CN102831274A - Correction method of distribution of quenching stress field of aluminum alloy thick plate - Google Patents

Abstract

The invention discloses a correction method of the distribution of a quenching stress field of an aluminum alloy thick plate. In the method, on the combination of X-ray surface stress calibration and finite element modeling and simulation, sample deformation deviation produced by the traditional layer cutting method during a stress test process is obtained, and the traditional computation model is corrected by a layered deformation compensation polynomial fitting function, thus, computation accuracy of the stress field in the layer cutting method is drastically improved, and the result comparison deviation is controlled to be +/-20MPa. The method comprises the steps of: obtaining and correcting a quenching heat transfer boundary condition, analyzing layer cutting deviation and correcting layer cutting computation model mathematically. The method has the advantages that by using the characteristics of multi test methods, the stress is accurately designed in a multi-layer complementary way, and the correction method based on multi test methods is provided; the defect that the traditional single test method has bad accuracy is overcome; and the deformation compensation function is introduced to correct the test deviation, so that higher test accuracy is achieved. The method is reliable, rational in analysis, notable in effect and environmental-protection, and can effectively improve the innerstress computation accuracy of the aluminum alloy thick plate and the like, thereby being suitable for industrialization and scientific research.

Description

The modification method of aluminium alloy thick plate quenching stress field distribution

Technical field

The invention discloses the modification method of aluminium alloy thick plate quenching stress field distribution, belong to the industrial test technical field.

Background technology

Aluminium alloy thick plate prepares process can be made and produce bigger unrelieved stress in the plate, thereby influence the following process manufacturing, so need test and subdue the stress distribution situation of slab and strength level.The test of its cut deal internal residual stress is the difficult point in the domestic Engineering Testing Technique always; There is not the ripe method of a cover to solve so far; Therefore how effectively to solve the accurate description problem of slab internal stress, preparation can produce the good society economic benefit to slab has very important effect.

The X-ray surface stress method of testing that has the international standard authentication at present in the stress test method; Because what diffraction was measured is the microstrain in the zone; Can directly ask for this regional stress intensity, method is directly reliable, has suitable precision; Therefore its degree of accuracy is very high as a result, generally be used in surface treatment processing, demarcating stress intensity.Efficient, high precision characteristics that nonlinear finite element method and software have; Prepare process finite element model (FEM) with this slab of setting up; Unique advantage aspect Analysis of Thick internal stress differentiation succession rule and characteristic distributions can be used as the main means of answering force inspecting and prediction.But; Some necessary hypothesis and too desirable parameter and boundary condition in the model; Make the numeric value analysis result calculated often differ bigger with actual; Particularly reflect material because of the plastic yield unevenness aspect that attribute difference showed, model also is difficult to accomplish, so result calculated and reality deviation to some extent.Layer is cut method (LRM) and is had advantage aspect the actual internal stress of description; Its precision can be able to guarantee through the scientific experiment method in theory; But experimental technique itself has certain limitation condition, so also there is difference in result of calculation and actual state, need revise equally and perfect.

To sum up; Every kind of stress test algorithm all has certain applicable situation and limitation, accurately obtain the distribution in the slab planted agent field of force, relies on a kind of method to accomplish the law illustration of slab stress field is described with distributing; Therefore; Rationally the characteristics of every kind of method of utilization make up effective slab stress field experiment test modification method, are the important technical of research slab stress distribution.

Summary of the invention

The object of the invention is to overcome the deficiency of prior art; Propose a kind of workable; Aluminium alloy thick plate quenching finite element model and slab layer are cut the experimental method computation model revise, thereby accurately obtain the modification method of the aluminium alloy thick plate quenching stress field distribution of slab stress field.

The modification method of aluminium alloy thick plate quenching stress of the present invention field distribution comprises the steps:

The first step: the data acquisition and processing (DAP) of the aluminium alloy thick plate hardened face coefficient of heat transfer

The aluminium alloy thick plate quenching is obtained the time-domain curve of lowering the temperature that quenches; Time-domain curve is dispersed with time step Δ τ; The initial temperature of supposing slab between each discrete regions is inside and outside consistent, resolves and finds the solution by formula (1), formula (2) coefficient of heat transfer hi, obtains the coefficient of heat transfer hi between the order discrete regions.

