Stamping parameter optimization control system based on motor stator and rotor processing

Abstract

The present invention discloses a stamping parameter optimization control system based on motor stator and rotor processing, which relates to the field of stamping control technology, including: determining key feature points according to motor stator and rotor design drawings, establishing a NURBS curve optimization model; determining curve control points, constructing an initial stamping curve; constructing a simulation space, simulating the initial stamping curve, and predicting the springback amount; judging whether the springback amount prediction result meets the springback constraint requirement, and optimizing the stamping curve when it does not meet the springback constraint requirement; simulating the optimized stamping curve, and predicting the springback amount, and making a springback judgment according to the springback amount prediction result, when the springback constraint requirement is met, determining the stamping curve, and generating stamping control parameters; when it does not meet the springback constraint requirement, continuing to adjust and optimize the control points of the stamping curve until the springback constraint requirement is met. The present invention solves the technical problem of low efficiency and precision of stamping processing in the prior art, and achieves the technical effect of improving the efficiency and precision of stamping processing.

Description

Technical Field

[0001] The invention relates to the technical field of stamping control, and in particular to a stamping parameter optimization control system based on motor stator and rotor processing. Background Art

[0002] With the continuous development of motor manufacturing technology and the growing market demand, the processing quality and production efficiency of motor stators and rotors, as the core components of motors, directly affect the overall performance and manufacturing cost of the motors. In the processing of motor stators and rotors, the stamping process is one of the most important links, which determines the accuracy of the shape, size and performance of the stators and rotors. However, with the increasing complexity of motor stator and rotor design and the continuous expansion of production scale, traditional stamping processing methods have gradually exposed problems such as slow response speed and difficulty in ensuring accuracy. Summary of the invention

[0003] The present application provides a stamping parameter optimization control system based on motor stator and rotor processing, which is used to solve the technical problems of low efficiency and precision of stamping processing in the prior art.

[0004] In view of the above problems, the present application provides a stamping parameter optimization control system based on motor stator and rotor processing.

[0005] The present application provides a stamping parameter optimization control system based on motor stator and rotor processing, the system comprising:

[0006] A key feature point determination unit, wherein the key feature point determination unit determines the key feature points according to the design drawings of the motor stator and rotor, and establishes a NURBS curve optimization model; a curve control point determination unit, wherein the curve control point determination unit determines the curve control points according to the NURBS curve optimization model, and constructs an initial stamping curve; a stamping parameter data acquisition unit, wherein the stamping parameter data acquisition unit constructs a simulation space based on the stamping process and the stamping parameter data acquired, including the blank holder force, the die gap, and the friction coefficient, wherein the simulation space includes a springback prediction module; a stamping curve simulation unit, wherein the stamping curve simulation unit simulates the initial stamping curve using the simulation space, and predicts the springback amount during the stamping process through the springback prediction module; and a first springback constraint requirement judgment unit. The first springback constraint requirement judgment unit judges whether the springback amount prediction result meets the springback constraint requirement. When it does not meet the springback constraint requirement, the control points of the NURBS curve are adjusted according to the springback amount prediction result to optimize the stamping curve; the second springback constraint requirement judgment unit, the second springback constraint requirement judgment unit simulates the optimized stamping curve using the simulation space, predicts the springback amount during the stamping process through the springback prediction module, and makes a springback judgment based on the springback amount prediction result. When the springback constraint requirement is met, the stamping curve is determined and the stamping control parameters are generated; the control point adjustment and optimization unit, when the springback constraint requirement is not met, continues to adjust and optimize the control points of the stamping curve until the springback amount prediction result corresponding to the stamping curve meets the springback constraint requirement.

[0007] The technical solution provided in this application has at least the following technical effects or advantages:

[0008] The present application determines key feature points and establishes a NURBS curve optimization model based on the design drawings of the motor stator and rotor; determines the curve control points based on the NURBS curve optimization model and constructs an initial stamping curve; constructs a simulation space based on the stamping process and the stamping parameter data collected, including the blank holder force, the die gap, and the friction coefficient, wherein the simulation space includes a springback prediction module; uses the simulation space to simulate the initial stamping curve, and predicts the springback amount during the stamping process through the springback prediction module; determines whether the springback amount prediction result meets the springback constraint requirement, and when it does not meet the springback constraint requirement, adjusts the control points of the NURBS curve according to the springback prediction result to optimize the stamping curve; uses the simulation space to simulate the optimized stamping curve, and predicts the springback amount during the stamping process through the springback prediction module, and makes a springback judgment according to the springback prediction result. When the springback constraint requirement is met, the stamping curve is determined and the stamping control parameters are generated; when the springback constraint requirement is not met, the control points of the stamping curve are continuously adjusted and optimized until the springback amount prediction result corresponding to the stamping curve meets the springback constraint requirement. The present invention solves the technical problems of low efficiency and precision of stamping processing in the prior art. By real-time monitoring of key parameters in the stamping process, the stamping process is accurately controlled, the stamping parameters are optimized and adjusted, and the technical effect of improving the efficiency and precision of the stamping process is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0010] Figure 1 A schematic diagram of the structure of a stamping parameter optimization control system based on motor stator and rotor processing provided in an embodiment of the present application;

[0011] Figure 2 A schematic flow chart of a stamping parameter optimization control method based on motor stator and rotor processing provided in an embodiment of the present application.

