CN117161160A - Accurate rolling method of numerical control four-roller plate bending machine - Google Patents
Accurate rolling method of numerical control four-roller plate bending machine Download PDFInfo
- Publication number
- CN117161160A CN117161160A CN202311296783.8A CN202311296783A CN117161160A CN 117161160 A CN117161160 A CN 117161160A CN 202311296783 A CN202311296783 A CN 202311296783A CN 117161160 A CN117161160 A CN 117161160A
- Authority
- CN
- China
- Prior art keywords
- plate
- roller
- rolling
- parameters
- bending machine
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 70
- 238000005452 bending Methods 0.000 title claims abstract description 66
- 238000005096 rolling process Methods 0.000 title claims abstract description 64
- 238000013178 mathematical model Methods 0.000 claims abstract description 10
- 238000012545 processing Methods 0.000 claims abstract description 7
- 239000000463 material Substances 0.000 claims description 36
- 230000008569 process Effects 0.000 claims description 24
- 230000008878 coupling Effects 0.000 claims description 19
- 238000010168 coupling process Methods 0.000 claims description 19
- 238000005859 coupling reaction Methods 0.000 claims description 19
- 238000012549 training Methods 0.000 claims description 17
- 238000010801 machine learning Methods 0.000 claims description 11
- 238000012360 testing method Methods 0.000 claims description 8
- 230000006870 function Effects 0.000 claims description 6
- 230000001105 regulatory effect Effects 0.000 claims description 6
- 238000005457 optimization Methods 0.000 claims description 5
- 230000001276 controlling effect Effects 0.000 claims description 4
- 238000012937 correction Methods 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000007493 shaping process Methods 0.000 claims description 3
- 238000005728 strengthening Methods 0.000 claims description 3
- 230000000007 visual effect Effects 0.000 claims description 3
- 230000001808 coupling effect Effects 0.000 claims description 2
- 238000013000 roll bending Methods 0.000 claims 4
- 206010063385 Intellectualisation Diseases 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 229910052751 metal Inorganic materials 0.000 description 2
- 239000002184 metal Substances 0.000 description 2
- 229910000831 Steel Inorganic materials 0.000 description 1
- 229910052782 aluminium Inorganic materials 0.000 description 1
- XAGFODPZIPBFFR-UHFFFAOYSA-N aluminium Chemical compound [Al] XAGFODPZIPBFFR-UHFFFAOYSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 230000007935 neutral effect Effects 0.000 description 1
- 229910001220 stainless steel Inorganic materials 0.000 description 1
- 239000010935 stainless steel Substances 0.000 description 1
- 239000010959 steel Substances 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P70/00—Climate change mitigation technologies in the production process for final industrial or consumer products
- Y02P70/10—Greenhouse gas [GHG] capture, material saving, heat recovery or other energy efficient measures, e.g. motor control, characterised by manufacturing processes, e.g. for rolling metal or metal working
Landscapes
- Control Of Metal Rolling (AREA)
Abstract
The application discloses a precise rolling method of a numerical control four-roller plate bending machine, which comprises the steps of obtaining performance parameters and technological parameters of a plate to be bent, inputting the performance parameters and the technological parameters into a mathematical model, and determining an initial roller position; the side roller moves to the latest roller position and starts rolling; performing image acquisition and processing fitting to obtain a real-time curvature radius through a machine vision system; if the deviation between the real-time curvature radius and the expected curvature radius of the plate is not within the allowable error range, correcting the position of the roller according to the deviation, and continuing rolling; after reaching the precision, storing all the accurate parameters of the rolling once as a group of data in a database; when a new round of plate rolling is started, when the data quantity does not reach a set value, the plate rolling is continuously performed by using a machine vision method, and the like.
Description
Technical Field
The application relates to the technical field of numerical control four-roller plate bending machines, in particular to an accurate rolling method of a numerical control four-roller plate bending machine.
