CN105975708A - Steel tube welding parameter optimization method based on numerical simulation and data analysis - Google Patents

Abstract

The invention relates to a steel tube welding parameter optimizing method based on numerical simulation and data analysis. The method comprises the steps of determining various welding conditions of a steel tube; building a finite element model, and adopting an eight-node hexahedral element for dividing a three-dimensional finite element model network; according to actual welding parameters, checking a heat source; performing simulation through a numerical simulation software Sysweld; after welding simulation, distribution of a residual stress field and a temperature field of a welding joint is analyzed, on the basis of the experiment principle of an uniform experiment, test points are sufficiently and uniformly dispersed within the experiment range, and the simplified testing scheme is determined; welding processes under different parameters are simulated in Sysweld simulation, MATLAB is used for data analysis, and the best technological condition is obtained. By means of the numerical simulation and data analysis technology, a large amount of test workpiece waste is avoided, workloads for parameter optimization are reduced, the optimization efficiency is improved, the parameter result accuracy determined through the method is high, and the method has great feasibility.

Description

A kind of optimization steel-pipe welding parametric technique based on numerical simulation Yu data analysis

Technical field

The present invention relates to a kind of optimization steel-pipe welding parametric technique based on numerical simulation Yu data analysis.

Background technology

Welding parameter is particularly important to steel-pipe welding, directly determines welding quality, needs welding parameters to carry out the most before welding Preferably.At present, optimizing steel-pipe welding parametric technique is by carrying out substantial amounts of workpiece test, to single parameter at previous experiences On the basis of constantly adjust, until filtering out more suitably parameter, process of the test needs to consume substantial amounts of workpiece, Experimentation cost is higher, and workload is the biggest, also cannot determine the reciprocal effect between each parameter simultaneously, affect integral solder quality.

In view of this, need to invent a kind of with low cost, simple to operate, the most comprehensively based on numerical simulation and data analysis Optimize steel-pipe welding parametric technique, replace the optimization method in the past based on the test of physics test specimen.

Summary of the invention

It is an object of the invention to overcome the defect of prior art, it is provided that a kind of base simple to operate, with low cost, the most comprehensive Optimization steel-pipe welding parametric technique in numerical simulation Yu data analysis.

To achieve these goals, the technical scheme that the present invention takes is as follows:

A kind of optimization steel-pipe welding parametric technique based on numerical simulation Yu data analysis, its key technology is, it includes as follows Step:

Step one, determine steel-pipe welding condition, including: ambient temperature, welding procedure, thickness of steel pipe, groove type, steel pipe Diameter and steel pipe material etc.；

Step 2, physical dimension according to actual weldment set up finite element in Visual-Environment Model, and use eight node hexahedral elements to divide three-dimensional finite element model grid；

Step 3, thermal source is checked with welding object according to the welding procedure in reality；

Step 4, carry out welding simulation by software Sysweld；

First compare with actual test specimen after step 5, welding analog and guarantee the accuracy of simulation, then the remnants analyzing weld seam should The field of force and the distribution in temperature field；

Step 6, experimental principle according to even test, select to affect some principal elements of welding result, and determine theirs Excursion, selects uniform designs table and gauge outfit design, finally determines short form test scheme；

Step 7, in Sysweld emulates, simulate the welding process under different tests parameter, weld seam under record different experimental conditions The microscopic structure composition of heat affected area after the maximum of middle Mises residual stress and cooling；

Step 8, utilize the MATLAB data to obtaining to carry out data analysis, finally give the process conditions of optimum.

Further, also include step 9, be applied to welding piece in kind, the accuracy of the result.

Compared with prior art, having the beneficial effect that acquired by the present invention:

The technology that the welding condition optimization method that the present invention proposes combines by utilizing numerical simulation and data analysis, it is to avoid The waste of lot of experiments workpiece, is decreased the workload that welding parameter optimizes, reduces experimentation cost, screened by this method The parametric results accuracy determined is high, and this method possesses bigger feasibility, it is easy to large-scale promotion.

Accompanying drawing explanation

Accompanying drawing 1 is the schematic diagram of embodiment 1 welding point；

Accompanying drawing 2 is the FEM (finite element) model that embodiment 1 is set up；

Accompanying drawing 3 is embodiment 1 60s, 70s, 140s and postwelding removes the effective stress field result of calculation (Von in clamping moment Mises stress result)；

Accompanying drawing 4 is the analysis of Residual Stress result of calculation figure after embodiment 1 is soldered；

Detailed description of the invention

Below in conjunction with detailed description of the invention, the present invention is carried out further details of narration.

Embodiment 1

Step one: the welding condition of record P91 high temperature main steam line: specification is 333mm × 30mm, and pipe diameter is 333mm, Thickness of pipe wall is 30mm.P91 steel is middle alloy heat resisting steel, has good non-oxidizability, preferable high temperature strength and resistance to hydrogen sulfide Corrosivity and there is preferable cold deformation performance.Due to the combination property that it is good, widely should obtain in thermal power station's construction With, it uses temperature at 500 DEG C～about 600 DEG C.This steel grade has been listed in ASEM and JIS standard.

