CN112685947B - Method and device for optimizing parameters of sheet material resilience model, terminal and storage medium - Google Patents
Method and device for optimizing parameters of sheet material resilience model, terminal and storage medium Download PDFInfo
- Publication number
- CN112685947B CN112685947B CN202110071095.6A CN202110071095A CN112685947B CN 112685947 B CN112685947 B CN 112685947B CN 202110071095 A CN202110071095 A CN 202110071095A CN 112685947 B CN112685947 B CN 112685947B
- Authority
- CN
- China
- Prior art keywords
- variable
- model
- characteristic
- newly added
- parameter
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Landscapes
- Bending Of Plates, Rods, And Pipes (AREA)
Abstract
The application provides a sheet material resilience model parameter optimization method, a device, a terminal and a storage medium, the application screens characteristic variables influencing bending resilience of sheet materials in a variable combination mode through an SFS variable screening mode, trains according to the screened variable combinations, and compares the parameters of test workpieces measured by different variable combinations with the parameters of reference workpieces, thereby determining the optimal modeling variable combination, and solves the technical problems that the process of analyzing the resilience of sheet materials is extremely complicated due to the fact that a large number of parameters need to be introduced in the existing finite element method, and the implementation is time-consuming.
Description
Technical Field
The application relates to the field of machining processes, in particular to a method for optimizing parameters of a sheet material resilience model.
Background
Sheet metal workpieces are common parts in industrial production, wherein the common sheet metal workpieces comprise V-shaped workpieces. The bending resilience of the metal plate workpiece is difficult to control effectively due to a plurality of factors influencing the bending resilience of the metal plate workpiece, the size of the workpiece, the mechanical property, the load condition and the like. To solve this highly complex nonlinear control problem, analytical methods commonly used in practical engineering applications currently include finite element methods. However, the process of analyzing the slab springback by using the existing finite element method is very complicated and time-consuming to implement due to a large number of influence parameters.
Therefore, how to improve the efficiency and the accuracy of the slab springback analysis becomes a technical problem which needs to be solved urgently by the technical personnel in the field.
Disclosure of Invention
The application provides a method, a device, a terminal and a storage medium for optimizing parameters of a sheet material springback model, which are used for solving the technical problems that the process of analyzing the springback of a sheet material is extremely complex and the implementation is time-consuming due to the fact that a large number of parameters need to be introduced in the existing finite element method.
In view of this, the first aspect of the present application provides a method for optimizing parameters of a sheet metal springback model, including:
s1: acquiring characteristic variables of plate processing, and establishing a characteristic variable set with an initial state as an empty set;
s2: determining a newly added variable from the characteristic variables by an SFS variable screening mode, and adding the newly added variable into the characteristic variable set;
s3: taking the characteristic variable set as model input, and performing model training and testing through an SVR (support vector regression) model to obtain test workpiece parameters of the current characteristic variable set;
s4: judging whether the current characteristic variable set contains all the characteristic variables, if so, executing the step S5, and if not, returning to the step S2;
s5: comparing each test workpiece parameter with a reference workpiece parameter one by one to respectively obtain a mean square error and a decision coefficient corresponding to each test workpiece parameter, wherein the reference workpiece parameter is a workpiece parameter obtained by a finite element method according to the characteristic variable or a workpiece parameter measured by actual processing;
s6: and determining a target characteristic variable set corresponding to the test workpiece parameter with the maximum difference according to the difference between the mean square error and the decision coefficient.
Optionally, the step S2 specifically includes:
s21: determining a newly added variable from the characteristic variables according to a preset SFS newly added variable screening formula, wherein the SFS newly added variable screening formula specifically comprises the following steps:
in the formula, X + As the newly added variable, J (Y) k +X + ) Objective function for variable screening, Y k Is the feature variable set;
s22: and adding the newly added variable into the feature variable set.
Optionally, after the step S6, the method further includes:
s7: and taking the characteristic variables in the target characteristic variable set as input variables, and inputting sample data to perform model training to obtain a plate material resilience model.
