Three-roller reducing skew rolling full-flow process development auxiliary platform

Through the three-roll variable diameter inclined rolling full process development auxiliary platform, the problems of poor representative data point layout and low manual post-processing efficiency in numerical simulation and process optimization of three-roll inclined rolling are solved, and efficient and high-precision process research and development are achieved. Simplify process optimization process and improve process data quality.

CN118981963AInactive Publication Date: 2024-11-19NINGBO UNIV
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202411134844.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-19
Publication Date
2024-11-19
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

During the existing three-roll inclined rolling numerical simulation and process optimization, there are problems such as poor representative data point layout, low manual post-processing efficiency and inaccurate proxy models, resulting in insufficient processing efficiency and accuracy.

Method used

Develop an auxiliary platform for the development of a three-roll variable diameter inclined rolling full-process process, including roll motion calculation module, material model calculation module, FEM post-processing auxiliary module, process database management module and process optimization module, using integrated neural network and high throughput The calculation method establishes a high-precision proxy model and optimizes the process index through the NSGA-II multi-objective optimization algorithm.

Benefits of technology

The efficiency and accuracy of numerical simulation of three-roll inclined rolling is improved, the front and post-processing process is simplified, the process data quality is improved, and the process optimization model is provided is provided, which shortens the process R&D cycle and saves design costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118981963A_ABST
    Figure CN118981963A_ABST
Patent Text Reader

Abstract

The invention provides a three-roller variable-diameter skew rolling full-flow process development auxiliary platform which comprises a roller motion calculation module used for continuously collecting roller motion data in the motion process of a roller, obtaining a time-speed control table and generating a. Csv file imported into simuact forming simulation software; the material model calculation module is used for performing hot working diagram modeling on the experimental data to obtain a J-C constitutive model; the FEM post-processing auxiliary module is used for measuring the size of the picture of the shaft part to obtain size data; the process database management module is used for storing the. Csv file, the J-C constitutive model and the size data into a process database; and the process optimization module is used for calling the process data in the process database and establishing a high-precision agent model, and performing process index optimization solution on the high-precision agent model through an NSGA-II multi-objective optimization algorithm to obtain a Pareto leading-edge solution set. The method has the beneficial effects that the processing efficiency and the model accuracy can be improved, and the efficiency and the precision of three-roller skew rolling numerical simulation are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of computer software systems, and in particular to a three-roller variable diameter skew rolling full-process process development auxiliary platform. Background Art

[0002] With the continuous development of modern engineering technology, especially in the field of transportation, hollow shafts have gradually emerged due to their unique structural characteristics and mechanical properties, becoming the focus of technical research and practical applications. As a new hollow shaft plastic forming technology, three-roller variable diameter skew rolling can achieve continuous processing. Uniform pressure distribution helps to reduce the internal stress in the finished product and can roll out a variety of different cross-sectional shapes. At present, the process research and development of three-roller variable diameter skew rolling hollow shafts is based on numerical simulation technology. In order to efficiently obtain high-precision process proxy models, it is very necessary to establish a three-roller variable diameter skew rolling full-process process development auxiliary platform.

[0003] With the rapid development of modern computer technology, the use of numerical simulation technology combining finite element method and optimization theory for high-quality manufacturing of hollow shafts has become a necessity. For the numerical simulation of hollow shafts, the optimal geometric parameter combination that meets the manufacturing accuracy is obtained through simulation and optimization calculation. This work has significant application value for high-quality hollow shaft manufacturing, shortening process development cycle, and saving design costs.

[0004] In the existing three-roller cross-rolling numerical simulation and process optimization, the commonly used process is manual pre-processing - orthogonal experimental design - manual post-processing - linear regression / surface fitting to establish an agent model - intelligent optimization algorithm to solve the optimal process. This process has the following defects: First, the orthogonal experimental arrangement of process data points ignores the potential optimal process points within the process space, which has the disadvantage of poor representativeness; Second, manual post-processing inevitably results in human errors and low efficiency; Third, it is difficult for traditional process proxy models to fit process parameters with high precision, and there is a problem of inaccuracy in the process proxy models.

