A mold production method for cold stamping 6G antenna parts

By collecting and analyzing the surface roughness and frequency band layout parameters of the cavity, combining finite element analysis and machine learning algorithms, a mold wear prediction model is constructed and the mold status is monitored in real time, which solves the problems of intensified mold wear and reduced service life, and realizes mold life Maximizes balance and optimizes antenna performance.

CN118839580BActive Publication Date: 2024-11-29JIANGSU HUACAN TELECOMM
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202411322415.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-23
Publication Date
2024-11-29
Estimated Expiration
2044-09-23

AI Technical Summary

Technical Problem

There are errors and residual stresses in the layout of the cavity surface roughness band of the mold for cold stamping 6G antenna components, resulting in increased mold wear and reduced service life, and lack of effective online monitoring and feedback control mechanisms.

Method used

The surface roughness and frequency band layout parameters of the cavity are collected through the surface morphology measuring instrument and spectrum analyzer, and the data mapping relationship is established. The residual stress distribution is simulated by finite element analysis, and the mold wear prediction model is constructed based on the machine learning algorithm, and the mold status is monitored in real time. Automatically adjust the frequency band layout parameters, establish a multi-scale correlation model, and optimize mold design and maintenance strategies.

Benefits of technology

It significantly improves the efficiency of mold design, manufacturing, use and maintenance, extends the mold life, ensures the stability of antenna performance, and realizes the quality control and intelligent optimization of the entire life cycle of the mold.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118839580B_ABST
    Figure CN118839580B_ABST
Patent Text Reader

Abstract

The present application provides a method for producing a mold for cold-stamping 6G antenna components, including: obtaining cavity surface roughness data and frequency band layout parameters of the mold for cold-stamping 6G antenna components, collecting relevant data through a surface morphology measuring instrument and a spectrum analyzer, and establishing a data mapping relationship between the cavity surface roughness and the frequency band layout, which is used for subsequent analysis of mold performance under the 6G frequency band; based on the acquired cavity surface roughness data and frequency band layout parameters, a finite element analysis method is used to simulate the distribution characteristics of residual stress on the cavity surface under different frequency band layouts, obtain a residual stress distribution cloud map, and determine the stress concentration area. The residual stress distribution cloud map intuitively displays the stress distribution and is used to identify potential high-risk areas for mold wear.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of 6G antenna component mold production, and in particular to a mold production method for cold stamping 6G antenna components. Background Art

[0002] There are certain errors and residual stresses in the frequency band layout of the cavity surface roughness of the mold for cold stamping 6G antenna parts, which have an important impact on mold wear. In actual production, the threshold of the cavity surface roughness and the residual stress distribution characteristics under different frequency band layouts are difficult to accurately control, resulting in increased mold wear and reduced service life. At the same time, improper selection of the type of cavity surface roughness frequency band layout will also introduce additional stress concentration and accelerate mold failure. In addition, the cavity surface roughness frequency band layout lacks an effective online monitoring and feedback control mechanism, making it difficult to adjust the frequency band layout parameters in real time, resulting in large fluctuations in cavity surface quality and difficult to predict mold wear status. The correlation between frequency band layout defects and cavity surface roughness is not clear, and it is difficult to establish a quantitative model for frequency band layout design and surface quality control, resulting in difficulty in mold design optimization. It is urgent to carry out defect mechanism research on cavity surface roughness frequency band layout, reveal the evolution law of frequency band layout defects and cavity surface roughness, establish a multi-scale correlation model of frequency band layout design, surface quality prediction and mold wear evaluation, and realize mold quality control and intelligent optimization throughout the life cycle. Summary of the invention

[0003] The present invention provides a method for producing a mold for cold stamping 6G antenna parts, which mainly includes:

[0004] Obtain the cavity surface roughness data and frequency band layout parameters of the mold for cold stamping 6G antenna parts, collect relevant data through a surface topography meter and a spectrum analyzer, and establish a data mapping relationship between the cavity surface roughness and the frequency band layout. This mapping relationship is used for subsequent analysis of mold performance under the 6G frequency band;

[0005] According to the obtained cavity surface roughness data and frequency band layout parameters, the finite element analysis method is used to simulate the distribution characteristics of residual stress on the cavity surface under different frequency band layouts, and the residual stress distribution cloud map is obtained to determine the stress concentration area. The residual stress distribution cloud map intuitively displays the stress distribution and is used to identify potential high-risk areas for mold wear;

[0006] Based on the cavity surface roughness data, frequency band layout parameters and residual stress distribution cloud map, a prediction model between frequency band layout parameters, cavity surface roughness and mold wear is trained and established through machine learning algorithms such as support vector machine or random forest. According to the model prediction results, the wear state and service life of the mold under the current frequency band layout and surface roughness are judged;

[0007] During the cold stamping process, an online monitoring system is used to collect the cavity surface roughness and frequency band layout parameters in real time. The mold wear status is dynamically evaluated by comparison and analysis with the prediction model. When the wear status exceeds the preset threshold, the feedback control mechanism is triggered. According to the output results of the prediction model, the frequency band layout parameters are automatically adjusted, such as adjusting the size or spacing of the antenna radiation unit to slow down the mold wear.

[0008] Aiming at the correlation between frequency band layout defects and cavity surface roughness, combined with online monitoring data and prediction model results, data mining technology such as association rule analysis is used to explore the evolution law of frequency band layout defects and surface roughness, and a multi-scale correlation model for frequency band layout design, surface quality prediction and mold wear evaluation is established. This model realizes cross-scale comprehensive analysis by integrating micro surface roughness, meso frequency band layout and macro mold wear data;

[0009] Integrate online monitoring data, prediction model results, and multi-scale correlation model analysis results into the mold life cycle management system. Through big data analysis and visualization technology, realize full-cycle monitoring and early warning of mold quality. Use multi-scale correlation models to optimize mold maintenance strategies, predict potential failures, and provide data support for mold design optimization.