$\frac{\mathrm{\φ}}{{\mathrm{\φ}}_0}=\frac{{T-T}_f}{T_0-T_f}=\underset{n=1}{\overset{\∞}{\mathrm{\Σ}}}\frac{{2\sin \mathrm{\λ}}_n\mathrm{\δ}}{{\mathrm{\λ}}_n\mathrm{\δ}+1/2\sin 2{\mathrm{\λ}}_n\mathrm{\δ}}{e}^{-{\left({\mathrm{\λ}}_n\mathrm{\δ}\right)}^2(\mathrm{a\τ}/{\mathrm{\δ}}^2)}\·\·\·\·\·\·\left(1\right)$

$\cot {\mathrm{\λ}}_n\mathrm{\δ}=\frac{{\mathrm{\λ}}_n\mathrm{\δ}}{h_i\mathrm{\δ}/{\mathrm{\λ}}_0}\·\·\·\·\·\·\left(2\right)$

Theoretical according to boiling heat transfer, high temperature section and low-temperature zone are less to surperficial effect of heat transfer, and sensitive temperature section (to aluminium alloy at 100 ℃-200 ℃) heat exchanging coefficient has the greatest impact, therefore, and can be according to heat exchange sensitivity and experience to h _iIt is average to carry out sectionally weighting by formula (3), obtains the function of the total coefficient of heat transfer h (T) of slab hardened face:

$h\left(T\right)=\underset{x=0}{\overset{i}{\mathrm{\Σ}}}{h}_x{\mathrm{\κ}}_x......\left(3\right)$

In formula (1), (2), (3):

T _f, T ₀, T, θ-be respectively temperature and excess enthalpy temperature in environment temperature, plate initial temperature, the plate;

A, k, h, τ, i-are respectively coefficient of heat emission, pyroconductivity, the coefficient of heat transfer, cool time, discrete number;

λ _n, δ-be respectively transcendental equation preceding n level separate, 1/2nd thicknesss of slab;

κ-weighting coefficient is got 0.05-0.3;

Second step: carry out slab quenching simulation modeling, obtain the theoretical stress field

Utilization MSC.Marc or ANSYS nonlinear Finite meta software carry out simulation modeling;

A, quenching temperature field simulation modeling:

(a) modeling physical dimension, the hot attribute of thing and starting condition should be consistent with experiment slab physical dimension, the hot attribute of thing and temperature condition of heat treatment;

(b) the surface film thermal conductance result who the first step is calculated imports in the models for temperature field, as slab quenching temperature field modeling boundary condition;

(c) because the slab regular shape by symmetric relation, is got 1/2nd length and carried out modeling, there is not constraint;

(d) at the sampling module of the middle part of model setting with the experimental sample consistent size, " life and death unit " technology is used in sampling;

(e) model all adopts 6 body 3D of 8 nodes solid element, and the thermal field cell type is No. 43.

According to the quenching temperature field model, obtain slab quenching temperature field;

B, quenching stress field simulation modeling:

(a) modeling physical dimension, the hot attribute of thing, grid dividing method, unit number, sampling method are consistent with the temperature field modeling;

(b) with the quenching temperature field as quenching stress field starting condition;

(c) because the slab regular shape by symmetric relation, is got 1/2nd length and carried out modeling and simulating, constraint condition is: three coordinate direction displacements of constraint on the elemental area of the slab plane of symmetry;

(d) model all adopts 6 body 3D of 8 nodes solid element, and the field of force is No. 7;

According to the finite element numerical computation rule, satisfy the condition of convergence by result of calculation, iterations, time step, marginal error parameter are set, calculate and obtain theoretical temperatures field and theoretical stress field result;

The 3rd step: quenching slab sample is carried out surface stress demarcate

The slab sample is heated to solid solubility temperature T ₀, quenching in the insulation back, carries out surface stress by the X-ray diffraction technology and demarcate, and the surface stress value of demarcating is carried out data statistics, obtains slab specimen surface stress mean value;

The 4th step: revise quenching realistic model thermal boundary condition, obtain the Reference Stress field

The 3rd surface stress mean value that obtains of step and second is gone on foot the surperficial theoretical stress value that obtains compares; According to comparison result adjustment coefficient of heat transfer function h (T); Make surperficial theoretical stress value and surface stress mean deviation less than ± 10MPa; Obtain adjusted coefficient of heat transfer function h (T), and then revise quenching realistic model thermal boundary condition, obtain the Reference Stress field;

The 5th step: measure slab experiment stress field by the layer method of cutting

The quenching slab sample of the 3rd step of milling successively gained obtains strain value successively, according to the strain-stress relation in the Elasticity, calculates slab plane X, Y both direction stress value c successively _x(t), c _y(t):