Optionally, the feature variables specifically include: the device comprises a plate length, a plate width, a plate thickness, an upper die fillet radius, an upper die pressing amount, mechanical parameters, an upper die length, a die unilateral gap and an upper die advancing speed.
Optionally, the test workpiece parameters and the reference workpiece parameters each include: the opening angle and the minimum bend rebound radius.
The application second aspect provides a sheet material resilience model parameter optimization device, includes:
the variable acquisition unit is used for acquiring characteristic variables of plate processing and establishing a characteristic variable set of which the initial state is an empty set;
the variable screening unit is used for determining a newly added variable from the characteristic variables through an SFS variable screening mode and adding the newly added variable to the characteristic variable set;
the test result acquisition unit is used for taking the characteristic variable set as model input, and performing model training and testing through an SVR (support vector regression) model to obtain the test workpiece parameters of the current characteristic variable set;
a screening cycle judgment unit, configured to determine whether the current feature variable set includes all the feature variables, if yes, execute a parameter comparison unit, and if not, execute a parameter screening unit;
the parameter comparison unit is used for comparing each test workpiece parameter with a reference workpiece parameter one by one to respectively obtain a mean square error and a decision coefficient corresponding to each test workpiece parameter, wherein the reference workpiece parameter is a workpiece parameter obtained by a finite element method according to the characteristic variables or a workpiece parameter measured by actual processing;
and the optimal parameter determining unit is used for determining a target characteristic variable set corresponding to the tested workpiece parameter with the maximum difference according to the mean square error and the difference of the decision coefficients.
Optionally, the variable screening unit specifically includes:
a newly added variable determining subunit, configured to determine a newly added variable from the feature variables according to a preset SFS newly added variable screening formula, where the SFS newly added variable screening formula specifically includes:
in the formula, X + For the newly added variable, J (Y) k +X + ) Objective function for variable screening, Y k Is the feature variable set;
and the newly added variable adding subunit is used for adding the newly added variable into the characteristic variable set.
Optionally, the method further comprises:
and the plate material resilience model building unit is used for taking the characteristic variables in the target characteristic variable set as input variables and inputting corresponding sample data to perform model training to obtain a plate material resilience model.
A third aspect of the present application provides a terminal, comprising: a processor and a memory;
the memory is used for storing program codes corresponding to the plate material resilience model parameter optimization method according to the first aspect of the application;
the processor is configured to execute the program code.
A fourth aspect of the present application provides a storage medium, in which program codes corresponding to the method for optimizing parameters of a slab springback model according to the first aspect of the present application are stored.
According to the technical scheme, the method has the following advantages:
the application provides a sheet material resilience model parameter optimization method, which comprises the following steps: s1: acquiring characteristic variables of plate processing, and establishing a characteristic variable set with an initial state as an empty set; s2: determining a newly added variable from the characteristic variables by an SFS variable screening mode, and adding the newly added variable into the characteristic variable set; s3: taking the characteristic variable set as model input, and performing model training and testing through an SVR (support vector regression) model to obtain test workpiece parameters of the current characteristic variable set; s4: judging whether the current characteristic variable set contains all characteristic variables, if so, executing the step S5, otherwise, returning to the step S2; s5: comparing each test workpiece parameter with a reference workpiece parameter one by one to respectively obtain a mean square error and a decision coefficient corresponding to each test workpiece parameter, wherein the reference workpiece parameter is a workpiece parameter obtained by a finite element method according to a characteristic variable or a workpiece parameter measured by actual processing; s6: and determining a target characteristic variable set corresponding to the tested workpiece parameter with the maximum difference according to the difference between the mean square error and the decision coefficient.
According to the method, the characteristic variables influencing the bending springback of the plate are subjected to variable combination screening in an SFS variable screening mode, training is carried out according to the screened variable combinations, and then the parameters of the test workpiece measured by different variable combinations are compared with the parameters of the reference workpiece, so that the optimal modeling variable combination is determined, and the technical problems that the process of analyzing the springback of the plate is extremely complex and the implementation is time-consuming due to the fact that a large number of parameters need to be introduced in the existing finite element method are solved.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without inventive exercise.
