Method and apparatus for predicting fatigue life of welded structure of vacuum electron beam welded superalloy

Abstract

The application discloses a high-temperature alloy vacuum electron beam welding structure fatigue life prediction method and device, and belongs to the technical field of aero-engines. The method comprises the following steps: fatigue test data of a target high-temperature alloy standard sample is acquired; the maximum area of an initial defect, a defect shape factor and crack source position information are extracted; a stress intensity factor range considering defect topography characteristics and a fatigue crack propagation threshold value of a material structure containing the initial defect are acquired, a damage tolerance theory formula is corrected, simulation data are acquired; fatigue test data and simulation data are fused to form an initial data set, a fatigue limit is calculated and expanded to the data set, the expanded data set is taken as input, a support vector regression method is used for training, hyperparameters are determined through Bayesian optimization, and a high-temperature alloy vacuum electron beam welding structure fatigue life prediction result is output. The method can well consider the influence of different topography defects on the fatigue performance of a welded joint.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of aero-engine structure safety, and particularly relates to a high-temperature alloy vacuum electron beam welding structure fatigue life prediction method and device. BACKGROUND

[0002] High-temperature alloys have excellent mechanical stability, oxidation resistance, corrosion resistance and other characteristics in high-temperature environments, and are widely used in different parts of aero-engine, such as turbine blades, turbine discs, combustion chamber casings and the like. For the combustion chamber casing, advanced welding processes such as laser welding, vacuum electron beam welding and brazing are generally used for welding the casing. In particular, for the vacuum electron beam welding process, the welding environment is generally under relatively vacuum conditions (vacuum degree: 10 -4 Pa-10 - 2 Pa). Under such welding conditions, compared with welding processes in air, the welding porosity defects will be much smaller in both quantity and size, and therefore, the welding quality of the combustion chamber casing structure obtained by using the vacuum electron beam welding process is generally better than that of other welding processes. Although the level of porosity welding defects is better than that of other welding processes, the GH4169 welded structure obtained by the vacuum electron beam welding process will inevitably produce welding defects such as welding micro-cracks, porosity and incomplete penetration. The introduction of these defects will lead to an increase in the dispersion of the fatigue performance of the GH4169 high-temperature alloy welded joint such as fatigue life and fatigue limit under alternating load conditions, thereby bringing new challenges to the fatigue life evaluation and fatigue reliability design of the casing welding structure.

[0003] In fact, not only the welding process, but also the additive manufacturing process, the casting process and the forging process, the materials or structures processed will inevitably have defects of different sizes and quantities inside. How to accurately predict the fatigue life of the material / structure containing initial defects is one of the most critical problems in engineering. Japanese scholar Murakami conducted fatigue performance tests on a large number of materials containing different types of initial defects, fitted the fatigue life results, fatigue limit and the maximum initial equivalent defect size of the material, and the Vickers hardness of the material, and based on the damage tolerance, obtained an empirical formula for predicting the fatigue limit of the material containing initial defects and an empirical formula for predicting the life based on the fatigue limit. The establishment of this set of methods provides an effective method for the fatigue performance evaluation of the material / structure containing initial defects in engineering. However, the method simply takes the projection area of the maximum initial defect perpendicular to the loading direction, and the maximum initial defect is taken as the core damage parameter for fatigue life evaluation of the material / structure. However, the most obvious disadvantage of the above formula is that it does not fully consider the influence of the defect geometric feature on the fatigue performance.

[0004] On the other hand, with the development of artificial intelligence technology, machine learning algorithms are increasingly widely used in the problem of fatigue performance evaluation of materials / structures, but the machine learning algorithm has a greater dependence on the quality and size of the input data. If the quality of the data set is not good enough, the material / structure fatigue performance evaluation result obtained by the machine learning method will also not be very good. For the fatigue problem of the structure in engineering practice, considering the difficulty and cost of the test, large-scale fatigue test cannot be carried out. SUMMARY

[0005] The purpose of the present application is to overcome the problems in the prior art and provide a high-temperature alloy vacuum electron beam welding structure fatigue life prediction method and device.

