Continuous stamping control method, system and device and storage medium

Abstract

The invention relates to the technical field of punching control, and discloses a continuous punching control method and system, an electronic device and a storage medium. The method comprises the steps that product data, defect data and material data are obtained, data preprocessing is carried out, a machine learning model is used for training, a stamping control model is constructed, and control parameters are optimized through a simulated annealing algorithm. By optimizing the stamping speed, pressure and temperature parameters, product defects can be effectively reduced, the yield is increased, the production efficiency is improved, the stability and the consistency of the stamping process are enhanced, and the method is suitable for stamping forming of various materials and complex processes.

Description

Technical Field

[0001] The present invention relates to the field of continuous stamping control technology, and in particular to a continuous stamping control method, system, device and storage medium. Background Art

[0002] At present, with the continuous progress of material science and automation technology, stamping technology has been further developed. Modern stamping equipment has greatly improved the degree of automation of production and the efficiency of production lines by introducing servo systems, CNC technology and intelligent control. The design and manufacture of stamping dies have also undergone a transformation from traditional manual manufacturing to CNC processing and rapid prototyping, which has greatly shortened the product development cycle and improved production flexibility.

[0003] In an existing stamping control technology, a die mounted on a press is used to apply pressure to the material, causing the material to separate or plastically deform, thereby obtaining a workpiece whose shape, size and performance meet the use requirements. The main equipment used is a punch press. Among them, room temperature refers to the room temperature of an industrial production environment. Stamping products are mainly used for the molding of parts, among which the more well-known products include switch accessories, gaskets, instrument panels, home appliance housings, decorative parts, car bumpers, and windshields.

[0004] However, during the stamping process, large defects often occur in stamped products, mainly manifested as bulges, indentations, damage, and wrinkles on the product surface, which seriously affects product quality and production efficiency. Summary of the invention

[0005] The present invention provides a continuous stamping control method, device, electronic equipment and storage medium to solve the problem of large defects such as protrusions, indentations, damages and wrinkles on the surface of stamped products that often occur during the stamping process.

[0006] In a first aspect, in order to solve the above technical problems, the present invention provides a continuous stamping control method, comprising: Acquiring initial data for continuous stamping control, the initial data including product data, defect data and material data; Performing data preprocessing on the initial data to obtain preprocessed data; Training a machine learning model based on the preprocessed data to obtain preliminary control parameters in the stamping process; According to the preliminary control parameters, a stamping control model is constructed to calculate the relationship between stamping speed, pressure and temperature and product quality, and obtain preliminary optimized stamping control parameters; The difference between the initially optimized stamping control parameters and the stamping control parameters in actual stamping production is compared. If the difference exceeds a preset threshold, the parameters in the stamping control formula are further optimized using a simulated annealing algorithm to obtain the final optimized stamping control parameters.

[0007] Preferably, performing data preprocessing on the initial data to obtain preprocessed data includes: For the missing data in the initial data, the mean and mode are used to fill the data to obtain complete data; For the complete data, outlier detection and filtering are performed using a box plot to obtain data after outlier processing; When the data after outlier processing exceeds the standardized interval, the data is linearly transformed through extreme value standardization to ensure that all data fall within the standardized interval and obtain pre-processed data that meets the requirements.

[0008] Preferably, the machine learning model is trained based on the preprocessed data to obtain preliminary control parameters in the stamping process, including: Based on the logistic regression model, the gradient descent algorithm is used to train the model and output the preliminary control parameters related to the basic settings of stamping speed, pressure and temperature; Based on the linear discriminant analysis model, a data projection direction of the linear discriminant analysis is generated, the preprocessed data is projected into a new feature vector, and preliminary control parameters related to product defect prediction are output for process adjustment; Based on the K nearest neighbor model, the K nearest neighbor algorithm is used to determine the weights, and the defect type data with the largest weight is output as the preliminary control parameters for adjusting the stamping parameters according to the defect type; The output results based on the logistic regression model, the linear discriminant analysis model and the K-nearest neighbor model are used together as preliminary control parameters in the stamping process to optimize the speed, pressure, temperature and defect correction in the stamping process.

[0009] Preferably, according to the preliminary control parameters, a stamping control model is constructed to calculate the relationship between stamping speed, pressure and temperature and product quality to obtain the preliminary optimized stamping control parameters, including: According to the preliminary control parameters, based on a linear regression model, a linear relationship between stamping speed, pressure, temperature and product quality is obtained; According to the linear relationship, based on the polynomial model, further adjusting the stamping speed, pressure and temperature to obtain adjusted stamping control parameters; When the adjusted stamping speed, pressure and temperature do not meet the quality requirements, the polynomial model is used again to make adjustments to obtain the initially optimized stamping control parameters.