[0010] Based on the integrated mold life cycle quality data and multi-scale correlation model, intelligent optimization algorithms such as genetic algorithms are used to optimize the frequency band layout design parameters. During the optimization process, the cavity surface quality, mold wear status and antenna performance requirements are considered. By adjusting the geometric parameters and arrangement of the antenna radiation unit, a balance is achieved between maximizing the mold life and optimizing the antenna performance, thereby realizing the intelligence and efficiency of mold design, manufacturing, use and maintenance.

[0011] Beneficial effects of the present invention:

[0012] In the present invention, the surface roughness of the cavity and the frequency band layout parameters are collected to establish a data mapping relationship, and the residual stress distribution is simulated by finite element analysis. Based on these data, a machine learning algorithm is used to construct a mold wear prediction model. During the cold stamping process, the mold state is monitored in real time. When the wear exceeds the threshold, the frequency band layout parameters are automatically adjusted to slow down the wear. At the same time, a multi-scale correlation model is established through data mining technology to achieve cross-scale comprehensive analysis. Finally, an intelligent optimization algorithm is used to optimize the frequency band layout design parameters based on the consideration of the cavity surface quality, mold wear state and antenna performance requirements, so as to achieve a balance between maximizing the mold life and optimizing the antenna performance. Through multi-dimensional data analysis and intelligent optimization, the present invention significantly improves the efficiency of mold design, manufacturing, use and maintenance, prolongs the mold life, and ensures the stability of antenna performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 Schematic diagram of a mold production method for cold stamping 6G antenna parts of the present invention Figure 1 .

[0014] Figure 2 Schematic diagram of a mold production method for cold stamping 6G antenna parts of the present invention Figure 2 . DETAILED DESCRIPTION

[0015] The following will describe the technical solutions in the embodiments of the present invention in detail in conjunction with the accompanying drawings in the embodiments of the present invention. The described embodiments are only part of the embodiments of the present invention.

[0016] like Figure 1-2 In this embodiment, a mold production method for cold stamping 6G antenna parts may specifically include:

[0017] Step S101, obtain the cavity surface roughness data and frequency band layout parameters of the mold for cold stamping 6G antenna parts, collect relevant data through a surface topography measuring instrument and a spectrum analyzer, and establish a data mapping relationship between the cavity surface roughness and the frequency band layout. This mapping relationship is used for subsequent analysis of mold performance under the 6G frequency band.

[0018] According to the cavity surface roughness data and frequency band layout parameters of the mold for cold stamping 6G antenna parts, the support vector machine algorithm was used to establish a nonlinear regression model of cavity surface roughness and frequency band layout, and the data mapping relationship between cavity surface roughness and frequency band layout was obtained.

[0019] According to the established data mapping relationship between the cavity surface roughness and the frequency band layout, the decision tree algorithm is used to classify and predict the mold performance under the 6G frequency band. If the prediction result meets the preset performance threshold, it is judged that the mold can be used to produce 6G antenna components, otherwise it is judged that the mold needs surface treatment. For molds that need surface treatment, the cavity surface is treated by chemical corrosion or mechanical polishing to obtain the processed cavity surface roughness data, which is input into the established data mapping model between the cavity surface roughness and the frequency band layout to predict its corresponding frequency band layout parameters.

[0020] Based on the predicted frequency band layout parameters, a genetic algorithm is used to optimize the structural dimensions of antenna components to obtain the antenna component structural parameters that meet the 6G frequency band requirements;

[0021] According to the optimized structural parameters of the antenna components, the corresponding cold stamping mold is processed by the CNC machining center, and the surface roughness of the mold cavity after processing is tested by a three-coordinate measuring machine. The test data is input into the data mapping model of the cavity surface roughness and the frequency band layout for verification; if the verification results show that the mold cavity surface roughness and the frequency band layout parameters meet the requirements, the mold is used to perform cold stamping of the 6G antenna components to obtain finished components; if the verification results show that the mold cavity surface roughness and the frequency band layout parameters do not meet the requirements, the cavity surface is returned and reprocessed until the requirements are met.

[0022] Specifically, we first collected the cavity surface roughness data and frequency band layout parameters of the mold for cold stamping 6G antenna parts, where the cavity surface roughness data included Ra and Rz parameters, and the frequency band layout parameters included center frequency and bandwidth. We used the support vector machine algorithm to train the collected data, established a nonlinear regression model of cavity surface roughness and frequency band layout, and mapped the cavity surface roughness to the frequency band layout space through the Gaussian kernel function to obtain the corresponding relationship between the two.

[0023] On this basis, the decision tree algorithm is used to classify and predict the mold performance. The performance thresholds are set as the frequency band layout deviation is less than 5% and the cavity surface roughness Ra is less than 8μm. If the prediction results meet the threshold requirements, the mold is determined to be able to be used to produce antenna components, otherwise the cavity surface needs to be processed.

[0024] For the molds that need to be processed, chemical etching or mechanical polishing is used to reduce the cavity surface roughness Ra to below 4μm, and then the processed roughness data is input into the established mapping model to predict its corresponding frequency band layout parameters.

[0025] Based on the prediction results, a genetic algorithm is used to optimize the structural dimensions of antenna components. For example, the vibrator height and feeding point position are used as optimization variables, and the frequency band coverage and radiation pattern performance are used as optimization targets. After 50 generations of evolution, the optimal structural parameters that meet the requirements of the 6G frequency band are obtained.

