CN113486587A - Seamless steel pipe perforation process model parameter prediction system based on Matlab - Google Patents

Seamless steel pipe perforation process model parameter prediction system based on Matlab Download PDF

Info

Publication number
CN113486587A
CN113486587A CN202110765429.XA CN202110765429A CN113486587A CN 113486587 A CN113486587 A CN 113486587A CN 202110765429 A CN202110765429 A CN 202110765429A CN 113486587 A CN113486587 A CN 113486587A
Authority
CN
China
Prior art keywords
diameter
roller
parameters
cone
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.)
Granted
Application number
CN202110765429.XA
Other languages
Chinese (zh)
Other versions
CN113486587B (en
Inventor
王清华
加世滢
胡建华
周新亮
赵铁林
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Taiyuan University of Science and Technology
Original Assignee
Taiyuan University of Science and Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Taiyuan University of Science and Technology filed Critical Taiyuan University of Science and Technology
Priority to CN202110765429.XA priority Critical patent/CN113486587B/en
Publication of CN113486587A publication Critical patent/CN113486587A/en
Application granted granted Critical
Publication of CN113486587B publication Critical patent/CN113486587B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B21—MECHANICAL METAL-WORKING WITHOUT ESSENTIALLY REMOVING MATERIAL; PUNCHING METAL
    • B21D—WORKING OR PROCESSING OF SHEET METAL OR METAL TUBES, RODS OR PROFILES WITHOUT ESSENTIALLY REMOVING MATERIAL; PUNCHING METAL
    • B21D28/00—Shaping by press-cutting; Perforating
    • B21D28/24—Perforating, i.e. punching holes
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/10—Geometric CAD
    • G06F30/12—Geometric CAD characterised by design entry means specially adapted for CAD, e.g. graphical user interfaces [GUI] specially adapted for CAD
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/044—Recurrent networks, e.g. Hopfield networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/084—Backpropagation, e.g. using gradient descent
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • General Physics & Mathematics (AREA)
  • Geometry (AREA)
  • General Engineering & Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Software Systems (AREA)
  • Computational Linguistics (AREA)
  • Computer Hardware Design (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Biophysics (AREA)
  • Mathematical Physics (AREA)
  • Biomedical Technology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Human Computer Interaction (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Architecture (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Mechanical Engineering (AREA)
  • Control Of Metal Rolling (AREA)
  • General Factory Administration (AREA)

Abstract

The invention relates to a parameter prediction system of a seamless steel tube perforation process model based on Matlab, belonging to the technical field of seamless steel tube perforation process model wall thickness prediction. The invention integrates the functions of data import, performance evaluation, model prediction, prediction result export and the like, reduces the workload, can predict the pipe type by using the BP neural network for known key influence parameters to guide the perforation production, ensures the production quality of the steel pipe and can ensure the prediction precision of the system during long-time monitoring.

