Method for predicting stress-strain curve under other strain rates based on existing data

Abstract

The invention discloses a method for predicting stress-strain curves under other strain rates based on existing data. The method comprises the following steps: a, converting an existing engineering stress-engineering strain curve into a real stress-real strain curve; b, intercepting a real stress-real strain curve of the yield-tensile section; c, scaling the real stress value of the real stress-real strain curve of the yield-tensile section; d, calculating a scaling factor corresponding to the strain rate; and e, scaling the true stress-true strain curve of the yield-tensile section corresponding to the static tensile test to obtain the true stress-true strain curve of the yield-tensile section corresponding to the strain rate. According to the method, stress-strain curves at other rates are accurately predicted on the basis of an existing high-speed tensile test, the defect that the test is carried out again due to the lack of test data of a certain strain rate is avoided, and therefore the detection cost and the detection time of the performance of the metal material are saved.

Description

Technical Field

[0001] The invention relates to a method for predicting stress-strain curves at other rates based on existing data, and belongs to the field of metal material performance detection. Background Art

[0002] As my country's requirements for energy conservation and emission reduction continue to increase, the automotive industry's demand for lightweighting is also increasing, and the impact of lightweighting on automobile safety is attracting more and more attention. As we all know, the deformation rate of automotive structural parts during the forming process is 0.1S -1 ~1S -1 , and the deformation rate of a car in an actual collision is as high as 100S -1 ~500S -1 The accuracy of vehicle collision safety prediction requires the support of accurate material dynamic constitutive model. Accurately obtaining stress-strain curves at different strain rates and establishing material cards that conform to the performance of the material itself are important guarantees for evaluating the collision safety of the entire vehicle.

[0003] In addition to the whole vehicle collision safety simulation, typical automotive assemblies, such as seats, B-pillars, etc., also need to undergo separate collision safety assessments, and these assessments require different strain rates. Different OEMs, especially different joint ventures, require different strain rates according to their own platform characteristics. High-speed stretching is an expensive and time-consuming test. After we complete a set of high-speed stretching tests, we often need another one or two strain rate stretching data. If we do a high-speed stretching test again, it will not only require a lot of test funds, but also take a long time. Therefore, it is very necessary to find a method to accurately predict the stress-strain curve at other strain rates based on the existing stretching data. Summary of the invention

[0004] The purpose of the present invention is to provide a method for predicting stress-strain curves at other strain rates based on existing data to address the shortcomings of the prior art, so as to accurately predict the stress-strain curves at unknown strain rates and save the detection cost and time of metal material properties.

[0005] To achieve the above object, the present invention adopts the following technical solution:

[0006] A method for predicting stress-strain curves at other strain rates based on existing data, the method comprising the following steps:

[0007] a. converting the engineering stress-engineering strain curves at different strain rates obtained from the existing multiple high-speed tensile tests including the quasi-static tensile test into true stress-true strain curves;

[0008] b. Determine the yield strength and tensile strength corresponding to each strain rate, intercept the true stress-true strain curve between the two, and obtain the true stress-true strain curve of the yield-tensile section;

[0009] c. The true stress value of the true stress-true strain curve of the yield-tensile section at each strain rate except the quasi-static tensile test is scaled using a certain scaling factor, so that the deviation between the area under the scaled true stress-true strain curve of the yield-tensile section and the area under the true stress-true strain curve obtained from the quasi-static tensile test does not exceed the set deviation limit;

[0010] d. Draw a relationship diagram between the scaling factor and the strain rate, and perform linear / nonlinear fitting to obtain a functional relationship between the scaling factor and the strain rate, and then use the functional relationship between the scaling factor and the strain rate to calculate the scaling factor corresponding to the required strain rate;

[0011] e. Use the scaling factor corresponding to the desired strain rate to scale the true stress value of the true stress-true strain curve of the yield-tensile section corresponding to the quasi-static tensile test to obtain the true stress-true strain curve of the yield-tensile section corresponding to the desired strain rate.

