Method and system for predicting propagation rate of fatigue crack across welding zone in arc additive remanufacturing

Abstract

The invention belongs to the technical field of computer-aided manufacturing, and relates to a method and system for predicting the propagation rate of fatigue cracks across welding zones in electric arc additive remanufacturing, and the method comprises the following steps: S1, manufacturing a sample; s2, performing in-situ SEM fatigue test on the sample; s3, obtaining a corresponding relation between the crack growth rate and the stress intensity factor range of the sample; s4, fitting to obtain a fatigue crack propagation rate model of the sample; s5, establishing a CZH sample fatigue crack growth rate prediction model; and S6, predicting the fatigue crack propagation rate of the sample across the welding zone. The system further comprises an electric hydraulic fatigue test system, a data acquisition module, a database, a fitting module and a prediction module. According to the method, by using the correction coefficient, the fatigue crack growth rate is predicted by using the fatigue crack growth rate parameter of the matrix material area.

Description

Technical Field

[0001] The present invention belongs to the technical field of computer-aided manufacturing, and in particular to the technical field of prediction of crack propagation in an arc additive manufacturing process, and in particular to a method and system for predicting fatigue crack propagation rate across a welding zone in arc additive manufacturing. Background Art

[0002] Over the past two decades, there has been an increasing demand to extend the service life of critical infrastructure and reduce the waste of high-value-added components through remanufacturing technology. The economic and environmental benefits of remanufacturing have been emphasized in various applications. WAAM provides an attractive remanufacturing method with the advantages of low tooling requirements, high deposition rate, flexible material selection and computerized operation, which greatly improves the efficiency and accuracy of repairing large components. Mild steel is widely used in critical remanufacturing applications such as pipelines in the energy and chemical industries. For components to be directly put into service after in-situ remanufacturing repair, it is usually required to meet or exceed the performance of the original components. Fatigue crack growth is one of the main failure modes of such components. Therefore, predicting the fatigue crack growth rate of WAAM remanufactured mild steel is of great significance to ensure the structural integrity of critical components and reduce the risk of fracture.

[0003] Existing studies have shown that the heat input of additive manufacturing (AM) technology and the thermal history of the building process directly affect the mechanical properties of the product. WAAM remanufacturing uses an arc as a heat source and a metal wire as a raw material to repair damaged parts in a layer-by-layer manner. During the remanufacturing process, a transition heat-affected zone (HAZ) is formed between the deposited material (i.e., the fusion zone, FZ) and the base material (BM) of the damaged part. The unique thermal history of WAAM can generate a HAZ with a different microstructure from that formed by conventional welding. The HAZ is mainly affected by one or two consecutively deposited WAAM layers. The temperature gradient and high heat input in the non-isothermal temperature cycle of the WAAM process can change the grain size and generate hard phases during the cooling process of the HAZ. Studies have shown that under high heat input and high cooling rate, in the HAZ of low-carbon dual-phase steel, the area close to the FZ is fully austenitized and subsequently transformed into martensite. Compared with the BM and FZ, the HAZ shows the greatest hardness, and tensile tests show that all specimens broke far away from the HAZ. Studies on the hardness of the HAZ of low-carbon bainitic steels have shown that the hardness increases with increasing cooling rate when the cooling rate exceeds 5°C / s. Phase transformations from bainite to austenite to martensite have also been observed in the HAZ of low-carbon low-alloy steel SA508 repaired by WAAM. In addition, non-isothermal temperature cycles can introduce a certain degree of inhomogeneity in the HAZ, such as C enrichment in austenite precipitation in low-carbon 9Ni steel due to the reheating history of multi-pass welding. The size and composition of the HAZ are greatly affected by the heat input and arc pulses of WAAM, and the low heat input and arc contraction caused by high-frequency pulses can reduce the root of the HAZ. Several studies have shown that HAZ materials of conventional welds generally exhibit lower fatigue resistance and hardness compared to the substrate; while other studies have also shown that fatigue resistance can be improved by precisely adjusting the heat input parameters. For example, a finer transformation microstructure can be obtained in the WAAM Ti-6Al-4V alloy to improve the fatigue limit by reducing the heat input and increasing the cooling rate. The study showed that by reducing the heat input, the harmful tensile stress on the surface of WAAM 316L stainless steel specimens can be reduced, thereby improving fatigue performance. However, for metallic materials, too low heat input can lead to increased porosity, thereby reducing resistance to fatigue crack growth. In addition, the tensile deformation behavior of HAZ, FZ and BM was studied using cross-welded tensile specimens made of WAAM low carbon steel S355. The results showed that the ultimate strength of FZ and HAZ was greater than that of BM, while the ductility of the former two was significantly lower than that of BM.