[0026] Using the optimized structural parameters, the corresponding cold stamping die is processed by a CNC machining center, and the machining accuracy is controlled within ±02mm. Then, a three-coordinate measuring machine is used to detect the surface roughness of the mold cavity after processing. The measurement step is 5mm. The detection data is input into the mapping model for verification. If the cavity surface roughness Ra is less than 4μm and the frequency band layout deviation is within 5%, the mold is used for cold stamping with a stamping pressure of 800kN and a holding time of 10s to obtain a finished antenna component with dimensional accuracy that meets the requirements. If the verification result does not meet the requirements, the cavity surface is reprocessed. After 3 to 5 iterations, a cold stamping die that meets the performance requirements is finally obtained to ensure the production quality of antenna components.

[0027] Step S102, based on the obtained cavity surface roughness data and frequency band layout parameters, the finite element analysis method is used to simulate the distribution characteristics of residual stress on the cavity surface under different frequency band layouts, obtain the residual stress distribution cloud map, and determine the stress concentration area. The residual stress distribution cloud map intuitively displays the stress distribution and is used to identify potential high-risk areas for mold wear.

[0028] Obtain the roughness data of the cavity surface, establish a three-dimensional model of the cavity surface, and map the roughness data to the model surface. According to the frequency band layout parameters, divide the three-dimensional model of the cavity surface into frequency bands and generate finite element models under different frequency band layouts. For each finite element model under the frequency band layout, set the material properties, boundary conditions and load conditions, and perform finite element analysis and calculation.

[0029] The finite element analysis results are used to obtain the distribution cloud map of residual stress on the cavity surface under different frequency band layouts, which intuitively displays the stress distribution. Through the stress distribution cloud map, the stress concentration area is identified to determine whether the stress concentration level exceeds the allowable stress of the material. If the stress concentration level exceeds the allowable stress of the material, the area is identified as a potential high-risk area for mold wear. Based on the analysis results of high-risk areas for mold wear under different frequency band layouts, the optimal frequency band layout scheme is determined to guide the optimization of the frequency band layout on the cavity surface.

[0030] Specifically, the cavity surface is first scanned using a three-coordinate measuring machine to obtain surface roughness data. The scanning interval is 1 mm, and a set of point cloud data containing X, Y, Z coordinates and corresponding roughness Ra values ​​is obtained.

[0031] Then, the point cloud data was imported using the reverse engineering software Imageware, and a three-dimensional CAD model of the cavity surface was generated through surface fitting. The roughness Ra value was then mapped to the CAD model surface to generate a roughness distribution map of the cavity surface. According to the structural characteristics of the product, frequency bands were divided on the three-dimensional model of the cavity surface, and six frequency band areas were divided, corresponding to different functional areas of the product. The meshing tool in the CAD software was used to mesh each frequency band area and generate finite element models of the six frequency band areas, with a mesh unit size of 1 mm.

[0032] In the finite element analysis software ANSYS, the cavity material is defined as Cr12MoV mold steel, with an elastic modulus of 210 GPa, a Poisson's ratio of 3, and a yield strength of 1000 MPa.

[0033] Apply an injection pressure of 3MPa on the cavity surface, apply an ejection force of 3000N on the bottom surface of the cavity, and set the friction coefficient between the cavity surface and the plastic part to 3.

[0034] Finite element analysis was run to obtain the Von Mises stress distribution cloud map of the six frequency bands. The stress distribution can be intuitively seen through the stress cloud map. The stress concentration area was identified and it was found that the stress was the largest at the center of the cavity. The maximum stress value was 980MPa, which was close to the material yield strength of 1000MPa. This area was determined to be a potential high-risk area for mold wear.

[0035] Based on the analysis results of the six frequency bands, it is found that the stress concentration of frequency band division scheme 2 is the smallest, and the maximum stress value is 850MPa, which is lower than the yield strength of the material. Therefore, frequency band division scheme 2 is determined to be the optimal scheme, which can effectively reduce the risk of mold wear.

[0036] Step S103, based on the cavity surface roughness data, frequency band layout parameters and residual stress distribution cloud map, a prediction model between frequency band layout parameters, cavity surface roughness and mold wear is trained and established through machine learning algorithms such as support vector machines or random forests. According to the model prediction results, the wear state and service life of the mold under the current frequency band layout and surface roughness are judged.

[0037] The cavity surface roughness data, frequency band layout parameters and residual stress distribution cloud map data are obtained as the training data set for the machine learning algorithm. The support vector machine and random forest machine learning algorithms are used for training based on the training data set to establish a prediction model between frequency band layout parameters, cavity surface roughness and mold wear.

[0038] According to the established prediction model, the frequency band layout parameters and surface roughness data of the current cavity are input to predict the wear state of the mold. If the predicted wear state exceeds the preset threshold, it is judged that the mold has reached its service life and needs maintenance or replacement; otherwise, it is judged that the mold can continue to be used. Obtain the cavity residual stress distribution cloud map and extract the residual stress distribution characteristics as a supplementary input of the prediction model to improve the accuracy of mold wear state prediction. By analyzing the prediction results, the key factors affecting mold wear are determined, such as the key frequency bands in the frequency band layout parameters and the critical value of surface roughness, providing a basis for optimizing mold design.

[0039] According to the mold wear status prediction results and key influencing factors analysis, formulate mold maintenance strategies, such as regular inspection of surface roughness, optimization of frequency band layout, extending mold service life and improving production efficiency.

[0040] Specifically, in order to predict the wear state of the mold, it is first necessary to collect the cavity surface roughness data, band layout parameters and residual stress distribution cloud map data. Among them, the surface roughness data can be measured by a roughness meter, the band layout parameters can be extracted from the mold design drawings, and the residual stress distribution cloud map can be obtained by simulation calculation using finite element analysis software.

