Membrane punch forming process generation method and system based on data driving

Abstract

The invention provides a diaphragm punch forming process generation method and system based on data driving, and the method comprises the steps: constructing a test model which is obtained through a first mapping relation, and obtaining punch forming information according to the test model, the punch forming information comprising punching die information and process information; obtaining an experiment result obtained by performing an experiment on a diaphragm prepared through the punch forming information and an output parameter obtained by inputting the first input parameter into the test model and outputting a parameter matrix composed of residual stress, rebound amount and warping degree by the test model; calculating an adjustment parameter, and performing iterative optimization on the first mapping relation and the first input parameter through the adjustment parameter to obtain a simulation model; and generating a diaphragm stamping process scheme through the simulation model. According to the method, the first mapping relation and the first input parameter are corrected by using the adjustment parameter, the test model is optimized, and the prediction precision of the model, the stability of the diaphragm stamping process and the diaphragm forming quality are improved.

Description

Technical Field

[0001] The present application relates to the field of diaphragm manufacturing technology, and in particular to a data-driven diaphragm stamping process generation method and system. Background Art

[0002] The diaphragm is the component in the pressure sensor that directly senses external pressure changes. When external pressure acts on the diaphragm, the diaphragm will deform, and this deformation is the basis for subsequent signal conversion. If the material, thickness or waveform of the diaphragm is not designed properly, it will cause uneven or inaccurate deformation, thus affecting the measurement results of the sensor.

[0003] For the stamping process of the diaphragm, a stamping experiment is carried out under the selected process parameters to record the forming conditions of the diaphragm, such as the degree of deformation, surface quality, dimensional accuracy, etc., and then precision measuring instruments such as a confocal microscope are used to measure and detect the formed diaphragm to evaluate its performance indicators.

[0004] However, the method of determining the process plan through stamping experiments, due to the complexity of material properties and the diversity of processing technology, leads to low stability of the diaphragm stamping process, low efficiency and low diaphragm forming quality. Summary of the invention

[0005] The present application provides a data-driven diaphragm stamping process generation method and system to solve the problems of low stability, low efficiency and low diaphragm forming quality of the diaphragm stamping process.

[0006] In a first aspect, the present application provides a data-driven diaphragm stamping process generation method, comprising:

[0007] Constructing a test model, wherein the test model is obtained through the first mapping relationship;

[0008] Acquire stamping information according to the test model, wherein the stamping information includes stamping die information and process information;

[0009] Acquire experimental results and output parameters, wherein the experimental results are the results of experiments on the diaphragm prepared by the stamping forming information, and the output parameters are the parameter matrix composed of residual stress, springback and warpage output by the test model when the first input parameter is input into the test model;

[0010] Calculating adjustment parameters to iteratively optimize the first mapping relationship and the first input parameter by using the adjustment parameters to obtain a simulation model;

[0011] A diaphragm stamping process plan is generated through the simulation model.

[0012] In some feasible embodiments, constructing a test model includes:

[0013] Establishing a first mapping relationship;

[0014] Get the first input parameter;

[0015] A test model is constructed according to the first mapping relationship and input parameters.

[0016] In some feasible embodiments, the first mapping relationship at least includes a dynamic equation, an elastic-plastic mechanics equation, a material model, a scale effect, and a forming mechanism;

[0017] The first input parameter includes at least the diaphragm wave number, the diaphragm wave depth, the strip thickness, the punch stroke, the punch speed, the punch pressure and the hardness of the polyurethane soft mold.

[0018] In some feasible embodiments, the calculating adjustment parameter includes:

[0019] Obtaining a first residual stress, a first springback amount, and a first warpage of the experimental results, and outputting parameters of a second residual stress, a second springback amount, and a second warpage;

[0020] Calculate adjustment parameters, the adjustment parameters including stress adjustment parameters, springback adjustment parameters and warpage adjustment parameters, the stress adjustment parameter is the difference between the first residual stress and the second residual stress, the springback adjustment parameter is the difference between the first springback and the second springback, and the warpage adjustment parameter is the difference between the first warpage and the second warpage.