Description

Seamless steel pipe perforation process model parameter prediction system based on Matlab
Technical Field
The invention belongs to the technical field of seamless steel tube perforation process model wall thickness prediction, and particularly relates to a parameter prediction system of a Matlab-based seamless steel tube perforation process model.
Background
With the rapid development of economy in China, the production of seamless steel tubes plays an important role in numerous fields of economic production in China, and the cross piercing process is most widely used in the production of seamless steel tubes. In order to improve the quality of products, reduce the cost required for production and increase the production rate of products, a number of new rolling software and management software have been developed, such as: the software for synchronously monitoring the wall thickness of the steel tube in the technical field of steel rolling can meet the requirement of high-precision seamless steel tube rolling.
In the production of seamless steel pipes by steel rolling science and technology, the former steel rolling production is manually operated by virtue of production experience, and now, due to the mature use of a mathematical model, the steel rolling production tends to be semi-automatically or automatically controlled by a computer. In a seamless steel tube production unit, a puncher punches a solid tube blank into a hollow blank tube, and is one of main production equipment in seamless steel tube production, so that the reasonably planned adjustment of the rolling process of the puncher is the fundamental guarantee for making high-quality steel tubes, and the reasonable process adjustment of the puncher is an important way for reducing the wall thickness quality defect of finished steel tubes. The most suitable parameters are determined, and the uniform wall thickness of the perforated capillary is ensured, which is very important for the steel rolling production.
MATLAB software is a high-level technical computing language and an interactive environment for algorithm development, data visualization, data analysis and numerical calculation, integrates a plurality of powerful functions of numerical analysis, matrix calculation, scientific data visualization, modeling and simulation of a nonlinear dynamic system and the like into an easy-to-use window environment, and provides a comprehensive solution for scientific research, engineering design and a plurality of scientific fields which need to carry out effective numerical calculation. The APP is an interface for information communication between a user and equipment by using a building module on a graphical interface, and a desired effect is achieved. The image identification is an important information processing technology, the digital image processing technology is applied to the prediction of the wall thickness of the seamless steel tube process model, the processing precision is high, the processing content is rich, the complex nonlinear processing can be carried out, the flexible flexibility is realized, and the processing content can be changed only by changing software.
Thus, the present invention was accomplished under the Matlab APP tool. The system is combined with Matlab programming to realize the establishment training and prediction of a network model, and provides a stable and reliable prediction method of the wall thickness of the seamless steel pipe perforation process model based on MATLAB APP, which has the advantages of simple design, high detection speed and high flexibility.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides a parameter prediction system of a seamless steel tube perforation process model based on Matlab, a prediction research system of each parameter of the seamless steel tube perforation process model based on a Matlab APP platform and prediction of the tube shape of the seamless steel tube by utilizing Matlab and adopting a BP neural network.
The invention is realized by the following technical scheme.
A parameter prediction system of a seamless steel tube perforation process model based on Matlab comprises a data import and export unit for realizing the calculation of process parameters, and a BP neural network for the training, verification and comprehensive evaluation of prediction effects of the tubular parameters of the seamless steel tube, and specifically comprises the following steps:
(1) establishing a MATLAB APP master-computer interactive interface, wherein the master-computer interactive interface consists of 8 input modules of blank information, roller design, kinematic parameters, top design, guide plate design, process parameters, tube type prediction and force and energy parameters, and can jump to a corresponding interface by writing a callback function of a corresponding menu in a code view and clicking the corresponding menu; according to the invention, a platform product is designed by utilizing Matlab APP, an interface does not need to be redesigned according to the production requirements of different units, the workload is reduced, the system can be used for verification to adjust parameters needing to be changed when finished products with different specifications are produced, and the influence on the control precision caused by errors due to adoption of empirical adjustment is reduced;
(2) each input module respectively establishes a corresponding sub-human-computer interaction interface, each sub-human-computer interaction interface is respectively provided with a corresponding design parameter input and output module, a Matlab function uigetfile () is used for acquiring a file path and a file name of a target Excel form to be imported and exported, a function xlsread () is used for reading data in the target form, and then a system imports the obtained process parameters corresponding to the perforating equipment and the seamless steel pipe type parameters into the target form through the function xlsrrite ();
(3) five parameters affecting the tube type were taken: the method comprises the following steps of taking the forward extension, the roll spacing, the guide plate spacing, the blank diameter and the plug diameter as the input of a BP neural network, training, verifying and predicting the technological parameters of the seamless steel tube by adopting the BP neural network to obtain the tubular prediction result of the perforated tubular billet, and comprises the following steps:
s1: the number of network layers: 1 layer of each of the input layer, the hidden layer and the output layer;
s2: initializing the network: establishing a seamless steel pipe cross-rolling perforated pipe type prediction model by using newff (), determining the number of neurons in an input layer and an output layer according to input and output variables, wherein the network structure is 5-10-2; setting a training function, a training algorithm and a normalization interval;
s3: setting the iteration frequency of the network as 2000 times, the learning rate as 0.1, and setting the allowable deviation of the outer diameter as 0.1 and the allowable deviation of the wall thickness as 0.1 according to the requirement of the national standard on the size deviation;