[0010] Preferably, the stamping control parameters after the preliminary optimization are compared with the stamping control parameters in the actual stamping production. If the difference exceeds a preset threshold, the parameters in the stamping control formula are further optimized using a simulated annealing algorithm to obtain the final optimized stamping control parameters, including: Obtain the initial solution of the feasible solution space, initialize the initial temperature, initial state, temperature decay step and acceptance probability threshold; According to the initial solution, calculating the objective function value of the initial solution; Calculating the objective function value of the new solution based on the initial solution and a new solution randomly selected from the feasible solution space; According to the objective function value, based on the acceptance probability formula, a simulated annealing algorithm is used to determine whether to accept the new solution. If accepted, the new solution is used as the new current solution; otherwise, the original current solution is maintained; According to the current solution, the temperature is updated, and then a cyclic optimization is performed to obtain the final optimized stamping control parameters.

[0011] Preferably, the method of determining whether to accept a new solution based on the objective function value and the acceptance probability formula by using a simulated annealing algorithm, and if accepted, using the new solution as a new current solution, otherwise, keeping the original current solution, includes: The acceptance probability of the simulated annealing algorithm is obtained by the following formula: in, x is the current solution selected by the simulated annealing algorithm in the current iteration, Is the new solution to be compared to see if it can replace the current solution, F(x) is the objective function value, is the objective function value of the new solution, T is the current temperature.

[0012] Preferably, the temperature is updated according to the current solution, and then cyclic optimization is performed to obtain the final optimized stamping control parameters, including: According to the current temperature, update the next set of temperatures to obtain new temperatures; According to the new temperature, compare the objective function value and transfer the state to obtain the optimized parameters; According to the stamping product quality data obtained in real time, when the optimized parameters meet the requirements, the optimal parameters of output speed, pressure and temperature are determined by the Pareto optimization method to obtain the final optimized stamping control parameters.

[0013] In a second aspect, the present invention provides a continuous stamping control system, comprising: Data acquisition module, used to acquire quality data, defect data and material data of stamping products; A data preprocessing module is used to perform missing data completion, outlier processing and data standardization operations according to the quality data to obtain preprocessed data; A model optimization module, used to perform a simulated annealing algorithm optimization operation according to the preprocessed data and the defect data to obtain an optimized formula; The result output module is used to input the optimized formula into a pre-trained machine learning model to obtain the target stamping parameter results.

[0014] In a third aspect, the present invention further provides an electronic device, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements any one of the above-mentioned continuous stamping control methods when executing the computer program.

[0015] In a fourth aspect, the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute any one of the continuous stamping control methods described above.

[0016] Compared with the prior art, the present invention has the following beneficial effects: First, the present invention realizes precise control of multiple key parameters such as stamping speed, pressure, and temperature through a data-driven intelligent optimization method. Unlike the prior art that relies on manual experience adjustment, the present invention can analyze and iteratively optimize real-time data in the production process by introducing machine learning models and simulated annealing algorithms, scientifically adjust process parameters, and ensure that the quality and production efficiency of stamping products are optimized. At the same time, by combining stamping defects and material properties, the present invention scientifically models the complex relationship between stamping speed, pressure, and temperature on the basis of optimizing the calculation model, effectively reducing product defects such as bulges, indentations, and wrinkles, and improving product qualification rate and production efficiency.

[0017] Secondly, the present invention adopts a modular optimization system design, covering functional modules such as data acquisition, preprocessing, model optimization and result output. The modules work together to realize the full process automation from data acquisition to optimization result output, greatly reducing manual intervention and improving the adaptability and operating efficiency of the system. Compared with the traditional stamping control method, the present invention automatically generates the optimal stamping parameters through the optimization model, reducing the time required for manual adjustment and repeated experiments, making the production process more stable and reliable.

[0018] In addition, the dynamic adjustment and flexible adaptability of the present invention significantly enhance the adaptability of the system when facing different materials and complex working conditions. By performing real-time data analysis on new materials and adjusting stamping parameters, the present invention can ensure stable product quality during the production process, solving the problem that the prior art is difficult to cope with complex environmental changes. In particular, under multiple iterative optimizations of stamping speed, pressure and temperature, the system can dynamically adjust stamping parameters, thereby significantly reducing the problem of unstable product quality caused by process fluctuations and improving the yield rate.

[0019] Furthermore, the present invention can achieve multiple iterations of optimization and accurately control the balance between various parameters in the stamping process through the application of simulated annealing algorithm. In actual production, the present invention can scientifically optimize the stamping speed, pressure and temperature parameters through the optimization calculation model according to product quality data and material characteristics, making the stamping process more efficient and accurate, and reducing production costs.