[0021] In some feasible embodiments, the iterative optimization of the first mapping relationship and the first input parameter by adjusting the parameter includes:

[0022] Acquire a first driving matrix, wherein the first driving matrix is ​​a matrix composed of a residual stress adjustment parameter, a springback adjustment parameter, and a warpage adjustment parameter;

[0023] Correcting the first mapping relationship by using the first driving matrix to obtain a second mapping relationship, and correcting the first input parameter by using the first driving matrix to obtain a second input parameter;

[0024] When the number of iterations is greater than or equal to the first number, the iterative optimization is ended by the following formula:

[0025]

[0026] In the formula, δ is the standard value, R is the residual stress, Q is the springback, S is the warpage, and n is n, n-1, n-2, n-3, n-4, n-5.

[0027] In some feasible embodiments, the correcting the first mapping relationship by using the first driving matrix to obtain the second mapping relationship includes:

[0028] Establishing a third mapping relationship, where the third mapping relationship is a mapping relationship between the first input parameter and the output parameter;

[0029] Performing a sensitivity analysis according to the third mapping relationship to obtain key parameters, wherein the sensitivity analysis is used to evaluate the degree of influence of the first input parameter on the output result;

[0030] Generate a sorted list of the key parameters, wherein the sorted list is a parameter list generated after sorting by sensitivity coefficients;

[0031] Extracting a priority parameter, wherein the priority parameter is a first input parameter with a high sorting priority in the sorting list;

[0032] The first mapping relationship is corrected by using the first driving matrix and the priority parameter to obtain a second mapping relationship.

[0033] In some feasible embodiments, the calculating adjustment parameter includes:

[0034] Acquiring setting information of the diaphragm, wherein the setting information includes diameter information and thickness information;

[0035] Matching a scene interval according to the setting information to assign weights to residual stress, springback, and warpage through the scene interval, wherein the scene interval includes interval limitations for diameter information and thickness information;

[0036] A weighted error is calculated according to the output parameters and the weights, and an adjustment parameter is calculated according to the weighted error.

[0037] In a second aspect, the present application provides a data-driven diaphragm stamping process generation system, comprising:

[0038] A construction unit, configured to construct a test model, wherein the test model is obtained through a first mapping relationship;

[0039] an acquisition unit, configured to acquire stamping information according to the test model, wherein the stamping information includes stamping die information and process information; and to acquire experimental results and output parameters, wherein the experimental results are the results of experiments on the diaphragm prepared by the stamping information, and the output parameters are the parameter matrix composed of residual stress, springback and warpage output by the test model when the first input parameter is input into the test model;

[0040] a processing unit, configured to calculate an adjustment parameter, so as to iteratively optimize the first mapping relationship and the first input parameter by using the adjustment parameter to obtain a simulation model;

[0041] A generating unit is used to generate a diaphragm stamping process plan through the simulation model.

[0042] It can be seen from the above technical scheme that the present application provides a data-driven diaphragm stamping process generation method and system, the method comprising: constructing a test model, the test model is obtained through a first mapping relationship, and stamping information is obtained according to the test model, the stamping information includes stamping die information and process information; obtaining experimental results and output parameters, the experimental results are the results obtained by conducting experiments on diaphragms prepared by the stamping information, and the output parameters are the parameter matrix composed of residual stress, springback and warpage output by the test model when the first input parameter is input into the test model; calculating adjustment parameters, so as to iteratively optimize the first mapping relationship and the first input parameter through the adjustment parameters to obtain a simulation model; generating a diaphragm stamping process plan through the simulation model. The method uses the adjustment parameters to correct the first mapping relationship and the first input parameter, gradually optimizes the test model, improves the prediction accuracy of the model and the stability of the diaphragm stamping process, and thus improves the diaphragm forming quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solution of the present application, the drawings required for use in the embodiments are briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0044] Figure 1 A schematic flow chart of a method for generating a diaphragm stamping process based on data drive provided in an embodiment of the present application;

[0045] Figure 2 A schematic diagram of iterative optimization provided in an embodiment of the present application;

[0046] Figure 3 A schematic diagram of the structure of a data-driven diaphragm stamping process generation system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0047] The following embodiments are described in detail, and examples thereof are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following embodiments do not represent all implementations consistent with the present application. They are only examples of systems and methods consistent with some aspects of the present application as detailed in the claims.

[0048] The diaphragm is the component in the pressure sensor that converts the pressure acting on it into an electrical signal. The diaphragm is a thin, flexible sheet of material that can deform when subjected to external pressure, so that the calculated signal is proportional to the measured pressure.