FIG. 1 is a schematic flow chart of a first embodiment of a method for optimizing parameters of a sheet metal springback model provided by the present application;
FIG. 2 is a schematic flowchart illustrating a second embodiment of a method for optimizing parameters of a sheet metal springback model provided by the present application;
FIG. 3 is a schematic flow chart of an SFS-SVM algorithm of the sheet material springback model parameter optimization method provided by the present application;
FIG. 4 is a schematic structural diagram of a first embodiment of a sheet material springback model parameter optimization device provided by the present application;
FIG. 5 is a cloud of Mises stress distribution of the blank when the upper die is pressed down to the state of complete contact of the blank in the loading and holding steps;
FIG. 6 is a schematic view of the sheet after the upper die is removed and unloaded and the sheet rebounds;
figure 7 is a schematic representation of the control after the panel rebounds after the upper die is removed and unloaded.
Detailed Description
The embodiment of the application provides a method, a device, a terminal and a storage medium for optimizing parameters of a sheet material springback model, and is used for solving the technical problems that the process of analyzing the springback of a sheet material is extremely complex and time-consuming to implement due to the fact that a large number of parameters need to be introduced in the existing finite element method.
In order to make the objects, features and advantages of the present invention more apparent and understandable, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the embodiments described below are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments in the present application without making any creative effort belong to the protection scope of the present application.
Referring to fig. 1 and 3, a first embodiment of the present application provides a method for optimizing parameters of a sheet metal springback model, including:
step 101: and acquiring characteristic variables of plate processing, and establishing a characteristic variable set with an initial state as an empty set.
It should be noted that, first, a characteristic variable set is constructed, which is used to store the acquired characteristic variables of the sheet processing, and the characteristic variable set is used as an input variable for model training in subsequent steps, and specifically, the initially established characteristic variable set is an empty set.
More specifically, the obtained feature variables specifically include: sheet length L, sheet width B, sheet thickness t, upper die fillet radius r, upper die pressing amount e and mechanical parameter sigma s E, the length d of the upper die, the single-side clearance c of the die and the advancing speed v of the upper die.
Step 102: and determining a newly added variable from the characteristic variables by an SFS variable screening mode, and adding the newly added variable into the characteristic variable set.
It should be noted that, when step 102 is executed for the first time, a newly added variable needs to be determined from the feature variables acquired in step 101, and the newly added variable is added to the feature variable set, and if step 102 is not executed for the first time, on the basis of the feature variable set in which the screened variables are stored, a newly added variable is continuously determined from the feature variables that are not screened, and the newly added variable is added to the feature variable set.
Step 103: and taking the characteristic variable set as model input, and performing model training and testing through the SVR model to obtain the test workpiece parameters of the current characteristic variable set.
It should be noted that, the SVR model is trained by using the feature variable set obtained in step 102 as a model input, and the trained model is used for testing to obtain the test workpiece parameters of the current feature variable set.
More specifically, the workpiece parameters of this embodiment specifically include: opening angle and minimum bend rebound radius
Step 104: and judging whether the current characteristic variable set contains all the characteristic variables, if so, executing the step 105, and if not, returning to the step 102.
If all the feature variables acquired in step 101 are already included in the current feature variable set, the loop is stopped, and step 105 may be started, otherwise, the loop returns to step 102.
Step 105: and comparing the parameters of each test workpiece with the parameters of the reference workpiece one by one to respectively obtain the mean square error and the decision coefficient corresponding to the parameters of each test workpiece.
It should be noted that, each test workpiece parameter is compared with the reference workpiece parameter one by one, and the mean square error and the decision coefficient of each group of test workpiece parameters compared with the reference workpiece parameter are respectively calculated.