the BP neural network model can reflect the training effect of the BP neural network in real time by utilizing a relative error histogram of actual values of wall thickness and outer diameter and a predicted value and a comparison line graph of the actual values of the wall thickness and the outer diameter and the predicted value;
(4) and loading and storing the design parameters in the sub human-computer interaction interface in an excel form.
Further, in the step (1), the main human-computer interaction interface can independently open any sub human-computer interaction interface.
Further, in the step (2):
the blank information module in the sub-human-computer interaction interface comprises input windows of the following parameters: date, material, pipe blank diameter, blank length, tapping temperature, capillary diameter, capillary wall thickness, capillary length, and blank information parameters can be generated into a table form.
The roller design module in the sub-human-computer interaction interface comprises an input window of the following parameters: the pipe blank diameter, the roller shape, the roller number, the rolling angle, the roller inlet cone generatrix inclination angle and the roller outlet cone generatrix inclination angle comprise the following parameter output windows: roll diameter, roll length, entrance roll profile taper angle, exit roll profile taper angle, entrance cone length, compression band length, exit cone length, roll end fillet;
the kinematic parameter module in the sub-human-computer interaction interface comprises the following parameter input windows: the tube blank diameter, the tubular billet wall thickness, the roller diameter, the roller rotating speed, the roller number, the feeding angle, the tangential sliding coefficient of the outlet section and the friction coefficient comprise the following parameter output windows: radial rolling reduction, axial slip coefficient, outlet speed, feeding speed, tube blank revolution and screw pitch of the blank before the top;
the top design module in the sub-human-computer interaction interface comprises the following parameter input windows: the device comprises a top material, a pipe blank diameter, a tubular billet wall thickness, a roller inlet cone generatrix inclination angle and a screw pitch, and comprises the following parameter output windows: the diameter of the nose part, the length of the nose part, the diameter of the top head, the length of the arc part of the top head, a perforation cone, a uniform wall cone, a reverse cone and the inclination angle of the generatrix of the cone of the top head;
the guide plate design module in the sub-human-computer interaction interface comprises the following parameter input windows: tube blank diameter, hollow billet wall thickness, feed angle, roll diameter, roll entry cone generating line inclination, top diameter, pitch, including following parameter output window: the guide plate inlet inclined plane inclination angle, the guide plate outlet inclined plane inclination angle, the guide plate inlet cone width, the guide plate outlet cone width, the inlet groove depth and the outlet groove depth;
the process parameter module in the sub-human-computer interaction interface comprises the following parameter input windows: the blank information comprises the diameter of the tube blank, the diameter of the capillary and the wall thickness of the capillary, the roller information comprises an inlet cone angle and an outlet cone angle, and the plug information comprises the diameter of the plug and the length of the plug; the following parameter output windows are included: roll spacing, guide plate spacing, and top extension;
the force parameter module in the sub-human-computer interaction interface comprises the following parameter input windows: the pipe billet diameter, the hollow billet wall thickness, the screw pitch, the feed angle, the rolling angle, the roller inlet cone angle, the roller outlet cone angle, the roller diameter, the inclination angle of the top cone bus, the top extension, the outlet speed, the reduction gearbox transmission ratio, the gear base transmission efficiency, the spindle transmission efficiency, the motor rotating speed and the material deformation resistance comprise the following parameter output windows: the device comprises a rolling force, a top axial force, a total axial force of each roller, a guide plate axial resistance, a torque required for rotating the rollers, a bending torque, a guide plate tangential resistance torque to the rollers, a torque required by one roller, a roller bearing torque, an idling torque, a motor torque and a motor power.
Further, in the step S2, an empirical formula is adopted
Figure BDA0003150856100000031
And determining the number of the hidden layer neurons, wherein m is the number of the hidden layer neurons, a and b are the number of the input layer neurons and the output layer neurons respectively, and c is a regulation constant between 1 and 10.
Further, in the step S2, five parameters of the forward extension, the roll gap, the guide plate gap, the billet diameter and the plug diameter are selected as input variables of the predictive model of the cross-piercing pipe type, and the wall thickness and the outer diameter are selected as output variables. The cross piercing is the first step of metal deformation in seamless steel pipe production, and the geometric dimension of the hollow billet is mainly determined by the outer diameter and the wall thickness. The deformation of a rolled piece is large during punching, according to the geometric relation of a deformation area and considering a bounce formula, the traditional tubular pipe mathematical model is as follows:
the outer diameter value is as follows:
Figure BDA0003150856100000041
wall thickness value:
Figure BDA0003150856100000042
wherein the roll gap b is (1-delta) dzAccording to empirical data, the compression quantity delta at a compression zone is 15-16%, and delta is 15.5%; l is the guide plate spacing in mm; d is the diameter of the roller, and the unit is mm; dz is the diameter of the tube blank, and the unit is mm; lp is the length of the arc part of the plug, and the unit is mm; y is the amount of protrusion in mm; beta is a feed angle, and the unit is degree; alpha is alpha2Is the roller outlet cone angle, and the unit is degree; δ n is the plug diameter in mm.
Figure BDA0003150856100000043
And
Figure BDA0003150856100000044
the values of the outer diameter and the wall thickness, respectively, are in mm.
In the mechanism model, the rigidity and the bounce value of the puncher are considered in the calculation of the outer diameter and the wall thickness, and the bounce value is obtained by experience, so that errors are inevitably generated.
According to the invention, a BP neural network is adopted to establish a tubular prediction model, too many input variables can complicate the network structure, and the selection is not comprehensive enough, so that the model precision is reduced. Selecting key parameters influencing tube shape as input vectors of BP neural network training, and using the outer diameter and the wall thickness as output vectors of the BP neural network training; and determining an ideal BP neural network prediction model according to the number and the output vector of the key parameters and the multiple times of training of the BP neural network and the comprehensive evaluation of the reference training effect.