[0049] When the diaphragm in the pressure sensor is subjected to external pressure, such as the pressure of gas or liquid, its shape will change slightly. This change in shape will cause strain on the surface of the diaphragm, thereby affecting the electrical components connected to it, such as the electrical signals of strain gauges, piezoelectric sensors, etc.

[0050] Depending on the application scenarios and requirements, the diaphragm materials and technologies used in pressure sensors vary. Diaphragms can be divided into many types, such as metal diaphragms, silicon diaphragms, ceramic diaphragms, and polymer diaphragms. The manufacturing process is also different for different types of diaphragms. For example, metal diaphragms need to be formed by stamping, which can produce diaphragms with precise size and shape.

[0051] In the application of pressure sensors, there are high requirements for the stability of the sensor core and the manufacturing quality of the components. As the elastic element in the pressure sensor, the quality and performance of the diaphragm directly affect the working efficiency and accuracy of the sensor. For the manufacturing process of the diaphragm, due to the complexity of material properties and the diversity of processing technology, there is a certain gap between the establishment of theoretical models and actual production.

[0052] Due to various factors, such as fluctuations in the quality of raw materials, insufficient precision of production equipment, and differences in the technical level of operators, the consistency of the finished diaphragm products is poor, affecting the stability of the final product. Even in the high-precision manufacturing process, the performance parameters of the diaphragm, such as sensitivity, linearity, hysteresis, etc., may deviate greatly due to the above-mentioned reasons.

[0053] In addition, for the stamping process of the diaphragm, a stamping experiment is carried out under the selected process parameters, the forming conditions of the diaphragm are recorded, and then the formed diaphragm is measured and tested using precision measuring instruments to evaluate its performance indicators. However, due to the complexity of material properties and the diversity of processing technology through stamping experiments, the diaphragm stamping process has low stability, low efficiency, and low diaphragm forming quality.

[0054] In order to solve the problems of low stability, low efficiency and low quality of diaphragm forming in the diaphragm stamping process, some embodiments of the present application provide a data-driven diaphragm stamping process generation method, such as Figure 1 As shown, including:

[0055] S100: Build a test model.

[0056] The test model is a model constructed based on an algorithm or a mapping relationship and is used to predict or simulate stamping forming results. In this embodiment, the test model is obtained through a first mapping relationship, which at least includes a dynamic equation, an elastic-plastic mechanics equation, a material model, a scale effect, and a forming mechanism.

[0057] The dynamic equation describes the law of motion under the action of force. In the process of diaphragm stamping, the dynamic equation is used to describe the interaction between the punch and the diaphragm, including dynamic parameters such as velocity and acceleration. The elastic-plastic mechanics equation describes the elastic deformation and plastic deformation behavior of the material under the action of force. In the process of diaphragm stamping, the elastic-plastic mechanics equation is used to describe the deformation behavior of the diaphragm under the action of force, including elastic deformation and plastic deformation.

[0058] The material model describes the physical and mechanical properties of the material, including elastic modulus, yield strength, Poisson's ratio, etc. In the diaphragm stamping process, the material model is used to describe the characteristics of the diaphragm material, which affects the stress and strain distribution during the molding process. The scale effect describes the difference in the behavior of the material at different scales, including microscale and macroscale. In the diaphragm stamping process, the scale effect is used to consider the deformation behavior of the diaphragm at different scales, especially the special effects at the micro-nano scale.

[0059] The forming mechanism describes the physical and chemical changes of materials during the forming process. In the diaphragm stamping process, the forming mechanism is used to understand the deformation and damage mechanism of the diaphragm during the stamping process, including plastic flow, crack propagation, etc.

[0060] Through dynamic equations, elastic-plastic mechanics equations, material models, scale effects and forming mechanisms, a comprehensive mathematical model can be established to accurately describe the physical and mechanical behaviors of the diaphragm stamping process.

[0061] To establish a test model, in some embodiments, a first mapping relationship F is first established, and then a first input parameter is obtained, and a test model is constructed according to the first mapping relationship and the input parameter. The first input parameter is an initial parameter for constructing a test model, and includes at least a diaphragm wave number m, a diaphragm wave depth h, a strip thickness a, a punch stroke l, a punch speed v, a punch pressure p, and a polyurethane soft mold hardness H. The first input parameter is an input of the test model, and is used to predict the output result during the diaphragm stamping process.