[0020] Finally, the system integration and operation convenience of the present invention provide great convenience for users. Through integration with electronic devices, designers can automatically execute the stamping control method through the system and obtain the optimization results in real time to ensure the efficiency and stability of production. The intelligent optimization algorithm and flexible system deployment of the present invention make it have a wide range of application value in different production environments.

[0021] In summary, the present invention demonstrates significant technical advantages in intelligent optimization, automated control, parameter adjustment, system integration and ease of operation. It can significantly improve the efficiency, product quality and economic benefits of stamping production and has broad industrial application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a schematic flow chart of a continuous stamping control method provided by a first embodiment of the present invention; Figure 2 It is a structural schematic diagram of a continuous stamping control method provided in the second embodiment of the present invention. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0024] Reference Figure 1 The first embodiment of the present invention provides a continuous stamping control method, comprising the following steps: S11, obtaining initial data for continuous stamping control, the initial data including product data, defect data and material data; S12, performing data preprocessing on the initial data to obtain preprocessed data; S13, training a machine learning model according to the preprocessed data to obtain preliminary control parameters in the stamping process; S14, constructing a stamping control model according to the preliminary control parameters, calculating the relationship between stamping speed, pressure and temperature and product quality, and obtaining preliminary optimized stamping control parameters; S15, comparing the difference between the stamping control parameters after preliminary optimization and the stamping control parameters in actual stamping production. If the difference exceeds a preset threshold, the parameters in the stamping control formula are further optimized using a simulated annealing algorithm to obtain the final optimized stamping control parameters.

[0025] In step S11, initial data for continuous stamping control is obtained, and the initial data includes product data, defect data and material data, including: Obtain product data: Product data includes physical parameters and geometric information related to stamped products. The above product data usually includes the length, width, thickness and shape of the product, which is used to determine the specific process requirements during the stamping process. Accurately obtaining the above product data ensures that subsequent process adjustments can meet the precise forming requirements of the product.

[0026] Obtaining defect data: Defect data is used to evaluate quality problems that may occur during the stamping process. By analyzing production history data and real-time inspection results, the defect data obtained includes product surface bulges, indentations, breakages, and wrinkles. Accurately obtaining defect data helps to identify potential defects in the process at an early stage and provide a basis for subsequent process optimization.

[0027] Obtain material data: Material data refers to detailed information related to the performance of stamping materials, including the tensile strength, yield strength, material thickness and metallographic structure of the material. The material data can help determine the appropriate pressure, speed and temperature to be applied during the stamping process, thereby ensuring production stability and product quality.

[0028] In step S12, the initial data is preprocessed to obtain preprocessed data, including: Step S121, for the missing data in the initial data, the mean and mode are used to fill the data to obtain complete data; Step S122, using a box plot to perform outlier detection and filtering on the complete data to obtain data after outlier processing; Step S123, when the data after outlier processing exceeds the standardized interval, the data is linearly transformed through extreme value standardization to ensure that all data falls within the standardized interval and obtain pre-processed data that meets the requirements.

[0029] In step S121, the missing data in the initial data are filled with the mean and mode to obtain complete data, including: Mean filling: For the numerical data in the initial data, such as stamping speed, pressure, and temperature parameters, the mean filling method is used, that is, the average value of all available similar data is filled into the missing data items to ensure the continuity and integrity of the data set.

[0030] Mode filling: For the categorical data in the initial data, such as product defect type or material category, the mode filling method is used. The rationality and consistency of the categorical data are ensured by calculating the most frequently occurring value in the categorical data in the initial data and filling in the missing items.

[0031] It should be noted that the data filling method can effectively fill the gaps caused by missing data and ensure that subsequent data analysis and model training can be carried out based on a complete data set.

[0032] In step S122, the complete data is subjected to outlier detection and filtering using a box plot to obtain outlier processed data, including: Detect outliers with box plots: Visually analyze the degree of dispersion of data by constructing box plots to determine whether there are outliers in the data. Box plots can clearly show the distribution of data and possible abnormal data points. Outliers are usually points far away from the quartile range of the data.

[0033] Outlier filtering: For detected outliers, adopt reasonable filtering strategies, such as removing extreme outliers or adjusting data according to actual needs, to ensure that the data used for subsequent analysis and model training are more representative and accurate.

[0034] In step S123, when the data after outlier processing exceeds the standardized interval, the data is linearly transformed by extreme value standardization to ensure that all data falls within the standardized interval and obtain preprocessed data that meets the requirements, including: Extreme value standardization: The preprocessed data is converted to the interval range [0, 1] by using the extreme value standardization method. Extreme value standardization uses linear transformation to scale the data according to the ratio of the minimum value to the maximum value, so that the distribution of the data is within a predetermined range, ensuring that all data can be processed under a unified standard.

[0035] Furthermore, the above linear transformation can be calculated by the following formula: in, is the original data value, is the standardized data value, and are the minimum and maximum values ​​in the data set, respectively.