Further, in the step S2, the transfer function used by each layer is tansig and purelin, and is trained by using logsig function; the training algorithm is set as a back propagation algorithm, a quasi-Newton back propagation algorithm, an elastic back propagation algorithm or a gradient descent back propagation algorithm; the normalization interval is set to [ -1,1 ].
Compared with the prior art, the invention has the beneficial effects that:
1. based on MATLAB APPDESIGER Chinese language environment, design nimble design user operation interface, need not redesign whole APP system to different steel pipe materials, only need compile configuration file can, avoid the transformation to current system in manufacturing and testing process, reduced work load, have actual meaning.
2. The known key influence parameters can utilize a BP neural network to predict the tube shape so as to guide the perforating equipment to adjust process parameters before production, and the tube shape precision during system production can be ensured.
3. The method integrates functions of data import, performance evaluation, model prediction, prediction result export and the like, can meet the requirement of scientific research, and has commercial application value.
Drawings
FIG. 1 is a schematic diagram of a BP neural network topology;
FIG. 2 is a flowchart of a BP neural network based tube type prediction training;
Detailed Description
The following examples are intended to illustrate the invention but are not intended to limit the scope of the invention. Unless otherwise specified, the examples follow conventional experimental conditions.
The Matlab-based seamless steel tube piercing process model parameter prediction system shown in fig. 1 and 2 comprises the following steps:
(1) establishing an MATLAB APP master-computer interactive interface, wherein the master-computer interactive interface consists of 8 input modules of parameter information, roller design, kinematic parameters, top design, guide plate design, process parameters, tube type prediction and force and energy parameters, and can jump to a corresponding interface by writing a callback function of a corresponding menu in a code view and clicking the corresponding menu;
(2) each input module respectively establishes a corresponding sub-human-computer interaction interface, each sub-human-computer interaction interface is respectively provided with a corresponding design parameter input and output module, a Matlab function uigetfile () is used for acquiring a file path and a file name of a target Excel form to be imported and exported, a function xlsread () is used for reading data in the target Excel form, and a system imports the obtained process parameters corresponding to the perforating equipment and the seamless steel pipe type parameters into the target form through the function xlsrrite () after calculation;
(3) five parameters affecting the tube type were taken: the method comprises the following steps of taking the forward extension, the roll spacing, the guide plate spacing, the blank diameter and the plug diameter as the input of a BP neural network, training, verifying and predicting the technological parameters of the seamless steel tube by adopting the BP neural network to obtain the tubular prediction result of the perforated tubular billet, and comprises the following steps:
s1: the number of network layers: 1 layer of each of an input layer, a hidden layer and an output layer is arranged;
s2: initializing the network: establishing a seamless steel pipe cross-rolling perforated pipe type prediction model by using newff (), determining the number of neurons in an input layer and an output layer according to input and output variables, wherein the network structure is 5-10-2; setting a training function, a training algorithm and a normalization interval;
s3: setting the iteration frequency of the network as 2000 times, the learning rate as 0.1, and setting the allowable deviation of the outer diameter as 0.1 and the allowable deviation of the wall thickness as 0.1 according to the requirement of the national standard on the size deviation;
(4) and loading and storing the design parameters in the sub human-computer interaction interface in an excel form.
The invention is based on the prediction of an MATLAB seamless steel pipe perforation process model, utilizes the MATLAB APP and codes of all sub-modules to design and complete a human-computer interface interaction system of the seamless steel pipe perforation process model prediction, and collects data of a certain steel mill for prediction and verification.
The APP Designer comprises buttons, text areas, editing fields and the like, an interface is built by adopting a component library, after interface layout is completed, an MATLAB can automatically create an m-code file according to the control type, and a system can store the fig file and the m file which are required to be in the same folder and are respectively used for storing a design view and a code view of the interface; on the basis, a callback function required to be realized is compiled for the submodule, data obtained by calculation in Matlab is written into a target table by using xlsread, data in an excel file is read by xlsread, a path and a file name of the file are obtained by uiputfile, selected file information is obtained by uigetfile, and fulllfile is created and combined into a complete file name by using information of each part of the file. The calculation result of the output parameter is realized by establishing a traditional formula in the code view.
The general interface mainly comprises a user name, a password and user login. And 7 modules of roll design, kinematic parameters, plug design, guide plate design, process parameters, pipe type prediction and force parameters are arranged above the main interface, and the parameters can be directly entered into the parameter information module after successful login.
And by writing a callback function of the corresponding menu in the code view, clicking the corresponding menu can jump to the corresponding interface.
The parameter information interface inputs the acquired data into the corresponding blank information module and the corresponding capillary information module, a data reading button is arranged, the data is displayed in a product information reading interface, a callback function is compiled for a data storage button, and the known information is stored in an Excel sheet1 form.
The roll design parameter setting interface sets a two-roll cone, sets a load data button to program a callback function to load the data saved in the Excel sheet1 into the roll design parameter module. Setting a calculation button, compiling a traditional formula to calculate the diameter of the roller, the length of the roller, the taper angle of the inlet roller, the taper angle of the outlet roller, the length of the inlet cone, the length of the compression band, the length of the outlet cone and the round angle of the roller end, displaying the calculated data in a roller parameter calculation interface, setting a data storage button, and storing the calculated data and the names of the data in Excel sheet2 in a one-to-one correspondence manner. And setting a return button, and returning to the roller design parameter interface from the interface, wherein the return button can return to the menu interface.