[0041] After collecting enough data, it is randomly divided into a training set and a test set. The training set is used to train the machine learning model, and the test set is used to evaluate the model's predictive performance. During the training process, the grid search method can be used to optimize the model's hyperparameters, such as the kernel function type of the support vector machine and the number of decision trees of the random forest, to improve the model's predictive accuracy.

[0042] By comparing the prediction results of the test set with the actual wear status, the model's evaluation indicators, such as mean square error and determination coefficient, are calculated to evaluate the model's prediction performance. When the model's prediction performance reaches the expected goal, it can be applied to actual production.

[0043] In the application process, it is necessary to collect the frequency band layout parameters and surface roughness data of the cavity in real time and input them into the prediction model to obtain the predicted value of the wear state of the mold. At the same time, it is also necessary to obtain the residual stress distribution cloud map of the cavity and extract the residual stress distribution characteristics, such as the maximum residual stress and residual stress gradient, as a supplementary input of the prediction model to improve the prediction accuracy.

[0044] According to the prediction results, corresponding mold maintenance strategies can be formulated. For example, when the predicted wear state exceeds the preset threshold, the mold can be maintained or replaced in time to avoid the problem of product quality degradation caused by excessive mold wear. By analyzing the prediction results, key factors affecting mold wear can also be determined, such as the key frequency band in the frequency band layout parameters and the critical value of surface roughness. For example, by comparing the mold wear state under different frequency band layout parameters, it is found that when the energy density of a certain frequency band exceeds 5 J / mm², the wear rate of the mold increases significantly. Therefore, the energy density of this frequency band can be controlled below 5 J / mm² to extend the service life of the mold.

[0045] Step S104: During the cold stamping process, the online monitoring system is used to collect the surface roughness of the cavity and the frequency band layout parameters in real time, and the mold wear state is dynamically evaluated by comparing and analyzing with the prediction model. When the wear state exceeds the preset threshold, the feedback control mechanism is triggered, and the frequency band layout parameters are automatically adjusted according to the output results of the prediction model, such as adjusting the size or spacing of the antenna radiation unit to slow down the mold wear.

[0046] The roughness data and band layout parameters of the mold surface during the cold stamping process are obtained, and the obtained data are input into the pre-established prediction model for comparison and analysis to obtain the evaluation results of the mold wear state. It is judged whether the evaluation result exceeds the preset threshold. If it exceeds the threshold, the feedback control mechanism is triggered. According to the output results of the prediction model, the optimization algorithm is used to determine the band layout parameters that need to be adjusted, such as the size or spacing of the antenna radiation unit. The association model between the band layout parameters and mold wear is established through the machine learning algorithm. Based on the association model, the intelligent optimization algorithm is used to determine the optimal band layout parameter adjustment scheme to minimize mold wear. The optimized band layout parameters are transmitted to the CNC system to control the CNC machine tool to automatically adjust the size and spacing of the antenna radiation unit, change the band layout, and slow down the wear rate of the mold. The influence of the adjusted band layout parameters on the mold wear is monitored in real time. If the wear state still exceeds the threshold, the band layout parameters are optimized again until the mold wear state is lower than the preset threshold. According to different mold materials and cold stamping process parameters, multiple prediction models and association models are established in advance. According to the actual production situation, the optimal model is dynamically switched and selected to improve the accuracy of prediction and control. Big data analysis technology is used to mine and analyze the massive amount of mold wear data and frequency band layout parameter data collected, extract implicit correlation rules and evolution trends, and provide data support for the optimization of prediction models and correlation models.

[0047] Specifically, during the cold stamping process, a high-precision laser displacement sensor can be used to measure the roughness data of the mold surface in real time, and a vector network analyzer can be used to obtain the frequency band layout parameters, including the size and spacing of the antenna radiation unit. The collected roughness data and frequency band layout parameters are input into the prediction model based on the support vector machine (SVM) algorithm, and the mold wear state evaluation results are obtained through comparison and analysis. If the evaluation result exceeds the preset threshold (such as the surface roughness Ra is greater than 8μm), the feedback control mechanism is triggered. According to the output results of the SVM prediction model, the particle swarm optimization (PSO) algorithm is used to determine the frequency band layout parameters that need to be adjusted, such as adjusting the size of the antenna radiation unit from 10mm×10mm to 8mm×8mm, and the spacing from 5mm to 8mm. The association model between the frequency band layout parameters and mold wear is established by the convolutional neural network (CNN) algorithm. Based on the association model, the genetic algorithm (GA) is used to determine the optimal frequency band layout parameter adjustment scheme to minimize mold wear. The optimized frequency band layout parameters are transferred to the CNC system to control the CNC machine tool to automatically adjust the size and spacing of the antenna radiation unit, change the frequency band layout, and slow down the wear rate of the mold. The influence of the adjusted frequency band layout parameters on the mold wear is monitored in real time. If the wear state still exceeds the threshold, the frequency band layout parameters are optimized again until the mold wear state is lower than the preset threshold. For different mold materials (such as high-speed steel, cemented carbide) and cold stamping process parameters (such as stamping speed, pressure), multiple prediction models and association models are established in advance. According to the actual production situation, the optimal model is dynamically switched and selected to improve the accuracy of prediction and control. The Hadoop big data platform is used to mine and analyze the massive mold wear data and frequency band layout parameter data collected, extract the implicit association rules and evolution trends, and provide data support for the optimization of prediction models and association models.

[0048] Step S105, for the correlation between frequency band layout defects and cavity surface roughness, combined with online monitoring data and prediction model results, through data mining technology such as association rule analysis, the evolution law of frequency band layout defects and surface roughness is mined, and a multi-scale correlation model for frequency band layout design, surface quality prediction and mold wear evaluation is established. This model realizes cross-scale comprehensive analysis by integrating micro surface roughness, meso frequency band layout and macro mold wear data.