[0006] The high-temperature alloy vacuum electron beam welding structure fatigue life prediction method of the present application comprises obtaining fatigue test data of a target high-temperature alloy standard specimen;

[0007] The fatigue fracture of the target high-temperature alloy standard specimen is analyzed to extract the maximum area of the initial defect, the defect shape factor and the crack source position information; based on the extracted maximum area of the initial defect, the defect shape factor and the crack source position information, the stress intensity factor range considering the defect topography characteristics is obtained based on statics simulation and crack propagation simulation, the quantitative relationship between the stress intensity factor range and the defect shape factor is established, the damage tolerance theory formula is corrected, and simulation data is obtained;

[0008] The fatigue test data and the simulation data are fused to form an initial data set, the fatigue limit is calculated based on the corrected damage tolerance theory formula and expanded to the data set, and the expanded data set is used as input. Using the support vector regression method, the hyperparameters are determined by Bayesian optimization, and the high-temperature alloy vacuum electron beam welding structure fatigue life prediction result is output.

[0009] Preferably, the defect shape factor f is the ratio of the length of the short half-axis of the defect to the length of the long half-axis of the defect, and f ∈(0, 1].

[0010] Preferably, when obtaining the simulation data, first, a finite element analysis model of the target high-temperature alloy standard specimen is established, then statics simulation analysis of the target high-temperature alloy standard specimen at the peak load is performed, then the finite element analysis model output file is imported into the crack propagation model, a crack of a specific shape and size is inserted, the simulation analysis result of the stress intensity factor corresponding to the defect tip is obtained, the topography influence coefficient is determined by fitting the relationship between different shape factors and stress intensity factors, and the damage tolerance theory formula is corrected.

[0011] Preferably, in determining the data set after the expansion, the fatigue life influencing factors are subjected to Spearman rank correlation analysis first to determine the main factors influencing the fatigue life, and the determined main factors influencing the fatigue life are taken as the input data set.

[0012] Preferably, the input data set includes temperature, stress amplitude, equivalent defect size, Vickers hardness, fatigue life obtained from fatigue tests, and physical parameters obtained from simulation analysis according to fatigue test data, wherein the physical parameters include stress intensity factor range and fatigue limit.

[0013] Preferably, the corrected damage tolerance formula is:

[0014] (1)

[0015] (2)

[0016] (3)

[0017] In the formula, ∆K th is the fatigue crack propagation threshold value of the material structure containing initial defects, C is a defect position influence coefficient, for internal defects C is 1.56, for subsurface defects, C is generally 1.43, and HV is Vickers hardness, is the equivalent size of the defect, R is the stress ratio, f is the defect shape factor, z 1 is the length of the short semi-axis of the defect; z 1 is the length of the short semi-axis of the defect; Δ K def is the stress intensity factor range considering the defect topography, i.e. K def,max - K def,min , S is the influence coefficient of the stress intensity factor and the defect topography obtained by numerical simulation, C 1 and m are material constants, which can be obtained by optimization solution of fatigue test result data, a is the crack size, N is the fatigue life.

[0018] The application also discloses an aviation engine high-temperature alloy vacuum electron beam welding structure fatigue life prediction device based on the fatigue life prediction method.

[0019] The receiving module is used for receiving an input data set of a target standard sample, wherein the input data set comprises fatigue test data and physical parameters simulated and analyzed according to the fatigue test data, and the physical parameters comprise a stress intensity factor range and a fatigue limit.

[0020] The prediction module is used for predicting the fatigue life of the vacuum electron beam welded structure of the high-temperature alloy by using a support vector regression model and the input data set.

[0021] The present application further discloses an electronic device, comprising at least one processor and a memory.

[0022] The memory stores computer execution instructions.

[0023] The at least one processor executes the computer execution instructions stored in the memory, so that the at least one processor executes the high-temperature alloy vacuum electron beam welded structure fatigue life prediction method.

[0024] The present application further discloses a computer readable storage medium, wherein the computer readable storage medium stores computer execution instructions, and when a processor executes the computer execution instructions, the high-temperature alloy vacuum electron beam welded structure fatigue life prediction method is realized.