The kinematic parameter setting interface is provided with a data loading button, and parameters required for calculating output parameters are loaded to the kinematic parameter interface from Excel sheet1 and sheet 2. Setting a calculation button, compiling a traditional formula to calculate the radial rolling reduction, the axial slip coefficient, the outlet speed, the feeding speed, the tube blank revolution and the screw pitch of the blank before the plug, displaying the calculated data in a kinematic parameter calculation interface, setting a data storage button, and storing the calculated data and the names of the data in Excel sheet3 in a one-to-one correspondence manner. And setting a return button, and returning to the kinematic parameter interface from the interface, wherein the kinematic parameter interface setting return button can return to the menu interface.
The plug design parameter setting interface is provided with a data loading button, and parameters required by calculating output parameters are loaded to the plug design parameter interface from Excel sheet1, sheet2 and sheet 3. Setting a calculation button, compiling a traditional formula to calculate the nasal diameter, the nasal length, the plug diameter, the length of the arc part of the plug, the inclination angle of the perforation cone, the uniform wall cone, the reverse cone and the bus of the plug cone, displaying the calculated data in a plug design calculation interface, setting a data storage button, and storing the calculated data and the names of the data in Excel sheet4 in a one-to-one correspondence manner. And setting a return button, and returning to the plug design parameter interface from the interface, wherein the plug design parameter interface setting return button can return to the menu interface.
And a data loading button is arranged on the guide plate design parameter setting interface, and parameters required by calculating output parameters are loaded to the guide plate design parameter interface from Excel sheet1, sheet2, sheet3 and sheet 4. Setting a calculation button, compiling a traditional formula to calculate the inclination angle of the guide plate inlet inclined plane, the inclination angle of the guide plate outlet inclined plane, the width of the guide plate inlet cone, the width of the guide plate outlet cone, the depth of the inlet groove and the depth of the outlet groove, displaying the calculated data in a guide plate design calculation interface, setting a data storage button, and storing the calculated data and the names of the data in Excel sheet5 in a one-to-one correspondence manner. And a return button is arranged, the interface can return to the guide plate design parameter interface, and the guide plate design parameter interface can return to the menu interface.
And a data loading button is arranged on the process parameter setting interface, and parameters required by calculating output parameters are loaded into the process parameter interface from an Excel sheet1, a sheet2 and a sheet 4. Setting a calculation button, compiling a traditional formula to calculate the roller distance, the guide plate distance and the plug extension, displaying the calculated data in a technological parameter calculation interface, setting a storage button, and storing the calculated data and the names of the data in an Excel sheet6 in a one-to-one correspondence manner. And setting a return button, returning to the process parameter interface from the interface, and returning to the menu interface by the process parameter interface setting return button.
And a data loading button is arranged on a traditional formula parameter setting interface in the pipe type prediction, and parameters required by calculating output parameters are loaded into the traditional formula parameter interface from an Excel sheet1, a sheet2, a sheet3, a sheet4 and a sheet 6. Setting a calculation button, compiling a traditional formula to calculate the length, the outer diameter and the wall thickness of the pipe, displaying the calculated data in a tubular calculation interface of the traditional formula, setting a data storage button, storing the calculated data and the names of the data in an Excel sheet7 in a one-to-one correspondence manner, setting a return button, and returning to the traditional formula parameter setting interface from the interface, wherein the return button for setting the traditional formula parameter interface can return to a menu interface.
The prediction interface of the tubular prediction BP neural network comprises three modules: loading a data set, and importing training and testing data; setting a network structure 5-10-2; the training functional area sets iteration times, a learning rate, training target settings, a training function and a training algorithm. Setting an emptying button; the network training button displays the training result to a BP neural network prediction result interface, a result export button is arranged, a return button can return to the BP neural network prediction interface from the interface, and the BP neural network prediction interface is provided with a return button which can return to a menu interface.
And predicting the outer diameter and the wall thickness of the capillary according to the trained prediction model. And setting a storage button, and storing the calculated data and the names thereof in an Excel sheet7 in a one-to-one correspondence manner. A return button is provided from which a return to the menu interface is possible.
The uniformity of the wall thickness is one of the main factors influencing the precision of the product, the factors influencing the wall thickness are many in the process of perforating the steel pipe, the relation between each factor and perforation is complex, whether the wall thickness is qualified or not is predicted according to the main parameters, and the parameters which can generate the product disqualification due to manual experience adjustment or unreasonable setting of certain parameters can be adjusted.
And a data loading button is arranged on the force and energy parameter setting interface, and parameters required by calculating output parameters are loaded into the force and energy parameter interface from Excel sheet1, sheet2, sheet3, sheet4 and sheet 6. Setting a calculation button, compiling a traditional formula to calculate the rolling force, the axial force of a top head, the total axial force of each roller, the axial resistance of a guide plate, the torque required by rotating the roller, the bending torque, the tangential resistance torque of the guide plate to the roller, the torque required by one roller, the torque of a roller bearing, the idle torque, the motor torque and the motor power, displaying the calculated result in the calculation of force energy parameters, setting a storage button, and storing the calculated data and the name thereof in an Excel sheet8 in a one-to-one correspondence manner. And setting a return button, and returning to the force and energy parameter interface from the interface, wherein the force and energy parameter interface setting return button can return to the menu interface.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the appended claims.