[0036] In step S13, the machine learning model is trained according to the preprocessed data to obtain preliminary control parameters in the stamping process, including: Step S131, based on the logistic regression model, a gradient descent algorithm is used to train the model, and preliminary control parameters related to the basic settings of stamping speed, pressure, and temperature are output; Step S132, based on the linear discriminant analysis model, generating a data projection direction of the linear discriminant analysis, projecting the preprocessed data into a new feature vector, and outputting preliminary control parameters related to product defect prediction for process adjustment; Step S133, based on the K nearest neighbor model, using the K nearest neighbor algorithm to determine the weight, output the defect type data with the largest weight as the preliminary control parameter for adjusting the stamping parameters according to the defect type; Step S134, the output results based on the logistic regression model, the linear discriminant analysis model and the K nearest neighbor model are used together as preliminary control parameters in the stamping process to optimize the speed, pressure, temperature and defect correction in the stamping process.

[0037] In step S131, based on the logistic regression model, a gradient descent algorithm is used to train the model, and preliminary control parameters related to the basic settings of stamping speed, pressure, and temperature are output, including: Logistic regression model: Logistic regression model is mainly used to deal with binary classification problems in stamping process, such as whether the product is qualified. The input data contains three key parameters: stamping speed, pressure and temperature. The model is trained to predict whether the product quality is qualified under these specific parameters.

[0038] Gradient descent algorithm: During the model training process, the gradient descent algorithm is used to continuously adjust the model parameters to minimize the loss function. After each iteration, the model updates the weights according to the gradient of the current parameters, and finally outputs the preliminary control parameters of stamping speed, pressure and temperature, which are used to set the basic conditions of the stamping process.

[0039] It should be noted that through the training of the logistic regression model, the system can output basic control parameters for different process requirements, laying the foundation for subsequent process optimization.

[0040] In step S132, based on the linear discriminant analysis model, the data projection direction of the linear discriminant analysis is generated, the preprocessed data is projected into a new feature vector, and preliminary control parameters related to product defect prediction are output for process adjustment, including: Linear discriminant analysis model: Linear discriminant analysis models are often used to solve multi-classification problems, especially when it is necessary to classify products according to the type of surface defects. The model analyzes the category labels and feature information in the preprocessed data to find the optimal projection direction in order to best distinguish samples of different defect categories. Linear discriminant analysis can not only reduce the data dimension, but also generate feature vectors with high discrimination ability by maximizing the difference between classes and minimizing the difference within classes, thereby providing a more reliable basis for subsequent classification and process adjustment.

[0041] Data projection: Based on the above analysis, the linear discriminant analysis model calculates the differences between different categories and generates an optimal projection direction to map the original high-dimensional data into a low-dimensional space. In this process, the model reduces the dimension while retaining the most valuable feature information for classification. The model maximizes the differences between categories and minimizes the differences within categories to find the projection axis that can best distinguish different defect categories, thereby ensuring the distinction of defect data in low-dimensional space. Through this projection method, complex high-dimensional data is converted into simplified feature vectors, which can effectively reflect different defect types and provide a reliable reference for parameter adjustment of the stamping process.

[0042] It should be noted that by using the linear discriminant analysis model, the system is able to predict and distinguish different defect types of products and adjust the stamping process parameters based on these prediction results.

[0043] In step S133, based on the K nearest neighbor model, the K nearest neighbor algorithm is used to determine the weights, and the defect type data with the largest weight is output as the preliminary control parameters for adjusting the stamping parameters according to the defect type, including: First, the system collects and analyzes historical data and uses the K-nearest neighbor model to calculate the distance between the current data point and its neighboring points in the historical data. Through this process, the system can identify defects in the current production process that are similar to those in previous production data. The core of the K-nearest neighbor model is to use these similarities to predict the most likely type of defects in the current product.

[0044] In the K-nearest neighbor algorithm, the system assigns a weight to each neighboring point, and the size of the weight is determined by the distance. Generally, the closer the point is, the greater the weight is, which has a more significant impact on the results during the prediction process. Finally, the model outputs the defect type data with the largest weight as a preliminary basis for adjusting the stamping process parameters.

[0045] Through the above method, the system can not only make accurate predictions with the help of historical data, but also make dynamic adjustments based on the current actual production situation, thereby improving the stability of the production process and reducing the incidence of product defects.

[0046] In step S134, the output results based on the logistic regression model, the linear discriminant analysis model and the K nearest neighbor model are used together as preliminary control parameters in the stamping process to optimize the speed, pressure, temperature and defect correction in the stamping process, including: Model fusion: The output results of the logistic regression model, linear discriminant analysis model and K nearest neighbor model work together as the preliminary control parameters of the stamping process. The logistic regression model provides basic process parameters, the linear discriminant analysis model predicts the defect category, and the K nearest neighbor model helps determine the defect type and adjustment plan.