[0012] The above method for predicting stress-strain curves at other strain rates based on existing data, and the specific method for converting the engineering stress-engineering strain curve into the true stress-true strain curve are:

[0013] Calculate true strain and true stress from engineering stress and engineering strain data:

[0014] ε 真实 =ln(1+ε 工程 )

[0015] σ 真实 =σ 工程 (1+ε 工程 )

[0016] Where ε 真实 is the true strain, ε 工程 is the engineering strain, σ 真实 is the true stress, σ 工程 is the engineering stress, using the calculated true strain ε 真实 and true stress σ 真实 Plot the true stress-true strain curve.

[0017] In the above method for predicting stress-strain curves at other strain rates based on existing data, when the true stress value of the true stress-true strain curve of the yield-tensile section at each strain rate except the quasi-static tensile test is scaled using a certain scaling factor, the deviation limit between the area under the scaled true stress-true strain curve of the yield-tensile section and the area under the true stress-true strain curve obtained from the quasi-static tensile test is set to 0.5%.

[0018] In the above method for predicting stress-strain curves at other strain rates based on existing data, the number of existing multiple high-speed tensile tests including quasi-static tensile tests is no less than 5 times.

[0019] The above method of predicting stress-strain curves at other strain rates based on existing data, when performing linear / nonlinear fitting on the relationship between scaling factor and strain rate, the fitting variance R 2 Not less than 0.98, if the fitting variance R 2 If the refitting variance R 2 If it is still less than 0.98, repeat the test.

[0020] The present invention can use a unique algorithm to accurately predict stress-strain curves at other rates on the basis of existing high-speed tensile tests, avoiding the disadvantage of re-conducting high-speed tensile tests due to lack of test data for one or two strain rates, thereby saving the detection cost and time of metal material properties. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0022] Figure 1 is the engineering stress-engineering strain curve obtained from the experiment of the embodiment;

[0023] Figure 2 is the true stress-true strain curve;

[0024] Figure 3 It is the true stress-true strain curve of the yield-tensile section;

[0025] Figure 4 is a plot of scaling factor versus true strain;

[0026] Figure 5 0.001S -1 , 1S -1 , 10S -1 Comparison chart of the fitting curve and the original data;

[0027] Figure 6 100S-1 、200S -1 、1000S -1 Comparison chart of the fitting curve and the original data;

[0028] Figure 7 It is 500S -1 The true stress-true strain curve of the yield-tensile section;

[0029] Figure 8 It is a flow chart of the present invention.

[0030] The symbols in this paper are as follows: 真实 is the true strain, ε 工程 is the engineering strain, σ 真实 is the true stress, σ 工程 is the engineering stress, s represents the area under the curve, is the scaling factor, is the strain rate, R 2 is the fitted variance. DETAILED DESCRIPTION

[0031] In view of the drawbacks of the prior art, the present invention provides a method for predicting stress-strain curves at other strain rates based on existing data, the method comprising the following steps:

[0032] a. Convert the engineering stress-engineering strain curves at different strain rates obtained from the existing multiple high-speed tensile tests including the quasi-static tensile test into true stress-true strain curves:

[0033] Calculate the true strain and true stress based on the engineering stress and engineering strain data obtained from the existing high-speed tensile test:

[0034] ε 真实 =ln(1+ε 工程 )

[0035] σ 真实 =σ 工程 (1+ε 工程 )

[0036] Where ε 真实 is the true strain, ε 工程 is the engineering strain, σ 真实 is the true stress, σ 工程 is the engineering stress, using the calculated true strain ε 真实 and true stress σ 真实 Plot the true stress-true strain curve;

[0037] b. Determine the yield strength and tensile strength corresponding to each strain rate, intercept the true stress-true strain curve between the two, and obtain the true stress-true strain curve of the yield-tensile section;

[0038] c. The true stress value of the true stress-true strain curve of the yield-tensile section at each strain rate except the quasi-static tensile test is scaled using a certain scaling factor. The scaling method is to multiply the true stress value (i.e., the ordinate value) of the true stress-true strain curve of the yield-tensile section by a scaling factor so that the deviation between the area under the scaled true stress-true strain curve of the yield-tensile section (i.e., the area enclosed by the true stress-true strain curve and the true strain axis) and the area under the true stress-true strain curve obtained by the quasi-static tensile test does not exceed a set deviation limit, which is set to 0.5% (the difference is greater than zero), and then all true stress-true strain curves (including the true stress-true strain curve obtained by the quasi-static tensile test and the scaled true stress-true strain curves of each yield-tensile section) are plotted in the same coordinate system;