P91 steel Welding Analysis: the welding of steel is relevant with the chemical composition of steel, the ambient temperature mechanical properties of steel and physical characteristic.Logical Cross Tables 1 and 2 and can more understand the characteristic of P91.

The chemical composition (Wt%) of table 1 P91 steel

C	S	P	Mn	Si	Ni	Cr	Mo	V	Nb
										0.078	0.0031	0.016	0.27	0.3	0.043	8.56	0.96	0.22	0.081

The ambient temperature mechanical properties of table 2 P91 steel

δb(MPa)	δ0.2(MPa)	δ (%)	Ak(J)	HB
					≥585	≥415	20	149L/min	≤250

Step 2: set up finite element in Visual-Environment according to the physical dimension of actual weldment Model, and use eight node hexahedral elements to divide three-dimensional finite element model grid；

Step 3: with welding object, thermal source is checked according to the welding procedure in reality

Welding procedure is as follows: welding wire is ER90S-B9, gage of wire 2.5mm, and welding rod is E9015-B9, core diameter 3.2mm, Its chemical composition is shown in Table 3.

Table 3 welding wire ER90S-B9, the chemical composition (mass fraction) of welding rod E9015-B9

Whole welding process divides 14 roads to complete.In welding process, interlayer temperature controls at 200～300 DEG C, welding point such as Fig. 1 Shown in, welding condition such as table 4:

Table 4 welding condition

What in this experiment, heat source model was chosen is double stripping mechanism, and it is calculated in order to the simulation realizing logarithm value.

Step 4, carry out welding simulation by software Sysweld；

Welding simulation is to complete based on special welding numerical simulation software Sysweld, to sealing of tube model given welding ginseng Carry out welding process emulation under several, result of calculation is carried out field of welding temperature, deformation and analysis on Stress Field, calculate grid model such as Shown in Fig. 2, different with the thermograde away from heat affected area according to heat affected area, nearly weld metal zone number of grid draw closeer, More sparse draw away from weld metal zone number of grid.

First compare with actual test specimen after step 5, welding analog and guarantee the accuracy of simulation, then the remnants analyzing weld seam should The field of force and the distribution in temperature field；

First compare with test specimen actual welding process and result after analogue simulation welds, it is ensured that the accuracy of simulation, What in this experiment, heat source model was chosen is double stripping mechanism, and it is calculated in order to the simulation realizing logarithm value, its mathematical table Reach formula as follows:

Front semielliptical heat flux distribution:

$q(x,y,z)=\frac{6\sqrt{3}{r}_1{Q}_1}{a_{1}bc\mathrm{\π}\sqrt{\mathrm{\π}}}\exp (-\frac{3x^2}{a_1^2}-\frac{3y^2}{b^2}-\frac{3z^2}{c^2}),x\≥0---\left(1\right)$

Rear semielliptical heat flux distribution:

$q(x,y,z)=\frac{6\sqrt{3}{r}_2{Q}_2}{a_{2}bc\mathrm{\&pi;}\sqrt{\mathrm{\&pi;}}}\exp (-\frac{3x^2}{a_2^2}-\frac{3y^2}{b^2}-\frac{3z^2}{c^2}),x<0---\left(2\right)$

In formula: q heat flux, J/m2s；

X, y, z are relative to the coordinate of source center；；

Hemisphere and the energy density of rear quarter, J/m3 before Q1, Q2 thermal source；

The parameter that a1, a2, b, c are relevant to molten bath.

R1, r2 are model front and rear part energy coefficient and r1+r2=2.

By to double stripping mechanism systematic analysis, the work of most critical is clear and definite two big class parameters, and the first kind is claimed For form parameter, including a1, a2, b, c etc., this kind of parameter after molten bath solidifies by measure the width (b) of weld seam, the degree of depth (c), The form parameter (a1, a2) of molten bath end arc crater determines.Equations of The Second Kind is referred to as the parameter that cannot measure, mainly Q1 and Q2 The two is used for characterizing the parameter of energy density.Under conditions of both knowing form parameter, Q1 is regarded as variable x, simultaneously by Q1, The ratio of Q2 regards as constant y, next utilizes the heat flux q value of above-mentioned formula key point a series of to the thermal source thermal treatment zone to give accurately Calculate, these values done accumulation process, and with test operation during selected heat input contrasted, calculate variable x's Size, next moves to again calculate link, the melting pool shape obtained via numerical computations and the concrete knot obtained by test Fruit is put together and compares.Misfit situation if belonging to, then to give necessary adjustment based on the size of constant y, enter the most again Row correlation computations, till meeting the condition that both are consistent, after parameters enters into a kind of basic suitably state, connects Get off to answer HSLA STEEL TO WELD to launch repeated multiple times verification, series of parameters is done necessary adjustment simultaneously.

By melting pool shape contrast during the molten bath that formed of correction after heat source model and actual welding, if the heat of obtain each layer of school The molten bath that source melting pool shape and actual welding are formed is the most identical, then can well be to pipeline welding by this group heat source model Cheng Jinhang numerical simulation, thus obtain temperature field and stress-strain result accurately.