Background
The numerical control four-roller plate bending machine is a machine for bending a metal plate into a certain curvature and is generally composed of an upper roller, a lower roller and two side rollers. The upper roller can be driven to rotate by a motor, the lower roller can be driven to move up and down by hydraulic pressure or a motor, and the side rollers can be regulated by hydraulic pressure to ensure the rolling precision. The numerical control four-roller veneer reeling machine is also provided with a numerical control system, and can realize automatic control through input parameters, thereby improving the production efficiency and the product quality. The numerical control four-roller plate bending machine is widely applied to industries of ships, wind power, buildings, petrifaction and the like and is used for manufacturing various metal plates, such as steel plates, aluminum plates, stainless steel plates and the like.
Although the numerical control four-roller plate bending machine can improve the production efficiency and the product quality to a certain extent, from a large number of practical application cases, the precision of the plate rolled by the numerical control four-roller plate bending machine still cannot meet the expected requirement, and the curvature radius size of the rolled plate needs to be measured manually by means of a tool and is improved by multi-pass rolling, so that the rolling method greatly wastes manpower and time.
And for the plates with known rolling material performance parameters and forming process parameters, the numerical control four-roller plate bending machine with certain structural parameters still repeatedly rolls for a plurality of times according to the traditional method, and the roller positions of the side rollers cannot be calculated according to a large amount of rolled plate material data, so that the degree of automation and intelligence can not reach the actual demands. A new rolling method is still needed to be applied to the numerical control four-roller plate bending machine, so that the rolling process is quicker and more efficient, and the rolling result is more accurate.
Disclosure of Invention
This section is intended to outline some aspects of embodiments of the application and to briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section as well as in the description of the application and in the title of the application, which may not be used to limit the scope of the application.
Therefore, the application aims to provide a numerical control four-roller plate bending machine accurate rolling method integrating machine vision and machine learning algorithm. By integrating the machine vision and the XGBoost-based intelligent coupling influence analysis side roller position prediction model method, the bending process of the numerical control four-roller plate bending machine is quicker and more accurate, a large amount of labor and time cost are saved, and automation and intellectualization of plate bending are realized.
In order to solve the technical problems, according to one aspect of the present application, the following technical solutions are provided:
a precise rolling method of a numerical control four-roller plate rolling machine comprises the following steps:
s1, acquiring performance parameters and forming process parameters of a plate material to be bent;
s2, inputting known parameters into a mathematical model to determine an initial side roller position;
s3, the upper computer controls hydraulic servo through the motion control card so that the side roller moves to the latest roller position;
s4, the upper roller of the four-roller plate bending machine rotates to drive the plate to be rolled and form a latest rolled plate part;
s5, performing image acquisition on the newly rolled plate part after rebound by using an area array CCD camera;
s6, performing image processing in an upper computer vision system, and fitting the real-time curvature radius of the latest coiled sheet material part;
s7, judging whether the deviation between the real-time curvature radius of the latest rolled sheet material part and the expected curvature radius is within an allowable error range;
s8, correcting the roller position of the side roller according to the deviation between the real-time curvature radius of the latest rolled sheet part and the expected curvature radius;
s9, storing the material performance parameters, the forming process parameters and the corresponding optimal roll position parameters of the rolled plate as a group of data in a database of an upper computer;
s10, when a new round of plate rolling is started, judging whether all parameter data amounts accumulated in a database meet set values;
s11, continuously rolling plates with different material performance parameters and forming process parameters by using a machine vision method;
s12, rolling a plate by using a machine learning method, constructing an XGBoost-based intelligent coupling influence analysis side roller position prediction model, and obtaining a trained and optimized roller position prediction model by taking 70% of a database as a training set and 30% as a test set;
s13, inputting the performance parameters and the forming process parameters of the plate material to be bent into a roll position prediction model, and outputting the roll position of the side roll by the model, so that the plate bending machine with certain structural parameters realizes one-time bending and forming of the plate material.
As a preferable scheme of the accurate rolling method of the numerical control four-roller plate bending machine, in the step S1, the performance parameters of the plate material to be bent comprise the plate yield limit sigma s The sheet thickness t and the sheet width b, and the forming process parameters comprise a sheet coiling feed speed V and a sheet expected curvature radius rho.
