A design method of a welded joint biaxial tensile fatigue test piece

Abstract

The application discloses a kind of welded joint biaxial tensile fatigue test piece design method, it is related to a kind of welded joint biaxial tensile fatigue test piece.The stress level of its test area is higher and uniform, test success rate is high and result is accurate.It is designed according to the following steps: step 1, geometric size optimization;Step 2, design biaxial ratio;Step 3, design weld and horizontal direction angle α;Step 4, process welded joint biaxial tensile fatigue test piece.The test piece realizes the effect reached in biaxial tensile fatigue test: first, fatigue crack is generated in the weld section of test area, and it is helpful to set different load level to carry out biaxial tensile fatigue test, so that test result is more accurate;Biaxial ratio γ that can simulate the stress state of actual casing weld dangerous position can be determined;Weld angle α that can simulate the shear stress and normal stress distribution of actual casing weld dangerous position can be determined.

Description

Technical Field

[0001] This invention relates to a biaxial tensile fatigue test specimen for welded joints, particularly for circumferential welds of thin-walled casings, which can be used to assess the multiaxial fatigue performance of welds under complex loads through biaxial tensile fatigue testing. Background Technology

[0002] In the field of aero-engines, various typical casings are often important pressure-bearing and life-limiting components. The thin-walled shell of the casing is typically welded to the mounting edge using processes such as electron beam welding and argon arc welding. Under service conditions, the circumferential weld of the thin-walled shell of the casing often bears complex cyclic loads such as internal pressure and axial force, exhibiting a typical biaxial stress state. To evaluate the multiaxial fatigue performance of the welded structure of the casing, it is necessary to design a biaxial tensile fatigue test specimen with welds.

[0003] Currently, most domestic studies employ bending-torsional fatigue testing to investigate the multiaxial fatigue performance of welded joints. Alternatively, cruciform specimens are used to study the biaxial fatigue crack propagation behavior of materials. For example:

[0004] In their paper "Comparison of Findley and MWCM multiaxial fatigue methods using the notch stress method on welded joints" (Procedia Structural Integrity, 2021, 1), Bibbo ND, Arora V., and Pedersen MM compared the Findley criterion and the improved MWCM multiaxial fatigue methods under proportional bending-torsional multiaxial loading. The applicability of the curve method to predict the multiaxial fatigue life of welded joints.

[0005] In their paper "Fatigue Crack Growth Behavior of CP-Ti Cruciform Specimens with Mixed Mode I-II Crack under Biaxial Loading" (Materials, 2022, 3), Liu JiaYu, Bao WenJie, Zhao JiaYu, and Zhou ChangYu used cruciform specimens as the research object and proposed a new effective stress intensity factor in the presence of initial cracks to study the biaxial fatigue crack propagation behavior.

[0006] In summary, the multiaxial fatigue test specimens used by scholars both domestically and internationally are not suitable for fatigue life studies of welded joints under complex stress states, and there are no reports on the design of test specimens that can meet the multiaxial fatigue performance assessment of thin-walled shell welds. Therefore, this invention improves upon existing cross-shaped biaxial tensile fatigue test specimens and develops a design method for biaxial tensile fatigue test specimens of welded joints. Summary of the Invention

[0007] To address the above problems, this invention proposes a design method for biaxial tensile fatigue test specimens of welded joints, which has a high and uniform stress level in the test area, a high success rate, and accurate results.

[0008] The technical solution of the present invention is as follows: the test piece is cross-shaped and is formed by connecting two T-shaped thin plates by butt welding, and a straight weld 3 is formed between the two. A circular thinning area 2 is provided on both the front and back sides of the center of the test piece. The cross-shaped test piece has four clamping arms 1, and a transition groove is provided between adjacent clamping arms 1, thereby forming a clamping arm transition arc 4 between adjacent clamping arms 1.

[0009] Follow these steps to design:

[0010] Step 1: Optimize geometry;

[0011] Based on the AMO (Adaptive Multiple-Objective) iterative genetic algorithm in Ansys Workbench-Direct Optimization, the geometric dimensions of the test specimen are optimized using a centrally thinned cross-shaped biaxial specimen as the initial configuration.

[0012] Meanwhile, the maximum equivalent stress σ in the test area is used eq,max The maximum equivalent stress σ at the arc of the cross arm eq-A,max The ratio of the values, and the equivalent stress σ at the center of the test area. eq-C,max The maximum equivalent stress σ at the center thinning chamfer. eq-R,max The ratio is used as a constraint condition, and iterative calculations are performed to determine its specific geometric parameters. Verification shows that the stress level in the test area is significantly higher than in the non-test area, meaning fatigue cracks are more likely to initiate in the test area. This test specimen can provide support for evaluating the multiaxial fatigue performance of the circumferential weld of the thin-walled casing.

