system and medium. Background Art

[0002] Industrial molds are essential key tools for industrial manufacturing of industrial products. The quality of the molds is crucial to the quality of the manufactured industrial products and their application performance.

[0003] In order to improve the manufacturing efficiency of industrial molds and reduce the initial design costs of molds, mold design software is usually used to first design a mold 3D model of the mold required for manufacturing industrial products, and then the industrial mold is manufactured based on the designed mold 3D model, and the industrial product is manufactured using the set manufacturing process parameters and the industrial mold.

[0004] However, the existing industrial mold manufacturing methods have shortcomings: it is impossible to know the performance of the actual mold manufactured by the mold 3D model, and it is difficult to adjust the mold 3D model and adjust the industrial mold manufacturing process parameter set according to the actual performance of the mold, resulting in the inability to obtain a mold 3D optimization model that meets the application performance of industrial products, thus reducing the manufacturing quality and efficiency of industrial molds. Summary of the invention

[0005] The first technical problem to be solved by the present invention is to provide an industrial mold intelligent manufacturing for the above-mentioned prior art.

[0006] The second technical problem to be solved by the present invention is to provide an industrial mold intelligent manufacturing system for realizing the above-mentioned industrial mold intelligent manufacturing method in view of the above-mentioned prior art.

[0007] The third technical problem to be solved by the present invention is to provide a readable storage medium for realizing the above-mentioned industrial mold intelligent manufacturing method in view of the above-mentioned prior art.

[0008] The technical solution adopted by the present invention to solve the first technical problem is: an industrial mold intelligent manufacturing method, characterized in that it includes the following steps:

[0009] Step 1, respectively obtaining in advance a set of product design parameters required for designing an industrial product structure and a set of product performance parameters characterizing the performance of the industrial product;

[0010] Step 2, pre-constructing a mold 3D initial model of a mold required for manufacturing an industrial product and a mold model performance model corresponding to the performance of the mold 3D initial model; wherein the mold 3D initial model corresponds to the industrial product one-to-one, the mold model performance model has a corresponding mold design parameter set, the input parameters of the mold model performance model are all design parameters in the product design parameter set, and the output parameters of the mold model performance model are all performance parameters in the product performance parameter set;

[0011] Step 3, obtaining in advance a set of manufacturing process parameters, a set of actual values ​​of product design parameters, and a set of corresponding actual values ​​of product performance parameters when actually producing a number of the same industrial products; wherein the set of actual values ​​of product design parameters includes a number of actual value subsets of product design parameters, and the set of actual values ​​of product performance parameters includes a number of actual value subsets of product performance parameters, and the actual value subsets of product design parameters when actually producing each industrial product correspond one-to-one to the actual value subsets of product performance parameters of the industrial product produced;

[0012] Step 4, optimizing the initial 3D model of the mold based on the actual value set of the product design parameters and the actual value set of the product performance parameters to obtain the optimized 3D model of the mold;

[0013] Step 5, setting an industrial mold manufacturing process parameter set of an actual mold corresponding to the mold 3D optimized model, and using the industrial mold manufacturing process parameter set to simulate the mold 3D optimized model to obtain a corresponding simulated industrial product;

[0014] Step 6, performing simulation tests on the simulated industrial product under multi-modal scenario application conditions, and obtaining multiple sets of modal application performance simulation test values ​​of the simulated industrial product; wherein the sets of modal application performance simulation test values ​​correspond one-to-one to the modal scenarios of the simulation tests corresponding to the simulated industrial products, and the sets of modal application performance simulation test values ​​are formed by simulation test values ​​corresponding to several application performance parameters of the physical industrial product corresponding to the simulated industrial product under the actual application conditions of the corresponding modal scenarios;

[0015] Step 7: making a judgment based on the obtained multiple sets of modal performance simulation test values:

[0016] When the multiple modal performance simulation test value sets meet the preset application performance simulation test conditions, proceed to step 8; otherwise, adjust the process parameters in the current industrial mold manufacturing process parameter set, and set the adjusted industrial mold manufacturing process parameter set as the industrial mold manufacturing process parameter set of the mold corresponding to the actual mold manufacturing 3D optimized model, and then proceed to step 5;

[0017] Step 8: Use the mold design parameter set corresponding to the mold 3D optimized model as the optimal design parameter set of the industrial mold corresponding to the industrial product, and use the current industrial mold manufacturing process parameter set as the optimal manufacturing process parameter set of the industrial mold corresponding to the industrial product.

[0018] Improved, in the industrial mold intelligent manufacturing method, the industrial product is a milling tool, and the product design parameter set includes the number of tool teeth, tool rake angle, tool back angle, tool rotation angle, blade height and tool thickness of the milling tool; the product performance parameter set includes the cutting speed feed rate, cutting depth and cutting angle of the milling tool.