As a preferable scheme of the accurate rolling method of the numerical control four-roller plate bending machine, in the step S2, a mathematical model is as follows: p=y-Y 1
Wherein P is a side roller position parameter, Y is a hydraulic cylinder position corresponding to the side roller position in an initial state, Y 1 To reach the position of the hydraulic cylinder corresponding to the latest side roller position;
wherein g, k, mu and v are structural parameters of the plate bending machine and can be obtained from corresponding numerical control four-roller plate bending machine structural parameter manuals;
wherein the radius of the upper roller of the plate bending machine is R a The radius of the lower roller is R b Radius of left roller is R c The radius of the right roller is R d ,R a ＝R b ＝R 1 ,R c ＝R d ＝R 2 D and r are structural parameters of the plate bending machine and can be obtained from corresponding numerical control four-roller plate bending machine structural parameter manuals;
wherein ρ is r The inner diameter of the plate is bent before rebound;
γ＝δ-ε
L'＝(ρ r +t+R 1 )sinε
wherein K is e E is the elastic modulus of the plate material, K is the relative strengthening coefficient of the plate s Is a plate section shape factor, generally rectangular in section, K s ＝1.5，ρ i To bend the inner diameter of the plate after rebound, K e E is obtained by checking performance parameter tables of different materials;
where ρ is the desired radius of curvature of the sheet to be bent.
In the step S3, an industrial personal computer is used as an upper computer, and a program is developed in a Visual Studio, so that the purpose of controlling a motion control card is achieved, and the motion control card enables the side rollers to reach a preset position through controlling hydraulic servo of the left side roller and the right side roller.
As a preferable scheme of the accurate rolling method of the numerical control four-roller plate bending machine, in the step S5, a binocular area array CCD camera is used for collecting images of the plate under the irradiation of an adjustable light source.
As a preferable scheme of the accurate rolling method of the numerical control four-roller plate bending machine, in the step S6, the image is preprocessed, graying, binarization and morphological processing are carried out through an image processing program developed in an upper computer, then a difference operator is utilized to calculate discrete curvature, and finally a numerical curve is fitted to calculate the real-time curvature radius.
In the step S7, it is determined whether the deviation between the real-time radius of curvature of the latest rolled sheet and the expected radius of curvature is within an allowable error range, where the allowable error range is determined according to the actual conditions such as precision of the four-roller plate bending machine, actual rolling precision requirement, etc.; if the deviation is within the set allowable error range, the expected curvature radius is considered to be satisfied, and then step S9 is performed;
if the deviation exceeds the set allowable error range, the expected curvature radius is not met, and then step S8 is performed, the correction side roller position is calculated by the upper computer, then step S3, step S4, step S5 and step S6 are performed, and then judgment is performed again, and the iteration is repeated until the deviation is within the allowable range.
As a preferable scheme of the accurate rolling method of the numerical control four-roller plate bending machine, in the step S8, the deviation is corrected by continuously working through a side roller controlled by a motion control card, and once for each working, the real-time curvature radius of the plate is visually identified once until the deviation is within an allowable error range.
As a preferable scheme of the accurate rolling method of the numerical control four-roller plate bending machine, in the step S10, when the judgment is yes, the data quantity of each parameter must reach: 5 different yield limits sigma s 5 different plate thicknesses t, 5 different plate widths b,3 different feeding speeds V, 5 plate expected curvature radiuses rho and 50 different side roller position parameters P, otherwise, judging as no;
for data stored in the database, the data format is as follows:
as a preferred scheme of the accurate rolling method of the numerical control four-roller plate bending machine, in the step S12, when a machine learning algorithm is used, an XGBoost-based intelligent coupling influence analysis side roller position prediction model is constructed, an XGBoost integrated learning method is adopted, and a regression tree is used as a base model to carry out bending forming multi-factor coupling influence relationClassifying and returning; constructing a complex influence coupling analyzer (SC) based on a gradient lifting method through associated regression tree joint decisions, wherein the constructed multi-factor coupling influence objective function is as follows: p=sc (t, b, σ s ,V,ρ)
The SC is a shaping relation function under the coupling effect, the data characteristics of different orders of magnitude do not influence the model result, normalization processing is not needed, 70% of data in a database is used as a training set, the other 30% is used as a test set, a roll position prediction model based on XGBoost is used for carrying out parameter training, and the evaluation index of the model is the mean square error of a side roll position; the training of the roller position prediction is the training of small sample data, a regular term needs to be added, the weight can be used as a super parameter, and finally the super parameter is regulated through Bayesian optimization until the set iteration times are reached.
