CN106202727A - The method for designing of irregular cable drawing mould and system - Google Patents
The method for designing of irregular cable drawing mould and system Download PDFInfo
- Publication number
- CN106202727A CN106202727A CN201610545822.7A CN201610545822A CN106202727A CN 106202727 A CN106202727 A CN 106202727A CN 201610545822 A CN201610545822 A CN 201610545822A CN 106202727 A CN106202727 A CN 106202727A
- Authority
- CN
- China
- Prior art keywords
- mould
- elongation percentage
- section
- design
- area
- 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
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/17—Mechanical parametric or variational design
Landscapes
- Physics & Mathematics (AREA)
- Geometry (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Computational Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Metal Extraction Processes (AREA)
Abstract
The invention provides method for designing and the system of a kind of irregular cable drawing mould, described method is a kind of on emulation and artificial intelligence basis, the multi pass drawing that proposition finite element simulation, artificial neural network, genetic algorithm combine with software modeling joins the method that mould designs, and this method for designing can ensure that mould local and the load balancing of the overall situation.The present invention is by emulation, optimization and software design, the pass of drawing mould is optimized design, the performance of drawing mould and drawing equipment can not only be given full play to, and for high-speed automated drawing production line, while greatly extending die life, also reducing and change the time safeguarding mould, production efficiency is largely increased, particularly significant to the economic benefit of the whole metal wire of raising, tube-drawing industry.
Description
Technical field
The present invention relates to the method for designing of technical field of mold, in particular it relates to one is based on artificial intelligence and finite element
The method for designing of the irregular cable drawing mould of analysis and load balancing and system.
Background technology
Irregular cable, for circular cross section wire, is the various different geometries in addition to circular cross-section
Noncircular cross section wire.Special-shaped wire is widely used in the fields such as space flight, machinery, automobile, electronics industry and communication.Intelligent electric
Net is as one of seven great strategy new industries, the wires and cables industry development by leaps and bounds that will promote China.Meanwhile, new energy power station
Construction, the development of equipment manufacturing, the popularization etc. of track traffic the most greatly expanded the new application field of cable.Special-shaped wire
Material is widely used in concentric gallows empty conductor and electromagnetic wire, and compared with round wires concentric gallows empty conductor, molded line gallows empty conductor has
There is bigger cross-sectional area of conductor utilization rate, diameter about 10% when cross section is identical, can be reduced;During equal diameter, then can increase effective cross-section
20%~25%.For improving the transmittability of extra-high voltage transmission line, the cable conductor of larger cross-section must use milliken conductor, with
The conductor resistance caused because of " kelvin effect " and " kindred effect " when reducing alternating current transmission increases and the decline of transmission line capability.Point
Cut conductor and typically use special-shaped wire (such as " S-shaped, Z-shaped " and ladder-shaped), be to utilize owing to special-shaped wire has increase cross-sectional area of conductor
Rate, the conducting wire sag rate that reduces, antiseptic property are good, broken string infringement is little, self-damping performance strong, reduce wind load, reduce aerial condutor dance
The advantages such as dynamic occurrence probability.
Irregular cable the most especially developed country is the most widely used, and the most domestic production and application are also in rising
Step section, the index such as the machining accuracy of drawing die, ruggedness all remains to be further improved.The quality of irregular cable mould is high-quality
The guarantee that the special-shaped electric wire of amount produces, therefore the research of special-shaped drawing mould is to the development of China's wires and cables industry and joint
Can have great importance by consumption reduction.
At present, the main method using roller die drawing and fixing die drawing to pull out of the production of domestic shaped wire.Roller die drawing is real
Matter is drawing pulls out, with rolling combines, the process of rolling, and roller is to be rotated with frictional force therebetween, in pass by drawing part
Realize drawing part deformation.Compared with pulling out with fixing die drawing, the pulling capacity of roller die drawing is little, and energy consumption is low, and compression ratio is big, manufacturing limit
Height, low to lubricating requirement, die life is long, requires blank low (such as character such as the cross section deformation of blank, change in size, welding
Inequality and surface defect etc.), the elastic limit of finished product and yield strength improve, and section hardness distribution uniform, residual stress is little
Etc. advantage, but Product Precision is relatively fixed die drawing and is pulled out low, still needs to carry out follow-up size finishing and processes.The production of fixing mould generally uses
The method of mould is joined in wire drawing.The method of mould is joined in wire drawing, according to finished size, shape and material during metal wire drawing, determines circle
Base or the size of slab, determine drawing passage and every time size and shape, and the method determining the intermediate annealing time.But
The method using wire drawing to join mould draws molded line, more serious when the inhomogeneities ratio of workpiece deformation draws round wires, the residual stress of formation
Bigger, cause profile bending, size inaccuracy, localized cracks probability the biggest.Therefore, shaped wire drawing die is being carried out
During tool design, the deficiency of traditional design method should be overcome, solve size and the stress brought in traditional design by big data
Problem.
It is known that artificial intelligence is research, is developed for simulation, extends and extends the theory of intelligence of people, method, skill
Art and new science of technology of application system, it is intended to how research utilizes the modernization instrument germline systems such as computer to imitate
The intelligent behavior of the mankind.The analysis developing into creation data and information of artificial intelligence technology and process provide effective side
Method, adds the wing of intelligence to manufacturing technology.Artificial intelligence technology is particularly suitable for solving especially complex and uncertain asks
Topic, the links in manufacture process nearly all can extensively apply artificial intelligence technology.Neutral net is as collection brain science, nerve
The multi-disciplinary crossing research such as psychology and information science field, multiple sides such as pattern recognition, machine learning, specialist systems
Face is applied, and becomes the active area in artificial intelligence study.The present invention is directed to the feature of the stronger learning capacity of neutral net,
The software modeling of section design method based on continuously linear change, combines neutral net with finite element technique, is applied to
The design of shaped wire drawing mould not only simplify the design cycle of mould, it is ensured that the accuracy of Design of Dies size, and
The aspects such as the wearability of mould, service life are made to have the biggest lifting.
Finding by prior art documents, Zhong Su proposed one in 2011 and utilizes CAXA CAXA, if
The method of meter bracing wire mould middle transition hole tee section.In the case of drawing passes and final shaped cross-section shape determine, to circle
Tee section is transitioned into the die draft in molded line requirement cross section and is designed.Finally with the form of line transect, approximation has drawn each mistake
Cross hole tee section, and utilize CAXA software to be processed on WEDM, thus obtain more accurate pass.But
Being that this method for designing needs to study intermetallic composite coating flow behavior, modeling process is complicated, and is determining every a time
During the area at pass interface, use scaling and adjusted so that design parameter has existed certain error.
Publication No. 1979496A, the Chinese invention application of Application No. 200510047903.6, the open one of this application
Copper-alloy pipe-material casting-milling technology parameter design and the method optimized, with data base as design basis, neutral net is technological parameter
With the method for designing of technic index, genetic algorithm is process parameter optimizing means, and comprehensive integration neutral net, genetic algorithm have
The simulation of limit unit, EXPERIMENTAL DESIGN, CAD Parametric designing and database technology, in technological design and parameter optimization, design and optimize
The technological parameter of copper-alloy pipe-material casting.This invention automaticity is high, can be suitably used for the machining deformation of copper-alloy pipe-material, makes to lack
Enrich Professional knowledge personnel less and also can make the processing technique of exact specification.But the method cannot be applicable to irregular cable draws
The design of withdrawing pattern tool.
Therefore, a kind of quick acquisition parameters optimization and the setting of irregular cable drawing mould more accurately and reliably how are designed
Meter method has become the key of solution problem.