Specifically, the part after bending forming is scanned by a three-dimensional scanning measuring instrument and three-dimensional point cloud data is obtained, then the obtained data is imported into computer aided design software to complete modeling, finally the model is imported into finite element analysis software to obtain an opening angle alpha and a sheet fillet radius R of the forming part, or the solid sheet is processed according to the same processing parameters, and then the opening angle alpha and the sheet fillet radius R of the forming part are measured.
Step 106: and determining a target characteristic variable set corresponding to the test workpiece parameter with the maximum difference according to the difference between the mean square error and the decision coefficient.
Next, a target characteristic variable set corresponding to the test workpiece parameter having the largest difference is determined according to the difference between the mean square error and the decision coefficient, and the target characteristic variable set is used as an optimal model parameter combination.
More specifically, taking a processing result of a V-shaped workpiece as an example, based on values of 7 input parameters such as a sheet length L and a sheet width B mentioned in this embodiment, after a model is introduced into finite element analysis software, a model can be solved to obtain an opening angle α after a bent forming part rebounds, and a sheet fillet radius R, so that a data sample set for SFS-SVR algorithm training and testing can be obtained, the sample set in an experiment has 95 rows, as shown in table 1, table 2 gives variable correlation ranking conditions of 7 input variables such as the sheet length L and the sheet width B on prediction results of a target α and a target R of an SFS-SVR model, and the parameter description is shown in table 2, and as can be seen from table 2, the correlation rankings of the output results α and R in the 7 input variables are respectively a sheet thickness t, a single-side clearance c of a die, a fillet radius R of an upper die, and an upper die advancing speed V.
TABLE 1 SFS-SVM Algorithm training and testing data sample sets
TABLE 2 relevancy ranking results for input variables
Table 3 shows a detailed data table of an optimal variable set { t, c, R, v } obtained by SFS-SVM model dimension reduction analysis, and we see that when the dimension is 4, the characteristic variable subset consists of { t, c, R, v } (sheet thickness t, unilateral clearance c of the die, fillet radius R of the upper die and advancing speed v of the upper die), the algorithm has a higher model decision coefficient (R) 2 = 0.972213) and lower mean square error (MSE = 0.00411), the prediction accuracy of the algorithmic model is relatively high.
TABLE 3 characteristic variable screening example (test set)
And carrying out finite element numerical analysis on the V-shaped bending resilience, verifying the resilience test result, and carrying out the finite element analysis in three steps, namely a loading step, a maintaining step and an unloading step. Fig. 5 shows a cloud of the Mises stress distribution of the blank when the upper die is pressed down to the state of complete contact with the blank in the loading and holding step, clearly showing that the Mises stress at the round corner part has a little increasing trend in the unloading step compared with the loading step, which is mainly the result of the bending rebound effect, in addition, as can be seen from fig. 6, the plastic deformation area is mainly concentrated at the round corner, and other areas, especially the deformation of the upper end of the plate is smaller, which can even be seen as a rigid translation area, and fig. 6 shows a schematic diagram after the plate rebounds after the upper die is removed and unloaded, because the plate is elastically restored after the unloading due to the existence of the elastic deformation when the plastic deformation occurs inside the plate, the opening angle α after the rebounding, and the radius of the round corner of the plate is as shown in fig. 6.
Taking 80mm for the length L of the sheet, 60mm for the width B of the sheet, 3mm for the thickness t of the sheet, 10mm for the radius r of the fillet of the upper die, and taking the mechanical parameter sigma of the material s And E, taking 0.00412, taking 10mm for the unilateral clearance c of the die, taking 10mm/s for the upper die advancing speed V, and performing a V-shaped part bending experiment based on a conventional experiment die and equipment to obtain the opening angle alpha after the rebound of the curved forming part, wherein the fillet radius R of the plate is 44.6 degrees and 11.5mm respectively.