Compared with the prior art, the application has the following beneficial effects: the application solves the problems that the curvature radius size of the rolled plate needs to be measured manually by means of a tool and the roll position of the side roll cannot be calculated according to a large amount of rolled plate data for the plate with known rolled material performance parameters and forming process parameters, and provides a method for fusing machine vision and a machine learning algorithm to be applied to a numerical control four-roll plate bending machine; a large number of data sets are obtained by rolling the plate by adopting a machine vision method, and are stored in a database, wherein 70% of the data sets are used as training sets, the other 30% of the data sets are used as test sets, parameters in a roll position prediction model based on XGBoost are trained and optimized, and super parameters are regulated by utilizing Bayesian optimization, so that the accuracy and generalization capability of the model are improved, and one-time accurate rolling and forming of the plate when the performance parameters and the forming technological parameters of the rolled material are known by a plate rolling machine with certain structural parameters are realized, and automation and intellectualization of rolling and forming are realized.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following detailed description will be given with reference to the accompanying drawings and detailed embodiments, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained from these drawings without inventive faculty for a person skilled in the art. Wherein:
FIG. 1 is a flow chart of a precise rolling method of a numerical control four-roller plate bending machine integrating machine vision and a machine learning algorithm provided by an embodiment of the application;
FIG. 2 is a schematic diagram of a mathematical model plane geometry for determining an initial roll position according to an embodiment of the present application;
FIG. 3 is a schematic diagram showing changes of image acquisition areas of CCD cameras and roller positions of side rollers in a bending forming process of a four-roller plate bending machine according to an embodiment of the application;
FIG. 4 is a schematic diagram of a method for forming an XGBoost intelligent coupling influence analysis side roller position prediction model according to an embodiment of the application.
Detailed Description
In order that the above objects, features and advantages of the application will be readily understood, a more particular description of the application will be rendered by reference to the appended drawings.
Next, the present application will be described in detail with reference to the drawings, wherein the sectional view of the device structure is not partially enlarged to general scale for the convenience of description, and the drawings are only examples, which should not limit the scope of the present application. In addition, the three-dimensional dimensions of length, width and depth should be included in actual fabrication.
For the purpose of making the objects, technical solutions and advantages of the present application more apparent, embodiments of the present application will be described in further detail below with reference to the accompanying drawings.
The application provides a precise rolling method of a numerical control four-roller plate rolling machine integrating machine vision and a machine learning algorithm. By integrating the machine vision and the XGBoost-based intelligent coupling influence analysis side roller position prediction model method, the bending process of the numerical control four-roller plate bending machine is quicker and more accurate, a large amount of labor and time cost are saved, and automation and intellectualization of plate bending are realized.
The application provides a precise rolling method of a numerical control four-roller plate bending machine integrating machine vision and a machine learning algorithm, which is shown in figure 1 and mainly comprises the following steps:
s1, acquiring performance parameters and forming process parameters of a plate material to be bent;
s2, inputting known parameters into a mathematical model to determine an initial side roller position;
s3, the industrial personal computer (upper computer) controls hydraulic servo through the motion control card, so that the side roller moves to the latest roller position;
s4, the upper roller of the four-roller plate bending machine rotates to drive the plate to be rolled and form a latest rolled plate part;
s5, performing image acquisition on the newly rolled plate part after rebound by using an area array CCD camera;
s6, performing image processing in an upper computer vision system, and fitting the real-time curvature radius of the latest coiled sheet material part;
s7, judging whether the deviation between the real-time curvature radius of the latest rolled sheet material part and the expected curvature radius is within an allowable error range;
s8, correcting the roller position of the side roller according to the deviation between the real-time curvature radius of the latest rolled sheet part and the expected curvature radius;
s9, storing the material performance parameters, the forming process parameters and the corresponding optimal roll position parameters of the rolled plate as a group of data in a database of an upper computer;
s10, when a new round of plate rolling is started, judging whether all parameter data amounts accumulated in a database meet set values;
s11, continuously rolling plates with different material performance parameters and forming process parameters by using a machine vision method;
s12, rolling a plate by using a machine learning method, constructing an XGBoost-based intelligent coupling influence analysis side roller position prediction model, and obtaining a trained and optimized roller position prediction model by taking 70% of a database as a training set and 30% as a test set;
s13, inputting the performance parameters and the forming process parameters of the plate material to be bent into a roll position prediction model, and outputting the roll position of the side roll by the model, so that the plate bending machine with certain structural parameters realizes one-time bending and forming of the plate material.