[0047] Process optimization: Based on the model fusion, the system can more accurately adjust the stamping speed, pressure and temperature, and make corrections based on the predicted defect type. This ensures stability in the production process and reduces the probability of producing defective products.

[0048] In step S14, a stamping control model is constructed based on the preliminary control parameters, and the relationship between stamping speed, pressure and temperature and product quality is calculated to obtain the preliminary optimized stamping control parameters, including: Step S141, obtaining the linear relationship between stamping speed, pressure, temperature and product quality based on the linear regression model according to the preliminary control parameters; Step S142, further adjusting the stamping speed, pressure and temperature according to the linear relationship based on the polynomial model to obtain adjusted stamping control parameters; Step S143, when the adjusted stamping speed, pressure and temperature do not meet the quality requirements, the polynomial model is used again to make adjustments to obtain the initially optimized stamping control parameters.

[0049] In step S141, according to the preliminary control parameters, based on the linear regression model, the linear relationship between stamping speed, pressure, temperature and product quality is obtained, including: The linear regression model is used to determine the linear dependence between stamping speed, pressure, temperature and product quality. The model generates a linear function for predicting product quality by fitting actual production data. Through this model, it is possible to determine which stamping parameters have a greater impact on product quality and provide a basis for further optimizing these parameters.

[0050] Furthermore, the linear relationship generated above can be calculated by the following formula: Among them, Q represents product quality, V represents stamping speed, P represents stamping pressure, T represents stamping temperature, a, b, c are regression coefficients, and d is a constant term.

[0051] In step S142, according to the linear relationship, based on the polynomial model, the stamping speed, pressure and temperature are further adjusted to obtain adjusted stamping control parameters, including: Based on the linear regression model, the polynomial model is used to further describe and optimize the nonlinear effects of punching speed, pressure and temperature. The polynomial model captures the complex interactions between parameters by adding higher-order terms.

[0052] Furthermore, the above polynomial model can be calculated by the following formula: Among them, Q represents product quality, V represents stamping speed, P represents stamping pressure, T represents stamping temperature, a, b, c represent regression coefficients, d, e, f are interaction term coefficients, and g is a constant term.

[0053] It should be noted that the polynomial model can more accurately describe the nonlinear effect of stamping parameters on product quality. Especially when dealing with high-strength materials or complex production processes, the optimization of the polynomial model can significantly improve the forming accuracy and consistency of the product. By continuously adjusting the parameters in the polynomial, more accurate stamping control parameters can be generated.

[0054] In step S143, when the adjusted stamping speed, pressure, and temperature do not meet the quality requirements, the polynomial model is used again to make adjustments to obtain the initially optimized stamping control parameters, including: In actual production, if the adjusted stamping speed, pressure and temperature still do not meet the expected product quality standards, the polynomial model will be further used for iterative optimization. In view of product quality deviations, the stamping parameters are further adjusted until the product quality meets the expected requirements. Each iterative optimization step is based on the polynomial model. By feeding back real-time data in production, the parameters are continuously corrected, and finally the preliminary optimized stamping control parameters that meet the quality requirements are obtained.

[0055] In step S15, the difference between the initially optimized stamping control parameters and the stamping control parameters in actual stamping production is compared. If the difference exceeds a preset threshold, the parameters in the stamping control formula are further optimized using a simulated annealing algorithm to obtain the final optimized stamping control parameters, including: Step S151, obtaining an initial solution of the feasible solution space, initializing the initial temperature, initial state, temperature attenuation step length and acceptance probability threshold; Step S152, calculating the objective function value of the initial solution according to the initial solution; Step S153, calculating the objective function value of the new solution according to the initial solution and a new solution randomly selected from the feasible solution space; Step S154, according to the objective function value, based on the acceptance probability formula, using the simulated annealing algorithm to determine whether to accept the new solution, if accepted, the new solution is used as the new current solution, otherwise, the original current solution is maintained; Step S155, updating the temperature according to the current solution, and then performing cyclic optimization to obtain the final optimized stamping control parameters.

[0056] In step S151, an initial solution of the feasible solution space is obtained, and the initial temperature, initial state, temperature attenuation step and acceptance probability threshold are initialized, including: Obtain the initial solution of the feasible solution space: The feasible solution space represents the space of all possible solutions in the process of adjusting the stamping control parameters. A solution is randomly selected in the feasible solution space as the initial solution. The initial solution is used as the initial parameter setting of the simulated annealing algorithm and will be used as a reference in the subsequent optimization process.