[0062] Among them, the diaphragm wave number is the number of diaphragm waveforms, the diaphragm wave depth is the depth of the diaphragm waveform, the strip thickness is the thickness of the strip, the punch stroke is the moving distance of the punch, and the punch speed is the moving speed of the punch. The punch pressure is the pressure applied by the punch, and the polyurethane soft mold hardness is the hardness of the soft mold. The first input parameter is composed of a parameter matrix [m,h,a,l,v,p,H]T .

[0063] Exemplarily, by combining dynamic equations, elastoplastic mechanics equations, material models, scale effects and forming mechanisms, a first mapping relationship is established to determine the diaphragm wave number, diaphragm wave depth, strip thickness, punch stroke, punch speed, punching pressure and polyurethane soft mold hardness. Based on the first mapping relationship and the first input parameter, a mathematical model capable of predicting the diaphragm stamping process is constructed.

[0064] S200: Acquire stamping information according to the test model.

[0065] The stamping information includes stamping die information and process information. Based on the constructed test model, the stamping die information and the diaphragm stamping process information are generated respectively. The stamping die information includes various parameter information of the upper die and the lower die of the die, such as die shape, material, size, punch waveform size and other information; the process information includes stamping speed, stamping pressure, die temperature and other information.

[0066] S300: Obtaining experimental results and output parameters.

[0067] The experimental results are obtained by conducting experiments on diaphragms prepared by stamping forming information, that is, the finished molds and diaphragms are processed by stamping mold information and process information, and then the diaphragms are tested. After the experiment, the experimental results are recorded, including residual stress R, springback Q, and warpage S. The experimental results provide experimental data in the actual stamping process.

[0068] The output parameters are the parameter matrix [R, Q, S] composed of the residual stress, springback and warpage output by the test model after the first input parameters are input into the test model. T ,The output parameters provide the prediction results of the test model for comparison and analysis with the experimental results.

[0069] Since the input of the test model is [m,h,a,l,v,p,H] T , the output is [R,Q,S] T , then the test model can be expressed by the following formula:

[0070]

[0071] For the experimental results and output parameters, the residual stress is detected by XRD (X-ray diffraction) equipment. Residual stress is the stress state remaining inside the stamped diaphragm after testing. It is one of the indicators to measure the stamping quality of the diaphragm and affects the stability and service life of the diaphragm.

[0072] The diaphragm springback and warpage are detected by confocal microscopy. The springback is the shape change of the diaphragm due to the elastic recovery of the material after stamping. The springback is used to evaluate the stamping waveform accuracy and dimensional stability. The warpage is the degree of bending or distortion on the diaphragm surface due to uneven stress distribution inside the material or unreasonable mold design during the stamping process. The warpage directly affects the stability of the pressure sensor core, the appearance quality of the diaphragm and the packaging performance.

[0073] S400: Calculate adjustment parameters to iteratively optimize the first mapping relationship and the first input parameter through the adjustment parameters to obtain a simulation model.

[0074] The adjustment parameter is calculated based on the experimental results and the output parameters, and is used to correct the test model and the first input parameter. In some embodiments, the adjustment parameter is the difference between the experimental results and the output parameters. Specifically, the first residual stress, the first springback, and the first warpage of the experimental results, and the second residual stress, the second springback, and the second warpage of the output parameters are obtained, and then the adjustment parameter is calculated.

[0075] The adjustment parameters include a stress adjustment parameter, a springback adjustment parameter, and a warpage adjustment parameter, wherein the stress adjustment parameter is the difference between the first residual stress and the second residual stress, the springback adjustment parameter is the difference between the first springback and the second springback, and the warpage adjustment parameter is the difference between the first warpage and the second warpage. The first mapping relationship and the first input parameter are corrected using the calculated adjustment parameters, and the prediction accuracy of the test model and the stability of the process are gradually improved through iterative optimization.

[0076] The first driving matrix [R 1 ,Q 1 ,S 1 ] T , so as to correct the first mapping relationship and the first input parameter through the first driving matrix.

[0077] In some embodiments, a first driving matrix is ​​obtained, and then the first mapping relationship is corrected by the first driving matrix to obtain a second mapping relationship, and the first input parameter is corrected by the first driving matrix to obtain a second input parameter. When the number of iterations is greater than or equal to the first number, the first number is the number determined by the test model, and the first number is at least six times. Taking six times as an example, the residuals of the six times all converge to the standard value, and the iteration ends.