In some alternative embodiments, in step S1, the sheet material property parameter to be bent includes a sheet yield limit σ s The sheet thickness t and the sheet width b, and the forming process parameters comprise a sheet coiling feed speed V and a sheet expected curvature radius rho.
In the embodiment, referring to fig. 2, in step S2, the mathematical model is derived based on the planar geometric relationship among the upper roll, the lower roll, the side roll and the rolled sheet of the numerical control four-roll plate bending machine (side roll arc feeding); in order to simplify the mathematical model, the neutral layer is not changed in the bending forming process, the influence of the plate width factor is ignored, the influence of the gravity of the extending end of the plate is not considered, the plate does not slip in the bending process, the plate thickness is not changed in the bending forming process, and the side roller is set to be in an initial state when being in close contact with the upper roller; the mathematical model is as follows:
P＝Y-Y 1
wherein P is a side roller position parameter, Y is a hydraulic cylinder position corresponding to the side roller position in an initial state, Y 1 To reach the position of the hydraulic cylinder corresponding to the latest side roller position;
wherein g, k, mu and v are structural parameters of the plate bending machine and can be obtained from corresponding numerical control four-roller plate bending machine structural parameter handbooks,μ＝∠ADH，v＝∠BDK；
wherein the radius of the upper roller of the plate bending machine is R a The radius of the lower roller is R b Radius of left roller is R c The radius of the right roller is R d ,R a ＝R b ＝R 1 ,R c ＝R d ＝R 2 D and r are structural parameters of the plate bending machine, can be obtained from corresponding numerical control four-roller plate bending machine structural parameter handbooks,
wherein ρ is r To bend the inner diameter of the sheet before rebound,
γ＝δ-ε
wherein K is e E is the elastic modulus of the plate material, K is the relative strengthening coefficient of the plate s Is a plate section shape factor, generally rectangular in section, K s ＝1.5，ρ i To bend the inner diameter of the plate after rebound, K e E is obtained by checking performance parameter tables of different materials;
where ρ is the desired radius of curvature of the sheet to be bent.
In some optional embodiments, in step S3, the industrial personal computer is used as an upper computer, and a program is developed in the Visual Studio, so as to achieve the purpose of controlling the motion control card, and the motion control card controls the left and right side rollers in a hydraulic servo manner so that the side rollers reach a predetermined position.
In some alternative embodiments, referring to fig. 3, in step S5, the image of the sheet is acquired using a binocular area array CCD camera under the irradiation of an adjustable light source, and the CCD camera acquires the area after the sheet is bent and rebounded.
In some optional embodiments, in step S6, the image is preprocessed, grayed, binarized, and morphologically processed by an image processing program developed in the host computer, and then discrete curvature is calculated by using a difference operator, and finally a numerical curve is fitted to calculate a real-time curvature radius.
In some optional embodiments, in step S7, it is determined whether the deviation between the real-time radius of curvature of the latest rolled sheet portion and the desired radius of curvature is within an allowable error range, where the allowable error range is determined according to the actual situations such as accuracy of the four-roller plate bending machine, actual rolling accuracy requirement, and the like; if the deviation is within the set allowable error range, the expected curvature radius is considered to be satisfied, and then step S9 is performed; if the deviation exceeds the set allowable error range, the expected curvature radius is not met, and then step S8 is performed, the correction side roller position is calculated by the upper computer, then step S3, step S4, step S5 and step S6 are performed, and then judgment is performed again, and the iteration is repeated until the deviation is within the allowable range.
In some alternative embodiments, in step S8, the deviation is corrected by continuously advancing the side roller controlled by the motion control card, and visually recognizing the real-time curvature radius of the sheet once every time the side roller advances until the deviation is within the allowable error range.