[0025] The present application further discloses a computer program product, and the computer program is executed by a processor to realize the high-temperature alloy vacuum electron beam welded structure fatigue life prediction method.

[0026] Compared with the prior art, the present application has the beneficial effects that:

[0027] The present application firstly quantitatively determines the evolution law of the notch tip stress intensity factor under different shape factor ratios based on a numerical simulation technology, and further modifies the traditional Murakami life prediction model according to the damage tolerance theory and the empirical relationship between the defect morphology and the notch tip stress intensity factor, so that the influence of different morphology defects on the fatigue performance of the GH4169 electron beam welded joint can be well considered.

[0028] The present application combines the traditional fatigue life prediction problem with the machine learning method (SVR, RF, XGBoost), and uses the physical parameters related to the GH4169 welded joint obtained by numerical simulation to expand and update the input data set of the traditional machine learning method, and establishes a new physical guided machine learning method for the GH4169 welded joint. The method enhances the interpretability of the traditional machine learning method through the expansion of physical information, and also improves the generalization ability and life prediction accuracy of the traditional machine learning method. Through the method combined with the non-destructive testing technology in engineering, the fatigue life prediction of the welded and additive manufactured parts in the aero-engine can be effectively carried out. BRIEF DESCRIPTION OF DRAWINGS

[0029] Figure 1 The GH4169 electron beam welded joint fatigue specimen drawing for the embodiment of the present application.

[0030] Figure 2 The GH4169 electron beam welded joint fatigue specimen for the embodiment of the present application.

[0031] Figure 3 The GH4169 fatigue load-life scatter plot for the embodiment of the present application.

[0032] Figure 4 The GH4169 electron beam welded joint partial fatigue fracture analysis result for the embodiment of the present application.

[0033] Figure 5 The GH4169 electron beam welded joint finite element simulation analysis result for the embodiment of the present application.

[0034] Figure 6 The GH4169 electron beam welded joint stress intensity factor simulation analysis result for the embodiment of the present application.

[0035] Figure 7 The fitting curve of the defect morphology feature influence mechanism on the notch tip stress intensity factor for the embodiment of the present application.

[0036] Figure 8 The GH4169 electron beam welded joint fatigue life sensitivity analysis result for the embodiment of the present application.

[0037] Figure 9 The data-driven and physically guided data-driven method implementation flowchart for the embodiment of the present application.

[0038] Figure 10 The GH4169 welded joint fatigue life prediction result considering the welded defect morphology feature for the embodiment of the present application.

[0039] Figure 11The GH4169 electron beam welded joint fatigue life prediction result based on the conventional machine learning method for the embodiment of the present application.

[0040] Figure 12 The GH4169 electron beam welded joint fatigue life prediction result based on the physical guided machine learning method for the embodiment of the present application. DETAILED DESCRIPTION

[0041] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions of the embodiments of the present application will be described clearly and completely below with reference to the drawings of the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the described embodiments of the present application, all other embodiments obtained by a person of ordinary skill in the art without any inventive effort fall within the protection scope of the present application.

[0042] Unless otherwise defined, the technical terms or scientific terms used herein should be understood as the common meanings thereof by those having ordinary skills in the art to which the present application belongs. The terms "first", "second" and similar terms used in the present application do not denote any order, quantity or importance, but are used to distinguish different components. The terms "include" or "contain" and similar terms mean that the elements or objects appearing before the "include" or "contain" cover the elements or objects listed after the "include" or "contain" and their equivalents, and do not exclude other elements or objects. The terms "connect" or "connected" and similar terms are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. The terms "upper", "lower", "left", "right" and the like are only used to represent relative positional relationships, and when the absolute positions of the described objects are changed, the relative positional relationships may also be changed accordingly.

[0043] The present application mainly aims at the welding structures of an aero-engine combustion chamber case, a flame tube and the like, which are served under the conditions of high temperature and alternating load, and welding defects are inevitably introduced in the welding process. Defects of different sizes, positions and morphological characteristics have completely different influences on the fatigue performance of the welding structure, and the existing life prediction methods do not fully consider the influence of defect geometric morphological characteristics on the fatigue performance.