Use formula (4), (5) that these stress value matches are obtained slab stress field function again:

${\mathrm{\σ}}_x\left(z\right)=\underset{i=0}{\overset{n}{\mathrm{\Σ}}}{A}_i{P}_i.....\left(4\right)$

${\mathrm{\σ}}_y\left(z\right)=\underset{j=0}{\overset{m}{\mathrm{\Σ}}}{B}_j{P}_j.....\left(5\right)$

Wherein $P_i={\left(\frac{t}{z}\right)}^i,$ $P_j={\left(\frac{t}{z}\right)}^j$

T-divest layer thickness; The stress at c (z)-thickness of slab z place; Polynomial expression constant coefficient Ai undetermined, Bi; Z is the slab degree of depth; N, m are the polynomial expression order;

The 6th step: the theory deformation amount comparison that the 5th step actual measurement deflection is corresponding with the 4th step Reference Stress field, the deflection error numerical value that obtains dispersing fits to the stress error function (6) that changes with layer depth to each discrete deflection error numerical value:

$f\left(z\right)=\underset{i=0}{\overset{n}{\mathrm{\Σ}}}{K}_i{z}^i---\left(6\right)$

In the 5th step of stress error function (6) substitution gained experiment stress field function (4), (5), obtain revised stress field function (7), (8);

${\mathrm{\σ}}_x\left(z\right)=f\left(z\right)\·\underset{i=0}{\overset{n}{\mathrm{\Σ}}}{A}_i{P}_i---\left(7\right)$

${\mathrm{\σ}}_y\left(z\right)=f\left(z\right)\·\underset{j=0}{\overset{m}{\mathrm{\Σ}}}{B}_j{P}_j---\left(8\right).$

The modification method of aluminium alloy thick plate quenching stress of the present invention field distribution in the 4th step, when surperficial theoretical stress value during greater than surface stress mean value, amplifies coefficient of heat transfer function h (T), and the stress value deviation that makes both is less than ± 10MPa;

When surperficial theoretical stress value during less than surface stress mean value, dwindle coefficient of heat transfer function h (T), the stress value deviation that makes both is less than ± 10MPa; Obtain Reference Stress field under the simulated conditions with this.

The modification method of aluminium alloy thick plate quenching stress of the present invention field distribution, the thickness of said aluminium alloy thick plate is greater than 10mm.

The modification method of aluminium alloy thick plate quenching stress of the present invention field distribution, embedding thermopair obtains the time-domain curve of lowering the temperature that quenches at the aluminium alloy thick plate center.

The present invention is owing to adopt said method; At first calculate and obtain slab, comprise that mainly boundary condition-surface film thermal conductance is found the solution, the X-ray diffraction surface stress is demarcated (XRD) and nonlinear finite element modeling (FEM) technology realizes with reference to the planted agent field of force through finite element modeling.Then, layer is cut experimental technique and carry out error analysis and correction.Obtain layer through finite element model calculating and cut inflection curves; And the test strain curve that it and layer are cut experimental technique (LRM) compared; Carry out error analysis, assessment, in the true strain reduction process that stress relief causes, proposed successively to reduce the mathematical compensation method of deformation error with polynomial fitting f (z); Original layer is cut algorithm correction, realized accurately obtaining quenching slab stress field target.

The present invention adopts the combined stress test and error analysis technology of surface stress demarcation-finite element simulation-mechanical test experiment; Not only solved the accuracy problem of finite element modeling emulation; Mainly realized tradition stratum is cut the correction of method computation model, improved layer and cut the accuracy of testing result of calculation.

Advantage of the present invention is: but 1, the present invention's each temperature range that the coefficient calculations method under the original constant temperature is introduced into discretize resolves and find the solution; Under the enough little situation of discrete time; Can it be assumed to be temperature constant state, thereby can obtain the whole temperature-fall period coefficient of heat transfer.

2, tradition stratum is cut computing method and revise, new correction model is calculated on the accuracy in the slab stress field obtained good effect.

3, the demarcation of X-ray surface stress slab stress field simulation modeling and experiment test have been applied to neatly; The Stress Field Distribution that distinct methods is obtained has science comparison property; Realized surface stress-internal stress, Theoretical Calculation and experimental calculation are organically combined, guaranteed the reliability of this modification method.

4, this modification method applicability is good, and is all available for the aluminium alloy thick plate of structural Stress Field Distribution characteristics, as long as slab thickness is identical with heat-treat condition, just can obtain Stress Field Distribution accurately and quickly.

