A method, medium and device for predicting the grinding fatigue life of a TC4 titanium alloy

Abstract

The present application relates to the field of fatigue life prediction, in particular to a TC4 titanium alloy grinding fatigue life prediction method, medium and equipment. Including the following steps: according to the comprehensive defect characteristic value a0 of the workpiece to be predicted and the target stress concentration factor K t ’ The fatigue crack initiation life N of the workpiece to be predicted is generated i Combined with the stress concentration of the grinding mark in the workpiece to be predicted, the internal residual stress and the different characteristics reflected by the surface and section of the grinding mark, the influence on the fatigue life is optimized and improved to obtain the fatigue crack initiation life with higher precision.

Description

Technical Field

[0001] This invention relates to the field of fatigue life prediction, and in particular to a method, medium, and equipment for predicting the fatigue life of TC4 titanium alloy grinding. Background Technology

[0002] Titanium alloys, due to their low density, corrosion resistance, and high temperature resistance, are widely used in aerospace, marine engineering, and medical fields, and are a core material for manufacturing aero-engine blades. However, titanium alloys also have high strength and low elastic modulus, making them difficult to machine. This leads to problems such as rapid tool wear, severe thermal damage, and poor surface quality during machining. Belt grinding, a flexible contact process, can reduce vibration and deformation during machining, mitigate the risk of surface burns, and improve surface quality, making it an important method for machining titanium alloy components. However, titanium alloy components (aero-engine blades) must withstand high-frequency alternating loads during service, and surface defects (burns, microcracks) can significantly reduce fatigue life. Therefore, high surface integrity of titanium alloys is extremely important for extending the service life of components and improving their reliability in the aerospace field.

[0003] Currently, grooves (grind marks) are generally considered a type of surface defect, and the fatigue life of titanium alloy samples after grinding is closely related to the characteristics of micro-scratches (defects). However, existing technologies lack methods for predicting fatigue life from grinding marks on titanium alloy surfaces. Summary of the Invention

[0004] To address one of the aforementioned technical problems, the present invention adopts the following technical solution:

[0005] According to one aspect of the present invention, a method for predicting the grinding fatigue life of TC4 titanium alloy is provided, the method comprising the following steps:

[0006] Based on the average proportion of the wear area (A), the average proportion of the wear perimeter (L), the average aspect ratio (R), and the average fractal dimension (D) of the wear surface in the workpiece to be predicted, a comprehensive defect feature value a0 of the wear marks on the workpiece surface is generated. a0 satisfies the following condition: a0 = 0.66A - 0.51L + 0.21R - 0.72D + 2.58. A reflects the quantity and size of the target wear marks on the surface to be inspected; L reflects the quantity and size of the target wear marks on the surface to be inspected, and whether the target wear mark is a convex or concave function shape, and can also reflect the superposition of target wear marks on the surface to be inspected; R reflects whether the target wear mark on the surface to be inspected is long and thin or short and wide; D reflects the complexity of the boundary shape of the target wear mark on the surface to be inspected; the target wear mark is a wear mark with a depth greater than a preset depth.

[0007] Based on the measured cross-sectional depth h and cross-sectional radius of curvature ρ0 corresponding to the wear marks on the workpiece to be predicted, the initial stress concentration factor K of the wear marks on the workpiece surface to be predicted is generated. t ;K t The following conditions must be met: ;

[0008] Based on the maximum tensile stress σ actually borne by the workpiece under the predicted operating conditions max And the residual stress Rs inside the workpiece after grinding, for K t The correction is performed to generate the target stress concentration factor K for the wear marks on the workpiece surface to be predicted. t ’ ;K t ’ The following conditions must be met: K t ’ =[(σ max +Rs)×K t ] / σ max ;

[0009] Based on a0 and K t ’ Generate the fatigue crack initiation life N of the workpiece to be predicted. i N i The following conditions must be met:

[0010] ;

[0011] Where G is the shear modulus of the workpiece to be predicted, E is the elastic modulus of the workpiece to be predicted, and σ r Let Δσ be the fatigue strength of the workpiece to be predicted when the stress ratio is r. r Let ΔK be the tensile stress amplitude of the workpiece to be predicted when the stress ratio is r. th This is the fatigue crack propagation threshold.

[0012] According to a second aspect of the present invention, a non-transitory computer-readable storage medium is provided, which stores a computer program that, when executed by a processor, implements the above-described method for predicting the grinding fatigue life of TC4 titanium alloy.

[0013] According to a third aspect of the present invention, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described method for predicting the grinding fatigue life of TC4 titanium alloy.