[0049] Obtain band layout design parameters, cavity surface roughness measurement data, and mold wear monitoring data to construct a multi-scale data set. Preprocess the multi-scale data, including data cleaning, feature extraction, and data integration, to obtain a standardized data representation. Use association rule mining algorithms, such as Apriori or FP-growth, to discover association rules and patterns between band layout defects and surface roughness from standardized data. Based on the mined association rules, construct a prediction model for the evolution of band layout defects and surface roughness, such as a decision tree, support vector machine, or neural network. Apply the prediction model to online monitoring data to predict the evolution trend of cavity surface quality and band layout defects in real time. Comprehensively analyze the cross-scale associations between band layout, surface roughness, and mold wear, establish a multi-scale association model, and quantify the influence mechanism between different factors. Based on the multi-scale association model, optimize the band layout design scheme, formulate a mold maintenance strategy, improve the cavity surface quality, and extend the mold service life.

[0050] Specifically, the frequency band layout design parameters are first obtained through a high-precision three-dimensional scanner, such as the number of frequency bands is 8, the width of each frequency band is 10mm, the cavity surface roughness measurement data is measured by a roughness meter, and the surface roughness Ra value is 8μm. The mold wear monitoring data is collected in real time through an online monitoring system, including mold temperature and pressure parameters, to construct a multi-scale data set. The multi-scale data is preprocessed, and the minimum-maximum normalization method is used to clean and extract features of the data. The frequency band layout parameters, surface roughness and mold wear characteristics are extracted, and standardized data representation is obtained through data integration. The Apriori association rule mining algorithm is used, and the minimum support is set to 05 and the minimum confidence is set to 8. The association rules between frequency band layout defects and surface roughness are found from the standardized data, such as the rule that the larger the frequency band width, the higher the surface roughness. According to the mined association rules, the support vector machine algorithm is used to construct a prediction model for frequency band layout defects and surface roughness evolution. The ratio of training set to test set is 7:3, and the model accuracy reaches 85%. The prediction model is applied to online monitoring data to predict the evolution trend of cavity surface quality and band layout defects in real time. When the surface roughness exceeds 2μm, it is predicted that the band layout may have defects. The cross-scale association between band layout, surface roughness and mold wear is comprehensively analyzed, and a multi-scale association model is established. The gray correlation analysis method is used to calculate the gray correlation between band width, surface roughness and mold temperature, and the influence mechanism between different factors is quantified. The influence of band width on surface roughness is 7, and the influence of surface roughness on mold wear is 6. Based on the multi-scale association model, the genetic algorithm is used to optimize the band layout design scheme. The population size is set to 50, the evolutionary generation is 100, the crossover probability is 8, and the mutation probability is 1. The optimal band layout scheme is obtained, with 6 bands and a width of 12mm for each band. At the same time, a mold maintenance strategy is formulated. When the surface roughness exceeds 0μm, the mold is polished to extend the mold service life by 20%.

[0051] Step S106, integrate the online monitoring data, prediction model results, and multi-scale correlation model analysis results into the mold life cycle management system, and realize full-cycle monitoring and early warning of mold quality through big data analysis and visualization technology. Utilize the multi-scale correlation model to optimize mold maintenance strategies, predict potential failures, and provide data support for mold design optimization.

[0052] The real-time data of the mold production process, including temperature, pressure, and speed parameters, is obtained through the online monitoring system, and the data is transmitted to the full life cycle management system. The real-time monitoring data obtained is input into the prediction model, and the performance prediction results of the mold in the future are obtained through algorithm analysis and calculation to determine whether the mold has potential quality risks. If the prediction results show that there is a quality risk, the early warning mechanism is triggered, and the system automatically generates early warning information, which is displayed through a visual interface to prompt relevant personnel to take timely countermeasures. According to the prediction results and historical data, the relationship between various parameters of the mold and its performance and life is analyzed through the association model to obtain the key factors affecting the quality of the mold, providing a basis for subsequent mold design optimization. Using big data analysis technology, various data in the entire life cycle of the mold are mined and analyzed to find the association rules and patterns between the data, providing data support for optimizing the mold maintenance strategy. The design parameters, material properties, and process conditions of the mold are input into the association model, and the machine learning algorithm is used for training to establish a mold performance prediction model to guide mold design optimization. Based on the historical data and prediction results in the full life cycle management system, a heuristic algorithm is used to optimize the mold maintenance strategy, formulate a scientific maintenance plan, maximize the mold service life, and reduce maintenance costs.

[0053] Specifically, the online monitoring system is used to obtain real-time data during the mold production process. For example, the temperature sensor collects data once per second, the pressure sensor collects data once every 5 seconds, and the speed sensor collects data once every 1 second. The collected data is transmitted to the database of the full life cycle management system via industrial Ethernet. The system uses the support vector machine algorithm to analyze and calculate the real-time monitoring data, and predicts the performance change trend of the mold in the next 8 hours by comparing it with historical data. If the prediction result shows that the mold has a risk of exceeding the dimensional tolerance by more than 5% in the next 2 hours, the early warning mechanism is triggered. The system automatically generates early warning information and displays it in the form of pop-up windows and voice prompts through a visual interface, and pushes the early warning information to the mobile APP of relevant personnel. The system uses an association rule mining algorithm to analyze the correlation between mold material properties, design parameters, process conditions and their performance and life, and finds that the key factors affecting mold quality, such as the hardness of the mold material, are positively correlated with its service life, with a correlation coefficient of 85. The decision tree algorithm is used to establish a mold performance prediction model, and its service life and key quality characteristics are predicted by inputting the design parameters, material properties and process conditions of the mold. Genetic algorithms are used to optimize mold maintenance strategies. By setting mold maintenance cycles and maintenance content parameters and combining mold historical data and prediction results, the optimal mold maintenance plan is obtained, which can increase the average service life of the mold by 20% and reduce the annual maintenance cost by 15%.