[0013] Specifically:

[0014] First, a geometric model of the test piece is established based on its initial configuration and simulation calculations are performed. The material parameters used are consistent with those of the actual casing. The load is set to equiaxial loading, and the end face boundary conditions are set to allow free displacement along the loading direction, with zero displacement in other directions. In the initial configuration of the test piece, constants include the clamping end thickness, clamping end width, test area thickness, and contour dimensions. Other parameters are used as design variables that need to be optimized.

[0015] The design variables are used as initial input parameters, including the test area diameter D2, the transition arc diameter D3, the transition center position (x, y), and the test area chamfer R. The equivalent stress σ at the center of the test area is also considered. eq-o Maximum equivalent stress σ in the test area eq-C,max The maximum equivalent stress σ at the chamfer of the test area eq-R,max The maximum equivalent stress σ of the transition arc of the clamping arm eq-A,max As the target dependent variable, its specific location is as follows: Figure 3 As shown; then set the optimization parameters, that is, set two expected results in Direct Optimization (Direct Optimization is a direct optimization method for the design exploration module in the Ansys Workbench toolbox): (1)(σ eq-C,max -σ eq-A,max ) / σ eq-A,max >8%; (2)K t (σ eq-R,max / σ eq-o () < 1.06. Where K t This represents the stress concentration factor of the test area; based on the expected results, the range of variation of the selected design variables, optimization objectives and constraints are set respectively, and the specific values ​​are shown in Table 1 and Table 2;

[0016] Table 1 Optimization Range of Design Variables

[0017] test area diameter <![CDATA[28≤D2≤40]]> transition arc diameter <![CDATA[22≤D3≤50]]> Transition center position (28, 28) ≤ (x, y) ≤ (35, 35) chamfering of test area 5≤R≤10

[0018] Table 2 Optimization Objectives and Constraints for the Dependent Variable

[0019] <![CDATA[σ eq-o ]]> Maximize 800 —— —— <![CDATA[σ eq-A，max ]]> Minimize 0 Value≤Upper Bound 610 <![CDATA[σ eq-R，max ]]> Maximize 800 Value ≥ Lower Bound 662 <![CDATA[σ eq-C，max ]]> Maximize 800 —— ——

[0020] Finally, the Pareto solution set was obtained through adaptive multiple objectives algorithm optimization. The stress results of the candidate samples were compared, and the optimal solution was extracted as the biaxial tensile fatigue test specimen of the welded joint for subsequent analysis.

[0021] Step 2: Design the dual-axis ratio;

[0022] It is necessary to ensure that the stress state of the test area of ​​the test piece can simulate the stress state of the actual dangerous parts of the casing. First, finite element simulation calculations were carried out on a certain type of combustion chamber casing under internal pressure and axial force loads. The principal stresses, namely σ1, σ2, and σ3, of the dangerous parts of the weld with relatively high stress levels were analyzed. The results showed that under the above loads, σ3 = 0 at the casing weld, and only σ1 and σ2 were present. σ1 is along the casing axial direction, and σ2 is along the casing circumferential direction, which is a bidirectional stress state. Therefore, the design of the simulated test piece only needs to ensure that σ1 / σ2 in the test area is consistent with the actual dangerous area of ​​the casing weld.

[0023] Secondly, based on the biaxial tensile fatigue test specimen of the welded joint determined in step 1, under the x-axis load F x Under the condition of being fixed, let's discuss the biaxial ratio γ, i.e. the y-axis load F. y / x-axis load F x The change in stress state in the test area of ​​the specimen when the loading value changes from 0.1 to 1 is shown in the figure below. Figure 4 As shown; interpolation was used to ensure that the stress state in the test area of ​​the test piece was consistent with the stress state distribution in the critical parts of the actual casing weld, and the corresponding biaxial ratio γ was finally determined; as shown Figure 5 As shown. By reasonably designing the biaxial ratio γ, the biaxial stress state of the annular weld of the thin-walled shell of the lower casing under complex loads (internal pressure and axial force) can be simulated based on the biaxial tensile fatigue test specimen of the welded joint, and its multiaxial fatigue performance can be evaluated. It has a certain degree of universality.

[0024] Step 3: Design the angle α between the weld and the horizontal direction;

[0025] It is necessary to ensure that the distribution of shear stress and normal stress on the weld cross-section is consistent with the actual critical areas of the casing weld. Firstly, a finite element analysis is conducted on a certain type of combustion chamber casing to analyze the distribution of shear stress and normal stress in the critical areas of the weld with relatively high stress levels.