[0057] Initialization initial temperature: The temperature in the simulated annealing algorithm is a control parameter that determines the likelihood that the algorithm will accept a new solution. The initial temperature is usually set high to explore a wider solution space and avoid falling into a local optimal solution.

[0058] Get the initial state: The initial state represents the starting state of the simulated annealing algorithm, including the initial solution and initial temperature.

[0059] Get the temperature decay step size: The temperature gradually decreases during simulated annealing, and the step size determines the speed at which the temperature decreases each time. A smaller step size helps with more detailed search, but may lead to a slower optimization process; a larger step size speeds up the search, but may skip some potential optimal solutions.

[0060] Get the acceptance probability threshold: In each iteration, the simulated annealing algorithm determines whether to accept a new solution based on the acceptance probability formula. The acceptance probability threshold is a set critical value that controls the probability of a new solution being accepted.

[0061] In step S154, according to the objective function value, based on the acceptance probability formula, using the simulated annealing algorithm to determine whether to accept the new solution, if accepted, the new solution is used as the new current solution, otherwise, the original current solution is maintained, including: The acceptance probability of the simulated annealing algorithm is obtained by the following formula: in, xis the current solution selected by the simulated annealing algorithm in the current iteration, Is the new solution to be compared to see if it can replace the current solution, F(x) is the objective function value, is the objective function value of the new solution, T is the current temperature.

[0062] In step S155, the temperature is updated according to the current solution, and then cyclic optimization is performed to obtain the final optimized stamping control parameters, including: According to the current temperature, update the next set of temperatures to obtain new temperatures; According to the new temperature, compare the objective function value and transfer the state to obtain the optimized parameters; According to the stamping product quality data obtained in real time, when the optimized parameters meet the requirements, the optimal parameters of output speed, pressure and temperature are determined by the Pareto optimization method to obtain the final optimized stamping control parameters.

[0063] In order to facilitate the understanding of the present invention, some preferred embodiments of the present invention are further described below.

[0064] In one implementation, the continuous stamping control method of the present invention can be applied to the stamping forming process of high-strength materials. For high-strength steel or aluminum alloy materials, due to their higher tensile strength and yield strength, the stamping speed, pressure and temperature parameters need to be precisely controlled during the stamping process. The present invention analyzes and feeds back real-time data during the stamping process through a machine learning model, and optimizes the stamping process parameters in real time in combination with pre-processed material data, defect data and product data, thereby effectively avoiding cracks, creases and local deformation problems in the product. In particular, for high-strength materials, stamping is usually carried out under higher stress conditions. The multiple iterative optimization mechanism constructed by the present invention ensures the stability of stamping under high stress conditions and significantly improves the forming accuracy and consistency of the product.

[0065] In this implementation, the parameter optimization strategy based on the simulated annealing algorithm plays a key role. Through multiple rounds of simulated annealing optimization process, the present invention can dynamically adjust the stamping pressure and speed parameters under different temperature conditions to ensure that the material will not undergo local plastic deformation due to excessive temperature or excessive stress during the processing. By adjusting the stamping process parameters in real time, the present invention can be widely used in the stamping forming of high-strength steel and magnesium alloys.

[0066] In another implementation, the present invention can also be used for real-time control and optimization of a multi-step stamping production line. In the multi-step stamping process, the product needs to be gradually formed through multiple stamping stations, and the stamping parameters of each station will affect the final molding quality of the product. The present invention collects and analyzes the real-time data of each station, and uses a machine learning model to gradually optimize the parameters of each station to ensure the coordination and stability of the entire production line.

[0067] In this implementation, the system can make personalized parameter adjustments for the production data of each workstation, such as stamping speed, mold temperature, and material elongation. For example, in the stamping production process, if the pressure of a certain process exceeds the preset threshold, the system will automatically reduce the pressure of the process or adjust the material feed speed to ensure that the product can still meet the expected quality requirements in the subsequent process. At the same time, the present invention can also use the Pareto optimization method to integrate the production parameters of each workstation and find the optimal combination of production parameters to ensure the efficiency and quality of the entire stamping process.

[0068] In addition, in this implementation, the system can realize the interconnection of multi-station data, so that the stamping parameters of each station can be adjusted in conjunction with the production data of the previous and next stations. For example, if the product thickness deviation occurs during the stamping process of the previous process, the system will automatically optimize the stamping speed and pressure of the subsequent stations to compensate for the error caused by the previous process and ensure the molding accuracy of the final product. Through this station linkage optimization, the present invention shows extremely high adaptability and reliability in multi-step stamping production lines.

[0069] Furthermore, in some complex stamping scenarios, the present invention can further use deep learning algorithms, such as convolutional neural networks or recurrent neural network models, to deeply mine a large amount of historical data in the stamping process, extract hidden process feature patterns, and make real-time predictions and optimization adjustments based on these patterns. This implementation method is suitable for highly complex stamping processes and can effectively improve stamping efficiency and reduce defect rates.