[0054] Step S107, based on the integrated mold life cycle quality data and multi-scale association model, an intelligent optimization algorithm such as a genetic algorithm is used to optimize the frequency band layout design parameters. During the optimization process, the cavity surface quality, mold wear state and antenna performance requirements are considered, and the geometric parameters and arrangement of the antenna radiation unit are adjusted to achieve a balance between maximizing the mold life and optimizing the antenna performance, thereby realizing intelligent and efficient mold design, manufacturing, use and maintenance.

[0055] The quality data of the entire life cycle of the mold is obtained, including the cavity surface quality data and the mold wear status data, and a multi-scale correlation model is constructed to establish the correlation between the quality data and the mold life and antenna performance. According to the cavity surface quality data and the mold wear status data, the current use status and remaining life of the mold are judged. If the remaining life of the mold is lower than the preset threshold, the antenna band layout optimization process is triggered. The genetic algorithm intelligent optimization algorithm is adopted, and the geometric parameters and arrangement of the antenna radiation unit are used as optimization variables, and the mold life and antenna performance are used as optimization targets to perform multi-objective optimization solutions. In the optimization process, the cavity surface quality and the mold wear status are associated with the optimization variables through the mold multi-scale correlation model, so that the optimization process can take into account the mold quality factors. Through iterative optimization, a set of optimal antenna layout design parameters that meet the requirements of mold life and antenna performance are obtained, including the geometric parameters and arrangement of the antenna radiation unit. According to the optimized antenna layout design parameters, the mold is intelligently modified and redesigned, and a new mold processing program and process parameters are generated to guide the processing and manufacturing of the mold. During the mold use and maintenance stage, the mold quality data is continuously collected and the mold status is evaluated in real time. When the mold quality drops below the threshold, the optimization process is re-triggered to upgrade the mold and realize intelligent management of the entire life cycle of the mold.

[0056] Specifically, by arranging a sensor array on the mold surface, the mold cavity surface morphology data and wear data are collected in real time, and the wavelet analysis algorithm is used to extract and fuse the collected multi-scale quality data to construct a nonlinear mapping model between quality data and mold life and antenna performance. When the model predicts that the remaining life of the mold is less than 1000 injection cycles, the optimization process is automatically triggered. A multi-objective genetic algorithm is used, with the antenna radiation unit length, width geometric parameters and array arrangement density as optimization variables, and the mold life greater than 5000 injection cycles and antenna gain greater than 15dB as optimization goals. The mold quality characteristic parameters are included in the constraints to search for the optimal antenna design parameter combination. Based on the optimization results, the intelligent CAD / CAE system is used to automatically modify and redesign the mold structure, generate a digital processing program, and guide mold processing. At the same time, the mold quality status is continuously monitored. When the cavity surface roughness exceeds 6um or the mold wear exceeds 2mm, the optimization is restarted to achieve intelligent upgrade of the mold.

[0057] The above only lists some preferred embodiments of the present invention, but the present invention is not limited thereto, and many improvements and changes can be made. As long as the improvements and changes are made on the basis of the basic principles of the present invention, they should be regarded as falling within the protection scope of the present invention.

Claims

1. A method for producing a mold for cold stamping 6G antenna parts, characterized in that: include: Obtain the cavity surface roughness data and frequency band layout parameters of the mold for cold stamping 6G antenna parts, collect relevant data through a surface topography meter and a spectrum analyzer, and establish a data mapping relationship between the cavity surface roughness and the frequency band layout. This mapping relationship is used for subsequent analysis of mold performance under the 6G frequency band; According to the obtained cavity surface roughness data and frequency band layout parameters, the finite element analysis method is used to simulate the distribution characteristics of residual stress on the cavity surface under different frequency band layouts, and the residual stress distribution cloud map is obtained to determine the stress concentration area. The residual stress distribution cloud map intuitively displays the stress distribution and is used to identify potential high-risk areas for mold wear; Based on the cavity surface roughness data, frequency band layout parameters and residual stress distribution cloud map, a prediction model between frequency band layout parameters, cavity surface roughness and mold wear is trained and established through machine learning algorithms. According to the model prediction results, the wear state and service life of the mold under the current frequency band layout and surface roughness are judged; During the cold stamping process, an online monitoring system is used to collect the cavity surface roughness and frequency band layout parameters in real time, which are input into the prediction model for comparison and analysis to dynamically evaluate the mold wear state. When the wear state exceeds the preset threshold, the feedback control mechanism is triggered to automatically adjust the frequency band layout parameters according to the output results of the prediction model to slow down the mold wear. Aiming at the correlation between frequency band layout defects and cavity surface roughness, combined with online monitoring data and prediction model results, data mining technology is used to explore the evolution law of frequency band layout defects and surface roughness, and a multi-scale correlation model for frequency band layout design, surface quality prediction and mold wear evaluation is established. This model realizes cross-scale comprehensive analysis by integrating micro surface roughness, meso frequency band layout and macro mold wear data. Integrate online monitoring data, prediction model results, and multi-scale correlation model analysis results into the mold life cycle management system. Through big data analysis and visualization technology, realize full-cycle monitoring and early warning of mold quality. Use multi-scale correlation models to optimize mold maintenance strategies, predict potential failures, and provide data support for mold design optimization. Based on the integrated mold life cycle quality data and multi-scale association model, a genetic algorithm is used to optimize the frequency band layout design parameters. During the optimization process, the cavity surface quality, mold wear status and antenna performance requirements are considered. By adjusting the geometric parameters and arrangement of the antenna radiation unit, a balance is achieved between maximizing the mold life and optimizing the antenna performance, thereby realizing the intelligence and efficiency of mold design, manufacturing, use and maintenance.