[0004] The above studies show that the phase transformation and microstructure generation in the heat-affected zone of WAAM mild steel can be quite complex, and the microstructure in the HAZ can vary greatly due to the thermal history of WAAM. These phenomena are caused by the high deposition rate, excessive energy input, and large temperature gradients in the WAAM process. Although many studies have focused on the effect of microstructure in the HAZ of WAAM mild steel on mechanical properties (such as hardness and ultimate strength), the understanding of fatigue behavior in the HAZ (such as crack initiation and propagation) is still insufficient, and the detailed mechanism of macroscopic fatigue behavior is inconclusive due to the lack of sufficient research. Some recent studies have investigated the fatigue crack growth behavior and its microstructural mechanism in other widely used additive manufacturing materials. Although these studies are valuable, the results and interpretations cannot be applied to mild steel due to material differences. In particular, the width of the HAZ of WAAM remanufactured parts is usually several times that of the HAZ produced by non-wire-feed additive manufacturing methods using more concentrated heat sources (such as lasers). Crack growth in the HAZ of WAAM parts accounts for a larger proportion of the overall fatigue crack growth life, which cannot be ignored from the perspective of damage tolerance analysis. A related study showed that WAAM low carbon steel outperformed conventional hot rolled steel in fatigue crack growth resistance and exhibited very slight anisotropy. The study found that the interlocking morphology of the microstructure in WAAM low carbon steel plays a key role in enhancing fatigue crack growth resistance. A comparative study of WAAM low carbon ER70S-6 showed that the fatigue crack growth rate of WAAM was greater than that of the HAZ and BM of conventional welded S355 steel. The results of the above two studies indicate that fatigue crack growth rate is closely related to microstructure, and inferring fatigue properties of BM and FZ to HAZ without understanding the microstructural mechanisms and their effects may lead to unreliable conclusions. Although there are a large number of studies discussing the metallurgical characteristics of WAAM low carbon steel, it is rare to see studies investigating the changes in fatigue crack growth behavior across the weld zone from full WAAM to substrate. In addition, understanding the microstructural mechanisms underlying the differences between different locations is equally important for manufacturing WAAM materials with desired fatigue properties. Summary of the invention

[0005] The purpose of this method is to propose a method and system that can accurately predict the fatigue crack growth rate across the weld zone based on the fatigue crack growth rate of the base material zone of the WAAM additively remanufactured component.

[0006] The method is as follows: first, an alternative simulation part of a remanufactured part is manufactured using arc additive technology, and fatigue specimens are sampled and manufactured in the base material (BM), heat affected zone (HAZ), cross-weld zone (CZH) and fully additive zone (WAAM) of the simulation part, so that cracks can initiate and propagate in the specified area. After that, fatigue crack growth tests are carried out on these specimens using an in-situ SEM fatigue testing machine, and relevant data on fatigue crack growth rate are recorded. Next, the growth rates of these regions are fitted and calibrated using a crack growth rate model based on linear elastic fracture mechanics to obtain crack growth rate parameters. Finally, a cross-weld zone correction parameter is proposed based on the test data. Using the correction parameter method, the fatigue crack growth rate of the cross-weld zone is corrected using the rate parameter of the substrate zone to obtain the cross-weld zone rate correction coefficient. On this basis, the goal of predicting the cross-weld zone fatigue crack growth rate using the fatigue crack growth rate data of conventional materials or the data given in the material manual can be achieved.