In this embodiment, in step S10, when it is determined as yes, the data amount (set value) of each parameter must be: 5 different yield limits sigma s 5 different plate thicknesses t, 5 different plate widths b, the 3 different feed speeds V, the 5 desired radius of curvature ρ of the sheet, the 50 different side roller positions reference the data stored in the database in the following format:
in this embodiment, referring to fig. 4, in step S12, the prediction model refers to an XGBoost-based intelligent coupling influence analysis side roller position prediction model, and the model adopts an XGBoost integrated learning method, and uses a regression tree as a base model to perform classification regression of a bending forming multi-factor coupling influence relationship; constructing a complex influence coupling analyzer (SC) based on a gradient lifting method through associated regression tree joint decisions, wherein the constructed multi-factor coupling influence objective function is as follows:
P＝SC(t,b,σ s ,V,ρ)
where SC is a shaping relationship function under coupling.
In this embodiment, the data features of different orders of magnitude do not affect the model results, and normalization is not necessary. 70% of data in a database is used as a training set, the other 30% is used as a test set, a roll position prediction model based on XGBoost is subjected to parameter training, and an evaluation index of the model is mean square error of a side roll position; the training of the roller position prediction is the training of small sample data, a regular term needs to be added, the weight can be used as a super parameter, and finally the super parameter is regulated through Bayesian optimization until the set iteration times are reached.
The application solves the problems that the curvature radius size of the rolled plate needs to be measured manually by means of a tool and the roll position of the side roll cannot be calculated according to a large amount of rolled plate data for the plate with known rolled material performance parameters and forming process parameters, and provides a method for fusing machine vision and a machine learning algorithm to be applied to a numerical control four-roll plate bending machine; a large number of data sets are obtained by rolling the plate by adopting a machine vision method, and are stored in a database, wherein 70% of the data sets are used as training sets, the other 30% of the data sets are used as test sets, parameters in a roll position prediction model based on XGBoost are trained and optimized, and super parameters are regulated by utilizing Bayesian optimization, so that the accuracy and generalization capability of the model are improved, and one-time accurate rolling and forming of the plate when the performance parameters and the forming technological parameters of the rolled material are known by a plate rolling machine with certain structural parameters are realized, and automation and intellectualization of rolling and forming are realized.
Although the application has been described hereinabove with reference to embodiments, various modifications thereof may be made and equivalents may be substituted for elements thereof without departing from the scope of the application. In particular, the features of the disclosed embodiments may be combined with each other in any manner as long as there is no structural conflict, and the exhaustive description of these combinations is not given in this specification merely for the sake of omitting the descriptions and saving resources. Therefore, it is intended that the application not be limited to the particular embodiment disclosed, but that the application will include all embodiments falling within the scope of the appended claims.
Claims (10)
1. The accurate rolling method of the numerical control four-roller plate rolling machine is characterized by comprising the following steps of:
s1, acquiring performance parameters and forming process parameters of a plate material to be bent;
s2, inputting known parameters into a mathematical model to determine an initial side roller position;
s3, the upper computer controls hydraulic servo through the motion control card so that the side roller moves to the latest roller position;
s4, the upper roller of the four-roller plate bending machine rotates to drive the plate to be rolled and form a latest rolled plate part;
s5, performing image acquisition on the newly rolled plate part after rebound by using an area array CCD camera;
s6, performing image processing in an upper computer vision system, and fitting the real-time curvature radius of the latest coiled sheet material part;
s7, judging whether the deviation between the real-time curvature radius of the latest rolled sheet material part and the expected curvature radius is within an allowable error range;
s8, correcting the roller position of the side roller according to the deviation between the real-time curvature radius of the latest rolled sheet part and the expected curvature radius;
s9, storing the material performance parameters, the forming process parameters and the corresponding optimal roll position parameters of the rolled plate as a group of data in a database of an upper computer;
s10, when a new round of plate rolling is started, judging whether all parameter data amounts accumulated in a database meet set values;
s11, continuously rolling plates with different material performance parameters and forming process parameters by using a machine vision method;
s12, rolling a plate by using a machine learning method, constructing an XGBoost-based intelligent coupling influence analysis side roller position prediction model, and obtaining a trained and optimized roller position prediction model by taking 70% of a database as a training set and 30% as a test set;
s13, inputting the performance parameters and the forming process parameters of the plate material to be bent into a roll position prediction model, and outputting the roll position of the side roll by the model, so that the plate bending machine with certain structural parameters realizes one-time bending and forming of the plate material.