[0012] Explanation of the accompanying drawings: key feature point determination unit 11, curve control point determination unit 12, stamping parameter data acquisition unit 13, stamping curve simulation unit 14, first springback constraint requirement judgment unit 15, second springback constraint requirement judgment unit 16, control point adjustment and optimization unit 17. DETAILED DESCRIPTION

[0013] The present application provides a stamping parameter optimization control system based on motor stator and rotor processing, aiming to solve the technical problems of low efficiency and low precision of stamping processing in the prior art. By real-time monitoring of key parameters in the stamping process, the stamping process is precisely controlled, the stamping parameters are optimized and adjusted, and the technical effect of improving the efficiency and precision of the stamping process is achieved.

[0014] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0015] It should be noted that any variations of the terms "include" and "have" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices.

[0016] Embodiment 1

[0017] like Figure 1 As shown, the present application provides a stamping parameter optimization control system based on motor stator and rotor processing, which is used to perform the following Figure 2 The stamping parameter optimization control method based on motor stator and rotor processing shown in the figure, the system includes:

[0018] The key feature point determination unit 11 determines the key feature points according to the design drawings of the motor stator and rotor, and establishes a NURBS curve optimization model.

[0019] In an embodiment of the present application, key feature information is extracted from the design drawings of the motor stator and rotor. These feature information include preset key features such as winding slots, ventilation holes, and terminal blocks. Based on the design drawings, the preset key features are identified and located, and the coordinate data of the key feature points are determined. The specific positions of these key feature points in space are determined by the geometric information and dimensioning on the drawings. The coordinate data of the identified key feature points are used to determine the control points of the NURBS curve. According to these control points, the corresponding weight factors are configured to establish a NURBS curve optimization model.

[0020] Furthermore, in the system provided by the embodiment of the application, the key feature points are determined according to the design drawings of the stator and rotor of the motor, and the NURBS curve optimization model is established, including:

[0021] A key feature point coordinate determination unit, which determines the coordinate data of the key feature point according to the key feature identification information; a control point weight factor configuration unit, which determines the control point according to the coordinate data of the key feature point, configures the control point weight factor, and constructs a NURBS curve model; a process parameter acquisition unit, which acquires stamping process parameters, including stamping stroke, speed, acceleration, and pressure; a first fitting unit, which fits the NURBS curve based on the stamping process parameters and establishes the NURBS curve optimization model.

[0022] In the embodiment of the present application, first, CAD software such as AutoCAD and SolidWorks is used to read the design drawings of the motor stator and rotor. Key features are automatically detected and located by image processing and pattern recognition technology. The design drawings are preprocessed using image processing libraries such as OpenCV, including grayscale and binarization processing, and then edge detection algorithms and morphological operations are used to identify key features such as winding slots, vents, and terminals. According to the identification information of the preset key features, the coordinate data of the key feature points are determined. The geometric operation function in the CAD software is used to obtain the specific coordinates of the identified key feature points in the design drawings. Using the NURBS curve theory, the control points of the NURBS curve are determined based on the coordinates of the identified key feature points. The weight factor of each control point is manually configured through empirical data, or automatically determined by an optimization algorithm such as a genetic algorithm or a gradient descent algorithm. The key feature points are used as initial control points and input into the NURBS curve model, and the weight factors of each control point are configured according to the design requirements and process requirements. The initial NURBS curve model is constructed using the determined control points and the configured weight factors. Using mathematical modeling tools, such as MATLAB, a NURBS curve model is constructed according to the determined control points and weight factors.

[0023] Next, the stamping process parameters are collected, including stamping stroke, speed, acceleration, pressure, etc. These parameters are obtained in real time through the sensors of the stamping equipment. Based on the collected stamping process parameters, the NURBS curve is fitted. The control point data is converted into the actual initial stamping curve using curve fitting technology. During the curve fitting process, mathematical software or programming tools such as MATLAB are used to implement it, and preliminary optimization is performed in combination with the stamping process parameters. In the MATLAB environment, the least squares method is used for curve fitting, and the control points and weights of the NURBS curve are adjusted to make the curve best match the stamping parameters. The initial NURBS curve model is adjusted through iterative optimization methods to ensure that the curve shape meets the design requirements and has good process adaptability. During the optimization process, optimization techniques such as gradient descent and genetic algorithms are used to gradually adjust the position and weight of the control points to optimize the curve shape. The NURBS curve optimization model is established through the above process.

[0024] Furthermore, in the system provided by the embodiment of the application, the configuration control point weight factor includes:

[0025] An abnormal impact evaluation unit, wherein the abnormal impact evaluation unit evaluates the abnormal impact of the winding slots, ventilation holes, and terminals according to historical abnormal cases, and obtains abnormal evaluation values ​​composed of the positions of the winding slots, ventilation holes, and terminals; an initial weight factor configuration unit, wherein the initial weight factor configuration unit configures the initial weight factors of the control points of the winding slots, ventilation holes, and terminals based on the abnormal evaluation values; a demand curve data acquisition unit, wherein the demand curve data acquisition unit determines the design requirements according to the design drawings, and obtains the shape and size of the demand curve; a deviation calculation unit, wherein the deviation calculation unit performs curve fitting based on the initial weight factors, and calculates the deviation amount by fitting the curve using the shape and size of the demand curve; an optimization adjustment unit, wherein the optimization adjustment unit optimizes and adjusts the initial weight factors based on the deviation amount, until the deviation amount from the shape and size of the demand curve meets the minimum error requirement, and determines the control point weight factors.