[0007] The present invention discloses a method for predicting fatigue crack growth rate in arc additive manufacturing across welding zones, comprising the steps of:

[0008] S1, Preparation of samples

[0009] The simulation part was manufactured by arc additive technology, and samples were taken from the BM and CZH areas of the simulation part, and notches were prefabricated as test specimens;

[0010] S2, in-situ SEM fatigue testing of the specimens

[0011] An electro-hydraulic fatigue test system installed in a SEM vacuum chamber was used to perform in-situ SEM fatigue tests on the BM specimen and the CZH specimen, respectively; the number of cycles N and SEM images during the fatigue test were obtained, and the observed crack length D corresponding to the number of cycles N was obtained based on the SEM images;

[0012] S3, the corresponding relationship between the crack growth rate and the stress intensity factor range of the sample is obtained

[0013] According to the number of cycles N and the corresponding crack size a recorded in S2, the corresponding relationship between the fatigue crack growth rate da / dN and the stress intensity factor range ΔK is obtained; wherein the crack size a is obtained by adding the depth of the prefabricated notch and the observed crack length.

[0014] S4, the fatigue crack growth rate model of the sample is obtained by fitting

[0015] The fatigue crack growth rate model is

[0016]

[0017] Wherein, C and m are the first parameter and the second parameter related to the material;

[0018] Using the expression (3) after taking the logarithm of both sides, according to the corresponding relationship between the fatigue crack growth rate da / dN and the stress intensity factor range △K of the BM specimen and the CZH specimen in step S3, linear regression is used for estimation to obtain the lnC and m fitting parameter values ​​of the BM specimen and the CZH specimen respectively; and then the fatigue crack growth rate model fitting average line of the BM specimen and the CZH specimen in the logarithmic coordinate system is obtained;

[0019] S5, Establishment of a prediction model for fatigue crack growth rate of CZH specimens

[0020] According to the model fitting average line of the BM specimen and the CZH specimen in the logarithmic coordinate system obtained in step S4, the intercept difference lnα of the two model fitting average lines is obtained, and the fatigue crack growth rate prediction model of the CZH specimen is determined as:

[0021]

[0022] in, is the fatigue crack growth rate of the CZH specimen predicted by the fitting parameters of the BM specimen, C BM and m BM are the first parameter and the second parameter of the BM specimen obtained in step S4, α is the correction parameter, and lnα is the intercept difference between the fitting mean lines of the fatigue crack growth rate model of the BM specimen and the CZH specimen in the logarithmic coordinate system;

[0023] S6, prediction of fatigue crack growth rate of specimens across weld zones

[0024] When the experimental parameters of the in-situ SEM fatigue test in S2 are changed, it is only necessary to perform an in-situ SEM fatigue test on the BM specimen and obtain the new C according to expression (3): BM and m BM , put into expression (4), we can directly predict the fatigue crack growth rate across the weld zone.

[0025] Preferably, the specific steps of preparing the sample in S1 are:

[0026] S11, using a component with a full-length groove as a base material, using welding wire as a filler material, and using arc additive technology to deposit the filler material in the groove in a stacked manner along the z-axis direction to obtain a simulated component;

[0027] S12, sampling from the BM area and the CZH area of ​​the simulation part to obtain BM sampling and CZH sampling;

[0028] S13, for BM and CZH samples, wire-cut electric discharge cutting was used to create a notch on one side of the sample as a pre-crack:

[0029] S14, mechanical polishing and electrochemical polishing are performed on the BM sample and the CZH sample respectively to obtain the BM sample and the CZH sample.

[0030] Preferably, said S12 also includes sampling from the HAZ area and the WAAM area; and steps S13 and S14 are also performed on the HAZ sampling and WAAM sampling.