Summary of the invention
For defect of the prior art, it is an object of the invention to provide the method for designing of a kind of irregular cable drawing mould
And system, can be designed that the pass Changing Pattern both meeting gradually transition, meet again the wire-drawing die of wire drawing machine speed ratio parameter
Pass cross section.
According to a first aspect of the present invention, it is provided that the method for designing of a kind of irregular cable drawing mould, described method is in emulation
On artificial intelligence basis, finite element simulation, artificial neural network, genetic algorithm are combined with software modeling, and institute
The method of stating ensure that designed mould has partial load equilibrium and the feature of GSLB;
Described finite element simulation, is used for calculating two parts stress:
A part is: inlet wire sectional area is identical with elongation percentage, but each section of the drawing mould bore area that geometric parameter is different
Equivalent stress and the mean stress value of bore area;Relatively obtain each section of equivalent stress excursion minimum of bore area
The geometric parameter of mould, gained geometric parameter is under this inlet wire sectional area and elongation percentage the optimum mould meeting partial load equilibrium
Tool geometric parameter;
Another part is: after inlet wire sectional area and elongation percentage are varied multiple times, in the drawing mould of corresponding various geometric
Equivalent stress that hole surface is each section and the mean stress value of bore area;It is calculated different inlet wire sectional area through simulation optimization
With the optimum mould geometric parameter under elongation percentage;
Described neutral net, be build drawing mould bore area mean stress value and mother metal initial cross-section area and
The neutral net of mapping relations between elongation percentage;
Described genetic algorithm, for determining the span of mould drawing passes, and enters the elongation percentage under each passage
Row Optimization Solution;
Described software modeling, be the section design method deformed with continuously " linearly " as foundation, design whole drawing
The mould structure of process, and it is partitioned into the mould structure under each shunting time according to the design load of elongation percentage.
Preferably, described neutral net, is as input, optimum mould geometric parameters using inlet wire sectional area and elongation percentage
The mean stress value of several and corresponding mould bore area is trained as output, thus builds drawing mould bore area
Mean stress value and mother metal initial cross-section area and elongation percentage between the BP neural network topology structure of mapping relations.
Preferably, described partial load equilibrium, refer to: inlet wire sectional area is identical with elongation percentage but geometric parameter is different
The excursion of the equivalent stress that drawing mould bore area is each section, i.e. span are minimum.
Preferably, described GSLB, refer to: in the whole drawing process of multi pass drawing, under each shunting time
The excursion of mean stress value suffered by drawing mould bore area, i.e. span are minimum.
Concrete, the method for designing of described irregular cable drawing mould, comprise the steps:
The first step, select orthogonal experiment design method, drawing mould elongation percentage and initial cross-section amass identical under conditions of,
With the geometric parameter of drawing mould, including compressional zone half-angle, compressional zone length, a length of design variable in sizing area, carry out limited
Unit's emulation also calculates the equivalent stress of each section of drawing mould bore area under various geometric and averagely should of bore area
Force value;
Second step, for meeting the equilibrium of mould local load, compare under the various geometric in the first step in drawing mould
The span of the equivalent stress that hole surface is each section, the geometric parameter corresponding to drawing mould that span is minimum, it is this
Elongation percentage and initial cross-section amass under optimum Design of Dies geometric parameter;
3rd step, the elongation percentage changing drawing mould and initial cross-section amass, and repeat the simulation calculation of the first step, and then obtain
Different elongation percentage and initial cross-section amass the lower equal stress value of equivalent stress peace;
4th step, build nerve network system, and using inlet wire sectional area and elongation percentage as input, optimum mould geometry
The mean stress value of the mould bore area of parameter and correspondence is trained as output, builds drawing mould bore area
The BP neural network topology structure of mapping relations between mean stress value and mother metal initial cross-section area and elongation percentage;
5th step, with the safety coefficient of each shunting during multi pass drawing time drawing as optimization aim, with each point
The elongation percentage of passage is parameters optimization, uses genetic algorithm to determine the span of drawing passes as optimizer, and to difference
Under drawing passes, the elongation percentage in each shunting time solves;
6th step, under different drawing passes, in neutral net, obtain optimum by the elongation percentage design load in each shunting time
Design of Dies geometric parameter values and the mean stress value of bore area of correspondence;
7th step, for meeting mould GSLB, compare under the different drawing passes that the 6th step obtains each shunting time
Between the excursion of mean stress value, the mould drawing passes that mean stress excursion is minimum, be the optimum of mould
Drawing passes;
8th step, the section design method deformed with continuously " linearly ", as foundation, design the mould of whole drawing process
Structure, and it is partitioned into the mould structure under each shunting time according to the design load of elongation percentage.
It is highly preferred that in the 5th step, value K of described safety coefficient is between 1.40-2.00.
It is highly preferred that in the 8th step, specifically include following steps:
(1) rectangular lines finished size figure and blank dimension figure are drawn;
(2) ruled surface is drawn;
(3) each passage mold sections is extracted;
(4) extract Cross section Design figure, and the line transect of design is converted into the cross section curve being made up of line segment and circular arc.
According to a second aspect of the present invention, it is provided that the design department of a kind of irregular cable drawing mould for realizing said method
System, including:
Finite element simulation module, this module is used for calculating two parts stress, and a part is: inlet wire sectional area and elongation percentage phase
With, but the equivalent stress of different each section of the drawing mould bore area of geometric parameter and the mean stress value of bore area;Ratio
Compared with the geometric parameter of the mould obtaining each section of equivalent stress excursion minimum of bore area, gained geometric parameter is this inlet wire
The optimum mould geometric parameter of partial load equilibrium is met under sectional area and elongation percentage;Another part is: inlet wire is varied multiple times and cuts
After area and elongation percentage, the equivalent stress of each section of the drawing mould bore area of corresponding various geometric and bore area
Mean stress value;Optimum mould geometric parameter under simulation optimization is calculated different inlet wire sectional area and elongation percentage;
Neural network module, this module using inlet wire sectional area and elongation percentage as input, optimum mould geometric parameter with
And the mean stress value of the mould bore area of correspondence is trained as output, build averagely should of drawing mould bore area
The BP neural network topology structure of mapping relations between force value and mother metal initial cross-section area and elongation percentage;
Genetic algorithm module, utilizes genetic algorithm to determine the span of drawing passes, and to every under different drawing passes
The elongation percentage value in one shunting time is optimized and solves;
Software modeling module, the section design method that this module deformed with continuously " linearly ", as foundation, designs whole drawing
Pull out the mould structure of process, and be partitioned into the mould structure under each shunting time according to the design load of elongation percentage.
Preferably, described genetic algorithm module, under different drawing passes, by the elongation percentage design load in each shunting time god
The Design of Dies geometric parameter values of optimum and the mean stress value of the bore area of correspondence is obtained in network.
Preferably, for meeting mould GSLB, described genetic algorithm module calculates under different drawing passes each point
The excursion of the mean stress value between passage, compares and obtains the mould drawing passes that mean stress excursion is minimum, should
Drawing passes is the optimal design passage of mould.
Compared with prior art, the present invention has a following beneficial effect:
1, the present invention is with the optimum results of finite element as sample, training of human artificial neural networks, establishes drawing mould endoporus
The mean stress value on surface, the mapping relations between optimum die size parameter and mother metal initial cross-section area and elongation percentage;
2, the present invention uses genetic Optimization Algorithm, distributes the elongation percentage in multi pass drawing and is optimized;
3, present invention local suffered by mould and GSLB during multi pass drawing, and the one-tenth of front a time
Shape size is the compressional zone original dimension of rear passage, contacts with wire rod with the compressional zone of passage mould after ensureing simultaneously, subtracts as far as possible
The inhomogeneities of little drawing wires internal stress strain, thus design the pass Changing Pattern both meeting gradually transition, full again
The wire-drawing die pass cross section of foot wire drawing machine speed ratio parameter.