[0014] This invention has at least one of the following beneficial effects:

[0015] In this invention, the crack initiation model of Giga-fatigue life prediction of FV520B-I with surfaceroughness is used as the basis. Then, the model is optimized and improved by combining the stress concentration of the wear marks, the internal residual stress, and the different characteristics reflected by the surface and cross-section of the wear marks in the workpiece to be predicted, to obtain a more accurate fatigue crack initiation life.

[0016] Specifically, in this invention, the comprehensive defect feature value a0 of the surface wear marks on the workpiece to be tested is fitted using feature values ​​such as the average proportion of wear mark area A, the average proportion of wear mark perimeter L, the average aspect ratio R, and the average fractal dimension D of the wear mark. This allows for a more accurate reflection of the different characteristics of the wear marks on the surface of the workpiece.

[0017] Furthermore, the cross-sectional shape of the grinding marks affects the stress concentration on the workpiece surface, and the residual stress on the workpiece surface after grinding also affects the actual magnitude of the tensile and compressive stresses under actual operating conditions. Both stress concentration and residual stress affect the stress concentration factor, ultimately impacting the fatigue crack initiation life. Therefore, in this invention, the predicted cross-sectional depth h, cross-sectional radius of curvature ρ0, and the maximum tensile stress σ actually borne by the workpiece under operating conditions are used to predict the stress concentration factor. max The residual stress Rs inside the workpiece after grinding is used to correct the stress concentration factor of the surface wear marks on the workpiece, thereby obtaining a stress concentration factor that better matches the current workpiece condition, and further improving the accuracy of fatigue crack initiation life. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A flowchart of a method for predicting the grinding fatigue life of TC4 titanium alloy provided in an embodiment of the present invention;

[0020] Figure 2 The fatigue crack initiation life N of the workpiece to be predicted is provided in the embodiments of the present invention. i Schematic diagram of improvements;

[0021] Figure 3The fatigue crack propagation life N of the workpiece to be predicted is provided in the embodiments of the present invention. p Schematic diagram of improvements;

[0022] Figure 4 To detect the surface morphology of the ground specimen using a white light interferometer (MFT-5000). Detailed Implementation

[0023] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] As one possible embodiment of the present invention, such as Figure 1 As shown, a method for predicting the grinding fatigue life of TC4 titanium alloy is provided, the method including the following steps:

[0025] S100: Based on the average proportion of the wear area (A), the average proportion of the wear perimeter (L), the average aspect ratio (R), and the average fractal dimension (D) of the wear surface in the workpiece to be predicted, a comprehensive defect feature value a0 of the wear marks on the workpiece surface to be predicted is generated. a0 satisfies the following condition: a0 = 0.66A - 0.51L + 0.21R - 0.72D + 2.58. A reflects the quantity and size of the target wear marks on the surface to be inspected. L reflects the quantity and size of the target wear marks on the surface to be inspected, and whether the target wear marks are convex or concave functions; it can also reflect the superposition of target wear marks on the surface to be inspected. R reflects whether the shape of the target wear marks on the surface to be inspected is elongated or short and wide. D reflects the complexity of the boundary shape of the target wear marks on the surface to be inspected. Target wear marks are wear marks with a depth greater than a preset depth.

[0026] To quantitatively evaluate the impact of wear tracks on the fatigue life of titanium alloy surfaces, a comprehensive defect characteristic value a0 is generated to evaluate the wear tracks by considering the different characteristics of wear tracks on titanium alloy surfaces (i.e., A, L, R, and D).

[0027] Since a0 is similar to the roughness value in terms of dimensions, and the morphology is also a three-dimensional representation of the roughness.

[0028] Furthermore, in the Giga-fatigue life prediction of FV520B-I with surfaceroughness, based on the publicly available fatigue initiation life fundamental model using slip band dislocation theory and energy theory, Where 'a' represents the size of the initial fatigue crack, which is generally considered to be related to the size of inclusions and defects present in the material. Furthermore, the calculation method for 'a' given in this prior art is ultimately related to the surface roughness Ra, as shown in the following formula:

[0029] ;

[0030] Where 'a' is the defect size. Ra is the equivalent size of the surface defect, and Ra is the surface roughness.

[0031] Based on the above, this step uses roughness to fit the four parameters A, L, R, and D, thereby forming the calculation model of a0 in this embodiment. Meanwhile, since the four parameters A, L, R, and D in this embodiment can reflect the surface characteristics of the wear track from different aspects, they more closely reflect the defect (wear track) size characteristic value in this embodiment. Therefore, a0 is used to replace a. .