[0026] In the embodiment of the present application, historical abnormal cases are first collected, and database technology such as SQL database is used to store and manage historical abnormal data, which includes processing parameters, processing results and abnormal situations of each key feature point in each processing. Use SQL query to extract abnormal case data related to winding slots, vents and terminals from the database. Then clean and preprocess the collected historical abnormal data to ensure the accuracy and completeness of the data. Then perform frequency analysis to count the frequency of abnormal occurrence of each key feature point. By calculating the number of abnormal occurrences of each feature point in historical cases, evaluate the frequency of its abnormal occurrence. Then perform correlation analysis to evaluate the degree of abnormal impact of each key feature point. By calculating the correlation between the number of abnormal occurrences of the feature point and the overall processing quality, determine its abnormal impact. This process is completed by using statistical analysis tools. The results of frequency analysis and correlation analysis are combined to generate an abnormal evaluation value for each key feature point. The abnormal evaluation value is expressed as a comprehensive score to reflect the abnormal risk of the feature point. Combine the frequency and correlation analysis results by weighted average or product to form an abnormal evaluation value.

[0027] Next, the initial weight of each control point is calculated based on its abnormal evaluation value. Control points with high evaluation values ​​are assigned higher weights, while control points with low evaluation values ​​are assigned lower weights. The initial weight of each control point is proportionally allocated according to the size of the abnormal evaluation value. The higher the abnormal evaluation value, the weight of the control point should be increased accordingly.

[0028] Then, determine the design requirements according to the design drawings to obtain the shape and size of the demand curve. The design drawings contain detailed geometric shape and size requirements. The geometric analysis function of the CAD software is used to extract the curve shape and size information in the design drawings. Mark and identify key feature points in the CAD software, including winding slots, ventilation holes, terminal blocks, etc. Use the tools of the CAD software to mark these key feature points and record their specific geometric coordinates. Then use the measurement function of the CAD software to extract the shape and size information of the demand curve. The specific steps include selecting the demand curve, selecting the curve to be extracted in the CAD software, using the measurement tool to extract the key point coordinates of the curve, the curve length, the radius of curvature and other parameters, and recording the key dimensions of the curve, such as length, width, height, radius, etc.

[0029] Next, curve fitting is performed based on the initial weights, and the deviation of the fitting curve is calculated using the shape and size of the demand curve. Use mathematical modeling tools such as MATLAB to build a NURBS curve model. Input the coordinates of the control points and the corresponding initial weights, and generate the initial fitting curve through the NURBS algorithm. Next, use the shape and size information of the demand curve to compare the generated initial fitting curve with the demand curve. Extract the key size and shape parameters of the demand curve as a benchmark and compare them with the corresponding parameters of the initial fitting curve. The specific method includes calculating the coordinate differences of key points, the differences in geometric parameters such as curve length and curvature radius. Next, the deviation is calculated, and the deviation between the fitting curve and the demand curve is evaluated by numerical comparison and graphic overlay. This process uses numerical calculation tools such as MATLAB to calculate the coordinate differences of each key point, as well as the differences in geometric parameters such as curve length and curvature radius.

[0030] The deviation is used as the optimization target and the minimum error requirement is set. The optimization target is the sum of the coordinate differences of each key point, the difference in curve length, the difference in curvature radius, etc. Initialize the optimization parameters based on the initial weights and the calculated deviations. The weights are gradually adjusted through an iterative process to reduce the deviations. For the gradient descent algorithm, the gradient of the deviation is calculated for each iteration, and the weights are adjusted based on the gradient. For the genetic algorithm, each iteration generates a new combination of weights through selection, crossover, and mutation operations, and calculates the corresponding deviations. After each iteration, evaluate the deviation between the fitting curve and the demand curve corresponding to the current weight combination. If the deviation meets the minimum error requirement, terminate the optimization process and determine the final control point weights; otherwise, continue to iterate. When the optimization process terminates, determine the current weight combination as the control point weight.

[0031] Furthermore, in the system provided by the embodiment of the application, the optimizing and adjusting the initial weight factor based on the deviation includes:

[0032] An analytical positioning unit, wherein the analytical positioning unit performs analytical positioning based on the deviation amount, and determines the node control target deviation and curve smoothness evaluation information of the curve control point; a target optimization unit, wherein the target optimization unit establishes a node control target benefit evaluation function and a smoothness benefit evaluation function, and performs target optimization with the node control target deviation and the curve smoothness evaluation information respectively, to obtain a first optimization scheme and a second optimization scheme, wherein the first optimization scheme is the optimal node control target benefit, and the second optimization scheme is the optimal curve smoothness benefit; a benefit balancing unit, wherein the benefit balancing unit uses a benefit balancing algorithm to balance the benefits of the first optimization scheme and the second optimization scheme, and determines an optimization adjustment scheme for the initial weight factor.