Description of drawings

The aluminum alloy rolled slab quenching of the 7075T6 temperature lowering curve that accompanying drawing 1 obtains for the embodiment of the invention 1.

The aluminum alloy rolled slab quenching of the 7075T6 heat transfer coefficients distribution curve that accompanying drawing 2 obtains for the embodiment of the invention 1.

The X-ray surface measured stress that accompanying drawing 3 (a) obtains for the embodiment of the invention 1 distributes.

The revised simulated stress of the boundary condition that accompanying drawing 3 (b) obtains for the embodiment of the invention 1 distributes.

Slab internal stress emulation-layer that accompanying drawing 4 obtains for the embodiment of the invention 1 is cut experimental result.

Accompanying drawing 5 is cut process sample deformation synoptic diagram for the embodiment of the invention 1 actual layer.

Accompanying drawing 6 is the match of 30mm slab emulation-experiment stress-deviation in the embodiment of the invention 1.

Accompanying drawing 7 is cut the contrast of method result of calculation for 30mm slab internal stress emulation-modification level in the embodiment of the invention 1.

Accompanying drawing 8 is cut the contrast of method result of calculation for 40mm slab internal stress emulation-modification level in the embodiment of the invention 2.

Embodiment

Below in conjunction with concrete instance and accompanying drawing the present invention is specified.

To carry out the stress field test after the aluminum alloy rolled slab quenching of 7075T6 is example, slab size 1200 * 220 * 30 or 40 (mm), and test sample size 160 * 160 * 30 or 40 (mm), quenching technical is 480 ℃ of solid solution+15 ℃ water-baths.

Embodiment 1

The sample thickness of slab is 30mm.

(1) the slab quenching coefficient of heat transfer confirms.

Test slab heart portion quenching temperature lowering curve is as shown in Figure 1; Curve is dispersed with time step Δ τ=10s, consistent inside and outside hypothesis slab initial temperature between each discrete regions, carry out preliminary coefficient of heat transfer h parsing and find the solution; Because surface film thermal conductance is very big during the strong convection heat exchange; Therefore, for ease of data processing, can be earlier with each interval h that calculates _iCarry out weighted mean after rounding again, each key point data list 1.Fig. 2 is the temperature variant distribution curve of quenching slab surface film thermal conductance.

Table 1

Temperature (degree)	?480	460	400	210	120	70	50
								h(w/m 2℃)	0	8000	10000	14000	21000	13000	8000
Time (second)	0	10	20	30	40	50	60

According to the table 1 slab quenching cooling data and the coefficient of heat transfer, can get:

h(T)=8000×0.1+10000×0.2+14000×0.3+21000×0.3+13000×0.05+8000×0.05=14350w/m ²℃。

(2) utilize MSC.Marc nonlinear Finite meta software to carry out slab quenching simulation modeling and calculating, obtain slab quenching stress field.Modeling rule:

(a) quenching temperature field: the model boundary condition is the heat exchange of outside surface cell surface, forms three-dimensional Transient Heat Transfer pattern, and starting condition is a slab quenching heat treatment condition, and geometric model is chosen along long 1/2nd symmetries of plate, does not have constraint.

(b) stress field model: the displacement of three coordinate directions of constraint on the slab symmetrical cell face, quenching thermal field aftertreatment result imports the stress field computation model as starting condition.The setting of geometric model, grid dividing are consistent with the temperature field.

(c) owing to the slab regular shape, by symmetric relation, desirable 1/2nd length are carried out modeling and simulating, and model all adopts 6 body 3D of 8 nodes solid element, and the thermal field cell type is No. 43, and the field of force is No. 7;

(d) same slab model meshes division methods is the same with the unit number, and sampling module 160 * 160 * 30 (mm) is set, and " life and death unit " technology is used in sampling;

(e) physical parameter table 1, data all can be obtained by experiment, in model, are provided with.Wherein elastic modulus E, heat-conduction coefficient λ, hot yield limit σ _s, specific heat c such as table 2.

The basic thing thermal parameter of table 2

(3) surface stress is demarcated

The X-ray diffraction appearance that utilization meets international standard carries out the surface stress test to this quenching slab central area, and the slab surface stress distribution characteristics after the acquisition data statistics comprises that intensity and state are like table.Shown in Fig. 3 (a).