[0032] Specifically, based on the surface morphology data of TC4 titanium alloy after belt grinding collected in Table 1 below, we can use A, L, R, and D as independent variables and roughness as the dependent variable to perform multiple linear regression analysis to obtain the characteristic coefficient results. Finally, we get a0=0.66A-0.51L+0.21R-0.72D+2.58.

[0033] Table 1

[0034]

[0035] The mean A of the wear mark area ratio, the mean L of the wear mark perimeter ratio, the mean R of the wear mark aspect ratio, and the mean D of the wear mark fractal dimension are obtained according to the following method:

[0036] S110: Based on the topography information map representing the height information of the surface to be inspected, obtain the image area corresponding to each target wear mark in the topography information map. A target wear mark is a wear mark with a depth greater than a preset depth. The preset depth needs to be determined according to the actual application field. Since the definition of surface defects varies for different products, in some scenarios with lower precision requirements, a deeper wear mark is needed to be considered a target wear mark, while in some scenarios with higher precision requirements, a shallower wear mark will be considered a target wear mark. Therefore, the preset depth needs to be determined by technicians according to the actual application scenario. In this embodiment, the minimum wear mark depth that affects fatigue life can be used as the preset depth. The surface to be inspected is the surface of the workpiece to be predicted after grinding.

[0037] S110 includes:

[0038] S111: Input the topography information map into the target Yolov8 algorithm model to generate the image region corresponding to each target wear mark in the topography information map. Gaussian filtering is used in the training samples of the target Yolov8 algorithm model to remove the small dot-like wear mark regions in the topography information map.

[0039] S120: Based on the topography information map and the pixel information of the image area corresponding to each target wear mark, generate the average percentage of wear mark area and the average percentage of wear mark perimeter on the surface to be detected.

[0040] S130: Based on the pixel information of the image region corresponding to each target wear mark, generate the average aspect ratio and average fractal dimension of the wear mark on the surface to be detected.

[0041] The mean value D of the fractal dimension of the wear mark satisfies the following condition:

[0042] .

[0043] Where n is the total number of target wear marks on the surface to be inspected. 2 Let be the total number of boundary pixels in the image region corresponding to the i-th target wear mark. 1 It represents the total number of remaining pixels in the image region corresponding to the i-th target wear mark, excluding the boundary pixels.

[0044] The average percentage of the wear area A, the average percentage of the wear perimeter L, and the average aspect ratio R of the wear on the surface to be tested shall satisfy the following conditions:

[0045] . . .

[0046] Among them, S i S represents the total number of pixels in the image region corresponding to the i-th target abrasion mark. z P represents the total number of pixels in the topographic image. i P represents the total number of boundary pixels in the image region corresponding to the i-th target wear mark. z Q represents the total number of boundary pixels in the topography image. iw Q represents the total number of pixels in the width direction of the bounding rectangle of the image region corresponding to the i-th target abrasion mark. ic Let be the total number of pixels in the length direction of the bounding rectangle of the image region corresponding to the i-th target abrasion mark.

[0047] S200: Based on the measured cross-sectional depth h and cross-sectional curvature radius ρ0 corresponding to the wear mark cross-section in the workpiece to be predicted, the initial stress concentration factor K of the wear mark on the surface of the workpiece to be predicted is generated. t K t The following conditions must be met: .

[0048] On the cross-section of the workpiece after grinding (i.e., the grinding mark section or XOZ surface), stress concentration occurs in titanium alloy materials near the grinding mark location. The stress concentration effect varies depending on the depth of the grinding mark; generally, narrower and deeper grinding marks exhibit a greater stress concentration effect, while wider and shallower grinding marks show a reduced stress concentration effect. This stress concentration effect significantly influences both the initiation and propagation of fatigue cracks. Specifically, the stress concentration effect primarily affects the fatigue crack propagation threshold (ΔK) of the workpiece. th Therefore, this embodiment also introduces a stress concentration factor to measure its influence on crack initiation. This parameter is mainly related to the depth and radius of curvature of the wear mark. Therefore, this embodiment also uses the method in S200 to better reflect the stress concentration differences caused by the wear mark cross-section.

[0049] Specifically, this study introduces an initial stress concentration factor K. t To consider this notch effect, since the notch of the wear cross-section in this embodiment can be simplified to a spherical notch, K t The notch can be represented by its root radius and depth. This allows us to measure the magnitude of stress concentration effects based on the cross-sectional characteristics of the wear track.

[0050] The measured values ​​of the cross-sectional depth h and the cross-sectional radius of curvature ρ0 corresponding to the grinding marks in the workpiece to be predicted are obtained according to the following steps:

[0051] S201: Based on the topography information map representing the height information of the surface to be inspected, obtain the grayscale information of the image region corresponding to each target wear mark in the topography information map. A target wear mark is a wear mark with a depth greater than a preset depth. Different heights in the topography information map are represented by different colors. The surface to be inspected is the surface of the workpiece to be predicted after grinding.