[0031] Preferably, the S3, obtaining the corresponding relationship between the crack growth rate and the stress intensity factor range of the sample, is specifically:

[0032] According to the crack geometry, the stress intensity factor K affected by the tensile stress σ is calculated as follows

[0033]

[0034] Where a is the crack size, which is the sum of the depth of the prefabricated notch and the observed crack length; W is the width of the specimen section, and F(a / W) is the geometric correction factor function, expressed as:

[0035]

[0036] According to the number of cycles N and the corresponding crack size a recorded in S2, the fatigue crack growth rate da / dN is calculated by the adjacent two-point method, and the stress intensity factor range △K is obtained according to formula (1), ΔK = (1-R)K, where R is the load ratio, thereby obtaining the corresponding relationship between the fatigue crack growth rate da / dN and the stress intensity factor range △K.

[0037] Preferably, in said S5, in establishing a CZH specimen fatigue crack growth rate prediction model,

[0038] lnα=-1.3641

[0039] Therefore, the fatigue crack growth rate prediction model of CZH specimen is:

[0040]

[0041] in, is the fatigue crack growth rate of the CZH specimen predicted by the fitting parameters of the BM specimen, C BM and m BM are the first parameter and the second parameter of the BM sample obtained in step S4.

[0042] The present invention also discloses a fatigue crack growth rate prediction system for arc additive remanufacturing across welding zones, comprising the following parts: an electric hydraulic fatigue test system, a data acquisition module, a database, a fitting module and a prediction module, specifically:

[0043] The electro-hydraulic fatigue testing system is equipped with a SEM vacuum chamber for performing fatigue testing on the specimen;

[0044] The data acquisition module acquires the cycle number N and SEM images during the fatigue test, obtains the observed crack length D corresponding to the cycle number N according to the SEM images, and sends the acquired data to the database.

[0045] The database is used to store the experimental parameters in the electric hydraulic fatigue test system and the data collected by the data acquisition module.

[0046] The fitting module obtains the first parameter and the second parameter in the fatigue crack growth rate model and the intercept difference of the fitting moving averages of the BM specimen and the CZH specimen according to the data stored in the database, and sends them to the prediction module, specifically including: obtaining the corresponding relationship between the crack growth rate and the stress intensity factor range of the specimen according to the recorded number of cycles N and the corresponding crack size a; obtaining the first parameter and the second parameter related to the material in the fatigue crack growth rate model according to the corresponding relationship and the fatigue crack growth rate model; obtaining the intercept difference of the fitting moving averages of the BM specimen and the CZH specimen according to the fitted fatigue crack growth rate model.

[0047] The prediction module obtains a CZH sample fatigue crack growth rate prediction model based on the first parameter and the second parameter in the received BM sample fatigue crack growth rate model, as well as the intercept difference, and uses the CZH sample fatigue crack growth rate prediction model to predict the CZH sample fatigue crack growth rate.

[0048] Compared with the prior art, the present invention has the following beneficial effects:.

[0049] 1. The present invention establishes a fatigue crack growth rate prediction model for CZH specimens. By using the correction coefficient, the fatigue crack growth rate parameters of the base material zone are used to predict the fatigue crack growth rate. Further, on this basis, the goal of predicting the cross-weld zone crack growth rate using conventional material rates or data given in the material manual can be achieved.

[0050] 2. The present invention correlates the fatigue crack growth behavior of the substrate and the cross-weld zone, which helps in the safety assessment of WAAM repaired parts. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1a is a schematic diagram of the remanufacturing process of the present invention;

[0052] Figure 1b are four sampling positions of the fatigue specimen of the present invention;

[0053] Figure 1cThe fatigue specimen size diagram and physical diagram of the present invention;

[0054] Figure 2 The FCGR test data of the present invention are represented by discrete marks, and the model mean fitting moving average is represented by a solid line;

[0055] Figure 3 This is a flow chart of the method for predicting fatigue crack growth rate across welding zones in arc additive manufacturing of the present invention. DETAILED DESCRIPTION

[0056] The exemplary embodiments, features and aspects of the present invention will be described in detail below with reference to the accompanying drawings. The same reference numerals in the accompanying drawings represent elements with the same or similar functions. Although various aspects of the embodiments are shown in the accompanying drawings, the drawings are not necessarily drawn to scale unless otherwise specified.