[0005] In summary, in order to meet the requirements of efficient and high-precision process research and development, it is very important to develop a three-roller variable diameter skew rolling full-process process development auxiliary platform to promote the high-quality manufacturing of three-roller skew rolled hollow shafts. Summary of the invention

[0006] The technical problem to be solved by the present invention is to improve processing efficiency and model accuracy, and to improve the efficiency and accuracy of three-roller oblique rolling numerical simulation. In order to overcome the defects of the above-mentioned prior art (or related technology), the present invention provides a three-roller variable diameter oblique rolling full-process process development auxiliary platform.

[0007] The present invention provides a three-roller variable diameter skew rolling full process process development auxiliary platform, comprising: a roller motion calculation module, used for continuously collecting roller motion data of the roller during the motion process and obtaining a time-speed control table of the roller moving in the radial direction according to the roller motion data, and generating a .csv file imported into the simufact forming simulation software according to the time-speed control table; A material model calculation module, used for acquiring experimental data obtained from a thermal simulation experiment and performing thermal processing diagram modeling according to the experimental data to obtain a JC constitutive model; An FEM post-processing auxiliary module, used to obtain the shaft part image obtained by simulation, and measure the size of the shaft part image according to the external input measurement control parameters to obtain the corresponding size data; a process database management module, connected to the roller motion calculation module, the material model calculation module and the FEM post-processing auxiliary module, respectively, and used for storing the .csv file, the JC constitutive model and the dimensional data into a process database for calling; A process optimization module is connected to the process database management module, and is used to call the process data pre-stored in the process database, and use an integrated neural network and a high-throughput computing method to establish a high-precision proxy model of the process data, and use the NSGA-Ⅱ multi-objective optimization algorithm to optimize the process indicators of the established high-precision proxy model to obtain a Pareto frontier solution set for use.

[0008] Compared with the prior art, the three-roller variable diameter skew rolling full process development auxiliary platform provided in this application has the following advantages: The three-roll variable diameter skew rolling full-process process development auxiliary platform provided in the present application is easy to operate, and covers the entire process of three-roll variable diameter skew rolling numerical simulation model establishment and process optimization, and the analysis results are accurate; it provides a roll motion calculation module and a material model calculation module to realize the intelligent calculation of the roll motion process in the pre-processing and the rapid establishment of the material constitutive model, thereby improving the pre-processing efficiency and the accuracy of the constitutive model, thereby improving the efficiency and accuracy of the three-roll skew rolling numerical simulation; it provides an FEM post-processing auxiliary module to realize the function of fast and accurate measurement in post-processing, and the measurement method based on machine vision greatly improves the process data quality and post-processing efficiency; it provides a process database management module and a process optimization module to facilitate the management of process data, and provides a high-precision proxy model establishment strategy to improve analysis accuracy.

[0009] In a possible implementation manner, the roller motion calculation module includes: a data acquisition unit, used for acquiring the size parameters and initial position parameters of the roller as the roller motion data; A file generation unit is connected to the data acquisition unit, and is used to obtain the preset axial movement speed of the chuck during the rolling of the roll to obtain the time-speed control table, and solve the time-speed control table according to the roll motion data to generate the .csv file applied to the simufact forming software.

[0010] In a possible implementation, the material model calculation module includes: a first analysis unit, configured to obtain the experimental data obtained from the thermal simulation experiment and classify the data according to different strain rates and deformation temperatures as modeling parameters of the JC constitutive model; A second analysis unit is connected to the first analysis unit and is used to create the strain starting point and strain interval point of the JC constitutive model, and obtain stress-strain data under any strain condition to perform thermal processing diagram modeling under the strain condition to obtain the JC constitutive model.

[0011] In a possible implementation, the shaft part picture is a shaft part appearance picture or a shaft part wall thickness picture.

[0012] In a possible implementation manner, the FEM post-processing auxiliary module includes: A data acquisition unit, used to acquire the image of the shaft parts obtained by simulation; A third analysis unit is connected to the data acquisition unit, and is used to acquire and, based on the externally input measurement control parameters, amplify the input shaft part image to perform accurate boundary detection to obtain the dimension data.