[0019] Furthermore, in the industrialized mold intelligent manufacturing method, the construction process of the mold 3D initial model includes the following steps:

[0020] Step a1, selecting a qualified industrial product as a calibration product, and obtaining images of the calibration product at multiple viewing angles to form a multi-view image set of the calibration product;

[0021] Step a2, extracting a feature point information sequence representing the three-dimensional structural features of the calibrated product according to the formed multi-view image set; wherein the feature point information sequence is formed by extracting a plurality of feature point information, and the feature point information includes the position coordinates of the feature points;

[0022] Step a3, dividing all feature points in the feature point information sequence into slices to obtain a plurality of slice areas and a feature point information sequence corresponding to each slice area; wherein all feature points on the same surface are divided into one slice;

[0023] Step a4, obtaining a point cloud data set of the calibration product, and calculating the point cloud sparsity of the point cloud data corresponding to each slice area on the calibration product;

[0024] Step a5, extracting the point cloud sparsity of the same slice area and all the corresponding feature points in a traversal manner as the modeling contour parameters of the same slice area, and the modeling contour parameters extracted from all slice areas together form a modeling contour parameter set that characterizes the overall contour of the calibration product;

[0025] Step a6, constructing a mold 3D reference model according to the obtained modeling profile parameter set;

[0026] Step a7, performing contour optimization processing on the mold 3D reference model to obtain the final mold 3D initial model.

[0027] Further improved, in the industrial mold intelligent manufacturing method, in step a7, the process of performing contour optimization processing on the constructed mold 3D reference model includes the following steps:

[0028] Step a71, selecting adjacent shard area pairs and the boundary lines of each shard area pair in a traversal manner in all shard areas to form a boundary line set;

[0029] Step a72, respectively obtaining all feature points corresponding to each boundary line in the boundary line set, obtaining the attribution relationship between the feature points and the corresponding boundary lines, and forming a boundary line-feature point attribution relationship list from all the attribution relationships;

[0030] Step a73, respectively calculating the slope value of each boundary line at each characteristic point thereof, and respectively forming a slope value sequence corresponding to each boundary line; wherein the boundary line and the slope value sequence are in a one-to-one correspondence, and the slope values ​​in the slope value sequence are arranged according to the front and rear positions on the boundary line;

[0031] Step a74, based on the obtained slope value sequences, a curve curvature determination process is performed on the boundary line corresponding to each slope value sequence:

[0032] When the variance of any three adjacent slope values ​​in the slope value sequence is greater than the preset variance threshold, the process proceeds to step a75; otherwise, the boundary line corresponding to the slope value sequence is not processed;

[0033] Step a75, adjusting the position of the feature point corresponding to the middle slope value among the three adjacent slope values ​​on the boundary curve, and taking the slope value corresponding to the feature point after the position adjustment as the updated slope value;

[0034] Step a76, allowing the updated slope value to replace the original slope value of the feature point in the slope value sequence to form a new slope value sequence, and then going to step a74 again.

[0035] As a further improvement, in the industrial mold intelligent manufacturing method, in step 4, the process of optimizing the mold 3D initial model based on the actual value set of product design parameters and the actual value set of product performance parameters to obtain the optimized 3D mold model includes the following steps:

[0036] Step 41, dividing the product design parameter actual value set into a product design parameter training set and a product design parameter verification set;

[0037] Step 42, according to the divided product design parameter training set and product design parameter verification set, the product performance parameter actual value set is correspondingly divided into a product performance parameter training set and a product performance parameter verification set;

[0038] Step 43, inputting the product design parameter training set into the mold model performance model, and using the output of the mold model performance model as the corresponding product performance parameter prediction value set; wherein the actual values ​​of the product design parameters in the product design parameter training set correspond one-to-one to the predicted values ​​of the product performance parameters in the product performance parameter prediction value set;

[0039] Step 44, calculating the output performance loss ratio of the mold model performance model according to the product performance parameter verification set and the obtained product performance parameter prediction value set;

[0040] Step 45, making a judgment based on the output performance loss ratio of the obtained mold model performance model:

[0041] When the output performance loss ratio is less than the preset loss ratio threshold, the current mold 3D initial model is used as the mold 3D optimized model; otherwise, the process proceeds to step 46;

[0042] Step 46, reselect another qualified industrial product as the calibration product, and proceed to step a1.

[0043] Further improvement, in the industrial mold intelligent manufacturing method, the mold model performance model is obtained by performing line fitting processing on the product design parameter set and the product performance parameter set; wherein: the mold model performance model is constructed as follows:

[0044] τ i =γ(θ);1≤i≤M;θ={θ1,θ2,…,θ N};

[0045] Among them, the industrial product corresponding to the mold 3D initial model has M performances, τ i is the i-th performance of the industrial product corresponding to the mold 3D initial model, θ is the product design parameter set of the mold model performance model, θ1,θ2,…,θ N are N product design parameters in the product design parameter set.