[0033] In an embodiment of the present application, the deviation between the fitting curve and the demand curve is calculated by a numerical calculation tool. The coordinates of each key point of the fitting curve and the demand curve are compared, and the coordinate difference of each control point is calculated, and these differences are the node control target deviation. The calculated node control target deviation is analyzed and located to determine the specific deviation of each control point. Through the data analysis tool, the deviations of all control points are summarized and analyzed to determine which control points have the greatest impact on the overall curve. The weights of these control points are adjusted preferentially to reduce the deviation of the overall curve. In the process of analytical positioning, the smoothness of the curve is evaluated and curve smoothness evaluation information is generated. The difference between adjacent control points is calculated by a numerical calculation tool. A smaller difference between adjacent control points indicates that the curve is relatively smooth, and a larger difference between adjacent control points indicates that the curve has a larger change. The coordinate differences of adjacent control points are summarized and analyzed to generate curve smoothness evaluation information.

[0034] In order to make optimization adjustments, the node control target benefit evaluation function and the smoothness benefit evaluation function are established, and the node control target deviation and the curve smoothness evaluation information are used as the targets for optimization, respectively, to obtain the first optimization scheme and the second optimization scheme. Specifically, the node control target benefit evaluation function is first established. The goal of this function is to minimize the deviation of each control point, thereby improving the fitting accuracy of the curve. The node control target benefit evaluation function is defined as the sum of the squares of the deviations of each control point. This evaluation function is constructed using numerical calculation tools such as MATLAB. The node control target benefit evaluation function is expressed as the sum of the squares of the deviations of all control points, which can more sensitively reflect larger deviations and make corresponding adjustments. Next, the smoothness benefit evaluation function is established. The goal of this function is to maintain the smoothness of the curve and prevent the curve from becoming unsmooth due to overfitting. The smoothness benefit evaluation function is defined as the sum of the squares of the differences between adjacent control points. This evaluation function is constructed using numerical calculation tools such as MATLAB. The smoothness benefit evaluation function ensures the smooth transition of the curve by evaluating the differences between adjacent control points. After the evaluation function is constructed, the target optimization is performed to obtain the first optimization scheme and the second optimization scheme, respectively. The first optimization scheme is optimized based on the node control target benefit evaluation function, with the goal of minimizing the node control target deviation. The second optimization scheme is optimized based on the smoothness benefit evaluation function, with the goal of maintaining the smoothness of the curve. During the optimization process, optimization techniques such as gradient descent algorithm or genetic algorithm are used for iterative adjustment to ensure that the value of the evaluation function reaches the optimal value. Finally, the benefit balance algorithm is used to balance the benefits of the first optimization scheme and the second optimization scheme, and determine the optimization adjustment scheme of the initial weight factor. The benefit balance algorithm comprehensively considers the fitting accuracy and curve smoothness, balances the benefits of the control point weight factors of the first optimization scheme and the second optimization scheme, and obtains the final optimization adjustment scheme.

[0035] Furthermore, in the system provided in the embodiment of the application, the use of the interest balance algorithm to balance the interests of the first optimization scheme and the second optimization scheme also includes:

[0036] A precision constraint condition acquisition unit, the precision constraint condition acquisition unit acquires the precision constraint condition of each curve control point; a constraint coefficient determination unit, the constraint coefficient determination unit determines the constraint coefficient of each node control target based on the precision constraint condition; a constraint condition determination unit, the constraint condition determination unit determines the curve smoothing constraint node and smoothing constraint value according to the design drawing; a benefit configuration constraint condition configuration unit, the benefit configuration constraint condition configuration unit configures the benefit configuration constraint condition according to the constraint coefficient of the node control target, the curve smoothing constraint node and the smoothing constraint value, and adds it to the benefit balancing algorithm.

[0037] In an embodiment of the present application, the design drawings are read by CAD software such as AutoCAD and SolidWorks to extract the precision constraints of each curve control point. These conditions define the maximum deviation allowed for each control point during the fitting process. The constraint coefficient of each node control target is determined based on the precision constraint by a data analysis tool. The constraint coefficient reflects the importance of each control point in meeting the precision requirements. The design drawings are read by CAD software to determine the curve smoothing constraint nodes and smoothing constraint values. The smoothing constraint nodes refer to the control points that need to focus on maintaining smoothness during the curve fitting process, and the smoothing constraint values ​​are the specific requirements of these nodes in terms of smoothness. According to the constraint coefficients of the node control targets, the curve smoothing constraint nodes and the smoothing constraint values, the benefit configuration constraints are configured, and these conditions are added to the benefit balance algorithm. Specifically, an objective function containing these constraints is defined. This objective function combines the deviations of the node control targets and the smoothness of the curve, and the goal is to minimize these deviations and maintain the smoothness of the curve at the same time. The objective function is divided into a node control target deviation part and a smoothness part.

[0038] The node control target deviation part reflects the deviation of each control point. The goal is to minimize these deviations, which is expressed as the sum of the squares of the deviations of each control point. It is expressed as follows:

[0039] ;

[0040] in is the constraint coefficient of the control point, is the deviation of the ith control point, and n is the number of control points.