To sum up, the present invention, by simulation optimization, uses artificial neural network to be used for calculating pulling capacity optimal value, greatly reduces
FEM calculation amount;In conjunction with the global optimizing ability that genetic algorithm is powerful, elongation percentage assignment problem in multi pass drawing is entered
Row solves, and significantly reduces the workload of design process;Based on mould local, the principle of GSLB, it is ensured that respectively draw
Pull out energy stress equalization in passage drawing process, effectively slow down the abrasion of mould, extend the mold use life-span;By software
Modeling, is optimized design to the pass of drawing mould, can not only give full play to the performance of drawing mould and drawing equipment, and
For high-speed automated drawing production line, while greatly extending die life, also reduce replacing and safeguard mould
The time of tool, production efficiency is largely increased, the heaviest to the economic benefit of the whole metal wire of raising, tube-drawing industry
Want.
Accompanying drawing explanation
By the detailed description non-limiting example made with reference to the following drawings of reading, the further feature of the present invention,
Purpose and advantage will become more apparent upon:
Fig. 1 a~1d is the inventive method theory diagram;
Fig. 2 is that one embodiment of the invention draws group process schematic;
In Fig. 3, (a), (b), (c), (d) are that one embodiment of the invention rectangle draws molding to have design drawing;
In Fig. 4, (a), (b), (c) are that one embodiment of the invention shoe draws molding to have design drawing;
Fig. 5 is that one embodiment of the invention shoe wire rod draws group production schematic diagram;
Fig. 6 is present system structured flowchart.
Detailed description of the invention
Below in conjunction with specific embodiment, the present invention is described in detail.Following example will assist in the technology of this area
Personnel are further appreciated by the present invention, but limit the present invention the most in any form.It should be pointed out that, the ordinary skill to this area
For personnel, without departing from the inventive concept of the premise, it is also possible to make some deformation and improvement.These broadly fall into the present invention
Protection domain.
As it is shown in figure 1, the method for designing of a kind of irregular cable drawing mould, described method is a kind of in emulation and artificial intelligence
On energy basis, the multi pass drawing that proposition finite element simulation, artificial neural network, genetic algorithm combine with software modeling is joined
Mould design method, and this method for designing can ensure that mould local and the overall situation load balancing.Described method:
As shown in Figure 1a, described finite element simulation is used for calculating two parts stress:
A part is: inlet wire sectional area is identical with elongation percentage, but each section of the drawing mould bore area that geometric parameter is different
Equivalent stress and the mean stress value of bore area;Relatively obtain each section of equivalent stress excursion minimum of bore area
The geometric parameter of mould, gained geometric parameter is under this inlet wire sectional area and elongation percentage the optimum mould meeting partial load equilibrium
Tool geometric parameter;
Another part is: after inlet wire sectional area and elongation percentage are varied multiple times, in the drawing mould of corresponding various geometric
Equivalent stress that hole surface is each section and the mean stress value of bore area;It is calculated different inlet wire sectional area through simulation optimization
With the optimum mould geometric parameter under elongation percentage;
Described neutral net as shown in Figure 1 b, is that the mean stress value of structure drawing mould bore area is initial with mother metal
The neutral net of mapping relations between area of section and elongation percentage;
Described genetic algorithm as illustrated in figure 1 c, is used for determining the span of mould drawing passes, and under each passage
Elongation percentage be optimized and solve;
Described software modeling, as shown in Figure 1 d, be the section design method deformed with continuously " linearly " as foundation, design
Go out the mould structure of whole drawing process, and be partitioned into the mould structure under each shunting time according to the design load of elongation percentage.
The section design method deformed with continuously " linearly " as foundation, i.e. during multi pass drawing before the one-tenth of a time
Shape size is the compressional zone original dimension of rear passage, contacts with wire rod with the compressional zone of passage mould after ensureing simultaneously, subtracts as far as possible
The inhomogeneities of little drawing wires internal stress strain, thus design the pass Changing Pattern both meeting gradually transition, full again
The wire-drawing die pass cross section model of foot wire drawing machine speed ratio parameter.
Embodiment 1
As in figure 2 it is shown, the present embodiment provides the method for designing of a kind of irregular cable drawing mould, described method is by following
Step is carried out:
The first step, simulation optimization
First in the pre-processing module of ANSYS Workbench, set up the geometric model of mould and rectangular wire;Then
The stress-strain data of the coefficient of friction obtained by experiment and metal material is loaded in the middle of model, grid division
FEM (finite element) model to drawing process emulation;In simulation process, the bottom of drawing mould is applied in fixed constraint, at wire rod
Front end applies the axial displacement load of-60mm, is used for simulating the pulling capacity of applying, and drawing speed is set as 3m/s, simulation time
For 0.02s;The rectangular wire material selected is copper material, and the initial cross-section taking alignment material amasss and elongation percentage value, then selects orthogonal
Test design method, with geometric parameter (compressional zone half angle 10-18 °, compressional zone length 1-5mm, the sizing area of drawing mould
Length value 2-6mm) it is design variable, carry out finite element simulation and to calculate drawing mould bore area under various geometric each
The equivalent stress of section and the mean stress value of bore area.
Second step, the partial load equilibrium of mould
Relatively obtain the geometric parameter of the minimum mould of each section of equivalent stress excursion of bore area, gained geometric parameter
It is under this inlet wire sectional area and elongation percentage the optimum mould geometric parameter meeting partial load equilibrium.
3rd step, expansion Design of Dies sample
The initial cross-section that rectangular wire is varied multiple times amasss and elongation percentage value, repeats the modeling in the first step and post processing
Journey, then selects orthogonal experiment design method, carries out finite element simulation and calculates drawing mould endoporus table under various geometric
Equivalent stress that face is each section and the mean stress value of bore area, and filter out that different initial cross-section is long-pending and corresponding under elongation percentage
Optimum mould geometric parameter, obtain more Design of Moulds sample.
4th step, set up BP artificial neural network
Using inlet wire sectional area and elongation percentage as input, optimum mould geometric parameter and the mould bore area of correspondence
Mean stress value as output for training the nerve network system of foundation, build the mean stress of drawing mould bore area
The BP neural network topology structure of mapping relations between value and mother metal initial cross-section area and elongation percentage;
Its neural network structure is:
(1) input: 2 input parameter i.e. mother metal initial cross-sections amass and elongation percentage;
(2) hidden layer: two-layer, is respectively provided with 10 neurons and 5 neurons;
(3) output layer: 6 i.e. target function values of neuron;Hidden layer uses logsig as transmission function, and output layer is adopted
With prulin as transmission function;
The network model set up is carried out neural metwork training, sets minimum expectation error amount err_goal=0.001;
Set maximum cycle max_epoch=35000;Set the learning rate lr=0.01 revising weights;Through 1000 instructions
After white silk, training terminates.