The experiment of the comparison group is carried out according to an Abaqus finite element method, the nonlinear dynamics problem in plastic forming is solved based on an ABAQUS/Explicit module, the length and width directions of the plate (workpiece) are far larger than the thickness direction, so that the plastic deformation of the plate (workpiece) can be seen as a plane strain process, in the simulation forming process, an upper die and a lower die are taken as rigid bodies, the plate is a Lagrange elastic-plastic deformation body, the plate material is surface hardening alloy structural steel 20MoCr4, the Young modulus is 212Gpa, the Poisson ratio is 0.3, a thick-shell small strain unit S4R is adopted, and the grid type is an 8-node hexahedron linear reduction integral unit. The sheet size value range is as follows: the thickness t is within 2-5 mm, the length L is within 80-100 mm, the width B is within 40-80 mm, the opening angle alpha of the V-shaped plate after bending is changed within 42-46 degrees, the fillet radius R of the V-shaped plate is changed within 8-12 mm, and a finite element model of the plate is shown in figure 7.
Comparing data obtained according to the Abaqus finite element method, the SFS-SVM algorithm and the bending experiment provided by the application obtain the opening angle alpha and the sheet fillet radius R after the bending forming part rebounds, and the result is shown in Table 4, and as can be seen from the Table 4, compared with the bending experiment result, the alpha relative error calculated by the Abaqus finite element method is 4.7%, and the R relative error is 8.7%, while the alpha relative error adopting the SFS-SVM algorithm is 1.6%, and the R relative error is 6.9%, based on a 64-core CPU, the time required for carrying out one-time Abaqus finite element calculation is 1 hour, 13 minutes and 48 seconds, the time required for the SFS-SVM algorithm is 12 minutes and 13 seconds, the time cost is far lower than that of the Abaqus finite element calculation, and the precision can also be ensured.
TABLE 4 comparison of the results of the opening angles alpha and R measured by different experimental methods
The bending rebound of the V-shaped workpiece is difficult to predict due to many factors such as the size and mechanical properties of the workpiece. In the embodiment, the opening angle alpha and the bending springback radius R after the sheet material rebounds are used as objective functions, and a Support Vector Machine (SVM) model is deployed into a sequential forward screening algorithm (SFS) to efficiently screen out an optimal characteristic variable parameter subset, so that the constructed bending springback model has high prediction accuracy.
The above is a detailed description of a first embodiment of the method for optimizing parameters of a sheet metal springback model provided by the present application, and the following is a detailed description of a second embodiment of the method for optimizing parameters of a sheet metal springback model provided by the present application.
Referring to fig. 2, based on the first embodiment provided above, step 102 of this embodiment specifically includes:
step 1021: determining a newly added variable from the characteristic variables according to a preset SFS newly added variable screening formula, wherein the SFS newly added variable screening formula specifically comprises the following steps:
in the formula, X + To newly add a variable, J (Y) k +X + ) Objective function for variable screening, Y k A characteristic variable set is obtained;
step 1022: and adding the newly added variable into the characteristic variable set.
In addition, this embodiment further includes, after step 106: step 107;
step 107: and (4) taking the characteristic variables in the target characteristic variable set as input variables, and inputting sample data to perform model training to obtain a plate material resilience model.
In the embodiment, the opening angle alpha and the minimum bending springback radius R after the sheet material rebounds are used as double objective functions, a Support Vector Machine (SVM) model is deployed into a sequential forward screening algorithm (SFS) to efficiently screen out an optimal characteristic variable parameter subset, and the sheet material rebounding model is built according to the optimal characteristic variable parameter subset, so that the parameter complexity of model building is simplified, and the built bending springback model has high prediction accuracy.
The above is a detailed description of a second embodiment of the method for optimizing parameters of a sheet metal springback model provided by the present application, and the following is a detailed description of a first embodiment of the apparatus for optimizing parameters of a sheet metal springback model provided by the present application.