[0041] The smoothness part reflects the smoothness of the curve. The goal is to maintain the smoothness of the curve. It is expressed as the sum of the squares of the differences between adjacent control points. It is expressed as follows:

[0042] ;

[0043] in is the smoothness index of the jth control point, and n is the number of control points. Combining these two parts, we define the comprehensive objective function J, .in It is a balance coefficient used to weigh the importance of node control target deviation and curve smoothness, and is pre-set by technical experts.

[0044] The curve control point determination unit 12 determines the curve control points according to the NURBS curve optimization model and constructs an initial stamping curve.

[0045] Furthermore, in the system provided in the embodiment of the application, according to the NURBS curve optimization model, the curve control points are determined and the initial stamping curve is constructed, including:

[0046] A curve control point conversion unit, the curve control point conversion unit extracts the size change and shape information of the key features and converts them into curve control points; a curve adjustment unit, the curve adjustment unit performs curve adjustment on the curve control points based on the control point weight factors to determine the shape and path of the stamping curve; a second fitting unit, the second fitting unit takes the shape and path of the stamping curve as the fitting target, uses the stamping process parameters and the curve control points to perform fitting, and obtains the initial stamping curve.

[0047] In an embodiment of the present application, CAD software is used to read the design drawings and extract key features, such as the dimensional changes and shape information of winding slots, ventilation holes, and terminal blocks. This information includes the length, width, height, curvature, etc. of the features. The dimensional changes and shape information of the extracted key features are converted into the coordinates of the control points through data processing tools to obtain the curve control points. Next, these control points are adjusted based on the control point weights. The control point weights determine the degree of influence of each control point on the shape of the curve. According to the previously determined weights, the control points are weighted and their positions are adjusted. Based on the adjusted control points, the shape and path of the stamping curve are determined. Using curve fitting methods such as NURBS, a preliminary stamping curve is generated through numerical calculation tools.

[0048] Then, the shape and path of the stamping curve are used as the fitting target, and the stamping process parameters, such as the blank holder force, die clearance, friction coefficient, etc., are used to fit the curve control points. Numerical optimization algorithms, such as gradient descent and genetic algorithms, are used to iteratively calculate and adjust the position and weight of the control points to make the curve more in line with the actual stamping process requirements. Finally, through the above steps, the initial stamping curve is obtained.

[0049] The stamping parameter data acquisition unit 13 constructs a simulation space based on the stamping process and the acquired stamping parameter data, including the blank holder force, die clearance, and friction coefficient, wherein the simulation space includes a springback prediction module.

[0050] In an embodiment of the present application, the stamping parameter data during the stamping process is monitored and recorded in real time by sensors and data acquisition systems installed on the stamping equipment. These parameters include blank holder force, die gap and friction coefficient. The blank holder force is measured and recorded by a pressure sensor, the die gap is measured and recorded by a displacement sensor, and the friction coefficient is measured and recorded by a friction sensor. Next, the collected data is preprocessed, and the simulation space is constructed using the preprocessed stamping parameters. The simulation space is a virtual environment that can simulate the physical behavior during the stamping process. A simulation model is established using finite element analysis software, and the collected parameter data is input into the model for simulation. The simulation space contains a springback prediction module for predicting the springback of the material during the stamping process. Springback refers to the deformation recovery caused by the release of internal stress after the material is unloaded. The springback prediction module is based on finite element analysis and calculates the springback amount of the material through numerical simulation.

[0051] Furthermore, in the system provided in the embodiment of the application, the simulation space is constructed based on the stamping process and the stamping parameter data collected, including the blank holder force, the die gap, and the friction coefficient, wherein the simulation space includes a springback prediction module, including:

[0052] A data simulation model building unit, wherein the data simulation model building unit collects process data and physical structure of mechanical equipment in the stamping process, and uses a finite element platform to build a data simulation model, including a front-end explicit simulation module and a rear-end invisible solver, wherein the front-end explicit simulation module is used to simulate and adjust stamping process parameters and simulate stamping dynamic behavior, and the rear-end invisible solver calculates and processes the blank holder force, die gap, and friction coefficient through simulation results; a training data set construction unit, wherein the training data set construction unit collects simulation experiment data and actual processing monitoring data, and constructs a training data set, which includes blank holder force, die gap, friction coefficient, and springback identifier; a model training unit, wherein the model training unit uses the training data set to perform supervised training on a neural network model, and when the convergence target of the prediction result is reached, the springback prediction module is obtained, and the springback prediction module is fitted into the rear-end invisible solver.

[0053] In the embodiment of the present application, the process data in the stamping process are first collected. These data include blank holder force, die gap and friction coefficient, which are obtained by sensors. At the same time, the physical structure data of the mechanical equipment are collected. These data include the geometric dimensions of the stamping equipment, the shape of the die and the material properties. Use a 3D scanner or CAD software to model the stamping equipment and the die to obtain detailed geometric dimension data and material properties. Through measuring tools and material testing equipment, determine the shape of the equipment and the physical properties of the material, such as elastic modulus, yield strength, etc. Next, use a finite element platform such as ANSYS and ABAQUS to establish a data simulation model. The model includes a front-end explicit simulation module and a rear-end invisible solver. The front-end explicit simulation module is responsible for simulating the adjustment of stamping process parameters and the dynamic behavior of stamping. The specific steps include creating a geometric model of the stamping parts and the die, defining the physical properties of the material, such as elastic modulus, yield strength, etc., setting boundary conditions and loads, including blank holder force, die gap and friction coefficient, and meshing the geometric model to improve the simulation accuracy. In the front-end explicit simulation module, the simulation adjustment of the stamping process parameters and the simulation of the dynamic behavior of stamping are performed. Through simulation, the deformation behavior and stress distribution of the material during the stamping process are obtained. These simulation results will be passed as input to the post-implicit solver. The post-implicit solver performs further calculations based on the simulation results of the pre-explicit simulation module. The post-implicit solver uses numerical solution algorithms, such as the implicit solution method in finite element analysis, to calculate parameters such as blank holder force, die clearance and friction coefficient.