5th step, genetic algorithm ask total drawing passes span and each shunting time elongation percentage
During multi pass drawing, the safety coefficient of each shunting time drawing is as optimization aim, with prolonging of each shunting time
The rate of stretching is parameters optimization, uses genetic algorithm to determine the span of drawing passes as optimizer, and to different drawing passes
Under the elongation percentage in each shunting time solve;
The mother metal sectional area of the present embodiment is 70.89, elongation percentage is 7.8;5-10 road is selected according to existing drawing experience
The secondary optimization pass limits as the present embodiment drawing, uses genetic algorithm to calculate 5 passages respectively to 10 passage copper wires multiple tracks
The optimal result of secondary drawing, by the result calculated it can be seen that safety coefficient K increases, passage along with the increase of road times N
The fewest drawing process of number is more susceptible to fracture of wire phenomenon;Peep optimization result it appeared that:
When drawing passes number N takes 5, safety coefficient K is about 1.02;
When drawing passes number N takes 6, safety coefficient K is about 1.14;
When drawing passes number N takes 7, safety coefficient K is about 1.27;
When drawing passes number N takes 8, safety coefficient K is about 1.4;
When drawing passes number N takes 9, safety coefficient K is between 1.48-1.58;
When drawing passes number N takes 10, safety coefficient K is between 1.7-2.1;
From result of calculation, when drawing passes number N takes 8,9, safety coefficient K, all between 1.4-2.0, meets safety
The condition of drawing.Therefore, the optimum results of final drawing passes is 8 or 9 passages.
Wherein the parameter of genetic algorithm is provided that
Population scale takes 200;Crossover operator uses 2 arithmetic crossover, and crossover probability is 0.8;Mutation operator uses Gauss
Variation, mutation probability is 0.2;Maximum genetic algebra was 500 generations.
6th step, the GSLB of mould
When drawing passes is 8, mother metal initial cross-section is amassed 70.89 and each shunting time elongation percentage value (1.31,1.3,1.3,
1.3,1.25,1.28,1.34,1.26) as known conditions, neutral net is obtained each shunting suboptimal mould geometry
The mean stress value of the bore area of parameter value and correspondence (511.65MPa, 454.1MPa, 467.33MPa, 471.59MPa,
539.94MPa, 556.57MPa, 589.7MPa, 598.97MPa), result understand this passage lower mold bore area and averagely should
The excursion of power is: 454.1MPa-598.97MPa;
When drawing passes is 9, mother metal initial cross-section is amassed 70.89 and each shunting time elongation percentage value (1.28,1.26,
1.27,1.26,1.23,1.23,1.26,1.35,1.18) as known conditions, neutral net is obtained each shunting time
The mean stress value of excellent mould geometric parameter values and the bore area of correspondence (509.09MPa, 301.28MPa,
396.2MPa, 450.71MPa, 538.31MPa, 599.3MPa, 632.1MPa, 574.33MPa, 611.5MPa), from result
The excursion of this passage lower mold bore area mean stress is: 301.28MPa-632.1MPa;
Under relatively two kinds of drawing passes, the excursion of the mean stress of mould bore area, the STRESS VARIATION under 9 passages
Scope is less, so selecting 9 passages as the drawing passes of this rectangular wire drawing mould.
7th step, AutoCAD model
The each passage design parameter obtained according to optimization, utilizes AutoCAD software to be modeled:
(1) rectangular lines finished size figure and blank dimension figure are drawn: utilize CAD software to draw the end product rule of rectangular electromagnetic wire
Very little figure, sets up territory, face, finds out the barycentric coodinates in territory, face, draws the size of inlet wire blank, such as Fig. 3 at position of centre of gravity height 10mm
Shown in (a).
(2) ruled surface is drawn: choose upper and lower two sections (the finished size figure of rectangular lines and blank dimension figure) and carry out setting-out,
Obtain by circular cross-section gradually to the ruled surface of molded line transition, as shown in Fig. 3 (b).
(3) each passage mold sections is extracted: first calculate the target area conduct in middle transition cross section according to the speed ratio chosen
Postorder determines the foundation of depth of section.With slice interception of commands middle transition cross section so that it is the speed ratio requirement that area coinciding is certain
If (intercepting area and be more than target area, then reduce and intercept height;Vice versa), as shown in Fig. 3 (c).
(4) a newly-built figure layer, chooses all curved surfaces under new figure layer, uses xedges order to extract edge, obtains sample
Bar line cross section curve, is the Cross section Design figure of required each passage.It is line transect due to obtain herein, is using slow wire feeding
During Wire-cut Electrical Discharge Machining, it is impossible to by linear cutter software identification, therefore also the line transect of design should be converted into by line segment
The cross section curve (such as can use and be converted by Yan Xiu workbox software) formed with circular arc, as shown in Fig. 3 (d).
(5) by the height of the special-shaped drawing mould of change to control its compressional zone half-angle, and cut at slow wire feeding wire electric discharge
Processing is realized on cutting mill bed.
Embodiment 2
As shown in Figure 4, Figure 5, the present embodiment provides the method for designing of a kind of irregular cable drawing mould, passes through following steps
Carry out:
The first step, simulation optimization
First in the pre-processing module of ANSYS Workbench, set up mould and the geometric model of shoe wire rod.Then
The stress-strain data of the coefficient of friction obtained by experiment and metal material is loaded in the middle of model, grid division
FEM (finite element) model to drawing process emulation.In simulation process, the bottom of drawing mould is applied in fixed constraint, before wire rod
End applies the axial displacement load of-60mm, is used for simulating the pulling capacity of applying, and drawing speed is set as 3m/s, and simulation time is
0.02s.The rectangular wire material selected is low-carbon (LC) steel, and the initial cross-section taking alignment material amasss and elongation percentage value, is just then selecting
Hand over test design method, with geometric parameter (compressional zone half angle 4-10 °, compressional zone length 2-8mm, the sizing of drawing mould
Section length value 1-6mm) it is design variable, carry out finite element simulation and calculate drawing mould bore area under various geometric
The equivalent stress of each section and the mean stress value of bore area.
Second step, the partial load equilibrium of mould
Relatively obtain the geometric parameter of the minimum mould of each section of equivalent stress excursion of bore area, gained geometric parameter
It is under this inlet wire sectional area and elongation percentage the optimum mould geometric parameter meeting partial load equilibrium.
3rd step, expansion Design of Dies sample
The initial cross-section that shoe wire rod is varied multiple times amasss and elongation percentage value, repeats the modeling in the first step and post processing
Journey, then selects orthogonal experiment design method, carries out finite element simulation and calculates drawing mould endoporus table under various geometric
Equivalent stress that face is each section and the mean stress value of bore area, and filter out that different initial cross-section is long-pending and corresponding under elongation percentage
Optimum mould geometric parameter, obtain more Design of Moulds sample.
4th step, set up BP artificial neural network
Using inlet wire sectional area and elongation percentage as input, optimum mould geometric parameter and the mould bore area of correspondence
Mean stress value as output for training the nerve network system of foundation, build the mean stress of drawing mould bore area
The BP neural network topology structure of mapping relations between value and mother metal initial cross-section area and elongation percentage;
Its neural network structure is:
(1) input: 2 input parameter i.e. mother metal initial cross-sections amass and elongation percentage;
(2) hidden layer: two-layer, is respectively provided with 10 neurons and 5 neurons;
(3) output layer: 6 i.e. target function values of neuron.Hidden layer uses logsig as transmission function, and output layer is adopted
With prulin as transmission function.
The network model set up is carried out neural metwork training, sets minimum expectation error amount err_goal=0.001;
Set maximum cycle max_epoch=35000;Set the learning rate lr=0.01 revising weights.Through 1500 instructions
After white silk, training terminates.