Claims (6)

1. A parameter prediction system of a seamless steel tube perforation process model based on Matlab is characterized by comprising the following steps:
(1) establishing an MATLAB APP master-computer interactive interface, wherein the master-computer interactive interface consists of 8 input modules of parameter information, roller design, kinematic parameters, top design, guide plate design, process parameters, tube type prediction and force and energy parameters, and can jump to a corresponding interface by writing a callback function of a corresponding menu in a code view and clicking the corresponding menu;
(2) each input module respectively establishes a corresponding sub-human-computer interaction interface, each sub-human-computer interaction interface is respectively provided with a corresponding design parameter input and output module, a Matlab function uigetfile () is used for acquiring a file path and a file name of a target Excel form to be imported and exported, a function xlsread () is used for reading data in the target Excel form, and a system imports the obtained process parameters corresponding to the perforating equipment and the seamless steel pipe type parameters into the target form through the function xlsrrite () after calculation;
(3) five parameters affecting the tube type were taken: the method comprises the following steps of taking the forward extension, the roll spacing, the guide plate spacing, the blank diameter and the plug diameter as the input of a BP neural network, training, verifying and predicting the technological parameters of the seamless steel tube by adopting the BP neural network to obtain the tubular prediction result of the perforated tubular billet, and comprises the following steps:
s1: the number of network layers: 1 layer of each of an input layer, a hidden layer and an output layer is arranged;
s2: initializing the network: establishing a seamless steel pipe cross-rolling perforated pipe type prediction model by using newff (), determining the number of neurons in an input layer and an output layer according to input and output variables, wherein the network structure is 5-10-2; setting a training function, a training algorithm and a normalization interval;
s3: setting the iteration frequency of the network as 2000 times, the learning rate as 0.1, and setting the allowable deviation of the outer diameter as 0.1 and the allowable deviation of the wall thickness as 0.1 according to the requirement of the national standard on the size deviation;
(4) and loading and storing the design parameters in the sub human-computer interaction interface in an excel form.
2. The Matlab-based seamless steel tube piercing process model parameter prediction system of claim 1, wherein: in the step (1), the main human-computer interaction interface can independently open any sub human-computer interaction interface.
3. The Matlab-based seamless steel tube piercing process model parameter prediction system of claim 1, wherein: in the step (2):
the blank information module in the sub-human-computer interaction interface comprises input windows of the following parameters: date, material, pipe blank diameter, blank length, tapping temperature, capillary diameter, capillary wall thickness, capillary length, and blank information parameters can be generated into a table form.
The roller design module in the sub-human-computer interaction interface comprises an input window of the following parameters: the pipe blank diameter, the roller shape, the roller number, the rolling angle, the roller inlet cone generatrix inclination angle and the roller outlet cone generatrix inclination angle comprise the following parameter output windows: roll diameter, roll length, entrance roll profile taper angle, exit roll profile taper angle, entrance cone length, compression band length, exit cone length, roll end fillet;
the kinematic parameter module in the sub-human-computer interaction interface comprises the following parameter input windows: the tube blank diameter, the tubular billet wall thickness, the roller diameter, the roller rotating speed, the roller number, the feeding angle, the tangential sliding coefficient of the outlet section and the friction coefficient comprise the following parameter output windows: radial rolling reduction, axial slip coefficient, outlet speed, feeding speed, tube blank revolution and screw pitch of the blank before the top;
the top design module in the sub-human-computer interaction interface comprises the following parameter input windows: the device comprises a top material, a pipe blank diameter, a tubular billet wall thickness, a roller inlet cone generatrix inclination angle and a screw pitch, and comprises the following parameter output windows: the diameter of the nose part, the length of the nose part, the diameter of the top head, the length of the arc part of the top head, a perforation cone, a uniform wall cone, a reverse cone and the inclination angle of the generatrix of the cone of the top head;
the guide plate design module in the sub-human-computer interaction interface comprises the following parameter input windows: tube blank diameter, hollow billet wall thickness, feed angle, roll diameter, roll entry cone generating line inclination, top diameter, pitch, including following parameter output window: the guide plate inlet inclined plane inclination angle, the guide plate outlet inclined plane inclination angle, the guide plate inlet cone width, the guide plate outlet cone width, the inlet groove depth and the outlet groove depth;
the process parameter module in the sub-human-computer interaction interface comprises the following parameter input windows: the blank information comprises the diameter of the tube blank, the diameter of the capillary and the wall thickness of the capillary, the roller information comprises an inlet cone angle and an outlet cone angle, and the plug information comprises the diameter of the plug and the length of the plug; the following parameter output windows are included: roll spacing, guide plate spacing, and top extension;
the force parameter module in the sub-human-computer interaction interface comprises the following parameter input windows: the pipe billet diameter, the hollow billet wall thickness, the screw pitch, the feed angle, the rolling angle, the roller inlet cone angle, the roller outlet cone angle, the roller diameter, the inclination angle of the top cone bus, the top extension, the outlet speed, the reduction gearbox transmission ratio, the gear base transmission efficiency, the spindle transmission efficiency, the motor rotating speed and the material deformation resistance comprise the following parameter output windows: the device comprises a rolling force, a top axial force, a total axial force of each roller, a guide plate axial resistance, a torque required for rotating the rollers, a bending torque, a guide plate tangential resistance torque to the rollers, a torque required by one roller, a roller bearing torque, an idling torque, a motor torque and a motor power.
4. The Matlab-based seamless steel tube piercing process model parameter prediction system of claim 1, wherein: in the step S2, an empirical formula is adopted
Figure FDA0003150856090000021
And determining the number of the hidden layer neurons, wherein m is the number of the hidden layer neurons, a and b are the number of the input layer neurons and the output layer neurons respectively, and c is a regulation constant between 1 and 10.
5. The Matlab-based seamless steel tube piercing process model parameter prediction system of claim 1, wherein: in step S2, five parameters, i.e., the forward extension, the roll gap, the guide plate gap, the billet diameter, and the plug diameter, are selected as input variables of the predictive model of the cross-piercing tube type, and the wall thickness and the outer diameter are selected as output variables.
6. The Matlab-based seamless steel tube piercing process model parameter prediction system of claim 1, wherein: in step S2, the transfer function used by each layer is tansig and purelin, the logsig function is used for training, and the training algorithm is set as a back propagation algorithm, a quasi-newton back propagation algorithm, an elastic back propagation algorithm or a gradient descent back propagation algorithm; the normalization interval is set to [ -1,1 ].
CN202110765429.XA 2021-07-06 2021-07-06 Seamless steel pipe perforation process model parameter prediction system based on Matlab Active CN113486587B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110765429.XA CN113486587B (en) 2021-07-06 2021-07-06 Seamless steel pipe perforation process model parameter prediction system based on Matlab