[0078] Among them, the first mapping relationship F is corrected to the second mapping relationship F1, F1 is respectively the second residual stress R1, the second springback Q1 and the second warpage S1; the first input parameters are respectively corrected to the diaphragm wave number m1, the diaphragm wave depth h1, the strip thickness a1, the punch stroke l1, the punch speed v1, the punch pressure p1, the polyurethane soft mold hardness H1, and the output parameters at this time, that is, the output parameters converge to the primary driving matrix, that is, see the following formula:

[0079]

[0080] Based on the second mapping relationship F1 after a correction and the first input parameter matrix [m1,h1,a1,l1,v1,p1,H1] T ,by Figure 2 For example, repeat processes 2 to 4 to obtain the finished diaphragm after one iteration and the secondary data driving matrix, that is, the second driving matrix is ​​[R2, Q2, S2] T , and then obtain the updated second mapping relationship F2 and the second input parameter matrix [m2,h2,a2,l2,v2,p2,H2] T Repeat the iteration n times, that is, see the following formula:

[0081]

[0082] When the n-time data driving matrix meets the convergence criterion, that is, the residuals of the last five times converge to the standard value δ, the iteration ends, see the following formula:

[0083]

[0084] In the formula, δ is the standard value, R is the residual stress, Q is the springback, S is the warpage, and n is n, n-1, n-2, n-3, n-4, n-5.

[0085] By obtaining the first driving matrix and performing iterative optimization, the simulation model and input parameters of the diaphragm stamping process can be gradually optimized to improve the prediction accuracy of the model and the stability of the process.

[0086] To further improve the prediction accuracy of the model and the stability of the process, in some embodiments, a third mapping relationship is established, and a sensitivity analysis is performed based on the third mapping relationship to obtain key parameters, and then a sorted list is generated for the key parameters, wherein the sorted list is a parameter list generated after sorting by sensitivity coefficients; priority parameters with high sorting priority are extracted from the sorted list, and the first mapping relationship is corrected by the first driving matrix and the priority parameters to obtain the second mapping relationship.

[0087] The third mapping relationship is a mapping relationship between the first input parameter and the output parameter, and the mapping relationship between the first input parameter and the output parameter can be established by using regression analysis, neural network or other machine learning methods.

[0088] Sensitivity analysis is a method to evaluate the influence of the first input parameter on the output result. It can be done through local sensitivity analysis or global sensitivity analysis. Local sensitivity analysis, for example, first-order derivative method, difference method, etc. Global sensitivity analysis, for example, variance-based sensitivity analysis, identifies the key parameters that have the greatest influence on the output parameter. Sensitivity analysis calculates the sensitivity coefficient by changing the value of a certain parameter and observing the change in the output result.

[0089] For example, through local sensitivity analysis, each parameter is changed slightly, the change of the output result is recorded, and then the sensitivity coefficient is calculated, that is, the sensitivity coefficient S of the diaphragm wave number is calculated respectively. m , the sensitivity coefficient S of the membrane wave depth h , the sensitivity coefficient S of the strip thickness a , Sensitivity coefficient S of punch stroke l , the sensitivity coefficient of punch speed S v , Sensitivity coefficient of punching force S p And polyurethane soft mold hardness S H If it is found through sensitivity analysis that the diaphragm wave depth h has the greatest influence on the residual stress R, then the first driving matrix and the diaphragm wave depth h are used to correct the first mapping relationship.

[0090] The priority parameter may be the first input parameter with the highest priority in the sorted list, or the first input parameter with the highest priority in the sorted list, or an interaction term between parameters.

[0091] The diameter of the diaphragm is between 18 and 25 mm, and the thickness is relatively thin, between 25 and 30 μm. Since the quality of the diaphragm affects the core stability of the pressure sensor, different stamping processes can be set for the diaphragm according to different application environments. In some embodiments, the setting information of the diaphragm can be obtained, and then the scene interval can be matched according to the setting information to assign weights to the residual stress, rebound amount and warpage through the scene interval, and the weighted error is calculated according to the output parameters and the weight, and the adjustment parameters are calculated according to the weighted error.