[0070] In addition, the control system of the present invention can be combined with industrial Internet of Things technology to obtain the operating status data of the stamping equipment in real time through the sensor network, and monitor and predict the operating status of the equipment through big data analysis, and optimize the process parameters in advance to extend the service life of the equipment and improve production efficiency.

[0071] In summary, refer to Figure 2 The second embodiment of the present invention provides a continuous stamping control system, comprising: Data acquisition module, used to acquire quality data, defect data and material data of stamping products; A data preprocessing module is used to perform missing data completion, outlier processing and data standardization operations according to the quality data to obtain preprocessed data; A model optimization module, used to perform a simulated annealing algorithm optimization operation according to the preprocessed data and the defect data to obtain an optimized formula; The result output module is used to input the optimized formula into a pre-trained machine learning model to obtain the target stamping parameter results.

[0072] Preferably, the data acquisition module can monitor key process parameters in the stamping process in real time through a variety of sensors and acquisition devices, including but not limited to stamping speed, pressure, temperature, product surface quality and material properties. The data acquisition module collects relevant data through sensors integrated into the production equipment and transmits the above data to the data preprocessing module to ensure that the system can obtain the latest production information and real-time data.

[0073] Preferably, after receiving the data transmitted by the data acquisition module, the data preprocessing module first fills the missing data, and uses methods such as mean and mode filling to handle data missing situations to ensure data integrity. Subsequently, the system detects and processes outliers in the data through box plots, filtering out abnormal data that may affect the analysis results. Finally, the system performs extreme value normalization on the processed data, converting all data to a specified interval [0, 1] to ensure that different feature data have a consistent range, making the subsequent model training process more accurate and robust.

[0074] Preferably, the model optimization module trains the preprocessed data through a machine learning algorithm to generate preliminary control parameters for the stamping process. That is, the system generates optimal process parameters based on multiple algorithms such as logistic regression model, linear discriminant analysis model and K nearest neighbor model, combined with real-time data in the production process. The module also optimizes the preliminary control parameters through a simulated annealing algorithm, and gradually improves the accuracy of key parameters such as stamping speed, pressure, temperature, etc. through multiple rounds of iterations to ensure that product quality reaches the best level.

[0075] Preferably, the result output module receives the optimized process parameters and outputs them to the control system of the production equipment for real-time adjustment of the operating parameters in the stamping process. The module can also monitor and feedback the process adjustment results to ensure that the adjustment of the process parameters is consistent with the actual production requirements. In addition, the result output module can generate a process parameter report in the production process for reference and analysis by operators to optimize the entire production process.

[0076] Through the collaborative work of the above modules, the continuous stamping control system of the present invention can monitor and optimize various parameters in the stamping process in real time, significantly improve production efficiency and product quality, and reduce equipment wear and production costs.

[0077] It should be noted that a continuous stamping control device provided in an embodiment of the present invention is used to execute all process steps of a continuous stamping control method in the above embodiment, and the working principles and beneficial effects of the two correspond one to one, so they will not be repeated.

[0078] The embodiment of the present invention further provides an electronic device. The electronic device includes: a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a continuous stamping control program. When the processor executes the computer program, the steps in the above-mentioned continuous stamping control method embodiments are implemented, such as Figure 1 Alternatively, when the processor executes the computer program, the functions of each module / unit in the above-mentioned device embodiments are implemented, such as the model optimization module.

[0079] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer program in the electronic device.

[0080] The electronic device may be a computing device such as a desktop computer, a notebook, a PDA, and a smart tablet. The electronic device may include, but is not limited to, a processor and a memory. Those skilled in the art will appreciate that the above components are merely examples of electronic devices and do not constitute a limitation on the electronic device. The electronic device may include more or fewer components than the above components, or may combine certain components, or different components. For example, the electronic device may also include input and output devices, network access devices, buses, etc.

[0081] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the electronic device, and uses various interfaces and lines to connect various parts of the entire electronic device.

[0082] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the electronic device by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0083] Wherein, if the module / unit integrated in the electronic device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0084] It should be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.

[0085] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A continuous stamping control method, characterized in that: include: Acquiring initial data for continuous stamping control, the initial data including product data, defect data and material data; Performing data preprocessing on the initial data to obtain preprocessed data; Training a machine learning model based on the preprocessed data to obtain preliminary control parameters in the stamping process; According to the preliminary control parameters, a stamping control model is constructed to calculate the relationship between stamping speed, pressure and temperature and product quality, and obtain preliminary optimized stamping control parameters; The difference between the initially optimized stamping control parameters and the stamping control parameters in actual stamping production is compared. If the difference exceeds a preset threshold, the parameters in the stamping control formula are further optimized using a simulated annealing algorithm to obtain the final optimized stamping control parameters.