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110765429.XA CN113486587B (en) 2021-07-06 2021-07-06 Seamless steel pipe perforation process model parameter prediction system based on Matlab

Publications (2)

Publication Number Publication Date
CN113486587A true CN113486587A (en) 2021-10-08
CN113486587B CN113486587B (en) 2022-08-05

Family

ID=77941507

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110765429.XA Active CN113486587B (en) 2021-07-06 2021-07-06 Seamless steel pipe perforation process model parameter prediction system based on Matlab

Country Status (1)

Country Link
CN (1) CN113486587B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117055487A (en) * 2023-08-24 2023-11-14 北京科技大学 Two-roller skew rolling perforation parameter optimization method based on mechanism data hybrid driving
CN117259447A (en) * 2023-11-23 2023-12-22 常州润来科技有限公司 Self-adaptive adjustment method and system for copper pipe rotary rolling production parameters

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103886159A (en) * 2014-04-03 2014-06-25 东北大学 DEFORM finite element modeling for seamless steel tube cross piercing process
CN104070083A (en) * 2014-06-27 2014-10-01 东北大学 Method for measuring rotating speed of guiding disc of perforating machine based on integrated PCA-ELM (Principal Component Analysis)-(Extrem Learning Machine) method
CN105334269A (en) * 2015-10-19 2016-02-17 江苏大学 Pipeline defect type determination method based on neural network and guided wave characteristic database
US20160133843A1 (en) * 2013-04-04 2016-05-12 The Board Of Trustees Of The University Of Illinois Purification of carbon nanotubes via selective heating
CN107506565A (en) * 2017-10-16 2017-12-22 安徽工业大学 Method for building up based on neutral net alloy cast steel roll Expert System for Materials Design
CN108388702A (en) * 2018-01-30 2018-08-10 河南工程学院 Engineering ceramics electrical discharge machining effect prediction method based on PSO neural networks
CN111291513A (en) * 2020-01-22 2020-06-16 苏州大学 BP neural network-based differential plate rolling force prediction method
WO2020234546A1 (en) * 2019-05-20 2020-11-26 Vallourec Tubes France Method and terminal for generating an index of compatibility between two ends of two tubes, and tube provided with an angular marking