[0046] The technical solution adopted by the present invention to solve the second technical problem is: an industrial mold intelligent manufacturing system, which realizes the industrial mold intelligent manufacturing method, and is characterized in that the industrial mold intelligent manufacturing system includes:

[0047] The parameter acquisition module is configured to respectively acquire a set of product design parameters required for designing an industrial product structure and a set of product performance parameters characterizing the performance of the industrial product, and to acquire a set of manufacturing process parameters, a set of actual values ​​of product design parameters, and a corresponding set of actual values ​​of product performance parameters when a plurality of the same industrial products are actually produced; wherein the set of actual values ​​of product design parameters includes a plurality of actual value subsets of product design parameters, the set of actual value of product performance parameters includes a plurality of actual value subsets of product performance parameters, and the actual value subsets of product design parameters when each industrial product is actually produced correspond one to one to the actual value subsets of product performance parameters of the industrial product produced;

[0048] A model building module is configured to build a mold 3D initial model of a mold required for manufacturing an industrial product and a mold model performance model corresponding to the performance of the mold 3D initial model; wherein the mold 3D initial model corresponds to the industrial product one-to-one, the mold model performance model has a corresponding mold design parameter set, the input parameters of the mold model performance model are all design parameters in the product design parameter set, and the output parameters of the mold model performance model are all performance parameters in the product performance parameter set;

[0049] The simulation module is configured to use the industrial mold manufacturing process parameter set to simulate the manufacturing of the mold 3D optimized model to obtain a corresponding simulated industrial product; perform simulation tests on the simulated industrial product under multi-modal scenario application conditions to obtain multiple modal application performance simulation test value sets of the simulated industrial product;

[0050] The judgment processing module is configured to make the judgment processing according to the obtained multiple modal performance simulation test value sets.

[0051] The technical solution adopted by the present invention to solve the third technical problem is: a readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, any one of the industrial mold intelligent manufacturing methods described is implemented.

[0052] Compared with the prior art, the advantages of the present invention are: the industrial mold intelligent manufacturing method of the present invention obtains a product design parameter set and a product performance parameter set in advance, constructs a mold 3D initial model and its mold model performance model, optimizes the mold 3D initial model to obtain a mold 3D optimized model, sets an industrial mold manufacturing process parameter set for the actual mold corresponding to the mold 3D optimized model, and thereby manufactures a simulated industrial product and its corresponding multiple modal application performance simulation test value sets. When the multiple modal performance simulation test value sets meet the preset application performance simulation test conditions, the mold design parameter set corresponding to the mold 3D optimized model and the current industrial mold manufacturing process parameter set are respectively used as the industrial mold optimal design parameter set and the industrial mold optimal manufacturing process parameter set, thereby realizing the adjustment of mold model parameters based on model performance, and can predict in advance the performance of the actual mold manufactured by the mold 3D optimized model, so as to obtain a mold 3D optimized model that meets the application performance of industrial products, thereby realizing intelligent manufacturing of industrial molds and improving the manufacturing quality and efficiency of industrial molds. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 It is a schematic diagram of the process of the industrial mold intelligent manufacturing method in an embodiment of the present invention;

[0054] Figure 2 Schematic diagram of an industrialized mold intelligent manufacturing system in an embodiment of the present invention. DETAILED DESCRIPTION

[0055] The present invention is further described in detail below with reference to the accompanying drawings.

[0056] This embodiment provides an industrial mold intelligent manufacturing method, where the industrial mold is a milling tool mold. Figure 1 As shown, the industrial mold intelligent manufacturing method of this embodiment includes the following steps:

[0057] Step 1, respectively obtaining in advance a set of product design parameters required for designing an industrial product structure and a set of product performance parameters characterizing the performance of the industrial product; wherein, in this embodiment, the industrial product is a milling tool, and the set of product design parameters includes the number of tool teeth, tool rake angle, tool clearance angle, tool rotation angle, blade height, and tool thickness of the milling tool; and the set of product performance parameters includes the cutting speed feed rate, cutting depth, and cutting angle of the milling tool;

[0058] Step 2, pre-constructing a mold 3D initial model of a mold required for manufacturing an industrial product and a mold model performance model corresponding to the performance of the mold 3D initial model; wherein the mold 3D initial model corresponds to the industrial product one-to-one, the mold model performance model has a mold design parameter set corresponding to itself, the input parameters of the mold model performance model are all design parameters in the product design parameter set, and the output parameters of the mold model performance model are all performance parameters in the product performance parameter set;

[0059] It should be noted that, in this embodiment, the mold model performance model is obtained by performing line fitting processing on the product design parameter set and the product performance parameter set; wherein: the mold model performance model is constructed as follows:

[0060] τ i =γ(θ);1≤i≤M;θ={θ1,θ2,…,θ N};

[0061] Among them, the industrial product corresponding to the mold 3D initial model has M performances, τ i is the i-th performance of the industrial product corresponding to the mold 3D initial model, θ is the product design parameter set of the mold model performance model, θ1,θ2,…,θ N are N product design parameters in the product design parameter set;

[0062] Step 3, obtaining in advance a set of manufacturing process parameters, a set of actual values ​​of product design parameters, and a set of corresponding actual values ​​of product performance parameters when actually producing a number of the same industrial products; wherein the set of actual values ​​of product design parameters includes a number of actual value subsets of product design parameters, and the set of actual values ​​of product performance parameters includes a number of actual value subsets of product performance parameters, and the actual value subsets of product design parameters when actually producing each industrial product correspond one-to-one to the actual value subsets of product performance parameters of the industrial product produced;