[0044] The present application considers that the aero-engine combustion chamber case GH4169 electron beam welding structure will inevitably introduce micro-cracks, incomplete penetration, porosity and other welding defects during welding, further aiming at the phenomenon that the welding defects lead to the dispersion of the fatigue life of the GH4169 welded joint, fully considering the influence of the position and geometric appearance of the welding defects, based on the finite element simulation analysis technology and Franc 3D crack propagation analysis software, the influence of the welding defect morphology characteristics is quantitatively analyzed, the influence mechanism of the defect morphology characteristics on the defect tip stress intensity factor is determined, and a new fatigue life prediction model of the GH4169 welded joint is established. At the same time, support vector regression (SVR), random forest (RF), extreme gradient boosting (XGBoost) and other methods are used to train the fatigue data set (applied load, temperature, Vickers hardness, defect equivalent size) of the GH4169 welded joint, so as to predict the fatigue life of the GH4169 welded joint. On this basis, the physical parameters obtained in the numerical simulation process are used to establish a new data set, which is used to expand the original data set, and the expanded data set (applied load, temperature, Vickers hardness, defect equivalent size, defect area, stress intensity factor range, fatigue limit, stress intensity factor threshold) is obtained, wherein the stress intensity factor range is the value considering the defect morphology characteristics. Then, the support vector regression method and the expanded data set are used to establish a new method of physical-support vector regression algorithm combined GH4169 welded joint fatigue life prediction.

[0045] The high-temperature alloy vacuum electron beam welding structure fatigue life prediction method considering defect morphology characteristics provided by the present application mainly includes the following steps:

[0046] First step: use the vacuum electron beam welding (EBW) welding process to electron beam weld the GH4169 high-temperature alloy, based on the national standard and the aviation standard, carry out X-ray and fluorescent detection on the welded joint, and then carry out two-stage aging heat treatment after detection, to obtain the GH4169 electron beam welded joint under the standard heat treatment system. Then process standard tensile and fatigue samples through wire cutting, and the sample surface quality is consistent with the actual case parts, especially the surface quality of the weld seam part.

[0047] Second step: Based on national standards GB / T228.1-2010 "Metallic materials-tensile testing-Part 1: Method of test at room temperature", GB / T228.2-2015 "Metallic materials-tensile testing-Part 2: Method of test at elevated temperature", GB / T3075-2008 "Metallic materials-fatigue testing-axial force control method", etc., the static mechanical properties and fatigue properties of GH4169 welded joints at room temperature, high temperature (400℃, 500℃, 550℃) are tested respectively, and the fatigue test data obtained are plotted into S-N curve graph. At the same time, by using fracture analysis technology, SEM analysis is carried out on each fatigue failure fracture, and the fatigue failure mechanism and fatigue crack source of GH4169 electron beam welded joint are clarified, and the maximum initial defect area, shape, position and other information of each failure sample are counted.

[0048] Third step: finite element analysis is carried out on GH4169 electron beam welded sample, and the influence of defect morphology on fatigue fracture performance of GH4169 welded joint is clarified through numerical simulation, especially the influence of stress intensity factor of defect tip. Mainly through the combination of ABAQUS + Franc 3D, first, ABAQUS is used to carry out static simulation analysis of GH4169 welded joint at the peak load of each cyclic load, then the inp file output by ABAQUS is imported into Franc 3D crack propagation software, and a crack with specific shape and size is inserted to carry out simulation analysis of stress intensity factor of defect tip. This step is mainly aimed at defects with similar area and position, and the specific influence of defect morphology is analyzed. Specifically, the defect shape factor f , f is introduced f ∈(0,1], when f is 1, the defect is a standard circular shape. This step aims to deeply study the influence mechanism of slender-shaped defects, elliptical defects and circular defects on the tip stress intensity factor based on numerical simulation technology when the equivalent size, position and other characteristics are similar.