5th step, genetic algorithm ask total drawing passes span and each shunting time elongation percentage
During multi pass drawing, the safety coefficient of each shunting time drawing is as optimization aim, with prolonging of each shunting time
The rate of stretching is parameters optimization, uses genetic algorithm to determine the span of drawing passes as optimizer, and to different drawing passes
Under the elongation percentage in each shunting time solve;
The mother metal sectional area of the present embodiment is 152.84, elongation percentage is 6.7.According to existing drawing experience, select 5-10
Passage, as the optimization pass limits of the present embodiment drawing, uses genetic algorithm to calculate 5 passages respectively to 10 passage low-carbon (LC) steel wires
The optimal result of material multi pass drawing, by the result calculated it can be seen that safety coefficient K increases along with the increase of road times N
Adding, the drawing process that road number of times is the fewest is more susceptible to fracture of wire phenomenon.Peep optimization result it appeared that:
When drawing passes number N takes 5, safety coefficient K is about 1.01;
When drawing passes number N takes 6, safety coefficient K is about 1.09;
When drawing passes number N takes 7, safety coefficient K is about 1.14;
When drawing passes number N takes 8, safety coefficient K is about 1.32;
When drawing passes number N takes 9, safety coefficient K is 1.42;
When drawing passes number N takes 10, safety coefficient K is between 1.5-1.89.
From result of calculation, when drawing passes number N takes 9,10, safety coefficient K, all between 1.4-2.0, meets peace
The condition of full drawing.Therefore, the optimum results of final drawing passes is 9 or 10 passages.
Wherein the parameter of genetic algorithm is provided that
Population scale takes 300;Crossover operator uses 2 arithmetic crossover, and crossover probability is 0.8;Mutation operator uses Gauss
Variation, mutation probability is 0.2;Maximum genetic algebra was 800 generations.
6th step, the GSLB of mould
When drawing passes is 9, mother metal initial cross-section is amassed 152.84 and each shunting time elongation percentage value (1.15,1.18,
1.22,1.23,1.22,1.24,1.30,1.31,1.28) as known conditions, neutral net is obtained each shunting time
The mean stress value of excellent mould geometric parameter values and the bore area of correspondence (650.126MPa, 557.048MPa,
671.503MPa, 732.618MPa, 733.409MPa, 748.226MPa, 694.598MPa, 701.97MPa, 711.55MPa),
The excursion being understood this passage lower mold bore area mean stress by result is: 557.048MPa-748.226MPa;
When drawing passes is 10, mother metal initial cross-section is amassed 152.84 and each shunting time elongation percentage value (1.09,1.13,
1.18,1.23,1.24,1.23,1.25,1.24,1.26,1.27) as known conditions, neutral net is obtained each shunting
The mean stress value of the bore area of suboptimal mould geometric parameter values and correspondence (562.22MPa, 536.18MPa,
668.21MPa, 640.71MPa, 690.32MPa, 710.55MPa, 632.16MPa, 707.35MPa, 699.5MPa,
720.64MPa), the excursion of this passage lower mold bore area mean stress knowable to result it is: 536.18MPa-
720.64MPa;
Under relatively two kinds of drawing passes, the excursion of the mean stress of mould bore area, the stress under 10 passages becomes
Change scope is less, so selecting 10 passages as the drawing passes of this rectangular wire drawing mould.
7th step, AutoCAD model
The each passage design parameter obtained according to optimization, utilizes AutoCAD software to be modeled:
(1) rectangular lines finished size figure and blank dimension figure are drawn: utilize CAD software to draw the end product rule of rectangular electromagnetic wire
Very little figure, sets up territory, face, finds out the barycentric coodinates in territory, face, draws the size of inlet wire blank, such as Fig. 4 at position of centre of gravity height 25mm
Shown in (a).
(2) ruled surface is drawn: choose upper and lower two sections (the finished size figure of shoe line and blank dimension figure) and carry out setting-out,
Obtain by circular cross-section gradually to the ruled surface of molded line transition, as shown in Fig. 4 (b).
(3) each passage mold sections is extracted: first calculate the target area conduct in middle transition cross section according to the speed ratio chosen
Postorder determines the foundation of depth of section.With slice interception of commands middle transition cross section so that it is the speed ratio requirement that area coinciding is certain
If (intercepting area and be more than target area, then reduce and intercept height;Vice versa).
(4) a newly-built figure layer, chooses all curved surfaces under new figure layer, uses xedges order to extract edge, obtains sample
Bar line cross section curve, is the Cross section Design figure of required each passage.It is line transect due to obtain herein, is using slow wire feeding
During Wire-cut Electrical Discharge Machining, it is impossible to by linear cutter software identification, therefore also the line transect of design should be converted into by line segment
The cross section curve formed with circular arc, as shown in Fig. 4 (c).
(5) by the height of the special-shaped drawing mould of change to control its compressional zone half-angle, and cut at slow wire feeding wire electric discharge
Processing is realized on cutting mill bed.
Above-described embodiment shaped wire drawing mould method can design the pass Changing Pattern both having met gradually transition,
Meet again the wire-drawing die pass cross section of wire drawing machine speed ratio parameter.In actual mould process of manufacture, can give full play to
Drawing mould and the performance of drawing equipment, while greatly extending die life, also reduce in actual mould application
Changing the time safeguarding mould, production efficiency is largely increased, to the economic effect improving whole metal wire, tube-drawing industry
Benefit is particularly significant.
Embodiment 3
The present embodiment provides the method for designing of a kind of irregular cable drawing mould, is carried out by following steps:
The first step, simulation optimization
First in the pre-processing module of ANSYS Workbench, set up mould and the geometric model of S-shaped wire rod.Then will
The coefficient of friction obtained by experiment and the stress-strain data of metal material are loaded in the middle of model, and grid division is the most available
The FEM (finite element) model of drawing process emulation.In simulation process, the bottom of drawing mould is applied in fixed constraint, in the front end of wire rod
Applying the axial displacement load of-60mm, be used for simulating the pulling capacity of applying, drawing speed is set as 1m/s, and simulation time is
0.05s.The S-shaped material wire selected is copper material, and the initial cross-section taking alignment material amasss and elongation percentage value, then selects orthogonal test
Method for designing, with geometric parameter (compressional zone half angle 6-13 °, compressional zone length 4-10mm, the sizing area length of drawing mould
Value 3-8mm) it is design variable, carry out finite element simulation and calculate each section of drawing mould bore area under various geometric
The mean stress value of equivalent stress and bore area.
Second step, the partial load equilibrium of mould
Relatively obtain the geometric parameter of the minimum mould of each section of equivalent stress excursion of bore area, gained geometric parameter
It is under this inlet wire sectional area and elongation percentage the optimum mould geometric parameter meeting partial load equilibrium.
3rd step, expansion Design of Dies sample
The initial cross-section that S-shaped wire rod is varied multiple times amasss and elongation percentage value, repeats the modeling in the first step and last handling process,
Then select orthogonal experiment design method, carry out finite element simulation and to calculate drawing mould bore area under various geometric each
The equivalent stress of section and the mean stress value of bore area, and filter out that different initial cross-section is long-pending and under elongation percentage corresponding
Excellent mould geometric parameter, obtains more Design of Moulds sample.
4th step, set up BP artificial neural network
Using inlet wire sectional area and elongation percentage as input, optimum mould geometric parameter and the mould bore area of correspondence
Mean stress value as output for training the nerve network system of foundation, build the mean stress of drawing mould bore area
The BP neural network topology structure of mapping relations between value and mother metal initial cross-section area and elongation percentage;
Its neural network structure is:
(1) input: 2 input parameter i.e. mother metal initial cross-sections amass and elongation percentage;
(2) hidden layer: two-layer, is respectively provided with 10 neurons and 5 neurons;
(3) output layer: 6 i.e. target function values of neuron.Hidden layer uses logsig as transmission function, and output layer is adopted
With prulin as transmission function.