[0057] like Figure 1a As shown, a hot-rolled Q235 mild steel block with a full-length groove (mild steel is used as an example here, but the following method can be used for metal materials) is used to represent the damaged part for WAAM remanufacturing. ER70S-3 wire with a diameter of 1.6mm is used as the WAAM filler material. The WAAM process is based on tungsten inert gas (TIG) welding technology. The shielding gas (Ar) flows coaxially with the tungsten electrode at a constant flow rate of 15L / min. The standard parameters used are an arc current of 160A, a power of 3.15kW, a welding gun travel speed of 160mm / min, and a wire feed speed of 1.5m / min. The preheating temperature is 125℃, and the interlayer temperature is less than 250℃. During the WAAM remanufacturing process, the filler material is deposited in the groove in a stacked manner along the z-axis direction, using an oscillating path strategy in the XY plane.

[0058] The sample is Figure 1b The dog-bone specimens were taken from four different locations as shown. Specimens A, C and D were taken completely from WAAM, BM and HAZ, respectively. Specimen B was taken from the cross-weld zone (CZH), including WAAM, HAZ and BM materials. The shape and size of the dog-bone specimens used for in-situ SEM (in-situ scanning electron microscopy) fatigue crack growth tests are shown in Figure 2. Figure 1c As shown in Figure 1. A notch with a depth of 0.625 mm and a width of 0.2 mm was made on one side of the specimen using wire electric discharge cutting (EDM) as a pre-crack. Figure 1c shown.

[0059] The in-situ SEM fatigue test specimens were mechanically polished using 400 mesh, 1000 mesh, 2000 mesh, and 3000 mesh SiC waterproof sandpaper in sequence. In order to remove the surface deformation layer caused by mechanical polishing, the specimens were electrochemically polished. The electrochemical polishing was performed at -30 °C using a polishing solution consisting of 7% perchloric acid and 93% ethanol. The process was performed at a voltage of 25 V and a current density of 450 mA / cm 2 , lasting about 1 minute.

[0060] In-situ SEM fatigue tests were performed using an electrohydraulic fatigue testing system installed in the SEM vacuum chamber. The maximum load was determined by calculating the initial stress intensity factor range ΔK0, initial crack size, specimen width, and specimen thickness using equation (1). In order to make ΔK as wide as possible within the linear range, the initial stress intensity factor range ΔK0 was designed to be approximately Slightly larger than the experimentally reported threshold stress intensity factor range for the material. Given the load ratio R = 0.1, this initial ΔK value corresponds to a maximum load of approximately 800-1000N. Therefore, the first two specimens BM and CZH used a maximum load of 800N. In order to speed up the testing process, a slightly larger value of 1000N was used for the next three specimens BMT, HAZ and WAAM. Fatigue tests were performed at room temperature with a loading frequency of 10Hz in a sinusoidal loading mode.

[0061] During the fatigue test, loading can be paused at any time for SEM imaging. When paused, the cycle number and SEM image are recorded. The observed crack length on the SEM image is measured using imaging software. The observed crack length is the distance between the crack tip and the notch root perpendicular to the direction of the load force. The resulting observed crack length D vs. cycle number N data is used for crack growth analysis.

[0062] Four specimens were manufactured for in-situ SEM fatigue testing. Table 1 lists some experimental parameter information.