The slab center surface is carried out the X-ray surface stress demarcates, carry out obtaining surface stress calibration value=-194.5 ± 13.38MPa after data statistics is handled,

(4) revise quenching realistic model thermal boundary condition, obtain the Reference Stress field

The stress field surface stress calculated value that realistic model calculates is about=-158MPa.With the about 40MPa of X-ray surface stress calibration value deviation, greater than permissible variation ± 10MPa,, need to amplify the coefficient of heat transfer because theoretical value is less than normal, recomputate slab quenching temperature field and stress field.Theoretical according to boiling heat transfer, 100 ℃-200 ℃ these temperature section heat exchanging coefficients have the greatest impact, and therefore, can amplify the weighted value of this section.

h(T)=8000×0.05+10000×0.1+14000×0.35+21000×0.4+13000×0.05+8000×0.05=15750w/m ²℃。

The quenching coefficient of heat transfer has determined the size of slab stress field, therefore with reference to (two) step, the revised coefficient of heat transfer is reset slab quenching temperature field boundary condition, recomputates slab quenching calculated stress field.Obtain the simulation calculation surface stress=-200MPa, with X-ray surface stress calibration value comparison, meet the requirement of deviation less than ± 10MPa, therefore, confirm that the simulation calculation Stress Field Distribution meets the actual stress field distribution.Be depicted as correction front and back The model calculation like Fig. 3 (b).

(5) the layer method of cutting calculated the test of slab stress

Through carrying out the measurement of a strain stress behind the milling depth 2mm successively, anti-mean intensity and the state that pushes away this layer internal stress, thus match obtains the stress distribution of slab along layer depth.

${\mathrm{\ϵ}}_x\left(t\right)=\frac{2}{E}\underset{i=0}{\overset{n}{\mathrm{\Σ}}}\frac{(A_i-\mathrm{\υ}{B}_i)(i-1)}{(i+1)(i+2)}{(1-\mathrm{\α})}^i;$ ${\mathrm{\ϵ}}_y\left(t\right)=\frac{2}{E}\underset{i=0}{\overset{n}{\mathrm{\Σ}}}\frac{(B_i-\mathrm{\υ}{A}_i)(i-1)}{(i+1)(i+2)}{(1-\mathrm{\α})}^i$

normalization thickness wherein, dimensionless.E is an elastic modulus, and v is a Poisson ratio.Z is that corresponding layer depth polynomial expression undetermined coefficient A, the number 2i of B, layer cut number of times m, Matrix Solving formula:

{ε _m}=[c _m1，c _m2...c _mi]·{A _i-vB _i}

{A _i-vB _i}=[[C] ^T[C]] ^-1[C] ^T{ε _m}

[C] matrix is m * i, and the strain data of simultaneous vertical and horizontal can be obtained following formula A _i, B _iThereby, try to achieve the Stress Field Distribution curve with experimental technique.

${\mathrm{\σ}}_x\left(z\right)=\underset{i=0}{\overset{n}{\mathrm{\Σ}}}{A}_i{P}_i;$ ${\mathrm{\σ}}_y\left(z\right)=\underset{j=0}{\overset{m}{\mathrm{\Σ}}}{B}_j{P}_j$

The slab quenching stress be parabolic distribution be 180MPa ~-240MPa, with reference to simulated stress field distribution 120MPa ~-there is deviation in 200MPa, reflects Stress Field Distribution simulation calculation result and experimental calculation result's contrast like Fig. 4.

(6) Theoretical Calculation and experimental calculation stress field error and additional

Because discharging small deformation, internal stress cuts the limited flexural deformation stack that method experiment clamping produces with layer; Thereby cause testing the data and the theoretical value that record and have deviation; It is higher that theoretical analysis can be surveyed the deformation strain result, and this deviation has directly caused the bigger than normal of experiment stress field calculating.

Actual layer is cut the milling situation situation then as shown in Figure 5 in the experiment, and it is thin two to occur, the result of thick middle, experiment distortion cause of increased that Here it is.The deflection theory that the 30mm slab adopts the square layer of 160mm to cut sample is seen table 3 with the actual measurement deflection.Therefore can be according to different slab differences with the difference on the depth profile; Along method approximate the obtain correction-compensation function f (z) of its degree of depth through data fitting; And be incorporated into original layer and cut reduction successively in the computation model, to subdue because the error that clamping deformation brings.