Deliberated index in test is the maximum of Von Mises residual stress.Von Mises is a kind of yield criterion, surrender standard We are generally equivalent stress value then, and it follows mechanics of materials fourth strength theory.Von Mises stress is from alteration of form Weigh than the angle of energy, i.e. think when the complicated shape parts of material reaches a certain value, material yield.Stress solve complete with After, can read that each time load walks by post-processing module as a result, it is possible to obtain the stress distribution situation of each load step, Fig. 3 respectively 60s, 70s, 140s and postwelding remove the moment such as clamping effective stress field result of calculation (Von Mises should Power result).This group figure shows the dynamic situation of change of the equivalent stress in different heating moment and cooling moment in welding process.Can To find to increase over time, welding residual stress primarily occur ins nearly weld metal zone, and residual stress is along bead direction symmetrically Distribution, simultaneously it appeared that at weld metal zone metal molten, when therefore starting, the residual stress of weld metal zone is 0, nearly weld metal zone metal Receive thermal expansion, show bigger residual stress.Relatively low from weld seam remotely temperature, the residual stress of generation is relatively small, Receiving restriction in the position of clamping due to sweating heat deformation, there is local residual stress in clamping position, is cooled to room after having welded Temperature, welding residual stress is concentrated mainly on weld metal zone, nearly weld metal zone and setting position, and maximum residual stress is 700.2MPa, in Symmetrical.

Step 6, experimental principle according to even test, select to affect some principal elements of welding result, and determine theirs Excursion, selects uniform designs table and gauge outfit design, finally determines short form test scheme.

In the present embodiment, it is welded as object of study with SMAW.Due to welding current, arc voltage, speed of welding and groove gap On welding process heat input have important impact, therefore welding residual stress also can be had a great impact, thus select this four Individual factor is as object of study.According to uniform experiment design, select U₁₀(10⁸) table design experiment scheme, need 10 numerical value altogether Simulation test.Uniform experiment design factor level is as shown in table 5.

Table 5 uniform experiment design factor level

Step 7, the welding process simulated in table 5 under 10 groups of different tests parameters in Sysweld emulates, record different experiments Under the conditions of the microscopic structure composition of heat affected area after the maximum of Mises residual stress and cooling in weld seam.

Step 8, utilize the MATLAB data to obtaining to carry out data analysis, finally give the process conditions of optimum.

According to the maximum of the Mises residual stress that ten groups of tests obtain, and draw optimum ginseng by the data analysis of MATLAB Respectively speed of welding v=148mm/min, arc voltage U=20V, welding current I=140A, the slope that number is each factor Mouth gap L=4mm.

In order to optimum results is verified, optimum welding procedure is carried out welding residual stress analysis.Welding condition is weldering Meet speed v=148mm/min, arc voltage U=20V, welding current I=140A, groove gap L=4mm.Meter Calculate result as shown in Figure 4.Wherein σ y represents the axial welding residual stress of high-temperature pipe, and σ x represents the welding of high-temperature pipe hoop Residual stress, unit is MPa.Being found out by optimum combination welding procedure analysis of Residual Stress result, deliberated index Mises is remaining The maximum of stress is 515.1MPa, significantly less than the stress value measured by analytical table ten groups of tests.

Step 9, being applied to welding piece in kind, through checking, the maximum of actual test specimen Mises residual stress is 509.6MPa Show that optimizing welding process result is correct and effective obtained by this test design method.

The above embodiment is only the preferred embodiments of the present invention, and and the feasible enforcement of non-invention exhaustive.For ability For the those skilled in the art of territory, on the premise of without departing substantially from the principle of the invention and spirit, it done any is obviously changed Dynamic, within all should being contemplated as falling with the claims of the present invention.

Claims (2)

1. an optimization steel-pipe welding parametric technique based on numerical simulation Yu data analysis, it is characterised in that it includes walking as follows Rapid:

Step one, determine steel-pipe welding condition, including: ambient temperature, welding procedure, thickness of steel pipe, groove type, steel pipe Diameter and steel pipe material etc.；

Step 2, physical dimension according to actual weldment set up finite element in Visual-Environment Model, and use eight node hexahedral elements to divide three-dimensional finite element model grid；

Step 3, thermal source is checked with welding object according to the welding procedure in reality；

Step 4, carry out welding simulation by software Sysweld；

First compare with actual test specimen after step 5, welding analog and guarantee the accuracy of simulation, then the remnants analyzing weld seam should The field of force and the distribution in temperature field；

Step 6, experimental principle according to even test, select to affect some principal elements of welding result, and determine theirs Excursion, selects uniform designs table and gauge outfit design, finally determines short form test scheme；

Step 7, in Sysweld emulates, simulate the welding process under different tests parameter, weld seam under record different experimental conditions The microscopic structure composition of heat affected area after the maximum of middle Mises residual stress and cooling；

Step 8, utilize the MATLAB data to obtaining to carry out data analysis, finally give the process conditions of optimum.

A kind of optimization steel-pipe welding parametric technique based on numerical simulation Yu data analysis the most according to claim 1, it is special Levy and be, also include that step 9 is applied to welding piece in kind, the accuracy of the result.