Referring to fig. 4, a third embodiment of the present application provides a sheet material springback model parameter optimization device, including:
a variable acquisition unit 301, configured to acquire a characteristic variable of sheet processing and establish a characteristic variable set of which an initial state is an empty set;
a variable screening unit 302, configured to determine a newly added variable from the feature variables in an SFS variable screening manner, and add the newly added variable to the feature variable set;
the test result obtaining unit 303 is configured to perform model training and testing through the SVR model by using the feature variable set as a model input, and obtain a test workpiece parameter of the current feature variable set;
a screening loop determining unit 304, configured to determine whether the current feature variable set includes all feature variables, if yes, execute a parameter comparing unit 305, and if no, execute a variable screening unit 302;
a parameter comparison unit 305, configured to compare each test workpiece parameter with a reference workpiece parameter one by one, and obtain a mean square error and a decision coefficient corresponding to each test workpiece parameter, respectively, where the reference workpiece parameter is a workpiece parameter obtained by a finite element method according to a characteristic variable or a workpiece parameter measured by actual processing;
and the optimal parameter determining unit 306 is configured to determine, according to the difference between the mean square error and the decision coefficient, a target characteristic variable set corresponding to the test workpiece parameter with the largest difference.
Optionally, the variable screening unit 302 specifically includes:
a newly added variable determining subunit 3021, configured to determine a newly added variable from the characteristic variables according to a preset SFS newly added variable screening formula, where the SFS newly added variable screening formula specifically includes:
in the formula, X + To newly add a variable, J (Y) k +X + ) Objective function for variable screening, Y k A characteristic variable set is obtained;
a newly added variable adding subunit 3022, configured to add the newly added variable to the feature variable set.
Optionally, the method further comprises:
and a sheet material springback model building unit 307, configured to use the characteristic variables in the target characteristic variable set as input variables, and input corresponding sample data to perform model training, so as to obtain a sheet material springback model.
The above is a detailed description of a first embodiment of the sheet metal springback model parameter optimization device provided by the present application, and the following is a detailed description of a terminal and a storage medium provided by the present application.
A fourth embodiment of the present application provides a terminal, including: a processor and a memory;
the memory is used for storing program codes corresponding to the plate material resilience model parameter optimization method as mentioned in the first embodiment and the second embodiment of the application;
the processor is used for executing the program code.
A fifth embodiment of the present application provides a storage medium having stored therein program codes corresponding to the method for optimizing parameters of a slab springback model as mentioned in the first and second embodiments of the present application.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one type of logical functional division, and other divisions may be realized in practice, for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The terms "first," "second," "third," "fourth," and the like in the description of the application and the above-described figures, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the application described herein are, for example, capable of operation in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk, and various media capable of storing program codes.
The above embodiments are only used for illustrating the technical solutions of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions in the embodiments of the present application.
Claims (10)
1. A method for optimizing parameters of a sheet material springback model is characterized by comprising the following steps:
s1: acquiring characteristic variables of plate processing, and establishing a characteristic variable set of which the initial state is an empty set;
s2: determining a newly added variable from the characteristic variables by an SFS variable screening mode, and adding the newly added variable into the characteristic variable set;
s3: taking the characteristic variable set as model input, and performing model training and testing through an SVR (support vector regression) model to obtain test workpiece parameters of the current characteristic variable set;
s4: judging whether the current characteristic variable set contains all the characteristic variables, if so, executing the step S5, otherwise, returning to the step S2;
s5: comparing each test workpiece parameter with a reference workpiece parameter one by one to respectively obtain a mean square error and a decision coefficient corresponding to each test workpiece parameter, wherein the reference workpiece parameter is a workpiece parameter obtained by a finite element method according to the characteristic variable or a workpiece parameter measured by actual processing;
s6: and determining a target characteristic variable set corresponding to the tested workpiece parameter with the maximum difference according to the difference between the mean square error and the decision coefficient.
2. The sheet metal springback model parameter optimization method according to claim 1, wherein the step S2 specifically comprises:
s21: determining a newly added variable from the characteristic variables according to a preset SFS newly added variable screening formula, wherein the SFS newly added variable screening formula specifically comprises the following steps:
in the formula, X + As the newly added variable, J (Y) k +X + ) Objective function for variable screening, Y k Is the characteristic variableGathering;
s22: and adding the newly added variable into the feature variable set.