Table 3 sample milling process emulation-experiment deformation deviation

Layer is cut number of times	1	2	3	4	5	6	7
								Emulation u ε	408	795	1126	1460	1762	2040	2281
160 actual measurement u ε	408	840	1250	1630	1930	2240	2490
								The deviation ratio	13%	2%	7%	6%	5%	4%	3%

With the square sample of 160mm is example, because the strain-stress resilient relationship of distortion directly reflects stress-deviation by the strain deviation of being out of shape, therefore can revise original experimental calculation data through the mode that successively compensates stress-deviation.According to deviation profile characteristics such as Fig. 6, make up a correction polynomial function and come compensate:

$f\left(z\right)=\underset{i=0}{\overset{3}{\mathrm{\Σ}}}{K}_i{z}^i$

K ₀＝-0.0063；K ₁＝0.22；；K ₂＝-2.5；K ₃＝29.5

K is for revising the polynomial function constant coefficient, revised computation model:

${\mathrm{\σ}}_x\left(z\right)=f\left(z\right)\·\underset{i=0}{\overset{n}{\mathrm{\Σ}}}{A}_i{P}_i;$ ${\mathrm{\σ}}_y\left(z\right)=f\left(z\right)\·\underset{j=0}{\overset{m}{\mathrm{\Σ}}}{B}_j{P}_j$

Can try to achieve revised layer cutting method calculated stress distributes.Revised Stress Field Distribution and aforementioned simulation result are tending towards coincideing like Fig. 7, and this is distributed in the surface and X-ray surface stress calibration result is very identical, explains that this test modification method can effectively improve quenching slab stress field test accuracy.

Embodiment 2

The sample thickness of slab is 40mm.

(1) the slab quenching coefficient of heat transfer confirms.

This tests modification method to the 40mm slab, same heat-treat condition, and its cooling velocity, the coefficient of heat transfer and 30mm slab quench very approaching, like table 4.Curve is dispersed with time step Δ τ=10s, consistent inside and outside hypothesis slab initial temperature between each discrete regions, carry out preliminary coefficient of heat transfer h parsing and find the solution.

Table 4

Temperature (degree)	?480	460	410	230	130	75	50
								h(w/m 2℃)	0	8000	10000	14000	21000	13000	8000
Time (second)	0	10	20	30	40	50	60

According to the table 4 slab quenching cooling data and the coefficient of heat transfer, get the weighting coefficient identical with the 30mm slab, can get:

h(T)=8000×0.1+10000×0.2+14000×0.3+21000×0.3+13000×0.05+8000×0.05=14350w/m ²℃。

(2) utilize MSC.Marc nonlinear Finite meta software to carry out slab quenching simulation modeling and calculating, obtain slab quenching stress field.Modeling rule is identical with the 30mm slab, the difference of physical dimension when unique difference is modeling.

(3) surface stress is demarcated

The X-ray diffraction appearance that utilization meets international standard carries out the surface stress test to this quenching slab central area, and obtaining surface stress intensity is-157 ± 8MPa.

(4) revise quenching realistic model thermal boundary condition, obtain the Reference Stress field

The stress field surface stress calculated value that realistic model calculates is about=-120MPa.With the about 37MPa of X-ray surface stress calibration value deviation, greater than permissible variation ± 10MPa,, need to amplify the coefficient of heat transfer because theoretical value is less than normal, recomputate slab quenching temperature field and stress field.Theoretical according to boiling heat transfer, 100 ℃-200 ℃ these temperature section heat exchanging coefficients have the greatest impact, and therefore, can amplify the weighted value of this section.

h(T)=8000×0.05+10000×0.1+14000×0.35+21000×0.4+13000×0.05+8000×0.05=15750w/m ²℃。

The quenching coefficient of heat transfer has determined the size of slab stress field, therefore with reference to (two) step, the revised coefficient of heat transfer is reset slab quenching temperature field boundary condition, recomputates slab quenching calculated stress field.Obtain the simulation calculation surface stress and be about-160MPa,, meet the requirement of deviation, therefore, confirm that the simulation calculation Stress Field Distribution meets the actual stress field distribution less than ± 10MPa with X-ray surface stress calibration value comparison.

(5) the layer method of cutting calculated the test of slab stress

To cut experimental technique the same with 30mm slab layer; The 40mm slab quenching stress that obtains be parabolic distribution be 140MPa ~-200MPa; With with reference to simulated stress field distribution 98MPa ~-there is deviation in 160MPa, reflects Stress Field Distribution simulation calculation result and experimental calculation result's contrast like Fig. 4.