[0049] Fourth step: after clarifying the influence mechanism of defect morphology characteristics on the fatigue performance of GH4169 welded joint, a set of GH4169 welded joint fatigue life prediction method considering the initial geometric characteristics of defects can be established based on damage tolerance theory. The main damage parameters involved are area , , f, K max , K min ,∆ K th ,∆ K def , etc. The main formulas involved in the above parameters are as follows:

[0050] (1)

[0051] (2)

[0052] (3)

[0053] where, ∆K th is the fatigue crack growth threshold of the material structure with initial defects proposed by Murakami et al., C is the defect location influence coefficient, for internal defects C is 1.56, for subsurface defects, C is generally 1.43, and HV is Vickers hardness, R is the stress ratio, f is the defect shape factor, ∆K def is the effective stress intensity factor range considering the defect topography, i.e., K def,max - K def,min , S is the stress intensity factor and defect topography influence coefficient obtained by numerical simulation, C 1 and m are material constants, which can be obtained by optimizing the fatigue versus fatigue test results data, a is the crack size. Through the above formula, a GH4169 welded joint life prediction method considering the defect topography characteristics can be obtained based on the damage tolerance theory.

[0054] Step 5: Spearman rank correlation analysis of factors affecting the fatigue life of GH4169 welded joints. In order to clarify the factors affecting the fatigue life of GH4169 electron beam welded joints and their sensitivity, Spearman rank correlation analysis is needed to analyze the fatigue life of GH4169 welded joints and temperature T , test load σ , Vickers hardness HV, maximum initial defect area area , equivalent defect size 、 Δ K th , σ 0.1 , etc. Through Spearman rank correlation analysis, the main influencing factors of GH4169 welded joints can be preliminarily determined, which can determine the initial input data set for the next machine learning method. Spearman rank correlation analysis is a nonlinear correlation analysis, and its formula is as follows:

[0055] (4);

[0056] where y is the fatigue life of GH4169 welded joint, x i are temperature, applied load, Vickers hardness, defect area, stress intensity factor, etc.

[0057] Step 6: Use support vector regression (SVR), random forest (RF), extreme gradient boosting (XGBoost) and other methods to predict the fatigue life of GH4169 welded joints. First, an initial data set is established, and the main data includes test temperature T , test load σ , Vickers hardness HV, maximum initial defect area area , equivalent defect size , etc. GH4169 electron beam welded joints have a total of 75 fatigue test data at 400℃, 500℃ and 550℃. The training set and test set are divided in the ratio of 7:3, and the training set accounts for 70% of all data. Through the training data set, the hyperparameters of the support vector regression method are solved by Bayesian optimization, and the main hyperparameters include penalty parameter C , error tolerance range parameter ε , kernel function control parameter γ . After obtaining the hyperparameters of the SVR method using the training set, the hyperparameters are applied to the test data set to predict the fatigue life of GH4169 electron beam welded joints, and then the generalization ability of SVR, RF and XGBoost methods is verified.

[0058] Step 7: The stress intensity factor considering defect morphology, crack propagation threshold, fatigue limit and fatigue life obtained by improving the physical model obtained by numerical simulation are used as extended data set, which is expanded to the initial data set. Similarly, GH4169 electron beam welded joints have a total of 75 fatigue test data at 400℃, 500℃ and 550℃. The training set and test set are divided in the ratio of 7:3, and the training set accounts for 70% of all data. The difference from step 6 is that the expanded data set contains more data information with actual physical meaning. Through the updated physical-training data set, the hyperparameters of the support vector regression method are solved by Bayesian optimization, and the main hyperparameters include penalty parameter C , error tolerance range parameter ε , kernel function control parameter γAfter obtaining the hyperparameters of physical-driven SVR, RF, XGBoost and other methods by using the physical dataset, the hyperparameters are applied to the test dataset to predict the fatigue life of GH4169 electron beam welded joints, and then the generalization ability of the physical-guided PI-SVR, PI-RF, PI-XGBoost and other methods is verified.