[0063] Step 4, optimizing the initial 3D model of the mold based on the actual value set of the product design parameters and the actual value set of the product performance parameters to obtain the optimized 3D model of the mold;

[0064] Step 5, setting an industrial mold manufacturing process parameter set of an actual mold corresponding to the mold 3D optimized model, and using the industrial mold manufacturing process parameter set to simulate the mold 3D optimized model to obtain a corresponding simulated industrial product;

[0065] Step 6, performing simulation tests on the simulated industrial product under multi-modal scenario application conditions, and obtaining multiple sets of modal application performance simulation test values ​​of the simulated industrial product; wherein the sets of modal application performance simulation test values ​​correspond one-to-one to the modal scenarios of the simulation tests corresponding to the simulated industrial products, and the sets of modal application performance simulation test values ​​are formed by simulation test values ​​corresponding to several application performance parameters of the physical industrial product corresponding to the simulated industrial product under the actual application conditions of the corresponding modal scenarios;

[0066] Step 7: making a judgment based on the obtained multiple sets of modal performance simulation test values:

[0067] When the multiple modal performance simulation test value sets meet the preset application performance simulation test conditions, go to step 8; otherwise, adjust the process parameters in the current industrial mold manufacturing process parameter set, and set the adjusted industrial mold manufacturing process parameter set as the industrial mold manufacturing process parameter set of the mold corresponding to the mold 3D optimized model, and then go to step 5; wherein the preset application performance simulation test conditions here are: the standard deviation of the performance simulation test value of each modal performance simulation test value set is less than the corresponding preset standard deviation threshold; that is, the total number of scenes in the multimodal scene is Y, and the modal performance simulation test value set marker corresponding to the yth modal scene is S y , the modal performance simulation test value set S y The standard deviation of all modal performance simulation test values ​​in is marked as The preset standard deviation threshold is marked as The preset application performance simulation test conditions are as follows: Needs to be satisfied 1≤y≤Y;

[0068] Step 8: Use the mold design parameter set corresponding to the mold 3D optimized model as the optimal design parameter set of the industrial mold corresponding to the industrial product, and use the current industrial mold manufacturing process parameter set as the optimal manufacturing process parameter set of the industrial mold corresponding to the industrial product.

[0069] Specifically, in step 2 of this embodiment, the process of constructing the mold 3D initial model includes the following steps a1 to a17:

[0070] Step a1, selecting a qualified industrial product as a calibration product, and obtaining images of the calibration product at multiple viewing angles to form a multi-view image set of the calibration product; wherein, in this embodiment, the multiple viewing angles here include a front view, a rear view, a left view, a right view, a top view, and a bottom view of the calibration product;

[0071] Step a2, extracting a feature point information sequence representing the three-dimensional structural features of the calibration product according to the formed multi-view image set; wherein the feature point information sequence is formed by the extracted multiple feature point information, and the feature point information includes the position coordinates of the feature points; for example, the feature point information sequence extracted here to calibrate the three-dimensional structural features of the product includes at least one point on the front face of the milling tool, at least one point on the main rear face of the tool, at least one point on the secondary rear face of the tool, at least one point on the main cutting edge of the tool, and at least one point on the secondary cutting edge of the tool;

[0072] Step a3, dividing all feature points in the feature point information sequence into slices to obtain a plurality of slice areas and a feature point information sequence corresponding to each slice area; wherein all feature points on the same surface are divided into one slice;

[0073] Step a4, obtain the point cloud data set of the calibration product, and calculate the point cloud sparsity of the point cloud data corresponding to each slice area on the calibration product; wherein the point cloud sparsity here is characterized by the point cloud density; the corresponding implementation program (part) is as follows:

[0074] clear all;

[0075] close all;

[0076] clc;

[0077] pc = pcread('rabbit.pcd');

[0078] pc=pcdownsample(pc,'random',0.1); %Reduce the amount of data

[0079] pc_point = pc.Location'; %Get point cloud data

[0080] kdtree = vl_kdtreebuild (pc_point); % Use vlfeat to build kdtree

[0081] dissum = 0;

[0082] for i=1:length(pc_point)

[0083] p_cur = pc_point(:,i);

[0084] [index, distance] = vl_kdtreequery(kdtree, pc_point, p_cur, 'NumNeighbors', 2); % Find the nearest non-self point to the current point

[0085] dissum = dissum + sqrt (distance (2)); % distance sum

[0086] end

[0087] avg=dissum / length(pc_point);

[0088] Step a5, extracting the point cloud sparsity of the same slice area and all the corresponding feature points in a traversal manner as the modeling contour parameters of the same slice area, and the modeling contour parameters extracted from all slice areas together form a modeling contour parameter set that characterizes the overall contour of the calibration product;

[0089] Step a6, constructing a mold 3D reference model according to the obtained modeling profile parameter set;

[0090] Step a7, performing contour optimization processing on the mold 3D reference model to obtain the final mold 3D initial model.