2. The method for producing a mold for cold stamping 6G antenna parts according to claim 1, characterized in that: The method of obtaining the cavity surface roughness data and frequency band layout parameters of the mold for cold stamping 6G antenna components, collecting relevant data through a surface topography measuring instrument and a spectrum analyzer, and establishing a data mapping relationship between the cavity surface roughness and the frequency band layout, which is used for subsequent analysis of the mold performance under the 6G frequency band, includes: According to the cavity surface roughness data and frequency band layout parameters of the mold for cold stamping 6G antenna parts, the support vector machine algorithm was used to establish a nonlinear regression model of cavity surface roughness and frequency band layout, and the data mapping relationship between cavity surface roughness and frequency band layout was obtained. Based on the established data mapping relationship between cavity surface roughness and frequency band layout, a decision tree algorithm is used to classify and predict the mold performance under the 6G frequency band. If the prediction result meets the preset performance threshold, the mold is judged to be able to be used to produce 6G antenna parts. Otherwise, the mold is judged to need surface treatment. For molds that need surface treatment, the cavity surface is treated by chemical etching or mechanical polishing to obtain the surface roughness data of the processed cavity, which is then input into the established data mapping model of cavity surface roughness and frequency band layout to predict the corresponding frequency band layout parameters. Based on the predicted frequency band layout parameters, a genetic algorithm is used to optimize the structural dimensions of antenna components to obtain the antenna component structural parameters that meet the 6G frequency band requirements. According to the optimized antenna component structural parameters, the corresponding cold stamping die is processed by the CNC machining center, and the surface roughness of the processed die cavity is tested by a three-coordinate measuring machine. The test data is input into the data mapping model of the cavity surface roughness and the frequency band layout for verification. If the verification results show that the surface roughness of the mold cavity and the frequency band layout parameters meet the requirements, the mold is used to perform cold stamping of the 6G antenna components to obtain finished components; If the verification results show that the mold cavity surface roughness and frequency band layout parameters do not meet the requirements, then return to reprocess the cavity surface until it meets the requirements.

3. The method for producing a mold for cold stamping 6G antenna parts according to claim 1, characterized in that: According to the obtained cavity surface roughness data and frequency band layout parameters, the finite element analysis method is used to simulate the distribution characteristics of the residual stress on the cavity surface under different frequency band layouts, obtain the residual stress distribution cloud map, determine the stress concentration area, and the residual stress distribution cloud map intuitively displays the stress distribution, which is used to identify potential high-risk areas for mold wear, including: Obtain the cavity surface roughness data, establish the cavity surface three-dimensional model, and map the roughness data to the model surface; According to the frequency band layout parameters, the three-dimensional model of the cavity surface is divided into frequency bands to generate finite element models under different frequency band layouts; For the finite element model under each frequency band layout, set the material properties, boundary conditions and load conditions, and perform finite element analysis and calculation; The finite element analysis results are used to obtain the distribution cloud diagram of the residual stress on the cavity surface under different frequency band layouts, which can intuitively display the stress distribution; Through the stress distribution cloud map, the stress concentration area is identified to determine whether the stress concentration exceeds the allowable stress of the material; If the stress concentration exceeds the allowable stress of the material, the area is identified as a potential high-risk area for mold wear; Based on the analysis results of high-risk areas of mold wear under different frequency band layouts, the optimal frequency band layout scheme is determined to guide the optimization of the frequency band layout on the cavity surface.

4. The method for producing a mold for cold stamping 6G antenna parts according to claim 1, characterized in that: The method is based on the cavity surface roughness data, the frequency band layout parameters and the residual stress distribution cloud map, and trains and establishes a prediction model between the frequency band layout parameters, the cavity surface roughness and the mold wear through a machine learning algorithm. According to the model prediction results, the wear state and service life of the mold under the current frequency band layout and surface roughness are judged, including: Obtain cavity surface roughness data, frequency band layout parameters, and residual stress distribution cloud map data as training data sets for machine learning algorithms; Support vector machine and random forest machine learning algorithms were used to train based on the training data set to establish a prediction model between frequency band layout parameters, cavity surface roughness and mold wear; According to the established prediction model, the frequency band layout parameters and surface roughness data of the current cavity are input to predict the wear state of the mold; If the predicted wear status exceeds the preset threshold, it is judged that the mold has reached the end of its service life and needs maintenance or replacement; Otherwise, the mold can still be used; Obtain the residual stress distribution cloud map of the cavity and extract the residual stress distribution characteristics as a supplementary input of the prediction model to improve the accuracy of mold wear state prediction; By analyzing the prediction results, the key factors affecting mold wear are determined, including the key frequency band in the frequency band layout parameters and the critical value of surface roughness, providing a basis for optimizing mold design; According to the mold wear status prediction results and key influencing factors analysis, a mold maintenance strategy is formulated. The maintenance strategy is to regularly detect surface roughness and optimize frequency band layout to extend the mold service life and improve production efficiency.