[0013] In a possible implementation, the process database management module includes: a database storage unit, used to store at least one process database; a data storage unit, connected to the database storage unit, for storing the .csv file, the JC constitutive model and the dimensional data in the process database; A database creation unit is connected to the database storage unit and is used to add the process database to the database storage unit according to the creation instruction when receiving a creation instruction input from the outside.

[0014] In a possible implementation, the process database management module further includes a data modification unit connected to the database storage unit, and configured to modify or delete the process data in the process database according to the modification instruction when receiving a modification instruction input from the outside.

[0015] In a possible implementation, the process optimization module includes: a data reading unit, used for calling the process data pre-stored in the process database; a model building unit, connected to the data reading unit, and used to establish a process optimization agent model for obtaining the process data based on the integrated neural network; a model optimization unit, connected to the model building unit, for performing high-throughput calculation on the process optimization proxy model according to the prediction accuracy condition that the integrated neural network needs to satisfy, until the high-precision proxy model that satisfies the prediction accuracy condition is generated; A model solving unit is connected to the model optimization unit and is used to optimize the process indexes of the high-precision proxy model using the NSGA-Ⅱ multi-objective optimization algorithm to obtain the Pareto frontier solution set. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a structural principle diagram of the present invention; Figure 2 This is a schematic diagram of the main interface of the three-roller variable diameter skew rolling full process development auxiliary platform; Figure 3 It is a schematic diagram of the software interface of the roller motion calculation module; Figure 4 It is a schematic diagram of the calculation results of the roller motion calculation module; Figure 5 This is a schematic diagram of the software interface of the material model calculation module; Figure 6 Schematic diagram of input of material deformation conditions and constitutive modeling points; Figure 7 Schematic diagram of the prediction results of the modified JC constitutive model; Figure 8 Schematic diagram of the prediction error of the modified JC constitutive model; Fig. 9 Schematic diagram of text data for the modified JC constitutive model; Fig.10 Schematic diagram of input data for thermal process modeling; Fig.11 It is a schematic diagram of thermal processing; Fig.12 This is a schematic diagram of the software interface of the FEM post-processing auxiliary module; Fig.13 It is a schematic diagram for inputting measurement data and measurement control parameters; Fig.14 This is a schematic diagram of the outer diameter measurement results; Fig.15 This is a schematic diagram of the wall thickness measurement results; Fig.16 It is a schematic diagram of the software interface of the process database management module; Fig.17 It is a visualization diagram of the process database; Fig.18 Schematic diagram of process data writing / modification operation; Fig.19 This is a schematic diagram of the process data deletion operation; Fig. 20 and Fig.21 This is a schematic diagram of the process database deletion operation; Fig. 22 This is a schematic diagram of the software interface of the process optimization module; Fig.23 Read schematic diagrams into the process database; Fig.24 Schematic diagram of input for neural network architecture, neural network modeling parameters, and high-throughput computing parameters; Fig.25 It is a schematic diagram of the neural network prediction results of process index 1; Fig.26 It is a schematic diagram of the neural network prediction results of process index 2; Fig. 27 Schematic diagram of input and calculation for NSGA-II multi-objective optimization; Fig.28 This is a schematic diagram of multi-objective optimization results; Explanation of the accompanying drawings: 1. Roller motion calculation module; 11. Data acquisition unit; 12. File generation unit; 2. Material model calculation module; 21. First analysis unit; 22. Second analysis unit; 3. FEM post-processing auxiliary module; 31. Data acquisition unit; 32. Third analysis unit; 4. Process database management module; 41. Database storage unit; 42. Data storage unit; 43. Database creation unit; 44. Data modification unit; 5. Process optimization module; 51. Data reading unit; 52. Model building unit; 53. Model optimization unit; 54. Model solving unit. DETAILED DESCRIPTION

[0017] First, those skilled in the art should understand that these implementations are only used to explain the technical principles of the embodiments of the present application, and are not intended to limit the protection scope of the embodiments of the present application. Those skilled in the art can make adjustments to them as needed to adapt to specific application scenarios.