[0091] In the step a7 of this embodiment, the process of performing contour optimization processing on the constructed mold 3D reference model includes the following steps a71 to a76:

[0092] Step a71, selecting adjacent shard area pairs and the boundary lines of each shard area pair in a traversal manner in all shard areas to form a boundary line set;

[0093] Step a72, respectively obtaining all feature points corresponding to each boundary line in the boundary line set, obtaining the attribution relationship between the feature points and the corresponding boundary lines, and forming a boundary line-feature point attribution relationship list from all the attribution relationships;

[0094] Step a73, respectively calculating the slope value of each boundary line at each characteristic point, and respectively forming a slope value sequence corresponding to each boundary line; wherein the boundary line and the slope value sequence are in a one-to-one correspondence, and the slope values ​​in the slope value sequence are arranged according to the front and back positions on the boundary line; of course, before calculating the slope value of the boundary line at its characteristic point, it is necessary to obtain the curve equation of the boundary line in advance;

[0095] Step a74, based on the obtained slope value sequences, a curve curvature determination process is performed on the boundary line corresponding to each slope value sequence:

[0096] When the variance of any three adjacent slope values ​​in the slope value sequence is greater than the preset variance threshold, the process proceeds to step a75; otherwise, the boundary line corresponding to the slope value sequence is not processed;

[0097] For example, in this embodiment, the total number of boundary lines is marked as N, and the nth boundary line (curve) is marked as L.n , 1≤n≤N, curve L n The curve equation is marked as Curve L n The slope value sequence is marked as K, K = {k1, k2, ..., k m ,…,k M}, 1≤m≤M; k1 is the boundary line L n The slope at its characteristic point D1, k2 is the boundary line L n The slope at its characteristic point D2, k m The boundary line L n At its characteristic point D m The slope at k M The boundary line L n At its characteristic point D M The slope at Indicates the curve L n At feature point D m The horizontal coordinate at

[0098] In the judgment process, for curve L n , for three consecutive slope values ​​k m , k m+1 and k m+2 , once k m , k m+1 and k m+2 The variance of these three adjacent slope values Greater than the preset variance threshold When the boundary curve L n If there is an obvious bending change at the characteristic points corresponding to the three continuous refractive indices, it is an abnormal situation of the edge curve change, and the boundary curve L needs to be readjusted. n The bending degrees at the three characteristic points are used to optimize the contour of the mold 3D reference model; wherein the variance of the three adjacent slope values ​​is The calculation method is as follows:

[0099]

[0100] Step a75, adjusting the position of the feature point corresponding to the middle slope value among the three adjacent slope values ​​on the boundary curve, and taking the slope value corresponding to the feature point after the position adjustment as the updated slope value;

[0101] Step a76, allowing the updated slope value to replace the original slope value of the feature point in the slope value sequence to form a new slope value sequence, and then going to step a74 again.

[0102] In addition, it should be noted that in the above step 4, the process of optimizing the mold 3D initial model based on the actual value set of the product design parameters and the actual value set of the product performance parameters to obtain the optimized mold 3D model includes the following steps 41 to 46:

[0103] Step 41, dividing the product design parameter actual value set into a product design parameter training set and a product design parameter verification set;

[0104] Step 42, according to the divided product design parameter training set and product design parameter verification set, the product performance parameter actual value set is correspondingly divided into a product performance parameter training set and a product performance parameter verification set;

[0105] Step 43, inputting the product design parameter training set into the mold model performance model, and using the output of the mold model performance model as the corresponding product performance parameter prediction value set; wherein the actual values ​​of the product design parameters in the product design parameter training set correspond one-to-one to the predicted values ​​of the product performance parameters in the product performance parameter prediction value set;

[0106] Step 44, calculating the output performance loss ratio of the mold model performance model according to the product performance parameter verification set and the obtained product performance parameter prediction value set; wherein the output performance loss ratio of the mold model performance model is calculated as follows:

[0107]

[0108] in, is the output performance loss ratio of the mold model performance model, Q u is the uth product performance parameter value in the product performance parameter verification set. The product performance parameter value Q u One-to-one correspondence with the u-th product design parameter in the product design parameter training set, Q' u is the uth product performance parameter prediction value in the product performance parameter prediction value set, and the product performance parameter prediction value Q' u One-to-one correspondence with the u-th product design parameter in the product design parameter training set;

[0109] Step 45, making a judgment based on the output performance loss ratio of the obtained mold model performance model:

[0110] When the output performance loss ratio is less than the preset loss ratio threshold, the current mold 3D initial model is used as the mold 3D optimized model; otherwise, the process proceeds to step 46; for example, the preset loss ratio threshold here is Set to

[0111] Step 46, reselect another qualified industrial product as the calibration product, and proceed to step a1.