The network model set up is carried out neural metwork training, sets minimum expectation error amount err_goal=0.001;
Set maximum cycle max_epoch=35000;Set the learning rate lr=0.01 revising weights.Through 1500 instructions
After white silk, training terminates.
5th step, genetic algorithm ask total drawing passes span and each shunting time elongation percentage
During multi pass drawing, the safety coefficient of each shunting time drawing is as optimization aim, with prolonging of each shunting time
The rate of stretching is parameters optimization, uses genetic algorithm to determine the span of drawing passes as optimizer, and to different drawing passes
Under the elongation percentage in each shunting time solve;
The mother metal sectional area of the present embodiment is 85.10, elongation percentage is 7.4.According to existing drawing experience, select 5-9 road
The secondary optimization pass limits as the present embodiment drawing, uses genetic algorithm to calculate 5 passages respectively many to 9 passage S-shaped copper wires
The optimal result of passage drawing, by the result calculated it can be seen that safety coefficient K increases, road along with the increase of road times N
The drawing process that number of times is the fewest is more susceptible to fracture of wire phenomenon.Peep optimization result it appeared that:
When drawing passes number N takes 5, safety coefficient K is about 1.03-1.08;
When drawing passes number N takes 6, safety coefficient K is about 1.18;
When drawing passes number N takes 7, safety coefficient K is about 1.29;
When drawing passes number N takes 8, safety coefficient K is about 1.42-1.48;
When drawing passes number N takes 9, safety coefficient K is between 1.63;
When drawing passes number N takes 10, safety coefficient K is between 1.81-2.25.
From result of calculation, when drawing passes number N takes 8,9, safety coefficient K, all between 1.4-2.0, meets safety
The condition of drawing.Therefore, the optimum results of final drawing passes is 8 or 9 passages.
Wherein the parameter of genetic algorithm is provided that
Population scale takes 200;Crossover operator uses 2 arithmetic crossover, and crossover probability is 0.8;Mutation operator uses Gauss
Variation, mutation probability is 0.2;Maximum genetic algebra was 500 generations.
6th step, the GSLB of mould
When drawing passes is 8, mother metal initial cross-section is amassed 85.10 and each shunting time elongation percentage value (1.30,1.28,
1.29,1.30,1.27,1.25,1.32,1.27) as known conditions, neutral net is obtained the suboptimal mould in each shunting
Tool geometric parameter values and correspondence bore area mean stress value (612.11MPa, 553.20MPa, 637.28MPa,
590.19MPa, 648.70MPa, 699.37MPa, 710.71MPa, 640.33MPa), result understand this passage lower mold endoporus
The excursion of surface mean stress is: 553.20MPa-710.71MPa;
When drawing passes is 9, mother metal initial cross-section is amassed 85.10 and each shunting time elongation percentage value (1.28,1.25,
1.23,1.24,1.24,1.26,1.25,1.30,1.20) as known conditions, neutral net is obtained each shunting time
The mean stress value of excellent mould geometric parameter values and the bore area of correspondence (521.16MPa, 501.33MPa,
579.30MPa, 611.52MPa, 629.09MPa, 575.32MPa, 671.69MPa, 570.21MPa, 567.18MPa), by result
The excursion understanding this passage lower mold bore area mean stress is: 501.33MPa-671.69MPa;
Under relatively two kinds of drawing passes, the excursion of the mean stress of mould bore area, the STRESS VARIATION under 8 passages
Scope is less, so selecting 8 passages as the drawing passes of this rectangular wire drawing mould.
7th step, AutoCAD model
The each passage design parameter obtained according to optimization, utilizes AutoCAD software to be modeled:
(1) rectangular lines finished size figure and blank dimension figure are drawn: utilize CAD software to draw the end product rule of rectangular electromagnetic wire
Very little figure, sets up territory, face, finds out the barycentric coodinates in territory, face, draws the size of inlet wire blank at position of centre of gravity height 13mm.
(2) ruled surface is drawn: choose upper and lower two sections (the finished size figure of S-shaped line and blank dimension figure) and carry out setting-out,
Obtain by circular cross-section gradually to the ruled surface of molded line transition.
(3) each passage mold sections is extracted: first calculate the target area conduct in middle transition cross section according to the speed ratio chosen
Postorder determines the foundation of depth of section.With slice interception of commands middle transition cross section so that it is the speed ratio requirement that area coinciding is certain
If (intercepting area and be more than target area, then reduce and intercept height;Vice versa).
(4) a newly-built figure layer, chooses all curved surfaces under new figure layer, uses xedges order to extract edge, obtains sample
Bar line cross section curve, is the Cross section Design figure of required each passage.It is line transect due to obtain herein, is using slow wire feeding
During Wire-cut Electrical Discharge Machining, it is impossible to by linear cutter software identification, therefore also should be by the line transect of design by swallow show instrument
Case software is converted into the cross section curve being made up of line segment and circular arc.
(5) by the height of the special-shaped drawing mould of change to control its compressional zone half-angle, and cut at slow wire feeding wire electric discharge
Processing is realized on cutting mill bed.
Embodiment 4
As shown in Figure 6, corresponding to method described in above-described embodiment 1-3, the present embodiment provides for realizing said method
The design system of irregular cable drawing mould, including:
Finite element simulation module, this module is used for calculating two parts stress, and a part is: inlet wire sectional area and elongation percentage phase
With, but the equivalent stress of different each section of the drawing mould bore area of geometric parameter and the mean stress value of bore area;Ratio
Compared with the geometric parameter of the mould obtaining each section of equivalent stress excursion minimum of bore area, gained geometric parameter is this inlet wire
The optimum mould geometric parameter of partial load equilibrium is met under sectional area and elongation percentage;Another part is: inlet wire is varied multiple times and cuts
After area and elongation percentage, the equivalent stress of each section of the drawing mould bore area of corresponding various geometric and bore area
Mean stress value;Optimum mould geometric parameter under simulation optimization is calculated different inlet wire sectional area and elongation percentage;
Neural network module, this module using the inlet wire sectional area of described finite element simulation module and elongation percentage as input,
Optimum mould geometric parameter and the mean stress value of the mould bore area of correspondence are trained as output, build drawing
Between mean stress value and mother metal initial cross-section area and the elongation percentage of mould bore area, the BP neutral net of mapping relations is opened up
Flutter structure;
Genetic algorithm module, utilizes genetic algorithm to determine the span of drawing passes, and to every under different drawing passes
The elongation percentage value in one shunting time is optimized and solves, afterwards using long-pending to the drawing passes tried to achieve and mother metal initial cross-section as training
The input of the neural network module completed, the optimum solving the excursion minimum meeting mould bore area mean stress draws
Pull out passage;
Software modeling module, the section design method that this module deformed with continuously " linearly ", as foundation, designs whole drawing
Pull out the mould structure of process, and the design load of each shunting time elongation percentage tried to achieve according to genetic algorithm module is partitioned into each shunting time
Under mould structure.
In the technology and method that in said system, modules correspondence realizes, corresponding step is similar to, and does not repeats them here.
To sum up, the present invention, by emulation, optimization and software design, is optimized design to the pass of drawing mould, not only
The performance of drawing mould and drawing equipment can be given full play to, and for high-speed automated drawing production line, greatly
While extending die life, also reducing and change the time safeguarding mould, production efficiency is largely increased, whole to improving
Individual metal wire, tube-drawing industry economic benefit particularly significant.
Above the specific embodiment of the present invention is described.It is to be appreciated that the invention is not limited in above-mentioned
Particular implementation, those skilled in the art can make various deformation or amendment within the scope of the claims, this not shadow
Ring the flesh and blood of the present invention.
Claims (10)
1. the method for designing of an irregular cable drawing mould, it is characterised in that: on emulation and artificial intelligence basis, will have
The emulation of limit unit, artificial neural network, genetic algorithm combine with software modeling, and described method ensure that designed mould
There is partial load equilibrium and the feature of GSLB;
Described finite element simulation, is used for calculating two parts stress:
A part is: inlet wire sectional area is identical with elongation percentage, but different each section of the drawing mould bore area of geometric parameter etc.
The mean stress value of efficacy and bore area;Relatively obtain the mould that each section of equivalent stress excursion of bore area is minimum
Geometric parameter, gained geometric parameter be under this inlet wire sectional area and elongation percentage meet partial load equilibrium optimum mould several
What parameter;
Another part is: after inlet wire sectional area and elongation percentage are varied multiple times, the drawing mould endoporus table of corresponding various geometric
Equivalent stress that face is each section and the mean stress value of bore area;It is calculated different inlet wire sectional area through simulation optimization and prolongs
Stretch the optimum mould geometric parameter under rate;
Described neutral net, is the mean stress value building drawing mould bore area and mother metal initial cross-section area and extension
The neutral net of mapping relations between rate;
Described genetic algorithm, for determining the span of mould drawing passes, and carries out excellent to the elongation percentage under each passage
Change solves;
Described software modeling, be the section design method deformed with continuously " linearly " as foundation, design whole drawing process
Mould structure, and the mould structure being partitioned under each shunting time according to the design load of elongation percentage.
The method for designing of a kind of irregular cable drawing mould the most according to claim 1, it is characterised in that described nerve
Network, is as input, optimum mould geometric parameter and the mould bore area of correspondence using inlet wire sectional area and elongation percentage
Mean stress value be trained as output, thus the mean stress value building drawing mould bore area is initially cut with mother metal
The BP neural network topology structure of mapping relations between face area and elongation percentage.
The method for designing of a kind of irregular cable drawing mould the most according to claim 1, it is characterised in that described local
Load balancing, refers to: inlet wire sectional area is identical with elongation percentage but each section of drawing mould bore area that geometric parameter is different etc.
The excursion of efficacy, i.e. span are minimum.
The method for designing of a kind of irregular cable drawing mould the most according to claim 1, it is characterised in that the described overall situation
Load balancing, refers to: in the whole drawing process of multi pass drawing, averagely should suffered by the lower drawing mould bore area in each shunting time
The excursion of force value, i.e. span are minimum.
5. according to the method for designing of a kind of irregular cable drawing mould described in any one of claim 1-4, it is characterised in that institute
The method of stating comprises the steps:
The first step, select orthogonal experiment design method, drawing mould elongation percentage and initial cross-section amass identical under conditions of, to draw
The geometric parameter of withdrawing pattern tool, including compressional zone half-angle, compressional zone length, a length of design variable in sizing area, carries out finite element and imitates
Very, and calculate equivalent stress and the mean stress of bore area of each section of drawing mould bore area under various geometric
Value;
Second step, for meet mould local load equilibrium, compare drawing mould endoporus table under the various geometric in the first step
The span of the equivalent stress that face is each section, the geometric parameter corresponding to drawing mould that span is minimum, it is this extension
Rate and initial cross-section amass under optimum Design of Dies geometric parameter;
3rd step, the elongation percentage changing drawing mould and initial cross-section amass, and repeat the simulation calculation of the first step, and then obtain difference
Elongation percentage and initial cross-section amass under equivalent stress peace all stress values;
4th step, build nerve network system, and using inlet wire sectional area and elongation percentage as input, optimum mould geometric parameter
And the mean stress value of the mould bore area of correspondence is trained as output, build the average of drawing mould bore area
The BP neural network topology structure of mapping relations between stress value and mother metal initial cross-section area and elongation percentage;
5th step, with the safety coefficient of each shunting during multi pass drawing time drawing as optimization aim, with each shunting time
Elongation percentage be parameters optimization, use genetic algorithm to determine the span of drawing passes as optimizer, and to different drawings
Under passage, the elongation percentage in each shunting time solves;Afterwards using long-pending to the drawing passes tried to achieve and mother metal initial cross-section as training
The input of the nerve network system completed;
6th step, under different drawing passes, in nerve network system, obtain optimum by the elongation percentage design load in each shunting time
Design of Dies geometric parameter values and the mean stress value of bore area of correspondence;
7th step, for meeting mould GSLB, compare under the different drawing passes that the 6th step obtains between each shunting time
The excursion of mean stress value, the mould drawing passes that mean stress excursion is minimum, be the optimum drawing of mould
Passage;
8th step, the section design method deformed with continuously " linearly ", as foundation, utilize AutoCAD software design to go out whole drawing
The mould structure of process, and the design load of each shunting time elongation percentage tried to achieve according to genetic algorithm is partitioned into the mould under each shunting time
Lamps structure.
The method for designing of a kind of irregular cable drawing mould the most according to claim 5, it is characterised in that in the 5th step,
Value K of described safety coefficient is between 1.40-2.00.
The method for designing of a kind of irregular cable drawing mould the most according to claim 5, it is characterised in that in the 8th step,
Specifically include following steps:
(1) finished product dimensional drawing and blank dimension figure are drawn;
(2) ruled surface is drawn;
(3) each passage mold sections is extracted;
(4) extract Cross section Design figure, and the line transect of design is converted into the cross section curve being made up of line segment and circular arc.
8. for realizing a design system for the irregular cable drawing mould of method described in any one of the claims 1-7,
It is characterized in that: including:
Finite element simulation module, this module is used for calculating two parts stress, and a part is: inlet wire sectional area is identical with elongation percentage,
But the equivalent stress of each section of the drawing mould bore area that geometric parameter is different and the mean stress value of bore area;Relatively
To the geometric parameter of the mould of each section of equivalent stress excursion minimum of bore area, gained geometric parameter is this inlet wire cross section
Amass and meet under elongation percentage the optimum mould geometric parameter of partial load equilibrium;Another part is: inlet wire sectional area is varied multiple times
After elongation percentage, the equivalent stress of each section of the drawing mould bore area of corresponding various geometric and bore area average
Stress value;Optimum mould geometric parameter under simulation optimization is calculated different inlet wire sectional area and elongation percentage;
Neural network module, this module is using the inlet wire sectional area of described finite element simulation module and elongation percentage as input, optimum
Mould geometric parameter and the mean stress value of mould bore area of correspondence be trained as output, build drawing mould
The BP neutral net topology knot of mapping relations between mean stress value and mother metal initial cross-section area and the elongation percentage of bore area
Structure;
Genetic algorithm module, utilizes genetic algorithm to determine the span of drawing passes, and under different drawing passes each point
The elongation percentage value of passage is optimized and solves, afterwards using long-pending to the drawing passes tried to achieve and mother metal initial cross-section as having trained
The input of neural network module, solve and meet the optimum drawing road that the excursion of mould bore area mean stress is minimum
Secondary;
Software modeling module, the section design method that this module deformed with continuously " linearly ", as foundation, designs whole drawing
The mould structure of journey, and the design load of each shunting time elongation percentage tried to achieve according to genetic algorithm module is partitioned under each shunting time
Mould structure.
The design system of a kind of irregular cable drawing mould the most according to claim 8, it is characterised in that described heredity is calculated
Method module, under different drawing passes, the mould obtaining optimum by the elongation percentage design load in each shunting time in neutral net sets
The mean stress value of the bore area of meter geometric parameter values and correspondence.
The design system of a kind of irregular cable drawing mould the most according to claim 8, it is characterised in that for meeting mould
Tool GSLB, described genetic algorithm module calculates the change of the mean stress value under different drawing passes between each shunting time
Change scope, compares and obtains the mould drawing passes that mean stress excursion is minimum, and this drawing passes is the optimum of mould and sets
Meter passage.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610545822.7A CN106202727B (en) | 2016-07-12 | 2016-07-12 | Design method and system of special-shaped cable drawing die |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610545822.7A CN106202727B (en) | 2016-07-12 | 2016-07-12 | Design method and system of special-shaped cable drawing die |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106202727A true CN106202727A (en) | 2016-12-07 |
CN106202727B CN106202727B (en) | 2020-05-22 |
Family
ID=57477767
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610545822.7A Active CN106202727B (en) | 2016-07-12 | 2016-07-12 | Design method and system of special-shaped cable drawing die |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106202727B (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108090282A (en) * | 2017-12-18 | 2018-05-29 | 上海理工大学 | Slow wire feeding Wire-cut Electrical Discharge Machining gap multiple physical field coupling Simulation analysis method |
CN108509666A (en) * | 2017-02-27 | 2018-09-07 | 中国科学院金属研究所 | Fine aluminium conducting wire multi-pass cold drawing plastic deformation modes prediction technique based on ANSYS/LS-dyna |
CN110298070A (en) * | 2019-05-15 | 2019-10-01 | 重庆创速工业技术研究院有限公司 | A kind of standard punch fixing seat generates and location mode |
WO2022032905A1 (en) * | 2020-08-11 | 2022-02-17 | 江苏亨通高压海缆有限公司 | Profile single-strand die for direct-current submarine cable and design method for profile single-strand die |
CN114798779A (en) * | 2022-03-23 | 2022-07-29 | 中南大学 | Intelligent metal drawing device and method based on neural network |
WO2023010630A1 (en) * | 2021-08-04 | 2023-02-09 | 徐州徐工液压件有限公司 | Method for manufacturing cold drawing die for steel pipes |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1624396A1 (en) * | 2004-07-09 | 2006-02-08 | Yazaki Corporation | Method of calculating the shape of a wiring structure, calculating apparatus, and computer -readable recording medium |
CN105200391A (en) * | 2014-05-29 | 2015-12-30 | 上海交友钻石涂层有限公司 | Preparation method of diamond-coated drawing mold for water lubrication |
CN105444935A (en) * | 2015-12-18 | 2016-03-30 | 深圳供电局有限公司 | Portable porcelain column insulator stress detection device |
-
2016
- 2016-07-12 CN CN201610545822.7A patent/CN106202727B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1624396A1 (en) * | 2004-07-09 | 2006-02-08 | Yazaki Corporation | Method of calculating the shape of a wiring structure, calculating apparatus, and computer -readable recording medium |
CN105200391A (en) * | 2014-05-29 | 2015-12-30 | 上海交友钻石涂层有限公司 | Preparation method of diamond-coated drawing mold for water lubrication |
CN105444935A (en) * | 2015-12-18 | 2016-03-30 | 深圳供电局有限公司 | Portable porcelain column insulator stress detection device |
Non-Patent Citations (1)
Title |
---|
LIN Z 等: "Numerical and experimental investigation of trapezoidal wire cold drawing through a series of shaped dies", 《THE INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY》 * |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108509666A (en) * | 2017-02-27 | 2018-09-07 | 中国科学院金属研究所 | Fine aluminium conducting wire multi-pass cold drawing plastic deformation modes prediction technique based on ANSYS/LS-dyna |
CN108509666B (en) * | 2017-02-27 | 2021-05-25 | 中国科学院金属研究所 | Pure aluminum wire multi-pass cold drawing plastic deformation mode prediction method based on ANSYS/LS-dyna |
CN108090282A (en) * | 2017-12-18 | 2018-05-29 | 上海理工大学 | Slow wire feeding Wire-cut Electrical Discharge Machining gap multiple physical field coupling Simulation analysis method |
CN110298070A (en) * | 2019-05-15 | 2019-10-01 | 重庆创速工业技术研究院有限公司 | A kind of standard punch fixing seat generates and location mode |
CN110298070B (en) * | 2019-05-15 | 2023-12-08 | 成都数模码科技有限公司 | Standard punch fixing seat generation and distribution method |
WO2022032905A1 (en) * | 2020-08-11 | 2022-02-17 | 江苏亨通高压海缆有限公司 | Profile single-strand die for direct-current submarine cable and design method for profile single-strand die |
WO2023010630A1 (en) * | 2021-08-04 | 2023-02-09 | 徐州徐工液压件有限公司 | Method for manufacturing cold drawing die for steel pipes |
CN114798779A (en) * | 2022-03-23 | 2022-07-29 | 中南大学 | Intelligent metal drawing device and method based on neural network |
Also Published As
Publication number | Publication date |
---|---|
CN106202727B (en) | 2020-05-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106202727A (en) | The method for designing of irregular cable drawing mould and system | |
CN104978456B (en) | A kind of universal valve design optimization method | |
CN105488562B (en) | A kind of layout of polygon method based on multiple-factor particle cluster algorithm | |
CN100383796C (en) | Copper-alloy pipe-material casting-milling technology parameter designing and optimizing method | |
CN104077439A (en) | Numerical simulation method of novel high-strength steel spoke drawing punching combined process | |
Lambiase | Optimization of shape rolling sequences by integrated artificial intelligent techniques | |
Deng et al. | A blank optimization design method for three-roll cross rolling of complex-groove and small-hole ring | |
CN107427885A (en) | Blank shape determining method, blank, manufacturing press-molded products, impact forming method, computer program and storage medium | |
Liu et al. | Small data-driven modeling of forming force in single point incremental forming using neural networks | |
Heng et al. | Towards intelligent design optimization: Progress and challenge of design optimization theories and technologies for plastic forming | |
CN110598275A (en) | Wheel profile optimization method based on response surface modeling and improved particle swarm optimization | |
CN104915490A (en) | Method and device for pneumatically anti-designing motor train unit head type | |
Noorbakhsh et al. | Design and optimization of multi-stage manufacturing process of stirling engine crankshaft | |
Feng et al. | Research on the optimization mechanism of loading path in hydroforming process | |
Yang et al. | Preform design for forging and extrusion processes based on geometrical resemblance | |
Shallan et al. | Optimization of plane and space trusses using genetic algorithms | |
Megharaj et al. | Design and analysis of a forging die for manufacturing of multiple connecting rods | |
CN106001933B (en) | It is cut by laser the optimization method of trimming line | |
Equbal et al. | Grey-based Taguchi method for optimisation of hot forging of connecting rod | |
Hsiang et al. | Application of finite element method and artificial neural network to the die design of radial forging processes | |
CN103761378B (en) | A kind of missile cabin radial connection structure lightweight design system | |
CN109210001B (en) | A kind of design method of superplastic forming hollow twist blade | |
CN106777672A (en) | Metal material forging microstructure flexible measurement method based on adaptive expert system | |
Jun et al. | An investigation of the variable cross-section extrusion process | |
Naeimi et al. | Optimum designing of forging preform die for the H-shaped parts using backward deformation method and neural networks algorithm |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | 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 |