[0018] The present application is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0019] See also Figure 1The embodiment of the present application discloses a three-roll variable diameter skew rolling full-process process development auxiliary platform, which integrates a roll motion calculation module 1, a material model calculation module 2, a FEM post-processing auxiliary module 3, a process database management module 4 and a process optimization module 5, wherein the roll motion calculation module 1 includes a data acquisition unit 11 and a file generation unit 12, the material model calculation module 2 includes a first analysis unit 21 and a second analysis unit 22, the FEM post-processing auxiliary module 3 includes a data acquisition unit 31 and a third analysis unit 32, the process database management module 4 includes a database storage unit 41, a data storage unit 42, a database creation unit 43 and a data modification unit 44, and the process optimization module 5 includes a data reading unit 51, a model building unit 52, a model optimization unit 53 and a model solving unit 54; through the above modules and units, the efficiency of the pre- and post-processing of the three-roll skew rolling numerical simulation can be effectively improved, a high-precision process optimization model can be established, and an efficient and high-quality design solution can be provided for the three-roll skew rolling shaft parts process.

[0020] Continue to see Figure 1 This application takes the aluminum alloy three-roller oblique rolling variable diameter hollow shaft as an example, and specifically includes the following implementation steps: Step S1: Main interface construction Based on the MATLAB APPDEIGNER platform, this application proposes a three-roller variable diameter skew rolling full process development auxiliary platform that can be used for commercial software simufact forming pre- and post-processing and process optimization, including: S11, complete the finite element model establishment through solidworks and other 3D modeling software, import the obtained stl file into simufact forming and complete the meshing and boundary condition setting; S12, import the thermal simulation experiment data .xlsx file into the material model calculation module 2 to calculate the constitutive model and convert the calculation result into secondary development code and import it into the simufact forming flow stress subroutine interface; S13, importing the initial position of the roll and the preset size of the rolled piece into the roll motion calculation module 1 to complete the roll motion calculation and export the .csv file applied to the simufact forming motion control table; S14, importing the .png file of the rolled piece dimensions obtained by simufact forming simulation into the FEM post-processing auxiliary module 3 to perform high-precision and rapid measurement of the formed dimensions; S15, importing the measurement data obtained by the FEM post-processing auxiliary module 3 into the database through the process database management module 4, completing the establishment of the process database; S16, the process optimization module 5 completes a high-precision process proxy model by calling the process data in the process database and using high-throughput calculations, and performs multi-objective process optimization in combination with the NSGA-II algorithm.

[0021] See also Figure 2 The main interface of the three-roller variable diameter skew rolling full process process development auxiliary platform proposed in this application is as follows Figure 2 As shown, it includes the following functions: 1. Roller motion calculation, 2. Material model calculation, 3. FEM post-processing assistance, 4. Process database establishment and management, 5. Process optimization.

[0022] Continue to see Figure 1 When establishing the roll motion calculation module 1, use MATLAB language to create a user interaction interface. The implementation of the roll motion calculation module 1 requires inputting the initial position of the roll and the size of the rolled piece. By inputting the initial position of the roll and the change of the rolled piece size to adapt to the rolling of different shaft shapes, the buttons on the UI interface can realize automatic calculation and generate the .csv file of the roll motion control table required for simufact forming. The operation steps include: (1) Input the geometric model of the shaft, the size parameters and the initial position parameters of the roller into the roller motion calculation module 1. The input interface is as follows: Figure 3 As shown: 1. The user enters the design dimensions of the rolled piece in the design dimension panel; 2. The user enters the initial position of the roll, the roll speed and the axial moving speed of the rolling in the rolling parameter panel; (2) Calculate based on input parameters and export the .csv file for use in the Simufact Forming roll control table: 1. Press the "Start Calculation" button to display the calculation results in the "Solution Result Table"; 2. Click the "Export CSV File" button to generate a .csv file for the simufact forming roll control table; 3. The input parameters of the roller motion calculation module 1 and the generated "solution result table" are as follows: Figure 4 shown.