[0026] Secondly, based on the biaxial tensile fatigue test specimen of the welded joint determined in step 1, under the biaxial ratio loading condition determined in step 2, the shear stress and normal stress variation law of the designed weld path when the weld angle α changes from 0° to 90° is discussed, so that the shear stress and normal stress distribution of the weld path in the test area of ​​the test specimen is consistent with the shear stress and normal stress distribution of the actual casing weld in the dangerous part, thereby determining the final weld angle α.

[0027] Step 4: Machining the biaxial tensile fatigue test specimen of the welded joint;

[0028] The biaxial tensile fatigue test specimen of the welded joint is formed by butt welding two thin plates, and different welding processes such as electron beam welding and argon arc welding can be used. The welded joint includes a base metal zone, a heat-affected zone, and a fusion zone. Based on step 2, a circular thinning zone of uniform thickness is set on both sides of the center of the welded plate to ensure a high and uniform stress level in the test area.

[0029] Secondly, a circular chamfer was used for transition at the thickness change points in the test area, with a chamfer radius greater than 5mm to prevent stress concentration at the chamfer root. By appropriately setting the biaxial loading ratio γ and weld angle α, the stress state along the weld path of the test piece was ensured to be completely consistent with that at the weld of the actual component, thus achieving the purpose of evaluating weld performance. In the actual biaxial tensile fatigue test, the weld test area of ​​this test piece is the key part for examining the fatigue performance of the welded joint.

[0030] The above methods enable the test specimen to achieve the desired effects in biaxial tensile fatigue testing: First, the test specimen designed based on the AMO optimization method can ensure that the stress level in the test area is higher than that in the non-test area and relatively uniform, thereby enabling fatigue crack initiation at the weld section in the test area and facilitating the setting of different load levels for biaxial tensile fatigue testing, resulting in more accurate test results; Second, based on the test specimen optimized in the first step, the variation trend of the stress state in the test area of ​​the test specimen with different biaxial ratios γ is discussed, and the biaxial ratio γ that can simulate the stress state of the critical parts of the actual casing weld is determined by interpolation; Third, based on the test specimen optimized in the first step, under the biaxial ratio loading condition determined in the second step, the variation trend of the shear stress and normal stress distribution in the test area of ​​the test specimen with the weld angle α is discussed, and the weld angle α that can simulate the shear stress and normal stress distribution of the critical parts of the actual casing weld is determined.

[0031] This invention provides a design approach for biaxial tensile fatigue test specimens of welded joints. Under the premise of ensuring test success, by rationally designing the specimen's geometry, biaxial loading ratio, and weld machining angles, the weld path of the specimen can be made consistent with the stress state of any component of interest. Furthermore, based on parametric modeling in Workbench preprocessing, the optimized design process is more convenient and efficient, and the verified results show higher accuracy. In summary, this design method can achieve the goal of evaluating the multiaxial fatigue performance of welds. Attached Figure Description

[0032] Figure 1 This is a three-dimensional model of a biaxial tensile fatigue test specimen for a welded joint.

[0033] Figure 2 This is a schematic diagram of a biaxial tensile fatigue test specimen for a welded joint.

[0034] 1 is the clamping arm; 2 is the circular thinning zone; 3 is the weld; 4 is the transition arc of the clamping arm.

[0035] Figure 3 This is a schematic diagram of the equivalent stress location;

[0036] Figure 4 This is a schematic diagram of the principal stress locations;

[0037] Figure 5 This is a graph showing the change in stress state in the test area of ​​the specimen as the biaxial ratio changes from 0.1 to 1;

[0038] Figure 6 This is a Mises equivalent stress distribution diagram of the welded joint after biaxial loading in a finite element simulation of a specific embodiment;

[0039] Figure 7a This is a diagram showing the principal stress results of the weld path during loading in a specific embodiment;

[0040] Figure 7b This is a diagram showing the principal stress results at the critical location of the actual casing weld.

[0041] Figure 8a This is a diagram showing the distribution of shear stress and normal stress along the weld path under biaxial loading in a specific embodiment.

[0042] Figure 8b This is a diagram showing the distribution of shear stress and normal stress at dangerous locations in the actual casing weld. Detailed Implementation

[0043] To clearly illustrate the technical features of this patent, the following detailed description is provided through specific embodiments and in conjunction with the accompanying drawings.