[0059] First, the solid solution state of GH4169 high-temperature alloy is welded by vacuum electron beam welding technology, and the welding sample drawing and physical drawing are shown in Figure 1 、 Figure 2 According to the relevant test national standard, the high-temperature fatigue performance test of GH4169 welded joint at 400℃, 500℃ and 550℃ is carried out, and the test results are plotted as a scatter plot, as shown in Figure 3 Damage, Figure 3 It can be clearly found that at low stress level, part of the fatigue data has relatively large dispersion. Further, the fracture analysis of GH4169 welded joint fatigue failure is carried out, and part of the results are shown in Figure 4 Based on the fracture direction, it can be judged that the welding defect is the root cause of the increase of the dispersion of the fatigue life of GH4169 electron beam welded joint.

[0060] In order to further study the influence of defect morphology characteristics on the fatigue behavior of GH4169 electron beam welded joint, ABAQUS and Franc 3D software are combined to simulate and analyze GH4169 welded joints with different defect morphology characteristics, and the related results are shown in Figure 4 、 Figure 5 、 Figure 6 and Figure 7 Under the condition of the same initial area, with the change of the morphology characteristic parameter f ( z 1 / z2 ), the effective driving stress intensity factor range K def of the defect tip under each working condition is calculated by ABAQUS finite element analysis software and Franc 3D crack propagation analysis software. It can be found that the stress intensity factor of the defect tip changes greatly, therefore, it is necessary to consider the influence of defect morphology on the fatigue performance of GH4169 electron beam welded joint. Based on the damage tolerance theory and simulation analysis results, a new welding structure fatigue life prediction method considering defect morphology characteristics is established. Thus, the main factors affecting the fatigue life of GH4169 electron beam welded joint have been basically determined. In order to quantitatively determine the action mechanism between fatigue life and related influencing factors, Spearman rank correlation analysis is carried out on GH4169 welded joint and its influencing factors, and the results are shown in Figure 8The fatigue life sensitivity analysis results shown.

[0061] Finally, the fatigue life prediction of GH4169 electron beam welded joints based on traditional machine learning methods and physically guided machine learning methods, the implementation flowchart of the data-driven method of machine learning and physical guidance is as shown in Figure 9 Before that, we first use the fatigue life prediction method considering the defect topography to predict the fatigue life of GH4169 welded joints, and compare it with the test results, the results are as shown in Figure 10 At the same time, the fatigue test results of GH4169 electron beam welded joints at 400℃, 500℃, 550℃ and the physical parameters obtained by related numerical simulation analysis are established as shown in Table 1. Secondly, we use machine learning methods such as support vector regression (SVR), random forest (RF), extreme gradient boosting (XGBoost) to train, optimize the hyperparameters and finally predict the life of the test set. The relevant input data is the information in the second to sixth columns of Table 1, and the relevant prediction results are as shown in Figure 11 Finally, we expand the relevant physical information to the initial data set, and establish the physically guided machine learning method, namely PI-SVR method, which is used to expand the physical data set. The main physical data set is ∆ K def , ∆ σ 0.1 (the information in the ninth to tenth columns of Table 1), and the fatigue life of GH4169 welded joints predicted by considering the defect topography. The fatigue life prediction effect of GH4169 electron beam welded joints based on physically guided machine learning method is as shown in Figure 12 Further comparative analysis of Figure 10 , Figure 11 , Figure 12 shows that the prediction results of physically guided machine learning method are better than those of the life prediction method considering defect topography and traditional machine learning method. Among them, the physically guided random forest method (PI-RF) is better than the physically guided support vector regression method (PI-SVR) and the extreme gradient boosting method (PI-XGBoost) in terms of fatigue life prediction accuracy and model generalization ability.

[0062] Table 1 GH4169 electron beam welded joint related input data set

[0063]

[0064] While embodiments of the application have been shown and described, it is to be understood that the embodiments described are merely exemplary of the principles and application of the present application. Numerous modifications and adaptions can be effected without departing from the spirit and scope of the present application, which is not limited to the exact construction and arrangement described. It is intended, therefore, to cover all modifications and adaptions that fall within the scope of the claims and their equivalents.