[0023] Continue to see Figure 1When establishing the material model calculation module 2, use the MATLAB language to create a user interaction interface. The implementation of the material model calculation module 2 requires reading the true stress-strain data of the thermal simulation experiment. By entering the thermal simulation deformation temperature and strain rate in the UI interface, and entering the starting point and interval point of the constitutive model establishment in the modified JC constitutive model solution interface, the modified JC model and the corresponding text can be generated. The establishment of the thermal processing diagram requires reading the stress-strain data under a certain strain, and the thermal processing diagram based on the Murty-Rao theory can be established. The operation steps include: (1) Input the thermal simulation experimental data and thermal simulation deformation conditions into the material model calculation module 2. The input interface is as follows: Figure 5 As shown: 1. Organize the experimental data obtained from the thermal simulation experiment and classify them according to different strain rates and deformation temperatures; 2. Enter the experimental temperature and experimental strain rate as constitutive modeling parameters in the software interface; 3. Read the thermal simulation experiment data through the software interface. The format of the read experimental data is .xlsx; 4. After the data is read in, the UI will display a prompt saying "Read in successfully". Figure 6 As shown; (2) The modified JC constitutive model is intelligently calculated based on the following formula. It has an embedded deformation condition optimization algorithm that can automatically solve the modified JC constitutive model under the best reference conditions. The formula is as follows:

[0024] in, A 0 -A 3 、m 1 、m 2 , C represents the material constant; ( represents the strain rate, represents the reference strain rate); T represents the experimental temperature; T ref represents the reference experimental temperature; The specific steps include: 1. Enter the constitutive model strain starting point and interval point in the UI interface; 2. Click the "Calculate" button to generate a visual prediction model of the modified JC (MJC) model, such as Figure 7 As shown; 3. Click the "Calculate Error" button to generate the error distribution graph, such as Figure 8 As shown; 4. Click "Generate MJC model text" to automatically generate the MJC model text. You can perform secondary development on simufact forming based on the text pseudo code, such as Fig. 9 As shown; (3) The thermal processing diagram modeling function is developed based on the Murty-Rao theory. The theoretical formula is as follows:

[0025]

[0026]

[0027] in, P represents the energy absorbed in the plastic deformation of the material; G represents the power dissipated by plastic power; J represents the energy consumed by the microstructure evolution; m represents the strain rate sensitivity factor; The specific steps include: 1. Click the Thermal Processing Calculation Module interface ( Fig.10 ) in "Read Data" to select the stress data for building the thermal processing diagram for import; 2. After the data is imported, the interface will prompt "Read successful". Click the "Generate" button in the interface to generate the corresponding thermal processing diagram. Fig.11 shown.

[0028] Continue to see Figure 1 When establishing FEM post-processing auxiliary module 3, use MATLAB language to create a user interaction interface (such as Fig.12 As shown in the figure), the dimension measurement of FEM post-processing auxiliary module 3 is developed based on machine vision detection. By importing the simulation result image, efficient and accurate measurement of the shape size and wall thickness can be achieved. The operation steps include: (1) Import the measurement image: Click the "Import Image" button in the UI interface to select the file directory and import the .png image to be tested. After the import is successful, the "Read Successful" prompt will be displayed on the UI interface (such as Fig.13 shown); (2) Measurement parameter input In the "Diameter / Wall Thickness Measurement Accuracy Control Parameters" panel, you can enter the measurement parameters. The initial values ​​in the panel are the measurement parameters given by the software by default. The default parameters can be modified according to needs. (3) Computational control 1. Click the "Read Control Parameters" button on the UI interface to send the measurement parameters to the software. After the software successfully reads the measurement parameters, the "Read Status" text box will prompt "Read Successful"; 2. Click the "Calculate" button to start the corresponding measurement. The measurement results are as follows: Fig.14 , Fig.15 shown.

[0029] Continue to see Figure 1 , when establishing the process database management module 4, use MATLAB language to create a user interaction interface (such as Fig.16 As shown in the figure), the process database management module 4 is developed on the basis of the SQLite database, which can store, read and delete process data, and realize the real-time management of process data by the three-roller variable diameter skew rolling full process process development auxiliary platform, including: (1) Data storage visualization 1. The main interface of the software is arranged with process database storage visualization (such as Fig.17 shown); 2. Click the "Database Address" button in the main interface of the software to import the databases created in different directories; 3. Click the "Update Database" button to display the data in the database under the corresponding directory in the table on the UI interface to visualize the stored data; (2) Data writing / modification The writing and modification of data is carried out by indexing the process number. In the UI interface of the process database management module 4, the process number and the data to be written / modified under the process number are entered. By clicking the "Write Data" button, the writing / modification of the data under the corresponding process number can be completed. Clicking the "Update to Main Interface" button can update the data visualization table of the main interface (such as Fig.18 shown); (3) Data deletion Data is also deleted by indexing the process number. After entering the process number in the UI interface of the process database management module 4, click the "Delete Data" button to delete the corresponding data. Click the "Update to Main Interface" button to update the data visualization table on the main interface (such as Fig.19 shown); (4) Database deletion Click the "Delete Database" button on the main interface to delete the database. You will be prompted during the deletion process (such as Fig. 20 ), click "YES" in the prompt interface to delete. After the deletion is completed, the data in the database visual table in the main interface will automatically disappear and a prompt "Database has been deleted" will appear (as shown in the figure below). Fig.21 as shown).