3. The method for optimizing parameters of a slab springback model according to claim 1, wherein said step S6 is followed by the step of:
s7: and taking the characteristic variables in the target characteristic variable set as input variables, and inputting sample data to perform model training to obtain a plate material resilience model.
4. The method for optimizing parameters of a slab springback model according to claim 1, wherein the characteristic variables specifically include: the device comprises the following components of a plate length, a plate width, a plate thickness, an upper die fillet radius, an upper die pressing amount, mechanical parameters, an upper die length, a die unilateral clearance and an upper die advancing speed.
5. The sheet material springback model parameter optimization method of claim 1, wherein said test workpiece parameters and said reference workpiece parameters each comprise: opening angle and minimum bend rebound radius.
6. The utility model provides a sheet material resilience model parameter optimization device which characterized in that includes:
the variable acquisition unit is used for acquiring characteristic variables of plate processing and establishing a characteristic variable set of which the initial state is an empty set;
the variable screening unit is used for determining a newly added variable from the characteristic variables in an SFS variable screening mode and adding the newly added variable to the characteristic variable set;
the test result acquisition unit is used for taking the characteristic variable set as model input, and performing model training and testing through an SVR (support vector regression) model to obtain the test workpiece parameters of the current characteristic variable set;
a screening cycle judgment unit, configured to determine whether the current feature variable set includes all the feature variables, if yes, execute a parameter comparison unit, and if no, execute a parameter screening unit;
the parameter comparison unit is used for comparing each test workpiece parameter with a reference workpiece parameter one by one to respectively obtain a mean square error and a decision coefficient corresponding to each test workpiece parameter, wherein the reference workpiece parameter is a workpiece parameter obtained by a finite element method according to the characteristic variables or a workpiece parameter measured by actual processing;
and the optimal parameter determining unit is used for determining a target characteristic variable set corresponding to the test workpiece parameter with the maximum difference according to the mean square error and the difference of the decision coefficients.
7. The plate material resilience model parameter optimization device according to claim 6, wherein the variable screening unit specifically comprises:
a newly added variable determining subunit, configured to determine a newly added variable from the characteristic variables according to a preset SFS newly added variable screening formula, where the SFS newly added variable screening formula specifically includes:
in the formula, X + For the newly added variable, J (Y) k +X + ) Objective function for variable screening, Y k Is the feature variable set;
and the newly added variable adding subunit is used for adding the newly added variable into the characteristic variable set.
8. The apparatus for optimizing parameters of a slab springback model according to claim 6, further comprising:
and the plate material resilience model building unit is used for taking the characteristic variables in the target characteristic variable set as input variables and inputting corresponding sample data to perform model training to obtain a plate material resilience model.
9. A terminal, comprising: a processor and a memory;
the memory is for storing program code corresponding to a method of optimizing parameters of a slab springback model as claimed in any one of claims 1 to 5;
the processor is configured to execute the program code.