[0044] This invention involves joining two thin plates together by butt welding to form a straight weld seam, and then creating circular thinning zones on both sides of the welded plates at their center. A three-dimensional model of the biaxial tensile fatigue test specimen is shown below. Figure 1 As shown, the structure is described as follows: Figure 2 As shown, Figure 2 The area marked 3 is the testing region for the biaxial tensile fatigue test specimen of the welded joint. The transverse cross-sectional view of the specimen is shown below. Figure 2 As shown in Section AA, a magnified view of the test area is as follows: Figure 2As shown in DETAIL B, the enlarged view reveals a rounded chamfer transition between the test area and other parts. This is intended to ensure a relatively uniform stress level in the weld test area and prevent stress concentration. In the actual biaxial tensile fatigue test, the four clamping areas 1 of the test piece are connected to the four actuators of the biaxial fatigue testing machine. The test piece is clamped and fixed by the actuator grips, with the center line of the grips coinciding with the axes of the two axes of the test piece. The biaxial tensile load is applied to the test piece by operating the testing machine, and then the location of weld joint crack initiation and fatigue life in the test area of ​​the test piece are monitored.

[0045] To determine the feasibility of this test specimen, this example uses finite element method (FEM) software to perform biaxial tensile simulation calculations. Due to the geometric symmetry of the specimen, a quarter-model is used in the calculation. Based on the FEM calculations, the principal stresses at the critical location of the thin-walled shell weld are determined to be σ1 / σ2 = 550.8 MPa / 213.2 MPa = 2.58, and the shear stress and normal stress distribution is τ. n / σ n =9.6MPa / 550.5MPa = 0.0174, where the shear stress is negligible relative to the normal stress. By discussing the effect of the biaxial ratio γ on the stress state of the test area of ​​the specimen, it can be concluded that when γ = 0.65, the loads on the x-axis and y-axis are set to F... x =15119.24N, F y =9827.5N, the principal stress at the weld in the test area of ​​the test specimen is σ1 / σ2 = 550.8MPa / 215.6MPa = 2.56, which is consistent with the stress state of the critical part of the actual casing weld. By discussing the influence of the weld angle α on the distribution of shear stress and normal stress in the test area of ​​the test specimen, it can be found that when the weld angle is set to 90°, the distribution of shear stress and normal stress along the weld path in the test area of ​​the test specimen is τ n / σ n =0MPa / 550.8MPa=0, consistent with the actual shear stress and normal stress distribution at the critical location of the casing weld (the shear stress at the critical location of the thin-walled casing weld is negligible compared to the normal stress). Therefore, tensile loads were set on the end faces of the clamping areas in the positive x-axis and y-axis directions, with material properties consistent with the casing: E=199GPa, υ=0.3. The stress results along the 90° weld path were extracted for analysis. The Mises equivalent stress distribution cloud diagram is shown below. Figure 6 As shown in the figure, the color depth in the test area of ​​the specimen, i.e., the circular thinning zone 2, is significantly greater than that in other areas. The stress state along the 90° weld path in the test area is as follows. Figure 7a As shown. The distribution of shear stress and normal stress along the 90° weld path in the test area is as follows. Figure 8aAs shown in the figure. In summary, it is easy to see that the stress concentration areas of this test specimen meet the design objectives and requirements. Therefore, the design method for biaxial tensile fatigue test specimens of welded joints proposed in this invention is effective and feasible.

[0046] There are many specific ways to implement this invention. The above description is only a preferred embodiment of this invention. It should be noted that for those skilled in the art, several improvements can be made without departing from the principle of this invention, and these improvements should also be considered within the scope of protection of this invention.

Claims

1. A method for designing biaxial tensile fatigue test specimens for welded joints, characterized in that, Follow these steps to design: Step 1: Optimize geometry; Based on the AMO adaptive multi-objective iterative genetic algorithm in Ansys Workbench-Direct Optimization, the geometric dimensions of the test specimen are optimized using a centrally thinned cross-shaped biaxial specimen as the initial configuration. Meanwhile, the maximum equivalent stress σ in the test area is used eq,max The maximum equivalent stress σ at the arc of the cross arm eq-A,max The ratio of the values, and the equivalent stress σ at the center of the test area. eq-C,max The maximum equivalent stress σ at the center thinning chamfer. eq-R,max The ratio is used as a constraint condition, and iterative calculations are performed to determine its specific geometric parameters. It has been verified that the stress level in the test area of ​​the test piece is significantly higher than that in the non-test area, meaning that fatigue cracks are more likely to initiate in the test area. This test piece can provide support for evaluating the multiaxial fatigue performance of the annular weld seam of the thin-walled shell of the casing. Step 2: Design the dual-axis ratio; Based on the specimen geometry determined in step 1, the x-axis load F x Under the condition of being fixed, let's discuss the biaxial ratio γ, i.e. the y-axis load F. y / x-axis load F x The stress state of the test area of ​​the test specimen changes from 0.1 to 1; the interpolation method is used to make the stress state of the test area of ​​the test specimen consistent with the stress state distribution of the actual dangerous part of the casing weld, and finally the corresponding biaxial ratio γ is determined; Step 3: Design the angle α between the weld and the horizontal direction; Based on the test specimen geometry determined in step 1, under the biaxial ratio loading condition determined in step 2, the shear stress and normal stress variation law of the designed weld path when the weld angle α changes from 0° to 90° is discussed respectively, so that the shear stress and normal stress distribution of the weld path in the test area of ​​the test specimen is consistent with the shear stress and normal stress distribution of the actual casing weld in the dangerous part, thereby determining the final weld angle α. Step 4: Machining the biaxial tensile fatigue test specimen of the welded joint; The test piece is processed according to the geometric dimensions of the test piece determined in step 1, the corresponding biaxial ratio determined in step 2, and the weld seam and horizontal angle determined in step 3.