2. The continuous stamping control method according to claim 1, characterized in that: Performing data preprocessing on the initial data to obtain preprocessed data includes: For the missing data in the initial data, the mean and mode are used to fill the data to obtain complete data; For the complete data, outlier detection and filtering are performed using a box plot to obtain data after outlier processing; When the data after outlier processing exceeds the standardized interval, the data is linearly transformed through extreme value standardization to ensure that all data fall within the standardized interval and obtain pre-processed data that meets the requirements.

3. The continuous stamping control method according to claim 1, characterized in that: Based on the preprocessed data, the machine learning model is trained to obtain preliminary control parameters in the stamping process, including: Based on the logistic regression model, the gradient descent algorithm is used to train the model and output the preliminary control parameters related to the basic settings of stamping speed, pressure and temperature; Based on the linear discriminant analysis model, a data projection direction of the linear discriminant analysis is generated, the preprocessed data is projected into a new feature vector, and preliminary control parameters related to product defect prediction are output for process adjustment; Based on the K nearest neighbor model, the K nearest neighbor algorithm is used to determine the weights, and the defect type data with the largest weight is output as the preliminary control parameters for adjusting the stamping parameters according to the defect type; The output results based on the logistic regression model, the linear discriminant analysis model and the K-nearest neighbor model are used together as preliminary control parameters in the stamping process to optimize the speed, pressure, temperature and defect correction in the stamping process.

4. The continuous stamping control method according to claim 1, characterized in that: According to the preliminary control parameters, a stamping control model is constructed to calculate the relationship between stamping speed, pressure and temperature and product quality, and obtain the preliminary optimized stamping control parameters, including: According to the preliminary control parameters, based on a linear regression model, a linear relationship between stamping speed, pressure, temperature and product quality is obtained; According to the linear relationship, based on the polynomial model, further adjusting the stamping speed, pressure and temperature to obtain adjusted stamping control parameters; When the adjusted stamping speed, pressure and temperature do not meet the quality requirements, the polynomial model is used again to make adjustments to obtain the initially optimized stamping control parameters.

5. The continuous stamping control method according to claim 1, characterized in that: The difference between the stamping control parameters after the preliminary optimization and the stamping control parameters in the actual stamping production is compared. If the difference exceeds the preset threshold, the parameters in the stamping control formula are further optimized using the simulated annealing algorithm to obtain the final optimized stamping control parameters, including: Obtain the initial solution of the feasible solution space, initialize the initial temperature, initial state, temperature decay step and acceptance probability threshold; According to the initial solution, calculating the objective function value of the initial solution; Calculating the objective function value of the new solution based on the initial solution and a new solution randomly selected from the feasible solution space; According to the objective function value, based on the acceptance probability formula, a simulated annealing algorithm is used to determine whether to accept the new solution. If accepted, the new solution is used as the new current solution; otherwise, the original current solution is maintained; According to the current solution, the temperature is updated, and then a cyclic optimization is performed to obtain the final optimized stamping control parameters.

6. The continuous stamping control method according to claim 5, characterized in that: The method of determining whether to accept a new solution based on the objective function value and the acceptance probability formula by using a simulated annealing algorithm, and if accepted, using the new solution as a new current solution, or maintaining the original current solution, includes: The acceptance probability of the simulated annealing algorithm is obtained by the following formula: in, x is the current solution selected by the simulated annealing algorithm in the current iteration, Is the new solution to be compared to see if it can replace the current solution, F(x) is the objective function value, is the objective function value of the new solution, T is the current temperature.

7. The continuous stamping control method according to claim 6, characterized in that: The temperature is updated according to the current solution, and then cyclic optimization is performed to obtain the final optimized stamping control parameters, including: According to the current temperature, update the next set of temperatures to obtain new temperatures; According to the new temperature, compare the objective function value and transfer the state to obtain the optimized parameters; According to the stamping product quality data obtained in real time, when the optimized parameters meet the requirements, the optimal parameters of output speed, pressure and temperature are determined by the Pareto optimization method to obtain the final optimized stamping control parameters.

8. A continuous stamping control system, characterized in that: include: Data acquisition module, used to acquire quality data, defect data and material data of stamping products; A data preprocessing module is used to perform missing data completion, outlier processing and data standardization operations according to the quality data to obtain preprocessed data; A model optimization module, used to perform a simulated annealing algorithm optimization operation according to the preprocessed data and the defect data to obtain an optimized formula; The result output module is used to input the optimized formula into a pre-trained machine learning model to obtain the target stamping parameter results.

9. An electronic device, characterized in that: The method comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements the continuous stamping control method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the continuous stamping control method according to any one of claims 1 to 7.

Images