[0092] Among them, the setting information includes diameter information and thickness information. The scene interval includes the interval limit of the diameter information and the thickness information, which is used to assign weights to the residual stress, springback and warpage. Each scene interval corresponds to a different weight allocation strategy. In different scene intervals, different weights can be assigned to the residual stress, springback and warpage.

[0093] In the application, several scene intervals are defined according to the diameter and thickness information of the diaphragm. For example, scene 1, diameter 18mm-20mm, thickness 25μm-27μm, scene 2, diameter 21mm-23mm, thickness 27μm-30μm, scene 3, diameter 24mm-25mm, arbitrary thickness, the weight for residual stress is w1, the weight for rebound is w2, the weight for warpage is w3, and the weight distribution measurement is specified, for example, scene 1, w1=0.2, w1=0.5, w1=0.3, scene 2, w1=0.4, w1=0.3, w1=0.3, scene 3, w1=0.5, w1=0.25, w1=0.25.

[0094] After obtaining the output parameters, the weighted error is calculated. The weighted error is the product of the absolute difference between the experimental results and the output parameters and the weight. Finally, the products of the residual stress, springback and warpage are added to obtain a weighted error. The weighted error is used to calculate the adjustment parameters, where the adjustment parameters are the product of the parameters and the weighted error.

[0095] In each iteration, the adjustment parameters are used to optimize the first mapping relationship and the input parameters. The first mapping relationship is updated first, and then the first input parameters are updated. The new parameters are input into the model to verify the validity of the boundary conditions to ensure that the new model continues to meet the indicators and to determine whether the weighted error is reduced to an acceptable value. If not, return and repeat the above steps.

[0096] In summary, after iterative optimization, an optimized test model, namely, a simulation model, is obtained.

[0097] S500: Generate a diaphragm stamping process plan through a simulation model.

[0098] The simulation model is used to generate a final diaphragm stamping process solution, wherein the diaphragm stamping process solution may include a solution such as the first input parameter. The generated diaphragm stamping process solution can achieve the expected performance indicators in actual production, thereby improving production efficiency and product quality.

[0099] Based on the above-mentioned data-driven diaphragm stamping forming process generation method, some embodiments of the present application also provide a data-driven diaphragm stamping forming process generation system, such as Figure 3 As shown, including:

[0100] A construction unit, configured to construct a test model, wherein the test model is obtained through a first mapping relationship;

[0101] an acquisition unit, configured to acquire stamping forming information according to the test model, wherein the stamping forming information includes stamping die information and process information; and to acquire experimental results and output parameters, wherein the experimental results are the results of testing through experiments, and the output parameters are the parameter matrix composed of residual stress, springback and warpage output by the test model when the first input parameter is input into the test model;

[0102] a processing unit, configured to calculate an adjustment parameter, so as to iteratively optimize the first mapping relationship and the first input parameter by using the adjustment parameter to obtain a simulation model;

[0103] A generating unit is used to generate a diaphragm stamping process plan through the simulation model.

[0104] The present application provides a method and system for generating a diaphragm stamping process based on data-driven, the method comprising: constructing a test model, the test model is obtained through a first mapping relationship, and stamping information is obtained according to the test model, the stamping information includes stamping die information and process information; obtaining experimental results and output parameters, the experimental results are the results obtained by conducting experiments on diaphragms prepared by stamping information, and the output parameters are the parameter matrix composed of residual stress, springback and warpage output by the test model when the first input parameter is input into the test model; calculating adjustment parameters, so as to iteratively optimize the first mapping relationship and the first input parameter by adjusting the parameters to obtain a simulation model; generating a diaphragm stamping process plan through the simulation model. The method uses the adjustment parameters to correct the first mapping relationship and the first input parameter, gradually optimizes the test model, improves the prediction accuracy of the model and the stability of the diaphragm stamping process, and thus improves the diaphragm forming quality.

[0105] Similar parts between the embodiments provided in this application can be referenced to each other. The specific implementation methods provided above are only a few examples under the general concept of this application and do not constitute a limitation on the protection scope of this application. For those skilled in the art, any other implementation methods expanded based on the scheme of this application without creative work belong to the protection scope of this application.