[0030] Continue to see Figure 1 When establishing process optimization module 5, MATLAB language is used to create a user interaction interface (such as Fig. 22 As shown in the figure, the process optimization module 5 is embedded with integrated neural network, high-throughput computing and NSGA-Ⅱ multi-objective optimization algorithm, which can read process data from the process database, and use integrated neural network and high-throughput computing to establish a high-precision proxy model of process data, and use NSGA-Ⅱ multi-objective optimization algorithm to optimize the process indicators of the established proxy model, including: (1) Reading the process database Click the "Read Data" button on the UI interface to automatically connect to the process database. When the reading is completed, the "Status" text box prompts "Read Successful", and the upper and lower limits of the decision variables appear in the "Decision Variable Text Box". Fig.23 shown); (2) Process neural network modeling and high-throughput computing 1. Enter the "input dimension", "output dimension", "number of nodes in hidden layer 1" and "number of nodes in hidden layer 2" of the neural network in the UI interface to complete the preset of the neural network architecture; 2. Enter the "learning rate", "maximum number of iterations", "target error", "number of integrated models" and "maximum number of failures" in the neural network control panel to complete the neural network control parameter input; 3. Enter the convergence conditions for high-throughput neural network construction in "Allow MSE"; 4. Click the "Start Calculation" button to build a neural network; 5. Click the "High Throughput Calculation" button to build a high-throughput neural network. During the high-throughput neural network building process, the "MSE" text box will update the prediction accuracy MSE of the neural network in real time until the prediction accuracy MSE is less than the allowed MSE to stop high-throughput modeling (such as Fig.24 As shown in the figure), click the "Terminate High Throughput" button to immediately end the high-throughput neural network modeling; 6. The results of establishing a high-precision process proxy model are as follows Fig.25 and Fig.26 As shown; (3) Multi-objective optimization of process indicators based on NSGA-II algorithm 1. Enter the "maximum number of iterations", "population size", "crossover ratio" and "mutation ratio" of the neural network in the UI interface to complete the input of NSGA-II algorithm control parameters (such as Fig. 27 shown); 2. Click the "Start Calculation" button to start multi-objective optimization. The "Iteration" text box will update the number of iterations in real time. The "Status" text will prompt when the optimization is completed. The "Stop Calculation" button can stop the multi-objective optimization process at any time (such as Fig. 27 shown); 3. The results of multi-objective optimization are as follows Fig.28 shown.

[0031] In the description of the present application, the description with reference to the terms "one embodiment", "some embodiments", "in the present embodiment", "specific example", or "some examples" etc. means that the specific features, mechanisms, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, mechanisms, materials or characteristics described may be combined in a suitable manner in any one or more embodiments or examples. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0032] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A three-roller variable diameter oblique rolling full process development auxiliary platform, It is characterized in that include: a roller motion calculation module (1), used for continuously collecting roller motion data of the roller during its motion and obtaining a time-speed control table for radial movement of the roller according to the roller motion data, and generating a .csv file imported into simufact forming simulation software according to the time-speed control table; a material model calculation module (2), used for acquiring experimental data obtained from a thermal simulation experiment and performing thermal processing diagram modeling based on the experimental data to obtain a JC constitutive model; an FEM post-processing auxiliary module (3), used to obtain a shaft part image obtained by simulation, and to perform dimensional measurement on the shaft part image according to externally input measurement control parameters to obtain corresponding dimensional data; a process database management module (4), connected to the roller motion calculation module (1), the material model calculation module (2) and the FEM post-processing auxiliary module (3), respectively, and used for storing the .csv file, the JC constitutive model and the dimensional data in a process database for calling; A process optimization module (5) is connected to the process database management module (4) and is used to call the process data pre-stored in the process database, establish a high-precision proxy model of the process data using an integrated neural network and a high-throughput calculation method, and optimize the process indicators of the established high-precision proxy model using an NSGA-II multi-objective optimization algorithm to obtain a Pareto frontier solution set for use.