[0039] d. Draw a relationship diagram between the scaling factor and the strain rate, and perform linear / nonlinear fitting to obtain a functional relationship between the scaling factor and the strain rate, and then use the functional relationship between the scaling factor and the strain rate to calculate the scaling factor corresponding to the required strain rate;

[0040] e. Use the scaling factor corresponding to the desired strain rate to scale the true stress value of the true stress-true strain curve of the yield-tensile section corresponding to the quasi-static tensile test, that is, multiply the true stress value (i.e., the ordinate value) of the true stress-true strain curve of the yield-tensile section corresponding to the quasi-static tensile test by the scaling factor corresponding to the desired strain rate to obtain the true stress-true strain curve of the yield-tensile section corresponding to the desired strain rate.

[0041] The present invention requires that the number of existing high-speed tensile tests is not less than 5 times, which must include a quasi-static tensile test.

[0042] When performing a linear / nonlinear fit on a plot of scaling factor versus strain rate, the fitted variance R 2 It should not be less than 0.98. If the fitting variance is less than 0.98, it means that the distribution of points is relatively discrete and it is necessary to fit with other sets of data at the same strain rate. 2 If it is still less than 0.98, retesting is required.

[0043] Example

[0044] This example is made of HC550 / 980DPD+Z-1.0mm, with a speed of 500S -1The existing high-speed tensile test data is based on the stress-strain curve prediction method. The existing high-speed tensile test data is based on the GB / T228.1-2010 "Metallic Material Tensile Test Method" on a quasi-static tensile testing machine with a rate of 0.001S. -1 Quasi-static tensile test. According to ISO26203-2:2011 "Metallic materials - High strain rate tensile test - Part 2: Hydraulic servo and other test systems", the test is carried out on a high-speed testing machine at a rate of 1S -1 , 10S -1 、100S -1 、200S -1 、1000S -1 The engineering stress-engineering strain curve obtained by dynamic tensile test is as follows: Figure 1 The specific prediction steps are as follows:

[0045] a. According to ε 真实 =ln(1+ε 工程 ), σ 真实 =σ 工程 (1+ε 工程 ), (where ε 真实 is the true strain, ε 工程 is the engineering strain, σ 真实 is the true stress, σ 工程 is engineering stress) Figure 1 The engineering stress-engineering strain curve is converted into a true stress-true strain curve, such as Figure 2 As shown;

[0046] b. Determine the yield strength and tensile strength corresponding to each strain rate, as shown in Table 1, and select the true stress-true strain curve of the yield-tensile section, such as Figure 3 As shown;

[0047] Table 1 Yield strength and tensile strength at various strain rates

[0048] <![CDATA[Strain rate (S -1 )]]> 0.001 1 10 100 200 1000 Yield strength (MPa) 693.43 707.61 717.08 806.54 809.33 986.56 Tensile strength(MPa) 1035.64 1090.21 1108.46 1119.16 1129.82 1169.13

[0049] c. Set the strain rate in step b to 1S -1 , 10S -1 、100S -1 、200S -1 、1000S -1 The true stress value of the true stress-true strain curve is scaled by different scaling factors to ensure that the area under the scaled curve and the strain rate are 0.001S -1 The area under the curve does not differ by more than 0.5%. 真实 represents the true strain, σ 真实 represents the true stress, s represents the area under the curve, and The strain rate is 0.001S -1 The area under the true stress-true strain curve is 179.0766, and the strain rate is 1000S -1 The area below the true stress-true strain curve after scaling is 179.8813, and the difference between the two is 0.45%, which meets the requirement of not more than 0.5%. The final results after scaling according to this method are shown in Table 2;

[0050] Table 2 Scaling factors

[0051] Strain rate 0.001 1 10 100 200 1000 Scaling Factor 1 1.01321 1.03444 1.05063 1.05808 1.10295

[0052] d. According to Table 2, the relationship curve between scaling factor and strain rate is plotted, such as Figure 4 As shown, a nonlinear fit is performed to obtain the scaling factor and strain rate The relationship is:

[0053]

[0054] Fitted variance R 2 It is 0.9943, higher than 0.98, which meets the requirements;

[0055] According to the scaling factor and strain rate 1S is calculated by the relationship -1 , 10S -1 、100S -1 、200S -1 、1000S -1 The corresponding scaling factor is based on the quasi-static 0.001S -1 The true stress-true strain curve is scaled to obtain the true stress-true strain curve of the yield-tensile section and compared with the test data, such as Figure 5 and Figure 6 As shown, the reliability of this method can be proved.