[0063] Table 1

[0064]

[0065]

[0066] For the crack geometry of Figure 1(c), the stress intensity factor K is calculated based on the maximum tensile stress σ as follows:

[0067]

[0068] Where a is the crack size, which is obtained by adding the depth of the prefabricated notch and the observed crack length. The initial crack size is the prefabricated notch depth; W is the width of the specimen section, and F(a / W) is the geometric correction factor function, expressed as:

[0069]

[0070] Using the obtained crack length (a) and number of cycles (N) data, FCGR is calculated by the adjacent two-point method. The FCGR data results da / dN vs.ΔK are as follows Figure 2 As shown in , the horizontal axis is ΔK and the vertical axis is da / dN. Among them, da is the crack size increment, dN is the cycle increment, da / dN is the crack size increment per cycle, and ΔK is the stress intensity factor range. ΔK=(1-R)K, where R is the load ratio, set to 0.1, thus obtaining the corresponding relationship between the fatigue crack growth rate da / dN and the stress intensity factor range ΔK.

[0071] The fatigue crack growth rate model describes the relationship between the stable crack growth rate and the stress intensity factor, which is:

[0072]

[0073] Wherein, K and m are the first and second parameters related to the material;

[0074] The two parameters (lnC, m) are estimated using linear regression after taking the logarithm of both sides of expression (3). Table 2 gives the values ​​of the fatigue crack growth rate model fitting parameters lnC and m and the model residual standard deviation s 2 . Figure 2 The solid line in the figure shows the model fitting mean line, and the other markers are the original data of fatigue crack growth tests corresponding to different specimens.

[0075] Table 2

[0076]

[0077]

[0078] Figure 2 The results show that the fatigue crack growth rate (FCGR) of the cross weld zone (CZH) specimen is the lowest among all the tested specimens, indicating that crack growth is inhibited by certain microstructural features in the CZH. Therefore, the crack growth rate of the repaired component cannot be directly predicted using the rate of the base material area BM or the additive area WAAM region.

[0079] The cross-weld zone is usually a stress concentration area due to its non-uniform microstructural characteristics. Research on the fatigue growth behavior of the cross-weld zone helps to accurately assess the safety of components and prevent catastrophic failure. Fatigue crack growth rate research based on damage tolerance design methods is usually destructive and has limitations in predicting the life of service components. Therefore, this application introduces a correction factor into the classic Paris model to correlate the fatigue crack growth behavior of the substrate and the cross-weld zone, and predicts the fatigue crack growth rate of the cross-weld zone with the fatigue crack growth data of the substrate, which helps to assess the safety of WAAM repaired components.

[0080] Figure 2 The FCGR test data in show that the model fitting averages of the BM specimen and the CZH specimen have similar slopes in the logarithmic coordinate system (4.0968 and 4.1123, respectively), while the intercepts are significantly different (-22.1808 and -23.0167, respectively). Therefore, the correction coefficient α can be introduced into expression (3) to correlate the FCGR data of the two regions by the difference in the intercepts of the model fitting averages of the BM specimen and the CZH specimen. After using the correction parameters, the fatigue crack growth rate prediction model of the CZH specimen is obtained as follows:

[0081]

[0082] Among them C BM and m BM are the first and second parameters obtained after Paris mean fitting of the FCGR data of the BM specimen; is the fatigue crack growth rate of the CZH specimen predicted by the fitting parameters of the BM specimen. After taking the logarithm of both sides of expression (4), the expression of the model becomes:

[0083]

[0084] Similar to the expression (3) with logarithms on both sides, (lnα+lnC) in expression (5) is also the intercept of the Paris mean fitting line. Therefore, according to Figure 3 The difference in the intercepts of the Paris mean fitting lines of the CZH sample and the BM sample can be directly obtained as lnα = -1.3641. Therefore, expression (4) can be expressed as:

[0085]

[0086] By applying the fatigue crack growth rate prediction model of CZH specimens, the fatigue crack growth rate across the welding zone can be accurately predicted through the FCGR test data of the substrate, thereby achieving an accurate prediction of the fatigue crack growth rate of the WAAM repaired component.

[0087] The present invention also discloses a fatigue crack growth rate prediction system for arc additive remanufacturing across welding zones, comprising the following parts: an electric hydraulic fatigue test system 1, a data acquisition module 2, a database 3, a fitting module 4 and a prediction module 5, specifically:

[0088] The electro-hydraulic fatigue testing system 1 is equipped with a SEM vacuum chamber for performing fatigue testing on the specimen;

[0089] The data acquisition module 2 acquires the cycle number N and SEM images during the fatigue test, obtains the observed crack length D corresponding to the cycle number N according to the SEM images, and sends the acquired data to the database.