Claims

1. A method for predicting the fatigue life of a vacuum electron beam welded superalloy welded structure, characterized in that, The method comprises the following steps: obtaining fatigue test data of a target superalloy standard sample; analyzing the fatigue fracture of the target superalloy standard sample to extract the maximum area of the initial defect, the defect shape factor and the crack source position information; based on the extracted maximum area of the initial defect, the defect shape factor and the crack source position information, the stress intensity factor range considering the defect topography characteristics is obtained based on statics simulation and crack propagation simulation, the quantitative relationship between the stress intensity factor range and the defect shape factor is established, the damage tolerance theory formula is corrected, and simulation data is obtained; fusing the fatigue test data and the simulation data to form an initial data set, calculating the fatigue limit based on the corrected damage tolerance theory formula and expanding the data set, taking the expanded data set as input, using the support vector regression method, determining the hyperparameters through Bayesian optimization, and outputting the superalloy vacuum electron beam welding structure fatigue life prediction result; when obtaining the simulation data, first, a finite element analysis model of the target superalloy standard sample is established, then statics simulation analysis is performed on the target superalloy standard sample, then the finite element analysis model output file is imported into the crack propagation model, a crack with a specific shape and size is inserted, the simulation analysis result of the stress intensity factor of the corresponding defect tip is obtained, the topography influence coefficient is determined by fitting the relationship between different defect shape factors and stress intensity factors, and the damage tolerance theory formula is corrected; the corrected damage tolerance formula is: （1） （2） （3） wherein, ∆K th is the fatigue crack propagation threshold value of the material structure with initial defects, C is the defect location influence coefficient, for internal defects C is taken as 1.56, for subsurface defects, C is generally taken as 1.43, and HV is the Vickers hardness, is the equivalent defect size, R is the stress ratio, f is the defect shape factor, z 1 is the defect short semi-axis length; z 2 is the defect long semi-axis length; Δ K def is the stress intensity factor range considering the defect topography, i.e., K K def,max - K def,min ), S is the stress intensity factor and defect topography influence coefficient obtained by numerical simulation, C 1 and m are material constants, which can be obtained by optimizing the fatigue test result data, a is the crack size, N is the fatigue life.

2. The fatigue life prediction method for high-temperature alloy vacuum electron beam welded structures as described in claim 1, characterized in that, when determining the expanded data set, first, Spearman rank correlation analysis is performed on the fatigue life influencing factors to determine the main factors affecting the fatigue life, and the determined main factors affecting the fatigue life are taken as the input data set.

3. The fatigue life prediction method for high-temperature alloy vacuum electron beam welded structures as described in claim 2, characterized in that, The input data set includes temperature, stress amplitude, equivalent defect size, Vickers hardness, fatigue life obtained by fatigue test, and physical parameters obtained by simulating and analyzing the fatigue test data, wherein the physical parameters include stress intensity factor range and fatigue limit.

4. A fatigue life prediction apparatus for a high-temperature alloy vacuum electron beam welding structure using the fatigue life prediction method of claim 1, characterized by The method comprises the following steps: a receiving module configured to receive an input data set of a target standard sample, wherein the input data set includes fatigue test data and physical parameters obtained by simulating and analyzing the fatigue test data, and the physical parameters include stress intensity factor range and fatigue limit; a prediction module configured to predict the superalloy vacuum electron beam welding structure fatigue life by using a support vector regression model and the input data set.

5. An electronic device, comprising: The method comprises the following steps: at least one processor and a memory; the memory stores computer execution instructions; the at least one processor executes the computer execution instructions stored in the memory, so that the at least one processor executes the superalloy vacuum electron beam welding structure fatigue life prediction method according to any one of claims 1 to 3.

6. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer execution instructions, and when the processor executes the computer execution instructions, the superalloy vacuum electron beam welding structure fatigue life prediction method according to any one of claims 1 to 3 is realized.

7. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to realize the superalloy vacuum electron beam welding structure fatigue life prediction method according to any one of claims 1 to 3.

Images