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160133843A1 (en) * 2013-04-04 2016-05-12 The Board Of Trustees Of The University Of Illinois Purification of carbon nanotubes via selective heating
CN103886159A (en) * 2014-04-03 2014-06-25 东北大学 DEFORM finite element modeling for seamless steel tube cross piercing process
CN104070083A (en) * 2014-06-27 2014-10-01 东北大学 Method for measuring rotating speed of guiding disc of perforating machine based on integrated PCA-ELM (Principal Component Analysis)-(Extrem Learning Machine) method
CN105334269A (en) * 2015-10-19 2016-02-17 江苏大学 Pipeline defect type determination method based on neural network and guided wave characteristic database
CN107506565A (en) * 2017-10-16 2017-12-22 安徽工业大学 Method for building up based on neutral net alloy cast steel roll Expert System for Materials Design
CN108388702A (en) * 2018-01-30 2018-08-10 河南工程学院 Engineering ceramics electrical discharge machining effect prediction method based on PSO neural networks
WO2020234546A1 (en) * 2019-05-20 2020-11-26 Vallourec Tubes France Method and terminal for generating an index of compatibility between two ends of two tubes, and tube provided with an angular marking
CN111291513A (en) * 2020-01-22 2020-06-16 苏州大学 BP neural network-based differential plate rolling force prediction method

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
XIAO D 等: "Process monitoring and fault diagnosis for piercing production of seamless tube", 《MATHEMATICAL PROBLEMS IN ENGINEERING》 *
加世滢 等: "基于MATLAB应用程序组件的无缝钢管穿孔工艺软件开发", 《钢管》 *
施少华: "无缝钢管斜轧穿孔能耗和穿孔过程的建模方法研究", 《中国优秀博硕士学位论文全文数据库(硕士)工程科技Ⅰ辑》 *
杨坤等: "基于遗传神经网络的钢管壁厚预测模型的研究", 《枣庄学院学报》 *
林伟路: "AZ31镁合金无缝管斜轧穿孔工艺的研究", 《中国优秀博硕士学位论文全文数据库(硕士)工程科技Ⅰ辑》 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117055487A (en) * 2023-08-24 2023-11-14 北京科技大学 Two-roller skew rolling perforation parameter optimization method based on mechanism data hybrid driving
CN117055487B (en) * 2023-08-24 2024-04-16 北京科技大学 Two-roller skew rolling perforation parameter optimization method based on mechanism data hybrid driving
CN117259447A (en) * 2023-11-23 2023-12-22 常州润来科技有限公司 Self-adaptive adjustment method and system for copper pipe rotary rolling production parameters
CN117259447B (en) * 2023-11-23 2024-02-02 常州润来科技有限公司 Self-adaptive adjustment method and system for copper pipe rotary rolling production parameters

Also Published As

Publication number Publication date
CN113486587B (en) 2022-08-05

Similar Documents

Publication Publication Date Title
CN113486587B (en) Seamless steel pipe perforation process model parameter prediction system based on Matlab
CN111353656B (en) Steel enterprise oxygen load prediction method based on production plan
CN104217063A (en) Simulation design method for rolling hole patterns of steel rail
CN100445901C (en) Dynamic cost control method for industrial process of procedure based on AR(p)model
Kao et al. Development of an integrated CAD/CAE/CAM system on taper-tipped thread-rolling die-plates
CN101320031B (en) Austenitic stainless steel accurate steel strip performance prediction model and cold rolling process planning thereof
CN105929812A (en) Strip steel hot continuous rolling quality fault diagnosis method and device
CN115141926B (en) Method for generating billet target heating curve for guiding billet heating process control
CN113901725A (en) Intelligent prediction system and method for continuous rolling process of seamless steel pipe based on Matlab
CN113569343A (en) Method, device and system for controlling rolling thickness of plate and storage medium
Li et al. Integrated predictive maintenance approach for multistate manufacturing system considering geometric and non-geometric defects of products
Wang et al. Design and optimization of rolling mills pass based on parameterization and orthogonal test
CN101387868A (en) Strip mill model control system and control method for self-adapting different heating-furnace conditions
Peng et al. Digital twin-enabled production optimization for steel industry
Pourgol-Mohammad et al. A practical sensitivity analysis method for ranking sources of uncertainty in thermal–hydraulics applications
JP2023076513A5 (en)
Zhu et al. Calculation method and analysis of residual stress in the strip bending roller straightening process
CN112329089B (en) Hot rolling rod wire rod full-process temperature field calculation system based on rolling program table
Zhang et al. Feature based integrated intelligent sequence design for cold extrusion
TWM604243U (en) System for optimizing metal stamping process parameters
CN113806690A (en) Intelligent optimization method and system for steel bar blanking and storable medium
Kim et al. Development of an integrated computer-aided process planning system for press working products
Kang et al. A study on roll wear in the roll forming process
Hernandez Computational Evaluation of Complexity Parameters of Machining and Manufacturing Systems
Kumar et al. Trends and developments in intelligent computer aided design of progressive dies

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