[0090] The database 3 is used to store the experimental parameters in the electric hydraulic fatigue test system and the data collected by the data acquisition module.

[0091] The fitting module 4 obtains the first parameter and the second parameter in the fatigue crack growth rate model, and the intercept difference of the fitting moving averages of the BM specimen and the CZH specimen according to the data stored in the database, and sends them to the prediction module, specifically including: obtaining the corresponding relationship between the crack growth rate and the stress intensity factor range of the specimen according to the recorded number of cycles N and the corresponding crack size a; obtaining the first parameter and the second parameter related to the material in the fatigue crack growth rate model according to the corresponding relationship and the fatigue crack growth rate model; obtaining the intercept difference of the fitting moving averages of the BM specimen and the CZH specimen according to the fitted fatigue crack growth rate model.

[0092] The prediction module 5 obtains the CZH sample fatigue crack growth rate prediction model according to the first parameter and the second parameter in the received BM sample fatigue crack growth rate model, as well as the intercept difference, and uses the CZH sample fatigue crack growth rate prediction model to predict the CZH sample fatigue crack growth rate.

[0093] The embodiments described above are only descriptions of the preferred implementation modes of the present invention, and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A method for predicting fatigue crack growth rate across welding zone in arc additive manufacturing, characterized by: It includes the steps of: S1, Preparation of samples The simulation part was manufactured using arc additive technology, and samples were taken from the BM and CZH areas of the simulation part, and notches were prefabricated as test specimens; S2, in-situ SEM fatigue testing of the specimens An electro-hydraulic fatigue test system installed in a SEM vacuum chamber was used to perform in-situ SEM fatigue tests on the BM specimen and the CZH specimen, respectively; the number of cycles N and SEM images during the fatigue test were obtained, and the observed crack length D corresponding to the number of cycles N was obtained based on the SEM images; S3, the corresponding relationship between the crack growth rate and the stress intensity factor range of the sample is obtained According to the number of cycles N and the corresponding crack size a recorded in S2, the corresponding relationship between the fatigue crack growth rate da / dN and the stress intensity factor range ΔK is obtained; wherein the crack size a is obtained by adding the depth of the prefabricated notch and the observed crack length. S4, the fatigue crack growth rate model of the sample is obtained by fitting The fatigue crack growth rate model is Wherein, C and m are the first parameter and the second parameter related to the material; Using the expression (3) after taking the logarithm of both sides, according to the corresponding relationship between the fatigue crack growth rate da / dN and the stress intensity factor range ΔK of the BM specimen and the CZH specimen in step S3, linear regression is used for estimation to obtain the lnC and m fitting parameter values ​​of the BM specimen and the CZH specimen respectively; and then the fatigue crack growth rate model fitting average line of the BM specimen and the CZH specimen in the logarithmic coordinate system is obtained; S5, Establishment of a prediction model for fatigue crack growth rate of CZH specimens According to the model fitting average line of the BM specimen and the CZH specimen in the logarithmic coordinate system obtained in step S4, the intercept difference lnα of the two model fitting average lines is obtained, and the fatigue crack growth rate prediction model of the CZH specimen is determined as: in, is the fatigue crack growth rate of the CZH specimen predicted by the fitting parameters of the BM specimen, C BM and m BM are the first parameter and the second parameter of the BM specimen obtained in step S4, α is the correction parameter, and lnα is the intercept difference between the fitting mean lines of the fatigue crack growth rate model of the BM specimen and the CZH specimen in the logarithmic coordinate system; S6, predict the fatigue crack growth rate of the specimen across the weld zone When the experimental parameters of the in-situ SEM fatigue test in S2 are changed, it is only necessary to perform an in-situ SEM fatigue test on the BM specimen and obtain the new C according to expression (3): BM and m BM , put into expression (4), we can directly predict the fatigue crack growth rate across the weld zone.