2. The method according to claim 1, wherein in the step S1, the performance parameters of the plate material to be bent include a plate yield limit σ s The sheet thickness t and the sheet width b, and the forming process parameters comprise a sheet coiling feed speed V and a sheet expected curvature radius rho.
3. The method for precisely rolling a numerically controlled four-roll bending machine according to claim 1, wherein in step S2, the mathematical model is as follows: p=y-Y 1
Wherein P is a side roller position parameter, Y is a hydraulic cylinder position corresponding to the side roller position in an initial state, Y 1 To reach the position of the hydraulic cylinder corresponding to the latest side roller position;
wherein g, k, mu and v are structural parameters of the plate bending machine and can be obtained from corresponding numerical control four-roller plate bending machine structural parameter manuals;
wherein the radius of the upper roller of the plate bending machine is R a The radius of the lower roller is R b Radius of left roller is R c The radius of the right roller is R d ,R a ＝R b ＝R 1 ,R c ＝R d ＝R 2 D and r are structural parameters of the plate bending machine and can be obtained from corresponding numerical control four-roller plate bending machine structural parameter manuals;
wherein ρ is r The inner diameter of the plate is bent before rebound;
γ＝δ-ε
L'＝(ρ r +t+R 1 )sinε
wherein K is e E is the elastic modulus of the plate material, K is the relative strengthening coefficient of the plate s Is a plate section shape factor, generally rectangular in section, K s ＝1.5，ρ i To bend the inner diameter of the plate after rebound, K e E is obtained by checking performance parameter tables of different materials;
where ρ is the desired radius of curvature of the sheet to be bent.
4. The method for precisely rolling a numerically controlled four-roll bending machine according to claim 1, wherein in the step S3, an industrial personal computer is used as an upper computer, and a program is developed in a visual studio to control a motion control card, and the motion control card controls hydraulic servos of left and right side rolls to enable the side rolls to reach a predetermined position.
5. The method for precisely rolling a plate by using the numerical control four-roller plate rolling machine according to claim 1, wherein in the step S5, the image of the plate is acquired by using a binocular area array CCD camera under the irradiation of an adjustable light source.
6. The method for precisely rolling a numerically controlled four-roll bending machine according to claim 1, wherein in step S6, the image is preprocessed, grayed, binarized, and morphologically processed by an image processing program developed in an upper computer, discrete curvatures are calculated by using a difference operator, and a numerical curve is fitted to calculate a real-time radius of curvature.
7. The method for precisely rolling a plate by using a numerical control four-roller plate rolling machine according to claim 1, wherein in the step S7, it is determined whether the deviation between the real-time radius of curvature of the latest rolled plate portion and the expected radius of curvature is within an allowable error range, and the allowable error range is determined according to the actual conditions such as the precision of the four-roller plate rolling machine, the actual rolling precision requirement, and the like; if the deviation is within the set allowable error range, the expected curvature radius is considered to be satisfied, and then step S9 is performed;
if the deviation exceeds the set allowable error range, the expected curvature radius is not met, and then step S8 is performed, the correction side roller position is calculated by the upper computer, then step S3, step S4, step S5 and step S6 are performed, and then judgment is performed again, and the iteration is repeated until the deviation is within the allowable range.
8. The method for precisely rolling a plate by using a numerical control four-roller plate rolling machine according to claim 1, wherein in the step S8, the deviation is corrected by controlling the side roller to continuously advance through the motion control card, and once for each advance, the real-time curvature radius of the plate is visually identified once until the deviation is within the allowable error range.