10. A storage medium having stored therein program code corresponding to a method of optimizing parameters of a slab springback model as claimed in any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110071095.6A CN112685947B (en) | 2021-01-19 | 2021-01-19 | Method and device for optimizing parameters of sheet material resilience model, terminal and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110071095.6A CN112685947B (en) | 2021-01-19 | 2021-01-19 | Method and device for optimizing parameters of sheet material resilience model, terminal and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112685947A CN112685947A (en) | 2021-04-20 |
| CN112685947B true CN112685947B (en) | 2022-12-16 |
Family
ID=75458597
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110071095.6A Active CN112685947B (en) | 2021-01-19 | 2021-01-19 | Method and device for optimizing parameters of sheet material resilience model, terminal and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112685947B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115545086B (en) * | 2022-11-07 | 2023-04-18 | 海纳科德（湖北）科技有限公司 | Migratable feature automatic selection acoustic diagnosis method and system |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102982339A (en) * | 2012-10-31 | 2013-03-20 | 浙江大学 | Hyperspectral characteristic variable selection method |
| US8805043B1 (en) * | 2010-04-02 | 2014-08-12 | Jasjit S. Suri | System and method for creating and using intelligent databases for assisting in intima-media thickness (IMT) |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3026588A1 (en) * | 2014-11-25 | 2016-06-01 | Inria Institut National de Recherche en Informatique et en Automatique | interaction parameters for the input set of molecular structures |
| KR101954820B1 (en) * | 2016-09-12 | 2019-03-06 | 두산중공업 주식회사 | Apparatus and method for analyzing influence on fouling in sea water desalination of membrane |
-
2021
- 2021-01-19 CN CN202110071095.6A patent/CN112685947B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8805043B1 (en) * | 2010-04-02 | 2014-08-12 | Jasjit S. Suri | System and method for creating and using intelligent databases for assisting in intima-media thickness (IMT) |
| CN102982339A (en) * | 2012-10-31 | 2013-03-20 | 浙江大学 | Hyperspectral characteristic variable selection method |
Non-Patent Citations (2)
| Title |
|---|
| Dynamic reliability analysis using the extended support vector regression (X-SVR);Jinwen Feng等;《Mechanical Systems & Signal Processing》;20190701;第126卷;368-39 * |
| 基于响应面法的台阶式凸模冲裁工艺参数优化;刘静等;《塑性工程学报》;20160627(第03期);52-57+67 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112685947A (en) | 2021-04-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP5582211B1 (en) | Stress-strain relationship simulation method, springback amount prediction method, and springback analysis device | |
| KR101088115B1 (en) | Springback occurrence cause identifying method, springback influence degree display method, springback occurrence cause portion identifying method, springback measure position specifying method, their devices, and their programs | |
| JP4410833B2 (en) | Springback generation cause analysis method, apparatus, program and recording medium | |
| Bahloul et al. | A study on optimal design of process parameters in single point incremental forming of sheet metal by combining Box–Behnken design of experiments, response surface methods and genetic algorithms | |
| JP5967321B2 (en) | Stretch flange crack prediction method, stretch flange crack prediction apparatus, computer program, and recording medium | |
| CN104602836A (en) | Springback cause identification method and springback cause identification device | |
| JP4880043B2 (en) | Break prediction method, break prediction system, program, and recording medium | |
| CN112685947B (en) | Method and device for optimizing parameters of sheet material resilience model, terminal and storage medium | |
| Wiebenga | Robust design and optimization of forming processes | |
| CN115049176A (en) | Material performance evaluation method and device and computer equipment | |
| CN113435089B (en) | Board bending rebound prediction method based on Gaussian process | |
| US6731996B1 (en) | Method of using isotropic plasticity to model the forming of anisotropic sheet | |
| Li et al. | Identification of material parameters from punch stretch test | |
| Javanmard et al. | Meshless analysis of backward extrusion by natural element method | |
| Vorkov et al. | Data-driven prediction of air bending | |
| CA2349824C (en) | A method of modeling the forming of anisotropic sheet | |
| CN115446150A (en) | Steel plate straightening method, device, terminal equipment and storage medium | |
| JP6841295B2 (en) | Springback amount divergence factor part identification method and device | |
| Mkaddem et al. | Experimental and numerical optimisation of the sheet products geometry using response surface methodology | |
| JP6303815B2 (en) | Forging crack prediction method, forging crack prediction program, and recording medium | |
| CN108629110A (en) | The method that quick obtaining corrects metal powder compression moulding DPC Parameters of constitutive model | |
| JP4855279B2 (en) | Numerical analysis of pipe making process | |
| CN114065480A (en) | Screening method and device for performance data of part stamping material | |
| JP2022108560A (en) | Analysis device, parameter optimization method, and parameter optimization program | |
| Asgari et al. | Design of experiments and springback prediction for AHSS automotive components with complex geometry |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |