CN106363283A - Method for determining serial double-wire submerged arc welding numerical simulation heat source model parameters - Google Patents

Abstract

The invention discloses a method for determining serial double-wire submerged arc welding numerical simulation heat source model parameters. The method comprises the following steps that a double-ellipsoid heat source model with the heat flux density attenuated along a quadratic function in the depth direction is built; serial double-wire submerged arc welding heat flux density distribution is divided into three areas according to the interaction of a front wire heat source and a rear wire heat source; the heat source model is subjected to finite element inversion, and the corresponding heat source model parameters are obtained; the heat source model parameters are adjusted to a certain extent, parameter combinations are subjected to finite element numerical simulation, and multiple sets of weld width and weld depth sample values are obtained; a sensibility equation of the parameters and the welding speed of the double-ellipsoid heat source model on the weld width and the weld depth is determined; and the sensibility equation is obtained through fitting of the multiple sets of sample values. By means of the sensibility equation, the heat source model parameters corresponding to different welding process parameters can be directly predicted, the calculation workload of welding simulation is greatly reduced, the simulation precision is greatly improved, process tests are reduced, and the development cost is reduced.

Description

A kind of tandem double wire hidden arc welding numerical simulation heat source model determination method for parameter

Technical field

The invention belongs to welding value heat source model technical field is and in particular to a kind of tandem double wire hidden arc welding numerical value Simulation heat source model determination method for parameter.

Background technology

Submerged-arc welding (containing submerged arc overlay welding and ESW etc.) is that a kind of electric arc burns the method welded under welding flux layer. Due to welding quality stable, welding productivity is high, no arc light and flue dust are little the advantages of so as to become pressure vessel, pipeline section system Make, the main welding method in the important structure steel fabrication such as box beam column.In recent years although successively occur in that many plant efficiently, The new welding method of high-quality, but the application of submerged-arc welding is not still affected.Submerged-arc welding can according to welding wire number difference Be divided into monofilament submerged-arc welding and multiplewire submerged arc welding, relatively monofilament submerged-arc welding, double wire hidden arc welding have can improve speed of welding 30～ 40%th, welding quality is good, welding wire deposition rate and the higher feature of current utilization rate.

The primary problem solving of submerged-arc welding numerical simulation is the problem that submerged-arc welding heat source model parameter selects.Mariages buries at present Arc-welding heat source model parameter determination method is mainly tentative calculation, due to experience and the time restriction of research worker, it is difficult to ensure that thermal source The precision of model, increased development cost simultaneously again.Further, since it is not simple between two electric arcs of tandem double wire hidden arc welding Superposition, two electric arcs presence influence each other.And consider two interactional numerical simulation studies seldom, therefore lead to existing There is the submerged-arc welding heat source model parameters precision simulating using simple superposition in technology not high.

Content of the invention

It is an object of the invention to overcoming deficiency of the prior art, there is provided a kind of tandem double wire hidden arc welding numerical simulation Heat source model determination method for parameter, solves model parameter in prior art and determines the not high technical problem of high cost, precision.

For solving above-mentioned technical problem, the invention provides a kind of tandem double wire hidden arc welding numerical simulation heat source model parameter Determination method, it is characterized in that, comprise the following steps:

Step one: set up the double stripping mechanism that heat flow density decays along depth direction quadratic function, determine mariages mould The heat flux distribution function of type；

Attenuation function is quadratic function, and heat flux distribution function is:

In first half ellipsoid, heat flow density distribution function is:

$q(x,y,z,t)=\frac{6\sqrt{3}{f}_{fi}{q}_i}{{\mathrm{\π}}^{\frac{3}{2}}{a}_{fi}{b}_i{c}_i}\exp (-3(\frac{{(x-vt)}^2}{{\left(\frac{a_{fi}}{{\mathrm{cos\α}}_i}\right)}^2}+\frac{y^2}{{b_i}^2}+\frac{z^2}{{\left(\frac{c_i}{{\mathrm{cos\α}}_i}\right)}^2}\left)\right){\left(\frac{z-\frac{c_i}{{\mathrm{cos\α}}_i}}{\frac{c_i}{{\mathrm{cos\α}}_i}}\right)}^2$

Latter half ellipsoid heat flux distribution function is:

$q(x,y,z,t)=\frac{6\sqrt{3}{f}_{ri}{q}_i}{{\mathrm{\π}}^{\frac{3}{2}}{a}_{ri}{b}_i{c}_i}\exp (-3(\frac{{(x-vt)}^2}{{\left(\frac{a_{ri}}{{\mathrm{cos\α}}_i}\right)}^2}+\frac{y^2}{{b_i}^2}+\frac{z^2}{{\left(\frac{c_i}{{\mathrm{cos\α}}_i}\right)}^2}\left)\right){\left(\frac{z-\frac{c_i}{{\mathrm{cos\α}}_i}}{\frac{c_i}{{\mathrm{cos\α}}_i}}\right)}^2$

F in formula_fAnd f_rIt is respectively the distribution index in molten bath front and rear part for total input power, and f_f+f_r=2, q_iFor i-th The electric arc effective power of root wire, v is speed of welding；α_iFor the welding inclination angle of the i-th root wire, a_fi、a_riCorresponding for i-th silk The length of double stripping mechanism forward and backward hemisphere major semiaxis, b_iFor corresponding pair of semiminor axis of ellipsoid length of i-th silk, c_iFor i-th Corresponding pair of ellipsoid depth of root silk, in formulaPart is quadratic function decay, and t is welding process The time carrying out, i=1,2；

Step 2, according to the interaction of forward and backward two thermals source, tandem double wire hidden arc welding heat flux distribution is divided into three Individual region, determines the heat flow density in each region；

Tandem double wire hidden arc welding heat flux distribution can be divided into three areas by two double ellipsoid forward and backward hemisphere demarcation line Domain: region 1 heat flow density is double ellipsoid first halfs of front silk and double ellipsoid first half superpositions of rear silk；Region 2 hot-fluid is close Degree is double ellipsoid latter halfs of front silk and double ellipsoid first half superpositions of rear silk；Region 3 heat flow density is the double ellipse of front silk Ball latter half and double ellipsoid latter half superpositions of rear silk；

Step 3, sets up FEM (finite element) model, take the weldingvoltage of one group of submerged-arc welding coupling, welding current, speed of welding and Welding inclination angle, as simulation basic parameter, applies the thermal source load of each region heat flow density description determined above, to above heat Source model parameter carries out finite element inversion, obtains corresponding heat source model parameter；

Step 4, is adjusted with the amplitude setting to heat source model parameter obtained as above and speed of welding, based on each ginseng Array is closed and is obtained corresponding molten wide, fusion penetration sample values；Sensitivity analyses are carried out to each group sample values it is known that parameter a_fiWith a_riImpact to molten wide, fusion penetration is less, with double stripping mechanism parameter: b₁、b₂、c₁、c₂It is |input paramete with speed of welding v, With molten wide w, fusion penetration p as output parameter, corresponding sample is substituted into regression equation and draws sensitivity analyses result:

$w(b_1,c_1,b_2,c_2,v)=x_{1w}{b_1}^{x_{2w}}{c_1}^{x_{3w}}{b_2}^{x_{4w}}{c_2}^{x_{5w}}{v}^{x_{6w}}$

$p(b_1,c_1,b_2,c_2,v)=x_{1p}{b_1}^{x_{2p}}{c_1}^{x_{3p}}{b_2}^{x_{4p}}{c_2}^{x_{5p}}{v}^{x_{6p}}$

Wherein, x^2w、x^3w、x^4w、x^5w、x^6wFor b₁、c₁、b₂、c₂, the sensitivity coefficient to molten wide w for the v；x^2p、x^3p、x^4p、x^5p、 x^6pFor b₁、c₁、b₂、c₂, the sensitivity coefficient to molten wide p for the v；x_1w、x_1pIt is and q₁、q₂Relevant quadratic function f (q₁,q₂)、g(q₁, q₂)；

Step 5, is fitted to gained sensitivity equation and simplifies, detailed process is:

Take x_1w、x_1pFor q₁、q₂Relevant quadratic function f (q₁,q₂)、g(q₁,q₂), the expression formula of quadratic function is as follows:

f(q₁,q₂)=α₀+α₁q₁+α₂q₂+α₃q₁q₂+α₄q₁ ²+α₅q₂ ²

g(q₁,q₂)=β₀+β₁q₁+β₂q₂+β₃q₁q₂+β₄q₁ ²+β₅q₂ ²

Wherein, α_j, β_jFor pre- fitting constant, j=0,1,2,3,4,5；

Multigroup sample values are substituted in sensitivity equation to f (q₁,q₂)、g(q₁,q₂) carry out quadratic fit；Draw and simplify Sensitivity equation afterwards:；

Step 6, obtains heat source model parameter b by monofilament submerged-arc welding predictor formula₁、b₂Value, recycle simplification above Sensitivity prediction equation heat source model other specification.

Further, in step 3, in finite element inversion, Land use models search method calculates double stripping mechanism ginseng Number.

Further, in step 4, gained heat source model parameter and speed of welding are adjusted with 10% amplitude.

Further, sensitivity equation increases modifying factor δ x.

Compared with prior art, the beneficial effect that the present invention is reached is: the present invention is considering two thermals source interactions On the premise of, heat source density distribution is divided into three different regions of density function, is used as having with the distribution of this density function The thermal source load of limit unit simulation, can improve the precision of numerical simulation.For untested welding condition, by prediction Formula can be extended to result, and makes result of calculation serialization such that it is able to obtain any technique within the specific limits Heat source model parameter corresponding to parameter combination.The heat corresponding to different welding conditions can be directly obtained using the present invention Source model parameter, the tentative calculation workload of welding analog is substantially reduced, and the precision of simulation is substantially improved, and reduces engineer testing, Save development cost.

Brief description

Fig. 1 is the schematic flow sheet of the inventive method；

Fig. 2 is two thermal source interaction schematic diagrams of tandem double wire hidden arc welding；

Fig. 3 is single, tandem double wire hidden arc welding model molten wide to each 1.2 times of susceptibility test results of parameter；

Fig. 4 is single, tandem double wire hidden arc welding model fusion penetration to each 1.2 times of susceptibility test results of parameter；

Fig. 5 is to be contrasted with experimental result using predictor formula gained model parameter analog result.

Specific embodiment

The invention will be further described below in conjunction with the accompanying drawings.Following examples are only used for clearly illustrating the present invention Technical scheme, and can not be limited the scope of the invention with this.

As shown in figure 1, a kind of tandem double wire hidden arc welding numerical simulation heat source model determination method for parameter of the present invention, bag Include following steps:

Step one: set up the double stripping mechanism that heat flow density decays along depth direction quadratic function, determine mariages mould The heat flux distribution function of type；

In prior art, double stripping mechanism sets the first half of welding pool as 1/4 ellipsoid, latter half of It is allocated as another 1/4 ellipsoid, heat flow density is in Gaussian function normal distribution in semiellipsoid, core has maximum, From center to edge, exponentially curve declines.Through analysis, double stripping mechanism gained Pool and reality in prior art Welding pool pattern has larger difference in the depth direction.Therefore, the present invention is on the depth direction of double stripping mechanism Add attenuation function, wire shaped is fused come effective control by this attenuation function, reach the purpose controlling weld pool shape, from And it is preferably identical so that finite element analyses gained melting pool shape and actual melting pool shape is had.Attenuation function generally includes a letter Number, exponential function and quadratic function, for differential declines function through a large amount of attenuation function tentative calculations, find that quadratic function gained melts Pond shape is similar to experiment gained melting pool shape.Concrete modeling process is referring to the patent of invention of Application No. 2016110116037 " a kind of modeling method of tandem double wire hidden arc welding numerical simulation heat source model ", therefore heat flow density is along the decay letter of depth direction Number selects quadratic function, and the heat flux distribution function of mariages model is:

In first half ellipsoid, heat flow density distribution function is:

$q(x,y,z,t)=\frac{6\sqrt{3}{f}_{fi}{q}_i}{{\mathrm{\π}}^{\frac{3}{2}}{a}_{fi}{b}_i{c}_i}\exp (-3(\frac{{(x-vt)}^2}{{\left(\frac{a_{fi}}{{\mathrm{cos\α}}_i}\right)}^2}+\frac{y^2}{{b_i}^2}+\frac{z^2}{{\left(\frac{c_i}{{\mathrm{cos\α}}_i}\right)}^2}\left)\right){\left(\frac{z-\frac{c_i}{{\mathrm{cos\α}}_i}}{\frac{c_i}{{\mathrm{cos\α}}_i}}\right)}^2---\left(1\right)$

Latter half ellipsoid heat flux distribution function is:

$q(x,y,z,t)=\frac{6\sqrt{3}{f}_{ri}{q}_i}{{\mathrm{\π}}^{\frac{3}{2}}{a}_{ri}{b}_i{c}_i}\exp (-3(\frac{{(x-vt)}^2}{{\left(\frac{a_{ri}}{{\mathrm{cos\α}}_i}\right)}^2}+\frac{y^2}{{b_i}^2}+\frac{z^2}{{\left(\frac{c_i}{{\mathrm{cos\α}}_i}\right)}^2}\left)\right){\left(\frac{z-\frac{c_i}{{\mathrm{cos\α}}_i}}{\frac{c_i}{{\mathrm{cos\α}}_i}}\right)}^2---\left(2\right)$

F in formula_fAnd f_rIt is respectively the distribution index in molten bath front and rear part for total input power, and f_f+f_r=2, q_iFor i-th The electric arc effective power of root wire, wherein q=η ui；U is weldingvoltage, and i is welding current, and η is electric arc effective thermal efficiency system Number, value 0.77～0.9；V is speed of welding；α_iFor the welding inclination angle of the i-th root wire, a_fi、a_riCorresponding double ellipse for i-th silk The length of ball heat source model forward and backward hemisphere major semiaxis, b_iFor corresponding pair of semiminor axis of ellipsoid length of i-th silk, c_iFor i-th silk Corresponding pair of ellipsoid depth, in formulaPart is attenuation function (quadratic function), and t is to weld The time of Cheng Jinhang, i=1,2.

Understand, weldingvoltage, welding current, speed of welding and welding inclination angle are known welding condition, determine Double stripping mechanism shape needs to solve four parameters a_fi、a_ri、b_iAnd c_i.And the size of this four parameters will directly influence The distribution in temperature field in welding process, therefore, it is crucial for solving this four parameters.

Step 2, according to the interaction of forward and backward two thermals source, tandem double wire hidden arc welding heat flux distribution is divided into three Individual region, determines the heat flow density in each region；

It is not simple overlaying relation between tandem double wire hidden arc welding forward and backward silk thermal source, but exist interactional Relation.Two thermal source interaction schematic diagrams of tandem double wire hidden arc welding are as shown in Fig. 2 tandem double wire hidden arc welding heat flux distribution Three regions can be divided into by two double ellipsoid forward and backward hemisphere demarcation line: region 1 heat flow density is double ellipsoid first halfs of front silk Divide and double ellipsoid first halfs of rear silk are superimposed；Region 2 heat flow density is the double ellipse of double ellipsoid latter halfs of front silk and rear silk Ball first half is superimposed；Region 3 heat flow density is double ellipsoid latter halfs of front silk and double ellipsoid latter half superpositions of rear silk.

These three regions are only referred to the region of heat flow density effect, because the impact each other of front silk and rear silk, lead to Heat flow density (can be understood as different in these three Regional land surface heat fluxes) of different sizes.These three regions are by finite element mould The subprogram intending software writing thermal source program, thus heat source model subregion is loaded into carry out in FEM (finite element) model anti- Drill.

Step 3, sets up FEM (finite element) model, take the weldingvoltage of one group of submerged-arc welding coupling, welding current, speed of welding and Welding inclination angle, as simulation basic parameter, carries out finite element inversion to above heat source model parameter, obtains corresponding thermal source mould Shape parameter；

This process is described in detail with an embodiment, preferred dimension is the q345 steel plate of 18.4mm × 200mm × 300mm, Set up three-dimensional welding FEM (finite element) model and be simulated temperature field in abaqus platform；Define temperature governing equation and perimeter strip The weldment material properties parameter being related in part equation, including the density of mother metal and weld seam, material phase transformation latent heat, convection coefficient, The coefficient of heat conduction, specific heat capacity, radiation heat transfer coefficient, thermal coefficient of expansion, elastic modelling quantity and Poisson's ratio etc.；Set absolute zero and glass The graceful constant of Wurz.

FEM (finite element) model is carried out with non-uniform grid division: adopt unit size to be that welding is molten at Seam and heat effected zone / to eight/10ths of pond width, and arrived for 1/5th of welding pool width using unit size in mother metal periphery A quarter.

Take the weldingvoltage of one group of submerged-arc welding coupling, welding current, speed of welding as simulation basic parameter, before taking welding Silk electric current is 600a, and front silk weldingvoltage is 36v, and silk welding current is 500a afterwards, and silk weldingvoltage is 40v afterwards, speed of welding For 3.5m/min, as control experiment, front silk aclinal, rear wire bond connects inclination angle and is 15 °, and forward and backward distance between weldingwires is 55mm, thermal effect Rate coefficient η takes 0.9, f_fTake 1.1, f_rTake 0.9.Experiment gained molten wide w=23.96mm, fusion penetration p=8.96mm, this experimental data can Using slide gauge direct measurement weldment gained.Above parameter is substituted into heat flux distribution formula, three regions are applied respectively Plus the thermal source load of each region heat flow density description determined above；Carry out inverting using pattern search method of the prior art, Mentioned by this pattern search method also has in " accelerating step length inverting multiplewire submerged arc welding double stripping mechanism parameter " literary composition, Belong to prior art, here is omitted.By inverting, draw a group model parameter of optimum.Enter to according to pattern search method Obtained from row finite element modelling, numerous parametric statisticss are concluded and can be obtained, and take a_fi=b_i, a_ri=2.5b_i, draw model parameter It is respectively as follows:

b₁=8.6, c₁=5.8, a_f1=8.6, a_r1=21.5

b₂=8.1, c₂=5.4, a_f2=8.1, a_r2=20.25

The fusion penetration molten wide simulating molten bath, simulation gained fusion penetration molten wide and the actual fusion penetration measuring can be obtained based on above parameter The data of molten wide is less than 10% it is known that simulating gained fusion penetration molten wide and differing with the fusion penetration molten wide data of actual measurement as shown in table 1, Therefore above heat source model can be used for simulating the heat flux distribution of biserial mariages thermal source.

Table 1 analog result is contrasted with experimental result

Step 4, is adjusted with 10% amplitude to heat source model parameter obtained as above and speed of welding, will a_fi、 a_ri、b_i、c_iAdjust with v to 0.8,0.9,1.1,1.2 ... times of raw parameter, each parameter combination is simulated obtaining molten bath Molten wide, fusion penetration, obtain multigroup parameter and molten wide, the corresponding sample values of fusion penetration.Sensitivity analyses are carried out to each group sample values.

The shadow to fusion penetration, molten wide for the parameter of the relatively respective heat source model of monofilament submerged-arc welding, mariages (forward and backward silk) submerged-arc welding Ringing, thus drawing the larger Heat-Source Parameters of impact, carrying out sensitivity analyses.The heat source model of monofilament is also double-ellipsoid heat source mould Type.The sensitivity analyses of monofilament parameter are referring to a kind of patent of invention " monofilament submerged-arc welding numerical value of Application No. 2016110210664 Simulation heat source model determination method for parameter ".Carry out monofilament heat source model parameter and mariages parameter comparison herein, be in order to for after After the simplification in face, the parameter computing of formula provides foundation.

When comparing the impact to fusion penetration molten wide of front silk or rear silk single heat source model parameter, it is front silk or rear silk parameter list Solely change, that is, during front silk parameter adjustment, silk parameter keeps constant afterwards, afterwards during silk parameter adjustment, front silk parameter keeps constant.

When each parameter adjustment of model is to 1.2 times, silk model after silk before monofilament submerged-arc welding, double wire hidden arc welding, double wire hidden arc welding In the impact to molten wide for each parameter as shown in figure 3, in silk model after silk before monofilament submerged-arc welding, double wire hidden arc welding, double wire hidden arc welding The impact to fusion penetration for each parameter is as shown in Figure 4 it is known that parameter a_fiAnd a_riImpact to molten wide fusion penetration is less, double wire hidden arc welding b₁、 b₂Impact to molten wide or fusion penetration is close.With double stripping mechanism parameter: b₁、b₂、c₁、c₂It is |input paramete with speed of welding v, With molten wide w, fusion penetration p as output parameter, corresponding sample is substituted into regression equation and draws sensitivity analyses result:

$w(b_1,c_1,b_2,c_2,v)=x_{1w}{b_1}^{x_{2w}}{c_1}^{x_{3w}}{b_2}^{x_{4w}}{c_2}^{x_{5w}}{v}^{x_{6w}}---\left(3\right)$

$p(b_1,c_1,b_2,c_2,v)=x_{1p}{b_1}^{x_{2p}}{c_1}^{x_{3p}}{b_2}^{x_{4p}}{c_2}^{x_{5p}}{v}^{x_{6p}}---\left(4\right)$

Wherein, x_2w、x_3w、x_4w、x_5w、x_6wFor b₁、c₁、b₂、c₂, the sensitivity coefficient to molten wide w for the v；x_2p、x_3p、x_4p、x_5p、 x_6pFor b₁、c₁、b₂、c₂, the sensitivity coefficient to molten wide p for the v；x_1w、x_1pIt is and q₁、q₂(q₁、q₂It is respectively forward and backward silk electric arc effective Power) relevant function, take x_1w、x_1pFor q₁、q₂(electric arc effective power) relevant quadratic function f (q₁,q₂)、g(q₁,q₂).

Step 5, is simplified and matching to gained sensitivity equation；

Multigroup sample values are substituted into formula (3) (4) be fitted, can obtain sensitivity equation (5), (6):

$w=\frac{\mathrm{\δ}xf(q_1,q_2)}{{b_1}^{0.073}{c_1}^{0.3215}{b_2}^{0.082}{c_2}^{0.2936}{v}^{0.997}}---\left(5\right)$

$p=\frac{\mathrm{\δ}xg(q_1,q_2){c_1}^{0.7881}{c_2}^{0.4035}}{{b_1}^{0.161}{b_2}^{0.172}{v}^{0.968}}---\left(6\right)$

Wherein q₁、q₂It is respectively forward and backward silk electric arc effective power, δ x is modifying factor, due to sensitivity equation ignored A_f、a_rParameter to fusion penetration p, molten wide w impact, cast out parameter a_fi、a_riIndex coefficient, in order to increase the accurate of formula Property, part will be cast out and be approximately equal to modifying factor δ x.

Take x_1w、x_1pFor q₁、q₂(electric arc effective power) relevant quadratic function f (q₁,q₂)、g(q₁,q₂), quadratic function Expression formula is as follows:

f(q₁,q₂)=α₀+α₁q₁+α₂q₂+α₃q₁q₂+α₄q₁ ²+α₅q₂ ²(7)

g(q₁,q₂)=β₀+β₁q₁+β₂q₂+β₃q₁q₂+β₄q₁ ²+β₅q₂ ²(8)

Wherein, α_j, β_j(j=0,1,2,3,4,5) it is pre- fitting constant.Q can effectively be represented in quadratic function₁, q₂Mutual Q in impact relation, and quadratic function formula₁, q₂Output parameter is affected for peer-to-peer, tally with the actual situation.

Multigroup sample values are substituted in sensitivity equation to f (q₁,q₂)、g(q₁,q₂) carry out quadratic fit；Draw and simplify Sensitivity equation afterwards:

$w=\frac{\mathrm{\δ}x(109.52+6.268q_1+10.116q_2+0.331q_1{q}_2+0.06{q_1}^2+0.0696{q_2}^2)}{{b_1}^{0.073}{c_1}^{0.3215}{b_2}^{0.082}{c_2}^{0.2936}{v}^{0.997}}---\left(9\right)$

$p=\frac{\mathrm{\δ}x(-15.36+0.46q_1+0.418q_2+0.024q_1{q}_2+0.004{q_1}^2+0.0098{q_2}^2){c_2}^{0.7881}{c_1}^{0.4035}}{{b_1}^{0.161}{b_2}^{0.172}{v}^{0.968}}---\left(10\right)$

Wherein △ x is fitted to by multigroup sample valuesApproximate acquirement 0.98.

Step 6, the simplification drawing sensitivity equation is used for heat source model parameter prediction, will be pre- using abaqus software Survey parameter and be used for heat source model loading, obtain temperature field result, weld pool shape and accordingly defeated is drawn by temperature field result Go out parameter molten wide w, fusion penetration p result, with experiment show.

Before taking welding, silk electric current is 615a, and front silk weldingvoltage is 34v, and silk welding current is 480a afterwards, and rear wire bond connects electricity Press as 40v, speed of welding is 3.5m/min, front silk welding inclination angle, rear wire bond connects inclination angle and is 15 °, and forward and backward distance between weldingwires is 55mm, as confirmatory experiment, tests and measures weldment molten bath molten wide w=23.58mm, fusion penetration p=10.31mm, and according to Fig. 3, Fig. 4 It can be seen that silk parameter b before double wire hidden arc welding₁, rear silk parameter b₂Similar to the affecting laws of output parameter w, p, b₁、b₂Value Can approximately be obtained by monofilament submerged-arc welding predictor formula.The concrete solution procedure of monofilament submerged-arc welding predictor formula is referring to Application No. A kind of 2016110210664 patent of invention " monofilament submerged-arc welding numerical simulation heat source model determination method for parameter ".First test Measure molten wide, the fusion penetration numerical value of front silk, by front silk parameter b₁, welding current, weldingvoltage, speed of welding substitute into the molten of monofilament In wide, two predictor formulas of fusion penetration, calculate parameter b₁；Then the molten wide of silk, fusion penetration numerical value after experiment is measured, by rear silk Parameter b₂, welding current, weldingvoltage, speed of welding substitute into the molten wide of monofilament, in two predictor formulas of fusion penetration, calculate parameter b₂.

Again by parameter b calculating₁And b₂Substitute into predictor formula (9) (10) and calculate parameter c₁And c₂, take a_fi=b_i,a_ri =2.5b_i, thus show that model parameter is:

b₁=8.5, c₁=3.9, a_f1=8.5, a_r1=21.3

b₂=8.0, c₂=7.9, a_f2=8.0, a_r2=20.

Above parameter is substituted in double stripping mechanism, draws temperature field and molten bath shape with the simulation of abaqus software Shape.Simulate the molten wide obtaining, fusion penetration numerical value with experiment gained molten wide, fusion penetration numerical value as shown in table 2, experimental result and predictor formula The analog result error only 2.5% about of gained model parameter, illustrates that both goodnesses of fit are higher.Simulate the melting pool shape that obtains with As shown in figure 5, the in figure left side is experiment gained melting pool shape, the right is simulation gained molten bath to experiment gained melting pool shape comparison diagram Shape graph, as can be seen from the figure also has preferably identical, predictor formula obtains with experimental result on simulation gained Pool The checking of experiment.

Table 2 predictor formula gained analog result is contrasted with experimental result

The present invention, on the premise of considering that two thermals source interact, heat source density distribution is divided into density function different Three regions, the thermal source load of finite element modelling is used as with the distribution of this density function, the precision of numerical simulation can be improved. For untested welding condition, result can be extended by predictor formula, and make result of calculation serialization, So as to obtain the heat source model parameter corresponding to any combination of process parameters within the specific limits.Can using the present invention Directly obtain the heat source model parameter corresponding to different welding conditions, the tentative calculation workload of welding analog substantially reduced, And the precision of simulation is substantially improved, reduce engineer testing, save development cost.

The above is only the preferred embodiment of the present invention it is noted that ordinary skill people for the art For member, on the premise of without departing from the technology of the present invention principle, some improvement and modification can also be made, these improve and modification Also should be regarded as protection scope of the present invention.

Claims (4)

1. a kind of tandem double wire hidden arc welding numerical simulation heat source model determination method for parameter, is characterized in that, comprise the following steps:

Step one: set up the double stripping mechanism that heat flow density decays along depth direction quadratic function, determine mariages model Heat flux distribution function；

Attenuation function is quadratic function, and heat flux distribution function is:

In first half ellipsoid, heat flow density distribution function is:

$q(x,y,z,t)=\frac{6\sqrt{3}{f}_{fi}{q}_i}{{\mathrm{\π}}^{\frac{3}{2}}{a}_{fi}{b}_i{c}_i}\exp (-3(\frac{{(x-vt)}^2}{{\left(\frac{a_{fi}}{{\mathrm{cos\α}}_i}\right)}^2}+\frac{y^2}{{b_i}^2}+\frac{z^2}{{\left(\frac{c_i}{{\mathrm{cos\α}}_i}\right)}^2}\left)\right){\left(\frac{z-\frac{c_i}{{\mathrm{cos\α}}_i}}{\frac{c_i}{{\mathrm{cos\α}}_i}}\right)}^2$

Latter half ellipsoid heat flux distribution function is:

$q(x,y,z,t)=\frac{6\sqrt{3}{f}_{ri}{q}_i}{{\mathrm{\π}}^{\frac{3}{2}}{a}_{ri}{b}_i{c}_i}\exp (-3(\frac{{(x-vt)}^2}{{\left(\frac{a_{ri}}{{\mathrm{cos\α}}_i}\right)}^2}+\frac{y^2}{{b_i}^2}+\frac{z^2}{{\left(\frac{c_i}{{\mathrm{cos\α}}_i}\right)}^2}\left)\right){\left(\frac{z-\frac{c_i}{{\mathrm{cos\α}}_i}}{\frac{c_i}{{\mathrm{cos\α}}_i}}\right)}^2$

F in formula_fAnd f_rIt is respectively the distribution index in molten bath front and rear part for total input power, and f_f+f_r=2, q_iFor the i-th root bead The electric arc effective power of silk, v is speed of welding；α_iFor the welding inclination angle of the i-th root wire, a_fi、a_riCorresponding double ellipse for i-th silk The length of ball heat source model forward and backward hemisphere major semiaxis, b_iFor corresponding pair of semiminor axis of ellipsoid length of i-th silk, c_iFor i-th silk Corresponding pair of ellipsoid depth, in formulaPart is carried out for welding process for quadratic function decay, t Time, i=1,2；

Step 2, according to the interaction of forward and backward two thermals source, tandem double wire hidden arc welding heat flux distribution is divided into three areas Domain, determines the heat flow density in each region；

Tandem double wire hidden arc welding heat flux distribution can be divided into three regions: area by two double ellipsoid forward and backward hemisphere demarcation line Domain 1 heat flow density is double ellipsoid first halfs of front silk and double ellipsoid first half superpositions of rear silk；Before region 2 heat flow density is Double ellipsoid latter halfs of silk and double ellipsoid first half superpositions of rear silk；Region 3 heat flow density is that double ellipsoids of front silk are later half Part and double ellipsoid latter half superpositions of rear silk；

Step 3, sets up FEM (finite element) model, takes the weldingvoltage of one group of submerged-arc welding coupling, welding current, speed of welding as mould Intend basic parameter, apply the thermal source load of each region heat flow density description determined above, above heat source model parameter is carried out Finite element inversion, obtains corresponding heat source model parameter；

Step 4, is adjusted with the amplitude setting to heat source model parameter obtained as above and speed of welding, based on each parameter group Conjunction obtains corresponding molten wide, fusion penetration sample values；With double stripping mechanism parameter: b₁、b₂、c₁、c₂It is defeated with speed of welding v Enter parameter, with molten wide w, fusion penetration p as output parameter, corresponding sample substituted into regression equation and draws sensitivity analyses result:

$w(b_1,c_1,b_2,c_2,v)=x_{1w}{b_1}^{x_{2w}}{c_1}^{x_{3w}}{b_2}^{x_{4w}}{c_2}^{x_{5w}}{v}^{x_{6w}}$

$p(b_1,c_1,b_2,c_2,v)=x_{1p}{b_1}^{x_{2p}}{c_1}^{x_{2p}}{b_2}^{x_{4p}}{c_2}^{x_{5p}}{v}^{x_{6p}}$

Wherein, x^2w、x^3w、x^4w、x^5w、x^6wFor b₁、c₁、b₂、c₂, the sensitivity coefficient to molten wide w for the v；x^2p、x^3p、x^4p、x^5p、x^6pFor b₁、c₁、b₂、c₂, the sensitivity coefficient to molten wide p for the v；x_1w、x_1pIt is and q₁、q₂Relevant quadratic function f (q₁,q₂)、g(q₁,q₂)；

Step 5, is fitted to gained sensitivity equation and simplifies, detailed process is:

Take x_1w、x_1pFor q₁、q₂Relevant quadratic function f (q₁,q₂)、g(q₁,q₂), the expression formula of quadratic function is as follows:

f(q₁,q₂)=α₀+α₁q₁+α₂q₂+α₃q₁q₂+α₄q₁ ²+α₅q₂ ²

g(q₁,q₂)=β₀+β₁q₁+β₂q₂+β₃q₁q₂+β₄q₁ ²+β₅q₂ ²

Wherein, α_j, β_jFor pre- fitting constant, j=0,1,2,3,4,5；

Multigroup sample values are substituted in sensitivity equation to f (q₁,q₂)、g(q₁,q₂) carry out quadratic fit；Draw after simplifying Sensitivity equation；

Step 6, obtains heat source model parameter b by monofilament submerged-arc welding predictor formula₁、b₂Value, recycle simplification above sensitive Property prediction equation heat source model other specification.

2. the modeling method of a kind of tandem double wire hidden arc welding numerical simulation heat source model according to claim 1, its feature It is that, in step 3, in finite element inversion, Land use models search method calculates double stripping mechanism parameter.

3. the modeling method of a kind of tandem double wire hidden arc welding numerical simulation heat source model according to claim 1, its feature It is, in step 4, heat source model parameter obtained as above and speed of welding to be adjusted with 10% amplitude.

4. the modeling method of a kind of tandem double wire hidden arc welding numerical simulation heat source model according to claim 1, its feature It is, in step 5, to increase in sensitivity equation and be equal to parameter a_fi、a_riThe modifying factor δ x of index coefficient.