Claims

1. A data-driven diaphragm stamping process generation method, characterized in that: include: Constructing a test model, wherein the test model is obtained through the first mapping relationship; Acquire stamping information according to the test model, wherein the stamping information includes stamping die information and process information; Acquire experimental results and output parameters, wherein the experimental results are the results of experiments on the diaphragm prepared by the stamping forming information, and the output parameters are the parameter matrix composed of residual stress, springback and warpage output by the test model when the first input parameter is input into the test model; Calculating adjustment parameters to iteratively optimize the first mapping relationship and the first input parameter by using the adjustment parameters to obtain a simulation model; A diaphragm stamping process plan is generated through the simulation model.

2. The data-driven diaphragm stamping process generation method according to claim 1 is characterized in that: The construction of the test model comprises: Establishing a first mapping relationship; Get the first input parameter; A test model is constructed according to the first mapping relationship and input parameters.

3. The data-driven diaphragm stamping process generation method according to claim 2 is characterized in that: The first mapping relationship at least includes a dynamic equation, an elastic-plastic mechanics equation, a material model, a scale effect, and a forming mechanism; The first input parameter includes at least the diaphragm wave number, the diaphragm wave depth, the strip thickness, the punch stroke, the punch speed, the punch pressure and the hardness of the polyurethane soft mold.

4. The data-driven diaphragm stamping process generation method according to claim 3 is characterized in that: The calculation adjustment parameters include: Obtaining a first residual stress, a first springback amount, and a first warpage of the experimental results, and outputting parameters of a second residual stress, a second springback amount, and a second warpage; Calculate adjustment parameters, the adjustment parameters including stress adjustment parameters, springback adjustment parameters and warpage adjustment parameters, the stress adjustment parameter is the difference between the first residual stress and the second residual stress, the springback adjustment parameter is the difference between the first springback and the second springback, and the warpage adjustment parameter is the difference between the first warpage and the second warpage.

5. The data-driven diaphragm stamping process generation method according to claim 4 is characterized in that: The iterative optimization of the first mapping relationship and the first input parameter by using the adjustment parameter includes: Acquire a first driving matrix, where the first driving matrix is ​​a matrix composed of a residual stress adjustment parameter, a springback adjustment parameter, and a warpage adjustment parameter; Correcting the first mapping relationship by using the first driving matrix to obtain a second mapping relationship, and correcting the first input parameter by using the first driving matrix to obtain a second input parameter; When the number of iterations is greater than or equal to the first number, the iterative optimization is ended by the following formula: In the formula, δ is the standard value, R is the residual stress, Q is the springback, S is the warpage, and n is n, n-1, n-2, n-3, n-4, n-5.

6. The data-driven diaphragm stamping process generation method according to claim 5 is characterized in that: The correcting the first mapping relationship by using the first driving matrix to obtain a second mapping relationship includes: Establishing a third mapping relationship, where the third mapping relationship is a mapping relationship between the first input parameter and the output parameter; Performing a sensitivity analysis according to the third mapping relationship to obtain key parameters, wherein the sensitivity analysis is used to evaluate the degree of influence of the first input parameter on the output result; Generate a sorted list of the key parameters, wherein the sorted list is a parameter list generated after sorting by sensitivity coefficients; Extracting a priority parameter, wherein the priority parameter is a first input parameter with a high sorting priority in the sorting list; The first mapping relationship is corrected by using the first driving matrix and the priority parameter to obtain a second mapping relationship.

7. The data-driven diaphragm stamping process generation method according to claim 1 is characterized in that: The calculation adjustment parameters include: Acquiring setting information of the diaphragm, wherein the setting information includes diameter information and thickness information; Matching a scene interval according to the setting information to assign weights to residual stress, springback, and warpage through the scene interval, wherein the scene interval includes interval limitations for diameter information and thickness information; A weighted error is calculated according to the output parameters and the weights, and an adjustment parameter is calculated according to the weighted error.

8. A data-driven diaphragm stamping process generation system, characterized in that: include: A construction unit, configured to construct a test model, wherein the test model is obtained through a first mapping relationship; an acquisition unit, configured to acquire stamping information according to the test model, wherein the stamping information includes stamping die information and process information; and to acquire experimental results and output parameters, wherein the experimental results are the results of experiments on the diaphragm prepared by the stamping information, and the output parameters are the parameter matrix composed of residual stress, springback and warpage output by the test model when the first input parameter is input into the test model; a processing unit, configured to calculate an adjustment parameter, so as to iteratively optimize the first mapping relationship and the first input parameter by using the adjustment parameter to obtain a simulation model; A generating unit is used to generate a diaphragm stamping process plan through the simulation model.

Images