[0112] According to actual needs, the industrial mold intelligent manufacturing method of this embodiment also includes: pre-forming an industrial mold state prediction model to use the industrial mold state prediction model to predict the qualified state of the industrial mold to be manufactured. Specifically, the formation process of the industrial mold state prediction model here includes forming a mold qualified state prediction model and a mold unqualified state prediction model; wherein:

[0113] The process of forming the mold qualification prediction model includes the following steps:

[0114] Step b1, pre-acquire a set of equipment state parameters of the manufacturing equipment state corresponding to when the industrial mold production line manufactures the industrial mold; wherein the set of equipment state parameters includes a set of equipment state parameters in a qualified mold state corresponding to when the mold production line manufactures qualified molds and a set of equipment state parameters in an unqualified mold state corresponding to when the mold production line manufactures unqualified molds;

[0115] Step b2, digitizing the data in the device state parameter set to obtain a digitized device state parameter set that can be represented by numerical values;

[0116] Step b3, using the numerical device state parameter set as the numerical data screening standard, screening the device state parameter set in the mold qualified state and the device state parameter set in the mold unqualified state respectively, and obtaining the numerical data set of the device state in the mold qualified state and the numerical data set of the device state in the mold unqualified state respectively, and combining the numerical data set of the device state in the mold qualified state as the first numerical data set to be screened and combining the numerical data set of the device state in the mold unqualified state as the second numerical data set to be screened;

[0117] Step b4, with the goal of screening the mold manufacturing stage state data, respectively perform data screening processing on the first to-be-screened numerical data set and the second to-be-screened numerical data set to obtain the mold manufacturing stage state numerical data set corresponding to the mold qualified state and the mold manufacturing stage state numerical data set corresponding to the mold unqualified state;

[0118] Step b5, normalizing the data in the equipment state parameter set under the qualified mold state and the equipment state parameter set under the unqualified mold state, respectively, to obtain a first normalized data set and a second normalized data set; for example:

[0119] For the device state parameter set in the mold qualified state, the device state parameter set in the mold qualified state is marked as X, X = {X i}, X iis the i-th device state parameter value in parameter set X;

[0120]

[0121] Among them, x i is the device status parameter value X i The normalized value after normalization, X min is the minimum value of all device status parameter values ​​in parameter set X, X max is the maximum value of all device status parameter values ​​in parameter set X;

[0122] For example, for a set of equipment status parameters in a mold unqualified state, the set of equipment status parameters in a mold unqualified state is marked as Y, where Y={Y j},Y j is the j-th device state parameter value in parameter set Y;

[0123]

[0124] Among them, y j is the device status parameter value Y j The normalized value after normalization, Y min is the minimum value of all device status parameter values ​​in parameter set Y, max is the maximum value of all device status parameter values ​​in parameter set Y;

[0125] Step b6, respectively calculating the correlation between each data in the first normalized data set and the numerical data set of the mold manufacturing stage state in the set, and marking all the calculated correlations as a mold qualified state data correlation sequence;

[0126] Step b7, respectively calculating the correlation between each data in the second normalized data set and the numerical data set of the mold manufacturing stage state in the set, and marking all the calculated correlations as a mold unqualified state data correlation sequence;

[0127] Step b8, selecting all the correlations that are in the mold qualified state data correlation sequence and whose correlation values ​​are greater than a first preset correlation threshold, and using all the equipment state parameters corresponding to the selected all correlations as a set of mold qualified rate influencing factors that affect the mold qualified rate;

[0128] Step b9, training processing is performed based on the mold qualification rate influencing factor set and the mold manufacturing stage state data corresponding to the mold being in a qualified state, to obtain a mold manufacturing stage state data prediction model corresponding to the mold product being predicted;

[0129] Step a10, performing training processing based on the mold failure rate influencing factor set and the mold manufacturing stage state data corresponding to the mold being in an unqualified state, to obtain a mold manufacturing stage state data prediction model corresponding to the prediction of manufacturing unqualified molds;

[0130] Step b11, the obtained mold manufacturing stage status data prediction model corresponding to when the manufacturing plant manufactures qualified molds and the mold manufacturing stage status data prediction model corresponding to when the manufacturing plant manufactures unqualified molds are formed together into the industrial mold state prediction model.

[0131] This embodiment also provides an industrial mold intelligent manufacturing system to implement the industrial mold intelligent manufacturing method. Figure 2 As shown, the industrial mold intelligent manufacturing system includes:

[0132] The parameter acquisition module 1 is configured to respectively acquire a set of product design parameters required for designing an industrial product structure and a set of product performance parameters characterizing the performance of the industrial product, and to acquire a set of manufacturing process parameters, a set of actual values ​​of product design parameters, and a corresponding set of actual values ​​of product performance parameters when actually producing a number of the same industrial products; wherein the set of actual values ​​of product design parameters includes a number of actual value subsets of product design parameters, the set of actual values ​​of product performance parameters includes a number of actual value subsets of product performance parameters, and the actual value subsets of product design parameters when actually producing each industrial product correspond one-to-one to the actual value subsets of product performance parameters of the industrial product produced;

[0133] The model building module 2 is configured to build a mold 3D initial model of the mold required for manufacturing the industrial product and a mold model performance model corresponding to the performance of the mold 3D initial model; wherein the mold 3D initial model corresponds to the industrial product one-to-one, the mold model performance model has a corresponding mold design parameter set, the input parameters of the mold model performance model are all design parameters in the product design parameter set, and the output parameters of the mold model performance model are all performance parameters in the product performance parameter set;

[0134] The simulation module 3 is configured to use the industrial mold manufacturing process parameter set to simulate the mold 3D optimized model to obtain a corresponding simulated industrial product; perform simulation tests on the simulated industrial product under multi-modal scenario application conditions to obtain multiple modal application performance simulation test value sets of the simulated industrial product;

[0135] The judgment processing module 4 is configured to make the judgment processing according to the obtained multiple modal performance simulation test value sets.