9. The method for precisely rolling a numerically controlled four-roll bending machine according to claim 1, wherein in the step S10, when the determination is yes, the data amount of each parameter must be reached: 5 different yield limits sigma s 5 different plate thicknesses t, 5 different plate widths b,3 different feeding speeds V, 5 plate expected curvature radiuses rho and 50 different side roller position parameters P, otherwise, judging as no; for data stored in the database, the data format is as follows:
10. the method for precisely rolling a numerical control four-roller plate bending machine according to claim 1, wherein in the step S12, when a machine learning algorithm is used, an intelligent coupling influence analysis side roller position prediction model based on XGBoost is constructed, an XGBoost integrated learning method is adopted, and a regression tree is used as a base model to carry out classification regression of a bending forming multi-factor coupling influence relation; constructing a complex influence coupling analyzer (SC) based on a gradient lifting method through associated regression tree joint decisions, wherein the constructed multi-factor coupling influence objective function is as follows: p=sc (t, b, σ s ,V,ρ)
The SC is a shaping relation function under the coupling effect, the data characteristics of different orders of magnitude do not influence the model result, normalization processing is not needed, 70% of data in a database is used as a training set, the other 30% is used as a test set, a roll position prediction model based on XGBoost is used for carrying out parameter training, and the evaluation index of the model is the mean square error of a side roll position; the training of the roller position prediction is the training of small sample data, a regular term needs to be added, the weight can be used as a super parameter, and finally the super parameter is regulated through Bayesian optimization until the set iteration times are reached.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311296783.8A CN117161160A (en) | 2023-10-08 | 2023-10-08 | Accurate rolling method of numerical control four-roller plate bending machine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311296783.8A CN117161160A (en) | 2023-10-08 | 2023-10-08 | Accurate rolling method of numerical control four-roller plate bending machine |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117161160A true CN117161160A (en) | 2023-12-05 |
Family
ID=88933782
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311296783.8A Pending CN117161160A (en) | 2023-10-08 | 2023-10-08 | Accurate rolling method of numerical control four-roller plate bending machine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117161160A (en) |
-
2023
- 2023-10-08 CN CN202311296783.8A patent/CN117161160A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9227236B2 (en) | Method and device for profile bending | |
| EP2667984B1 (en) | Automated pipe forming press comprising a light source for measuring the internal contour of the pipe | |
| CN111487869B (en) | Straightening control method and device | |
| DE19653569C2 (en) | Process for the automated management of a straightening process | |
| US4531391A (en) | Adaptive method and apparatus for correcting deviations in the shape of objects | |
| US10913099B2 (en) | End grooving system and process for tubing | |
| CN110116146B (en) | Intelligent threading method for high-frequency welded pipe | |
| CN113714735B (en) | Auxiliary method suitable for controlling heat treatment deformation of large aluminum alloy plate | |
| CN117161160A (en) | Accurate rolling method of numerical control four-roller plate bending machine | |
| JP2515217B2 (en) | Method and apparatus for bending metallic material by bend-ing-roll | |
| EP4151329A1 (en) | Steel pipe roundness prediction model generation method, steel pipe roundness prediction method, steel pipe roundness control method, steel pipe manufacturing method, and steel pipe roundness prediction device | |
| EP4151330A1 (en) | Steel pipe roundness prediction method, steel pipe roundness control method, steel pipe production method, method for generating steel pipe roundness prediction model, and steel pipe roundness prediction device | |
| CN205519880U (en) | Radiator main fin decides make -up machine automatically | |
| CN111318579B (en) | Roughing camber control method based on data driving | |
| CN117019948A (en) | Intelligent roll bending forming system and method | |
| EP1817129B1 (en) | High-speed milling cutter system and method for producing metallic guide elements | |
| CN113642180B (en) | Online sensing method for spinning forming state | |
| EP2683503B1 (en) | Method and device for angle-precise edging or bending of sheet metal plates | |
| CN209901969U (en) | Seven-roller straightener with front end rolling device and function of improving straightening precision of titanium alloy plate | |
| Feoktistov et al. | Technique for the development of the technological process for the tightening of aircraft parts on CNC presses | |
| US11565297B2 (en) | Bending machine and method for controlling the bending machine | |
| CN112859735B (en) | JCO forming tubular online closed-loop control method, system and device | |
| CN109344525A (en) | Metal roll bending scheduler emulation mode based on Virtual Manufacture | |
| RU2626253C2 (en) | Method of shaping bicurved sheet parts | |
| RU2799579C1 (en) | Steel pipe ovality prediction method, steel pipe ovality control method, steel pipe manufacturing method, steel pipe ovality prediction model generation, and steel pipe ovality prediction device |
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 |