[0056] e. Using the scaling factor and strain rate The strain rate is calculated to be 500S -1 The scaling factor is 1.0618, and the quasi-static is 0.001S. -1 The true stress value of the true stress-true strain curve is scaled by 1.0618 times to obtain a strain rate of 500S -1 The true stress-true strain curve is as follows: Figure 7 shown.

[0057] Compared with the existing methods, the beneficial effects of the present invention are: on the basis of the existing high-speed tensile test, the present invention proposes a method for predicting stress-strain curves at other rates based on existing data, which can be used to accurately predict the stress-strain curve under unknown strain rates, avoiding the need to re-perform high-speed tensile tests due to the lack of one or two rates when making material cards. The present invention has the advantages of simple operation, low cost and high precision.

Claims

1. A method for predicting stress-strain curves at other strain rates based on existing data, characterized in that: The method comprises the following steps: a. converting the engineering stress-engineering strain curves at different strain rates obtained from the existing multiple high-speed tensile tests including the quasi-static tensile test into true stress-true strain curves; b. Determine the yield strength and tensile strength corresponding to each strain rate, intercept the true stress-true strain curve between the two, and obtain the true stress-true strain curve of the yield-tensile section; c. The true stress value of the true stress-true strain curve of the yield-tensile section at each strain rate except the quasi-static tensile test is scaled using a certain scaling factor, so that the deviation between the area under the scaled true stress-true strain curve of the yield-tensile section and the area under the true stress-true strain curve obtained from the quasi-static tensile test does not exceed the set deviation limit; d. Draw a relationship diagram between the scaling factor and the strain rate, and perform linear / nonlinear fitting to obtain a functional relationship between the scaling factor and the strain rate, and then use the functional relationship between the scaling factor and the strain rate to calculate the scaling factor corresponding to the required strain rate; e. Use the scaling factor corresponding to the desired strain rate to scale the true stress value of the true stress-true strain curve of the yield-tensile section corresponding to the quasi-static tensile test to obtain the true stress-true strain curve of the yield-tensile section corresponding to the desired strain rate.

2. The method of predicting stress-strain curves at other strain rates based on existing data according to claim 1, characterized in that: The specific method of converting the engineering stress-engineering strain curve into the true stress-true strain curve is: Calculate true strain and true stress from engineering stress and engineering strain data: e 真实 =ln(1+ε 工程 ) s 真实 =s 工程 (1+e 工程 ) Where ε 真实 is the true strain, ε 工程 is the engineering strain, σ 真实 is the true stress, σ 工程 is the engineering stress, using the calculated true strain ε 真实 and true stress σ 真实 Plot the true stress-true strain curve.

3. The method of predicting stress-strain curves at other strain rates based on existing data according to claim 1, characterized in that: When the true stress value of the true stress-true strain curve of the yield-tensile section at each strain rate except the quasi-static tensile test is scaled using a certain scaling factor, the deviation limit between the area under the scaled true stress-true strain curve of the yield-tensile section and the area under the true stress-true strain curve obtained from the quasi-static tensile test is set to 0.5%.

4. The method of predicting stress-strain curves at other strain rates based on existing data according to claim 1, characterized in that: The number of the existing multiple high-speed tensile tests including the quasi-static tensile test is no less than 5 times.

5. The method of predicting stress-strain curves at other strain rates based on existing data according to claim 1, characterized in that: When performing a linear / nonlinear fit on a plot of scaling factor versus strain rate, the fitted variance R 2 Not less than 0.98, if the fitting variance R 2 If the refitting variance R 2 If it is still less than 0.98, repeat the test.

Images