[0054] The simulation experiment data and actual processing monitoring data are collected through the historical database to construct a training data set. These data include the blank holder force, die gap, friction coefficient and its springback amount identification. The constructed training data set is used to conduct supervised training on the neural network model. When the convergence target of the prediction results is reached, the springback prediction module is obtained. The neural network model is trained using deep learning frameworks such as TensorFlow and Keras. The input parameters are the blank holder force, die gap and friction coefficient, and the output is the springback amount identification of the material. Through repeated training and optimization, the parameters of the neural network are adjusted to ensure that the model can accurately predict the springback of the material. When the neural network model reaches the convergence target of the prediction results, the springback prediction module is fitted into the post-invisible solver.

[0055] Furthermore, the system provided in the application embodiment also includes:

[0056] A sample data set construction unit, wherein the sample data set construction unit takes the blank holder force, die gap, and friction coefficient as targets respectively to construct sample data sets of each target parameter, wherein the sample data set is actual processing monitoring data of the current material extracted using the processing log data of the current motor stator and rotor, including stamping process parameters: stamping stroke, speed, acceleration, and pressure; a solver construction unit, wherein the solver construction unit takes the blank holder force, die gap, and friction coefficient as dependent variables and the stamping process parameters as independent variables, performs relationship fitting learning through the sample data set, and constructs a solver for each target parameter; a post-positioned invisible solver establishment unit, wherein the post-positioned invisible solver establishment unit establishes an association mapping between the solver and the target parameters, generates the post-positioned invisible solver, and fits it into the simulation space.

[0057] In the embodiment of the present application, the actual processing monitoring data of the current material is first extracted by reading the processing log data of the current motor stator and rotor. These data include stamping process parameters, stamping stroke, speed, acceleration and pressure. Data processing tools such as Excel are used to extract and preprocess these data to ensure the accuracy and consistency of the data.

[0058] Next, we construct sample data sets of target parameters, taking blank holder force, die gap and friction coefficient as targets. The sample data sets contain the stamping process parameters and target parameters recorded during the actual processing, namely, blank holder force, die gap and friction coefficient. Specifically, the stamping stroke, speed, acceleration and pressure data at each time point, as well as the corresponding blank holder force, die gap and friction coefficient, are extracted from the processing log.

[0059] Then, the blank holder force, die gap and friction coefficient are used as dependent variables, and the stamping stroke, speed, acceleration and pressure are used as independent variables, and the relationship fitting learning is performed through the sample data set. In this process, machine learning methods such as support vector machines are used for fitting learning to establish a relationship model between independent variables and dependent variables. In this way, a solver for each target parameter is constructed. The solver predicts the corresponding blank holder force, die gap and friction coefficient based on the input stamping process parameters.

[0060] After building the solver, establish an association mapping between the solver and the target parameters. Specifically, by binding the solver to the relevant modules in the simulation space, the simulation space can call the solver to predict the target parameters. For example, in the finite element simulation platform, the solver is integrated into the simulation process through a programming interface or script. In this way, during the simulation process, the simulation platform calls the solver to predict the corresponding blank holder force, die gap and friction coefficient according to the current stamping process parameters. Then, the solvers of each target parameter are combined to generate a post-invisible solver. Finally, the post-invisible solver is fitted into the simulation space so that it can work with the front-end explicit simulation module to calculate the blank holder force, die gap and friction coefficient.

[0061] The stamping curve simulation unit 14 simulates the initial stamping curve using the simulation space, and predicts the springback amount during the stamping process through the springback prediction module.

[0062] In the embodiment of the present application, the previously determined initial stamping curve is imported into the simulation space, and the initial stamping curve is dynamically simulated using the simulation space. During the simulation process, various physical effects in the stamping process are simulated, such as material deformation, die stress, blank holder force, etc. The springback prediction module is used to predict the springback amount in the stamping process based on the simulation results to obtain the springback amount prediction result. The springback amount prediction result includes the shape change or springback amount of the part after stamping.

[0063] The first springback constraint requirement judgment unit 15 judges whether the springback amount prediction result meets the springback constraint requirement. If not, the control points of the NURBS curve are adjusted according to the springback amount prediction result to optimize the stamping curve.

[0064] In the embodiment of the present application, the springback constraint requirement is a requirement predefined according to product design requirements and manufacturing standards. The springback amount prediction result is compared with the springback constraint requirement. When the springback amount prediction result does not meet the springback constraint requirement, the control points of the NURBS curve are adjusted through the interest balance algorithm to optimize the stamping curve.