(6) Theoretical Calculation and experimental calculation stress field error and additional

Owing to cause the sample elasticity deformation reason the same, as long as the sample heavy in section consistent size of experiment, clamping force is consistent; Can think that the error size that they cause is suitable basically, like this, can directly use existing correction-compensation function; Omit experiment link, improve testing efficiency.For the 40mm slab; Adopting the square layer of 160mm to cut sample experimentizes when calculating; Can directly use 30mm slab stress test correction-compensation function f (z) in first example, and be incorporated into layer and cut in the computation model reduction successively, to subdue because the error that clamping deformation brings.

$f\left(z\right)=\underset{i=0}{\overset{3}{\mathrm{\Σ}}}{K}_i{z}^i$

K ₀＝-0.0063；K ₁＝0.22；；K ₂＝-2.5；K ₃＝29.5

K is for revising the polynomial function constant coefficient, revised computation model:

${\mathrm{\σ}}_x\left(z\right)=f\left(z\right)\·\underset{i=0}{\overset{n}{\mathrm{\Σ}}}{A}_i{P}_i;$ ${\mathrm{\σ}}_y\left(z\right)=f\left(z\right)\·\underset{j=0}{\overset{m}{\mathrm{\Σ}}}{B}_j{P}_j$

Can try to achieve revised layer cutting method calculated stress and distribute, after the 40mm quenching slab stress modifier of acquisition be distributed as 95MPa ~-165MPa.Revised Stress Field Distribution and aforementioned simulation result are tending towards coincideing like Fig. 8, and this is distributed in the surface and X-ray surface stress calibration result is very identical.

Claims (4)

1. the modification method of aluminium alloy thick plate quenching stress field distribution comprises the steps:

The first step: the data acquisition and processing (DAP) of the aluminium alloy thick plate hardened face coefficient of heat transfer

The aluminium alloy thick plate quenching is obtained the time-domain curve of lowering the temperature that quenches; Time-domain curve is dispersed with time step Δ τ; The initial temperature of supposing slab between each discrete regions is inside and outside consistent, resolves and finds the solution by formula (1), formula (2) coefficient of heat transfer hi, obtains the coefficient of heat transfer hi between the order discrete regions.

$\frac{\mathrm{\φ}}{{\mathrm{\φ}}_0}=\frac{{T-T}_f}{T_0-T_f}=\underset{n=1}{\overset{\∞}{\mathrm{\Σ}}}\frac{{2\sin \mathrm{\λ}}_n\mathrm{\δ}}{{\mathrm{\λ}}_n\mathrm{\δ}+1/2\sin 2{\mathrm{\λ}}_n\mathrm{\δ}}{e}^{-{\left({\mathrm{\λ}}_n\mathrm{\δ}\right)}^2(\mathrm{a\τ}/{\mathrm{\δ}}^2)}\·\·\·\·\·\·\left(1\right)$

$\cot {\mathrm{\λ}}_n\mathrm{\δ}=\frac{{\mathrm{\λ}}_n\mathrm{\δ}}{h_i\mathrm{\δ}/{\mathrm{\λ}}_0}\·\·\·\·\·\·\left(2\right)$

To h _iIt is average to carry out sectionally weighting by formula (3), obtains the function of the total coefficient of heat transfer h (T) of slab hardened face:

$h\left(T\right)=\underset{x=0}{\overset{i}{\mathrm{\Σ}}}{h}_x{\mathrm{\κ}}_x......\left(3\right)$

In formula (1), (2), (3):

T _f, T ₀, T, θ-be respectively temperature and excess enthalpy temperature in environment temperature, plate initial temperature, the plate;

A, k, h, τ, i-are respectively coefficient of heat emission, pyroconductivity, the coefficient of heat transfer, cool time, discrete number;

λ _n, δ-be respectively transcendental equation preceding n level separate, 1/2nd thicknesss of slab;

κ-expression weighting coefficient;

Second step: carry out slab quenching simulation modeling, obtain the theoretical stress field

Utilization MSC.Marc or ANSYS nonlinear Finite meta software carry out simulation modeling;

A, quenching temperature field simulation modeling:

(a) modeling physical dimension, the hot attribute of thing and starting condition should be consistent with experiment slab physical dimension, the hot attribute of thing and temperature condition of heat treatment;

(b) the surface film thermal conductance result who the first step is calculated imports in the models for temperature field, as slab quenching temperature field modeling boundary condition;

(c) because the slab regular shape by symmetric relation, is got 1/2nd length and carried out modeling, there is not constraint;

(d) at the sampling module of the middle part of model setting with the experimental sample consistent size, " life and death unit " technology is used in sampling;

(e) model all adopts 6 body 3D of 8 nodes solid element, and the thermal field cell type is No. 43.