2. The method for predicting fatigue crack growth rate across welding zone in arc additive manufacturing according to claim 1 is characterized in that The specific steps of preparing the sample in S1 are as follows: S11, using a component with a full-length groove as a base material, using welding wire as a filler material, and using arc additive technology to deposit the filler material in the groove in a stacked manner along the z-axis direction to obtain a simulated component; S12, sampling from the BM area and the CZH area of ​​the simulation part to obtain BM sampling and CZH sampling; S13, for BM and CZH samples, wire-cut electric discharge machining was used to create a notch on one side of the sample as a pre-crack; S14, mechanical polishing and electrochemical polishing are performed on the BM sample and the CZH sample respectively to obtain the BM sample and the CZH sample.

3. The method for predicting fatigue crack growth rate across welding zone in arc additive manufacturing according to claim 2 is characterized in that The S12 also includes sampling from the HAZ area and the WAAM area; and steps S13 and S14 are also performed on the HAZ sampling and WAAM sampling.

4. The method for predicting fatigue crack growth rate across welding zone in arc additive manufacturing according to claim 1 is characterized in that S3, the corresponding relationship between the crack growth rate and the stress intensity factor range of the sample is obtained, specifically: According to the crack geometry, the stress intensity factor K affected by the tensile stress σ is calculated as follows Where a is the crack size, which is the sum of the depth of the prefabricated notch and the observed crack length; W is the width of the specimen section, and F(a / W) is the geometric correction factor function, expressed as: According to the number of cycles N and the corresponding crack size a recorded in S2, the fatigue crack growth rate da / dN is calculated by the adjacent two-point method, and the stress intensity factor range ΔK is obtained according to formula (1), ΔK = (1-R)K, where R is the load ratio, thereby obtaining the corresponding relationship between the fatigue crack growth rate da / dN and the stress intensity factor range ΔK.

5. The method for predicting fatigue crack growth rate across welding zone in arc additive manufacturing according to claim 1 is characterized in that S5, establishes a prediction model for fatigue crack growth rate of CZH specimens, lnα＝-1.3641 Therefore, the fatigue crack growth rate prediction model of CZH specimen is: in, is the fatigue crack growth rate of the CZH specimen predicted by the fitting parameters of the BM specimen, C BM and m BM are the first parameter and the second parameter of the BM sample obtained in step S4.

6. A system using the arc additive manufacturing cross-welding zone fatigue crack growth rate prediction method according to any one of claims 1 to 5, comprising the following parts: an electric hydraulic fatigue test system, a data acquisition module, a database, a fitting module and a prediction module, specifically: The electro-hydraulic fatigue testing system is equipped with a SEM vacuum chamber for performing fatigue testing on the specimen; The data acquisition module acquires the cycle number N and SEM images during the fatigue test, obtains the observed crack length D corresponding to the cycle number N according to the SEM images, and sends the acquired data to the database. The database is used to store the experimental parameters in the electric hydraulic fatigue test system and the data collected by the data acquisition module. The fitting module obtains the first parameter and the second parameter in the fatigue crack growth rate model and the intercept difference of the fitting moving averages of the BM sample and the CZH sample according to the data stored in the database, and sends them to the prediction module, specifically including: According to the recorded number of cycles N and the corresponding crack size a, the corresponding relationship between the crack growth rate and the stress intensity factor range of the specimen is obtained; According to the corresponding relationship and the fatigue crack growth rate model, the first parameter and the second parameter related to the material in the fatigue crack growth rate model are obtained; according to the fitted fatigue crack growth rate model, the intercept difference of the fitting moving average of the BM specimen and the CZH specimen is obtained. The prediction module obtains a CZH sample fatigue crack growth rate prediction model based on the first parameter and the second parameter in the received BM sample fatigue crack growth rate model, as well as the intercept difference, and uses the CZH sample fatigue crack growth rate prediction model to predict the CZH sample fatigue crack growth rate.

Images