2. The design method for a biaxial tensile fatigue test specimen of a welded joint according to claim 1, characterized in that, Step 1 is as follows: First, establish the geometric model of the test piece based on its initial configuration and perform simulation calculations. The material parameters used are consistent with those of the actual casing. The load is set to equiaxial loading, and the end face boundary conditions are set to allow free displacement along the loading direction, with zero displacement in other directions. In the initial configuration of the test piece, constants include the clamping end thickness, clamping end width, test area thickness, and contour dimensions. Other parameters are used as design variables that need to be optimized. The design variables are used as initial input parameters, including the test area diameter D2, the transition arc diameter D3, the transition center position (x, y), and the test area chamfer R. The equivalent stress σ at the center of the test area is also considered. eq-o Maximum equivalent stress σ in the test area eq-C,max The maximum equivalent stress σ at the chamfer of the test area eq-R,max The maximum equivalent stress σ of the transition arc of the clamping arm eq-A,max As the target dependent variable; then set the optimization parameters, that is, set two expected results in Direct Optimization: (σ eq-C,max -σ eq-A,max ) / σ eq-A,max >8%; K t (σ eq-R,max / σ eq-o )<1.06; where K t This represents the stress concentration factor of the test area; based on the expected results, the range of variation of the selected design variables, the optimization objective, and the constraints are set respectively. The Pareto solution set was finally obtained through optimization. The stress results of the candidate samples were compared, and the optimal solution was extracted as the biaxial tensile fatigue test specimen of the welded joint for subsequent analysis.

3. The design method for a biaxial tensile fatigue test specimen of a welded joint according to claim 1, characterized in that, Step 2 is as follows: First, a finite element simulation calculation is carried out on a certain type of combustion chamber casing under the action of internal pressure and axial force load. The principal stresses of the dangerous parts of the weld with relatively high stress level, namely σ1, σ2, and σ3, are analyzed. The results show that under the above load, σ3 = 0 at the casing weld, and only σ1 and σ2 are present. σ1 is along the casing axial direction, and σ2 is along the casing circumferential direction, which is a bidirectional stress state. Therefore, the design of the simulation test piece only needs to satisfy that σ1 / σ2 in the test area is consistent with the actual dangerous area of ​​the casing weld. Secondly, based on the biaxial tensile fatigue test specimen of the welded joint determined in step 1, under the x-axis load F x Under the condition of being fixed, let's discuss the biaxial ratio γ, i.e. the y-axis load F. y / x-axis load F x The stress state of the test area of ​​the test specimen changes from 0.1 to 1; the interpolation method is used to make the stress state of the test area of ​​the test specimen consistent with the stress state distribution of the actual dangerous part of the casing weld, and finally the corresponding biaxial ratio γ is determined.

4. The design method for a biaxial tensile fatigue test specimen of a welded joint according to claim 1, characterized in that, Step 3 specifically involves: First, conducting finite element analysis on a certain type of combustion chamber casing to analyze the distribution of shear stress and normal stress in the critical areas of the weld with relatively high stress levels; Secondly, based on the biaxial tensile fatigue test specimen of the welded joint determined in step 1, under the biaxial ratio loading condition determined in step 2, the shear stress and normal stress variation law of the designed weld path when the weld angle α changes from 0° to 90° is discussed, so that the shear stress and normal stress distribution of the weld path in the test area of ​​the test specimen is consistent with the shear stress and normal stress distribution of the actual casing weld in the dangerous part, thereby determining the final weld angle α.

Images