[0065] The second springback constraint requirement judgment unit 16 uses the simulation space to simulate the optimized stamping curve, predicts the springback amount during the stamping process through the springback prediction module, makes a springback judgment based on the springback amount prediction result, and when the springback constraint requirement is met, determines the stamping curve and generates stamping control parameters.

[0066] In the embodiment of the present application, the second springback constraint requirement judgment unit uses the simulation space to simulate the optimized stamping curve, and combines the springback prediction module to predict the springback amount of the simulated stamping process. The springback amount prediction result is compared with the springback constraint requirement. If the springback amount prediction result meets the requirement, it is considered that the stamping curve is appropriate. The stamping curve is determined to be the final solution, and the corresponding stamping control parameters are generated.

[0067] The control point adjustment and optimization unit 17, when the springback constraint requirement is not met, continues to adjust and optimize the control points of the stamping curve until the springback amount prediction result corresponding to the stamping curve meets the springback constraint requirement.

[0068] In an embodiment of the present application, when the springback prediction result does not meet the springback constraint requirement, the control point adjustment and optimization of the stamping curve is continued, and the control points of the stamping curve are continuously adjusted through iteration and feedback loops until the springback prediction result meets the required springback constraint requirement.

[0069] In the embodiments of the present application, in summary, the embodiments of the present application have at least the following technical effects:

[0070] The present application determines key feature points and establishes a NURBS curve optimization model based on the design drawings of the motor stator and rotor; determines the curve control points based on the NURBS curve optimization model and constructs an initial stamping curve; constructs a simulation space based on the stamping process and the stamping parameter data collected, including the blank holder force, the die gap, and the friction coefficient, wherein the simulation space includes a springback prediction module; uses the simulation space to simulate the initial stamping curve, and predicts the springback amount during the stamping process through the springback prediction module; determines whether the springback amount prediction result meets the springback constraint requirement, and when it does not meet the springback constraint requirement, adjusts the control points of the NURBS curve according to the springback prediction result to optimize the stamping curve; uses the simulation space to simulate the optimized stamping curve, and predicts the springback amount during the stamping process through the springback prediction module, and makes a springback judgment according to the springback prediction result. When the springback constraint requirement is met, the stamping curve is determined and the stamping control parameters are generated; when the springback constraint requirement is not met, the control points of the stamping curve are continuously adjusted and optimized until the springback amount prediction result corresponding to the stamping curve meets the springback constraint requirement. The present invention solves the technical problems of low efficiency and precision of stamping processing in the prior art. By real-time monitoring of key parameters in the stamping process, the stamping process is accurately controlled, the stamping parameters are optimized and adjusted, and the technical effect of improving the efficiency and precision of the stamping process is achieved.

[0071] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0072] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0073] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. Stamping parameter optimization control system based on motor stator and rotor processing, characterized in that: The stamping parameter optimization control system based on motor stator and rotor processing includes: A key feature point determination unit, wherein the key feature point determination unit determines the key feature points according to the design drawings of the motor stator and rotor, and establishes a NURBS curve optimization model; A curve control point determination unit, wherein the curve control point determination unit determines the curve control points according to the NURBS curve optimization model and constructs an initial stamping curve; A stamping parameter data acquisition unit, wherein the stamping parameter data acquisition unit constructs a simulation space based on the stamping process and the stamping parameter data acquired, including the blank holder force, the die clearance, and the friction coefficient, wherein the simulation space includes a springback prediction module; A stamping curve simulation unit, wherein the stamping curve simulation unit simulates the initial stamping curve by using the simulation space, and predicts the springback amount during the stamping process by using the springback prediction module; a first springback constraint requirement judgment unit, which judges whether the springback amount prediction result meets the springback constraint requirement, and when it does not meet the springback constraint requirement, adjusts the control point of the NURBS curve according to the springback amount prediction result to optimize the stamping curve; a second springback constraint requirement judgment unit, which simulates the optimized stamping curve using the simulation space, predicts the springback amount during the stamping process through the springback prediction module, makes a springback judgment based on the springback amount prediction result, and determines the stamping curve and generates stamping control parameters when the springback constraint requirement is met; A control point adjustment and optimization unit, wherein when the springback constraint requirement is not met, the control point adjustment and optimization unit continues to adjust and optimize the control points of the stamping curve until the springback amount prediction result corresponding to the stamping curve meets the springback constraint requirement; The method of determining key feature points and establishing a NURBS curve optimization model based on the design drawings of the motor stator and rotor includes: A key feature recognition unit, wherein the key feature recognition unit recognizes and locates preset key features according to the design drawings, wherein the preset key features include winding slots, ventilation holes, and wiring terminals; A key feature point coordinate determination unit, wherein the key feature point coordinate determination unit determines coordinate data of the key feature point according to the key feature identification information; A control point weight factor configuration unit, which determines the control point according to the coordinate data of the key feature point, configures the control point weight factor, and constructs a NURBS curve model; A process parameter acquisition unit, wherein the process parameter acquisition unit acquires stamping process parameters, including stamping stroke, speed, acceleration, and pressure; A first fitting unit is used to fit the NURBS curve based on the stamping process parameters to establish the NURBS curve optimization model.