5. The method for producing a mold for cold stamping 6G antenna parts according to claim 1, characterized in that: In the cold stamping process, the online monitoring system is used to collect the surface roughness of the cavity and the frequency band layout parameters in real time, and the parameters are input into the prediction model for comparison and analysis, and the wear state of the mold is dynamically evaluated. When the wear state exceeds the preset threshold, the feedback control mechanism is triggered, and the frequency band layout parameters are automatically adjusted according to the output results of the prediction model to slow down the wear of the mold, including: Obtain the roughness data and band layout parameters of the die surface during the cold stamping process, input the acquired data into the pre-established prediction model for comparison and analysis, and obtain the die wear status evaluation results; Determine whether the evaluation result exceeds the preset threshold. If so, trigger the feedback control mechanism. According to the output result of the prediction model, use the optimization algorithm to determine the frequency band layout parameters that need to be adjusted. The frequency band layout parameters include the size or spacing of the antenna radiation unit. The correlation model between the frequency band layout parameters and mold wear is established through machine learning algorithms. Based on the correlation model, an intelligent optimization algorithm is used to determine the optimal frequency band layout parameter adjustment scheme to minimize mold wear. The optimized frequency band layout parameters are transferred to the CNC system to control the CNC machine tool to automatically adjust the size and spacing of the antenna radiation unit, change the frequency band layout, and slow down the wear of the mold; Monitor the influence of the adjusted frequency band layout parameters on the mold wear in real time. If the wear state still exceeds the threshold, re-optimize the frequency band layout parameters until the mold wear state is lower than the preset threshold. According to different mold materials and cold stamping process parameters, multiple prediction models and correlation models are established in advance. According to the actual production situation, the optimal model is dynamically switched and selected to improve the accuracy of prediction and control; Big data analysis technology is used to mine and analyze the massive amount of mold wear data and frequency band layout parameter data collected, extract implicit correlation rules and evolution trends, and provide data support for the optimization of prediction models and correlation models.

6. The method for producing a mold for cold stamping 6G antenna parts according to claim 1, characterized in that: The cross-scale integrated analysis includes: Obtain band layout design parameters, cavity surface roughness measurement data, and mold wear monitoring data to construct a multi-scale data set; Preprocess multi-scale data, including data cleaning, feature extraction and data integration, to obtain standardized data representation; Using association rule mining algorithms Apriori or FP-growth, the association rules and patterns between frequency band layout defects and surface roughness are discovered from standardized data; Based on the mined association rules, a prediction model of frequency band layout defects and surface roughness evolution is constructed; Apply the prediction model to online monitoring data to predict the evolution trend of cavity surface quality and band layout defects in real time; Comprehensively analyze the cross-scale correlation between frequency band layout, surface roughness and mold wear, establish a multi-scale correlation model, and quantify the influence mechanism between different factors; Based on the multi-scale correlation model, the frequency band layout design scheme is optimized, the mold maintenance strategy is formulated, the cavity surface quality is improved, and the mold service life is extended.

7. The method for producing a mold for cold stamping 6G antenna parts according to claim 1, characterized in that: The online monitoring data, prediction model results, and multi-scale correlation model analysis results are integrated into the mold life cycle management system. Through big data analysis and visualization technology, full-cycle monitoring and early warning of mold quality are achieved. The multi-scale correlation model is used to optimize mold maintenance strategies, predict potential failures, and provide data support for mold design optimization, including: The online monitoring system can obtain real-time data during the mold production process, including temperature, pressure and speed parameters, and transmit the data to the full life cycle management system; Input the acquired real-time monitoring data into the prediction model, and obtain the performance prediction results of the mold in the future through algorithm analysis and calculation to determine whether the mold has potential quality risks; If the prediction results indicate that there is a quality risk, the early warning mechanism will be triggered and the system will automatically generate early warning information, which will be displayed through a visual interface to prompt relevant personnel to take timely response measures; According to the prediction results and historical data, the relationship between various mold parameters and their performance and life is analyzed through the correlation model, and the key factors affecting the mold quality are obtained, providing a basis for subsequent mold design optimization; Use big data analysis technology to mine and analyze various data throughout the mold life cycle, discover the correlation rules and patterns between the data, and provide data support for optimizing mold maintenance strategies; Input the mold design parameters, material properties and process conditions data into the correlation model, train it through the machine learning algorithm, and establish a mold performance prediction model to guide mold design optimization; Based on the historical data and prediction results in the full life cycle management system, heuristic algorithms are used to optimize the mold maintenance strategy and formulate scientific maintenance plans to maximize the mold service life and reduce maintenance costs.

8. The method for producing a mold for cold stamping 6G antenna parts according to claim 1, characterized in that: The method uses a genetic algorithm based on the integrated mold life cycle quality data and multi-scale association model to optimize the frequency band layout design parameters. During the optimization process, the cavity surface quality, mold wear status and antenna performance requirements are considered. By adjusting the geometric parameters and arrangement of the antenna radiation unit, a balance between maximizing the mold life and optimizing the antenna performance is achieved, thereby realizing intelligent and efficient mold design, manufacturing, use and maintenance, including: Obtain quality data of the entire life cycle of the mold, including cavity surface quality data and mold wear status data, build a multi-scale correlation model, and establish the correlation between quality data and mold life and antenna performance; According to the cavity surface quality data and mold wear status data, the current usage status and remaining life of the mold are judged. If the remaining life of the mold is lower than the preset threshold, the antenna frequency band layout optimization process is triggered; Genetic algorithm is used to solve multi-objective optimization problems, taking the geometric parameters and arrangement of antenna radiation units as optimization variables, and mold life and antenna performance as optimization targets. During the optimization process, the cavity surface quality and mold wear status are associated with the optimization variables through the mold multi-scale correlation model, so that the optimization process can take mold quality factors into account; Through iterative optimization, a set of optimal antenna layout design parameters that meet the requirements of mold life and antenna performance are obtained, including the geometric parameters and arrangement of the antenna radiation unit; According to the antenna layout design parameters obtained through optimization, the mold is intelligently modified and redesigned, and new mold processing procedures and process parameters are generated to guide the processing and manufacturing of the mold. During the mold use and maintenance stage, the mold quality data is continuously collected and the mold status is evaluated in real time. When the mold quality drops below the threshold, the optimization process is re-triggered and the mold is upgraded to achieve intelligent management of the entire life cycle of the mold.

Citation Information

Patent Citations

  • Method for predicting the wear life of stamping dies with full-time wear accumulation

    CN109033673A

  • Abrasion simulation method for sharp edge forming die of automobile fender

    CN115952597A