2. The three-roller variable diameter skew rolling full process development auxiliary platform according to claim 1, It is characterized in that The roller motion calculation module (1) comprises: a data acquisition unit (11), used for acquiring dimension parameters and initial position parameters of the rolling roller as the rolling roller motion data; A file generation unit (12) is connected to the data acquisition unit (11) and is used to obtain the preset axial movement speed of the chuck during rolling of the roll to obtain the time-speed control table, and solve the time-speed control table according to the roll movement data to generate the .csv file applied to the simufact forming software.

3. The three-roller variable diameter skew rolling full process process development auxiliary platform according to claim 1, It is characterized in that The material model calculation module (2) comprises: a first analysis unit (21), used for acquiring the experimental data obtained from the thermal simulation experiment and classifying them according to different strain rates and deformation temperatures as modeling parameters of the JC constitutive model; A second analysis unit (22) is connected to the first analysis unit (21) and is used to create a strain starting point and a strain interval point of the JC constitutive model, and to obtain stress-strain data under any strain condition to perform thermal processing diagram modeling under the strain condition to obtain the JC constitutive model.

4. The three-roller variable diameter skew rolling full process process development auxiliary platform according to claim 1, It is characterized in that The shaft parts picture is a shaft parts appearance picture or a shaft parts wall thickness picture.

5. The three-roller variable diameter skew rolling full process process development auxiliary platform according to claim 1, It is characterized in that The FEM post-processing auxiliary module (3) comprises: a data acquisition unit (31), used to acquire the image of the shaft part obtained by simulation; A third analysis unit (32) is connected to the data acquisition unit (31) and is used to acquire and, based on the externally input measurement control parameters, amplify the input shaft part image to perform accurate boundary detection to obtain the dimension data.

6. The three-roller variable diameter skew rolling full process process development auxiliary platform according to claim 1, It is characterized in that The process database management module (4) comprises: a database storage unit (41), used for storing at least one process database; a data storage unit (42), connected to the database storage unit (41), and used for storing the .csv file, the JC constitutive model and the dimensional data in the process database; A database creation unit (43) is connected to the database storage unit (41) and is used to add the process database to the database storage unit (41) according to the creation instruction when receiving a creation instruction input from the outside.

7. The three-roller variable diameter skew rolling full process process development auxiliary platform according to claim 6, It is characterized in that The process database management module (4) further comprises a data modification unit (44), connected to the database storage unit (41), for modifying or deleting the process data in the process database according to the modification instruction when receiving a modification instruction input from the outside.

8. The three-roller variable diameter skew rolling full process process development auxiliary platform according to claim 1, It is characterized in that The process optimization module (5) comprises: a data reading unit (51), used for calling the process data pre-stored in the process database; a model building unit (52), connected to the data reading unit (51), and used to establish a process optimization agent model for obtaining the process data based on the integrated neural network; a model optimization unit (53), connected to the model construction unit (52), and configured to perform high-throughput calculations on the process optimization proxy model according to the prediction accuracy conditions that the integrated neural network needs to satisfy that are input externally, until the high-precision proxy model that satisfies the prediction accuracy conditions is generated; A model solving unit (54) is connected to the model optimization unit (53) and is used to optimize the process indexes of the high-precision proxy model using the NSGA-II multi-objective optimization algorithm to obtain the Pareto frontier solution set.

Citation Information

Patent Citations

  • Feed speed optimization method based on main shaft data simulation

    CN110221580A

  • Material rheological behavior analysis and prediction method and system

    CN116913421A

  • Copper-alloy pipe-material casting-milling technology parameter designing and optimizing method

    CN1979496A