[0136] This embodiment also provides a readable storage medium. Specifically, the readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned industrial mold intelligent manufacturing method is implemented.

[0137] Although the preferred embodiments of the present invention are described in detail above, it should be clearly understood that various modifications and variations are possible for those skilled in the art. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An industrial mold intelligent manufacturing method, characterized in that: The steps include: Step 1, respectively obtaining in advance a set of product design parameters required for designing an industrial product structure and a set of product performance parameters characterizing the performance of the industrial product; Step 2, pre-constructing a mold 3D initial model of a mold required for manufacturing an industrial product and a mold model performance model corresponding to the performance of the mold 3D initial model; wherein the mold 3D initial model corresponds to the industrial product one-to-one, the mold model performance model has a mold design parameter set corresponding to itself, the input parameters of the mold model performance model are all design parameters in the product design parameter set, and the output parameters of the mold model performance model are all performance parameters in the product performance parameter set; Step 3, obtaining in advance a set of manufacturing process parameters, a set of actual values ​​of product design parameters, and a set of corresponding actual values ​​of product performance parameters when actually producing a number of the same industrial products; wherein the set of actual values ​​of product design parameters includes a number of actual value subsets of product design parameters, and the set of actual values ​​of product performance parameters includes a number of actual value subsets of product performance parameters, and the actual value subsets of product design parameters when actually producing each industrial product correspond one-to-one to the actual value subsets of product performance parameters of the industrial product produced; Step 4, optimizing the initial 3D model of the mold based on the actual value set of the product design parameters and the actual value set of the product performance parameters to obtain the optimized 3D model of the mold; Step 5, setting an industrial mold manufacturing process parameter set of an actual mold corresponding to the mold 3D optimized model, and using the industrial mold manufacturing process parameter set to simulate the mold 3D optimized model to obtain a corresponding simulated industrial product; Step 6, performing simulation tests on the simulated industrial product under multi-modal scenario application conditions, and obtaining multiple sets of modal application performance simulation test values ​​of the simulated industrial product; wherein the sets of modal application performance simulation test values ​​correspond one-to-one to the modal scenarios of the simulation tests corresponding to the simulated industrial products, and the sets of modal application performance simulation test values ​​are formed by simulation test values ​​corresponding to several application performance parameters of the physical industrial product corresponding to the simulated industrial product under the actual application conditions of the corresponding modal scenarios; Step 7: making a judgment based on the obtained multiple sets of modal performance simulation test values: When the multiple modal performance simulation test value sets meet the preset application performance simulation test conditions, proceed to step 8; otherwise, adjust the process parameters in the current industrial mold manufacturing process parameter set, and set the adjusted industrial mold manufacturing process parameter set as the industrial mold manufacturing process parameter set of the mold corresponding to the mold after the 3D optimization model is manufactured, and then proceed to step 5; Step 8: Use the mold design parameter set corresponding to the mold 3D optimized model as the optimal design parameter set of the industrial mold corresponding to the industrial product, and use the current industrial mold manufacturing process parameter set as the optimal manufacturing process parameter set of the industrial mold corresponding to the industrial product.

2. The industrial mold intelligent manufacturing method according to claim 1 is characterized in that: The industrial product is a milling tool, and the product design parameter set includes the number of tool teeth, tool rake angle, tool clearance angle, tool rotation angle, blade height and tool thickness of the milling tool; the product performance parameter set includes the cutting speed feed rate, cutting depth and cutting angle of the milling tool.

3. The industrial mold intelligent manufacturing method according to claim 2 is characterized in that: The construction process of the mold 3D initial model includes the following steps: Step a1, selecting a qualified industrial product as a calibration product, and obtaining images of the calibration product at multiple viewing angles to form a multi-view image set of the calibration product; Step a2, extracting a feature point information sequence representing the three-dimensional structural features of the calibrated product according to the formed multi-view image set; wherein the feature point information sequence is formed by extracting a plurality of feature point information, and the feature point information includes the position coordinates of the feature points; Step a3, dividing all feature points in the feature point information sequence into slices to obtain a plurality of slice areas and a feature point information sequence corresponding to each slice area; wherein all feature points on the same surface are divided into one slice; Step a4, obtaining a point cloud data set of the calibration product, and calculating the point cloud sparsity of the point cloud data corresponding to each slice area on the calibration product; Step a5, extracting the point cloud sparsity of the same slice area and all the corresponding feature points in a traversal manner as the modeling contour parameters of the same slice area, and the modeling contour parameters extracted from all slice areas together form a modeling contour parameter set that characterizes the overall contour of the calibration product; Step a6, constructing a mold 3D reference model according to the obtained modeling profile parameter set; Step a7, performing contour optimization processing on the mold 3D reference model to obtain the final mold 3D initial model.