According to the quenching temperature field model, obtain slab quenching temperature field;

B, quenching stress field simulation modeling:

(a) modeling physical dimension, the hot attribute of thing, grid dividing method, unit number, sampling method are consistent with the temperature field modeling;

(b) with the quenching temperature field as quenching stress field starting condition;

(c) because the slab regular shape by symmetric relation, is got 1/2nd length and carried out modeling and simulating, constraint condition is: three coordinate direction displacements of constraint on the elemental area of the slab plane of symmetry;

(d)Model all adopts 6 body 3D of 8 nodes solid element, and the field of force is No. 7;

According to the finite element numerical computation rule, satisfy the condition of convergence by result of calculation, iterations, time step, marginal error parameter are set, calculate and obtain theoretical temperatures field and theoretical stress field result;

The 3rd step: quenching slab sample is carried out surface stress demarcate

The slab sample is heated to solid solubility temperature T ₀, quenching in the insulation back, carries out surface stress by the X-ray diffraction technology and demarcate, and the surface stress value of demarcating is carried out data statistics, obtains slab specimen surface stress mean value;

The 4th step: revise quenching realistic model thermal boundary condition, obtain the Reference Stress field

The 3rd surface stress mean value that obtains of step and second is gone on foot the surperficial theoretical stress value that obtains compares; According to comparison result adjustment coefficient of heat transfer function h (T); Make surperficial theoretical stress value and surface stress mean deviation less than ± 10MPa; Obtain adjusted coefficient of heat transfer function h (T), and then revise quenching realistic model thermal boundary condition, obtain the Reference Stress field;

The 5th step: measure slab experiment stress field by the layer method of cutting

The quenching slab sample of the 3rd step of milling successively gained obtains strain value successively, according to the strain-stress relation in the Elasticity, calculates slab plane X, Y both direction stress value c successively _x(t), c _y(t):

Use formula (4), (5) that these stress value matches are obtained slab stress field function again:

${\mathrm{\σ}}_x\left(z\right)=\underset{i=0}{\overset{n}{\mathrm{\Σ}}}{A}_i{P}_i.....\left(4\right)$

${\mathrm{\σ}}_y\left(z\right)=\underset{j=0}{\overset{m}{\mathrm{\Σ}}}{B}_j{P}_j.....\left(5\right)$

Wherein $P_i={\left(\frac{t}{z}\right)}^i,$ $P_j={\left(\frac{t}{z}\right)}^j$

T-divest layer thickness; The stress at c (z)-thickness of slab z place; Polynomial expression constant coefficient Ai undetermined, Bi; Z is the slab degree of depth; N, m are the polynomial expression order;

The 6th step: the theory deformation amount comparison that the 5th step actual measurement deflection is corresponding with the 4th step Reference Stress field, the deflection error numerical value that obtains dispersing fits to the stress error function (6) that changes with layer depth to each discrete deflection error numerical value:

$f\left(z\right)=\underset{i=0}{\overset{n}{\mathrm{\Σ}}}{K}_i{z}^i---\left(6\right)$

In the 5th step of stress error function (6) substitution gained experiment stress field function (4), (5), obtain revised stress field function (7), (8);

${\mathrm{\σ}}_x\left(z\right)=f\left(z\right)\·\underset{i=0}{\overset{n}{\mathrm{\Σ}}}{A}_i{P}_i---\left(7\right)$

${\mathrm{\σ}}_y\left(z\right)=f\left(z\right)\·\underset{j=0}{\overset{m}{\mathrm{\Σ}}}{B}_j{P}_j---\left(8\right).$

2. according to the modification method of the said aluminium alloy thick plate quenching stress of claim 1 field distribution, it is characterized in that: in the 4th step,

When surperficial theoretical stress value during greater than surface stress mean value, amplify coefficient of heat transfer function h (T), the stress value deviation that makes both is less than ± 10MPa;

When surperficial theoretical stress value during less than surface stress mean value, dwindle coefficient of heat transfer function h (T), the stress value deviation that makes both is less than ± 10MPa; Obtain Reference Stress field under the simulated conditions with this.

3. according to the modification method of the said aluminium alloy thick plate quenching stress of claim 1 field distribution, it is characterized in that: the thickness of said aluminium alloy thick plate is greater than 10mm.

4. according to the modification method of the said aluminium alloy thick plate quenching stress of claim 1 field distribution, it is characterized in that: embedding thermopair obtains the time-domain curve of lowering the temperature that quenches at the aluminium alloy thick plate center.

Images