2. The stamping parameter optimization control system based on motor stator and rotor processing according to claim 1 is characterized in that: The configuration control point weight factor includes: An abnormal impact evaluation unit, wherein the abnormal impact evaluation unit evaluates the abnormal impact of the winding slots, ventilation holes, and wiring terminals according to historical abnormal cases, and obtains abnormal evaluation values ​​of the winding slots, ventilation holes, and wiring terminals at various positions; An initial weight factor configuration unit, which configures initial weight factors of each control point of the winding slot, the ventilation hole, and the connection terminal based on the abnormal evaluation value; A demand curve data acquisition unit, wherein the demand curve data acquisition unit determines the design requirements according to the design drawings and obtains the shape and size of the demand curve; A deviation calculation unit, wherein the deviation calculation unit performs curve fitting based on the initial weight factor, fits the curve using the shape and size of the demand curve, and calculates the deviation; An optimization and adjustment unit is used to optimize and adjust the initial weight factor based on the deviation until the deviation from the shape and size of the demand curve meets the minimum error requirement, thereby determining the control point weight factor.

3. The stamping parameter optimization control system based on motor stator and rotor processing according to claim 2 is characterized in that: According to the NURBS curve optimization model, the curve control points are determined and the initial stamping curve is constructed, including: A curve control point conversion unit, which extracts size changes and shape information of key features and converts them into curve control points; A curve adjustment unit, wherein the curve adjustment unit performs curve adjustment on the curve control points based on the control point weight factors to determine the shape and path of the stamping curve; The second fitting unit takes the shape and path of the stamping curve as a fitting target, and uses the stamping process parameters and curve control points to perform fitting to obtain the initial stamping curve.

4. The stamping parameter optimization control system based on motor stator and rotor processing according to claim 1, characterized in that: The simulation space is constructed based on the stamping process and the stamping parameter data collected, including the blank holder force, die clearance, and friction coefficient, wherein the simulation space includes a springback prediction module, including: A data simulation model building unit, wherein the data simulation model building unit collects process data and physical structure of mechanical equipment in the stamping process, and uses a finite element platform to build a data simulation model, including a front explicit simulation module and a rear invisible solver. The front explicit simulation module is used to simulate and adjust stamping process parameters and simulate stamping dynamic behavior, and the rear invisible solver calculates and processes blank holding force, die clearance, and friction coefficient based on simulation results; A training data set construction unit, wherein the training data set construction unit collects simulation experiment data and actual processing monitoring data to construct a training data set, which includes blank holder force, die clearance, friction coefficient and springback amount identification; A model training unit, wherein the model training unit performs supervised training on a neural network model using the training data set, and when a convergence target of a prediction result is reached, obtains the rebound prediction module, and fits the rebound prediction module into a post-invisible solver.

5. The stamping parameter optimization control system based on motor stator and rotor processing according to claim 4, characterized in that: Also includes: A sample data set construction unit, wherein the sample data set construction unit takes the blank holder force, die clearance, and friction coefficient as targets, and constructs sample data sets of each target parameter. The sample data set is actual processing monitoring data of the current material extracted using the processing log data of the current motor stator and rotor, including stamping process parameters: stamping stroke, speed, acceleration, and pressure; A solver construction unit, wherein the solver construction unit uses the blank holder force, die clearance, and friction coefficient as dependent variables and the stamping process parameters as independent variables, performs relationship fitting learning through the sample data set, and constructs a solver for each target parameter; A post-invisible solver establishment unit is used to establish an association mapping between the solver and the target parameter, generate the post-invisible solver, and fit it into the simulation space.

6. The stamping parameter optimization control system based on motor stator and rotor processing according to claim 2, characterized in that: The optimizing and adjusting the initial weight factor based on the deviation includes: An analytical positioning unit, which performs analytical positioning based on the deviation to determine a node control target deviation of a curve control point and curve smoothness evaluation information; A target optimization unit, wherein the target optimization unit establishes a node control target benefit evaluation function and a smoothness benefit evaluation function, and performs target optimization using the node control target deviation and the curve smoothness evaluation information, respectively, to obtain a first optimization scheme and a second optimization scheme, wherein the first optimization scheme is the optimal node control target benefit, and the second optimization scheme is the optimal curve smoothness benefit; The benefit balancing unit uses a benefit balancing algorithm to balance the benefits of the first optimization scheme and the second optimization scheme, and determines an optimization adjustment scheme for the initial weight factor.

7. The stamping parameter optimization control system based on motor stator and rotor processing according to claim 6, characterized in that: The method of using a benefit balance algorithm to balance the benefits of the first optimization scheme and the second optimization scheme also includes: A precision constraint condition acquisition unit, wherein the precision constraint condition acquisition unit acquires the precision constraint condition of each curve control point; A constraint coefficient determination unit, wherein the constraint coefficient determination unit determines the constraint coefficient of each node control target based on the accuracy constraint condition; A constraint condition determination unit, wherein the constraint condition determination unit determines a curve smoothing constraint node and a smoothing constraint value according to a design drawing; The benefit configuration constraint condition configuration unit configures the benefit configuration constraint condition according to the constraint coefficient of the node control target, the curve smooth constraint node, and the smooth constraint value to add the benefit configuration constraint condition into the benefit balance algorithm.

Images