4. The industrial mold intelligent manufacturing method according to claim 3 is characterized in that: In step a7, the process of performing contour optimization processing on the constructed mold 3D reference model includes the following steps: Step a71, selecting adjacent shard area pairs and the boundary lines of each shard area pair in a traversal manner in all shard areas to form a boundary line set; Step a72, respectively obtaining all feature points corresponding to each boundary line in the boundary line set, obtaining the attribution relationship between the feature points and the corresponding boundary lines, and forming a boundary line-feature point attribution relationship list from all the attribution relationships; Step a73, respectively calculating the slope value of each boundary line at each characteristic point thereof, and respectively forming a slope value sequence corresponding to each boundary line; wherein the boundary line and the slope value sequence are in a one-to-one correspondence, and the slope values ​​in the slope value sequence are arranged according to the front and rear positions on the boundary line; Step a74, based on the obtained slope value sequences, a curve curvature determination process is performed on the boundary line corresponding to each slope value sequence: When the variance of any three adjacent slope values ​​in the slope value sequence is greater than the preset variance threshold, the process proceeds to step a75; otherwise, the boundary line corresponding to the slope value sequence is not processed; Step a75, adjusting the position of the feature point corresponding to the middle slope value among the three adjacent slope values ​​on the boundary curve, and taking the slope value corresponding to the feature point after the position adjustment as the updated slope value; Step a76, allowing the updated slope value to replace the original slope value of the feature point in the slope value sequence to form a new slope value sequence, and then going to step a74 again.

5. The industrial mold intelligent manufacturing method according to claim 4 is characterized in that: In step 4, the process of optimizing the initial 3D model of the mold based on the actual value set of the product design parameters and the actual value set of the product performance parameters to obtain the optimized 3D model of the mold includes the following steps: Step 41, dividing the product design parameter actual value set into a product design parameter training set and a product design parameter verification set; Step 42, according to the divided product design parameter training set and product design parameter verification set, the product performance parameter actual value set is correspondingly divided into a product performance parameter training set and a product performance parameter verification set; Step 43, inputting the product design parameter training set into the mold model performance model, and using the output of the mold model performance model as the corresponding product performance parameter prediction value set; wherein the actual values ​​of the product design parameters in the product design parameter training set correspond one-to-one to the predicted values ​​of the product performance parameters in the product performance parameter prediction value set; Step 44, calculating the output performance loss ratio of the mold model performance model according to the product performance parameter verification set and the obtained product performance parameter prediction value set; Step 45, making a judgment based on the output performance loss ratio of the obtained mold model performance model: When the output performance loss ratio is less than the preset loss ratio threshold, the current mold 3D initial model is used as the mold 3D optimized model; otherwise, the process proceeds to step 46; Step 46, reselect another qualified industrial product as the calibration product, and proceed to step a1.

6. The industrial mold intelligent manufacturing method according to any one of claims 1 to 5, characterized in that: The mold model performance model is obtained by performing line fitting processing on the product design parameter set and the product performance parameter set; wherein: the mold model performance model is constructed as follows: t i =γ(θ);1≤i≤M;θ={θ1,θ2,…,θ N }; Among them, the industrial product corresponding to the mold 3D initial model has M performances, τ i is the i-th performance of the industrial product corresponding to the mold 3D initial model, θ is the product design parameter set of the mold model performance model, θ1,θ2,…,θ N are N product design parameters in the product design parameter set.

7. An industrial mold intelligent manufacturing system, which implements the industrial mold intelligent manufacturing method according to claim 1, characterized in that: The industrial mold intelligent manufacturing system includes: The parameter acquisition module is configured to respectively acquire a set of product design parameters required for designing an industrial product structure and a set of product performance parameters characterizing the performance of the industrial product, and to acquire a set of manufacturing process parameters, a set of actual values ​​of product design parameters, and a corresponding set of actual values ​​of product performance parameters when a plurality of the same industrial products are actually produced; wherein the set of actual values ​​of product design parameters includes a plurality of actual value subsets of product design parameters, the set of actual value of product performance parameters includes a plurality of actual value subsets of product performance parameters, and the actual value subsets of product design parameters when each industrial product is actually produced correspond one to one to the actual value subsets of product performance parameters of the industrial product produced; A model building module is configured to build a mold 3D initial model of a mold required for manufacturing an industrial product and a mold model performance model corresponding to the performance of the mold 3D initial model; wherein the mold 3D initial model corresponds to the industrial product one-to-one, the mold model performance model has a corresponding mold design parameter set, the input parameters of the mold model performance model are all design parameters in the product design parameter set, and the output parameters of the mold model performance model are all performance parameters in the product performance parameter set; The simulation module is configured to use the industrial mold manufacturing process parameter set to simulate the manufacturing of the mold 3D optimized model to obtain a corresponding simulated industrial product; perform simulation tests on the simulated industrial product under multi-modal scenario application conditions to obtain multiple modal application performance simulation test value sets of the simulated industrial product; The judgment processing module is configured to make the judgment processing according to the obtained multiple modal performance simulation test value sets.

8. A readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the industrial mold intelligent manufacturing method described in any one of claims 1 to 6 is implemented.

Images