[0052] S202: Based on the grayscale information corresponding to all target wear marks, obtain the pixel depth feature value and pixel width feature value of the wear mark on the surface to be detected. The pixel depth feature value is the mode of grayscale values ​​among all grayscale information. The pixel width feature value is the average pixel width of the image region corresponding to all target wear marks.

[0053] S203: Generate h corresponding to the wear mark section in the workpiece to be predicted based on the pixel depth feature value.

[0054] S204: Based on the pixel width feature value and h of the wear mark on the surface to be detected, generate ρ0 corresponding to the wear mark cross section in the workpiece to be predicted.

[0055] S204 specifically includes:

[0056] S214: Generate the cross-sectional width detection value w of the surface scratch to be detected based on the mapping relationship between the pixel width feature value and the actual width information.

[0057] S224: Based on h and w, generate ρ0 corresponding to the wear mark section in the workpiece to be predicted. ρ0 satisfies the following condition:

[0058] .

[0059] The methods for obtaining the average proportion of the wear area A, the average proportion of the wear perimeter L, the average aspect ratio R, and the average fractal dimension D of the wear surface corresponding to the wear mark in the workpiece, as well as the detection value h of the cross-sectional depth and the detection value ρ0 of the cross-sectional curvature radius corresponding to the wear mark in the workpiece to be predicted, can refer to the method disclosed in the patent CN119295467B, "A method, storage medium and device for detecting the wear mark quality of a ground workpiece surface".

[0060] S300: Based on the maximum tensile stress σ actually borne by the workpiece under the predicted operating conditions. max And the residual stress Rs inside the workpiece after grinding, for K t The correction is performed to generate the target stress concentration factor K for the wear marks on the workpiece surface to be predicted. t ’ K t ’ The following conditions must be met: K t ’ =[(σ max +Rs)×K t ] / σ max .

[0061] In the crack initiation stage, besides the characteristic state of the surface wear cross-section after material processing, the residual stress on the workpiece surface is also a key factor affecting its fatigue life. Residual stress already exists in the workpiece before fatigue crack initiation. During subsequent use, the stress generated by external loads and the residual stress generated after processing are considered to be superimposed; that is, the residual stress after processing can resist or reinforce the stress generated by external loading. Therefore, the actual stress borne by the workpiece may be greater than or less than the actual loaded stress.

[0062] Meanwhile, since the stress concentration factor describes the change in local stress state caused by geometric discontinuities in the material (wear marks), it essentially reflects the local stress situation. Residual stress, by altering the actual local stress state, works synergistically with the stress concentration factor to determine the stress amplitude in the critical fatigue region (wear mark region). Therefore, the stress concentration factor can serve as a bridge to measure the impact of residual stress on fatigue life. Thus, in this embodiment, residual stress is also used to measure K...t Make corrections.

[0063] Since greater tensile stress makes the workpiece more prone to fatigue damage, this embodiment uses the maximum tensile stress (σ) actually borne by the workpiece. max +Rs), and the maximum tensile stress σ actually applied. max The ratio between them can be used to further increase K. t This is corrected to more closely approximate the actual stress concentration state of the workpiece, i.e., K. t ’ .

[0064] Meanwhile, since stress concentration effect mainly affects the fatigue crack propagation threshold (ΔK) of the workpiece th The specific relationship is: ΔK th ’ =ΔK th / K t ’ ; where ΔK th ΔK is the original fatigue crack propagation threshold. th ’ This is the corrected fatigue crack propagation threshold. In this embodiment, the corrected ΔK is... th ’ Substitution back, .

[0065] In addition, considering the N disclosed in the prior art i The workpiece processing conditions corresponding to the basic model differ somewhat from those in this embodiment. Therefore, the N in the prior art also needs further clarification. i The coefficient "20" in the denominator of the formula is appropriately adjusted to "50" to further improve the accuracy of fatigue life prediction, resulting in the final value. .

[0066] S400: Based on a0 and K t ’ Generate the fatigue crack initiation life N of the workpiece to be predicted. i N i The following conditions must be met:

[0067] .

[0068] Where G is the shear modulus of the workpiece to be predicted, E is the elastic modulus of the workpiece to be predicted, and σ r Let Δσ be the fatigue strength of the workpiece to be predicted when the stress ratio is r. r Let ΔK be the tensile stress amplitude of the workpiece to be predicted when the stress ratio is r. thThis is the fatigue crack propagation threshold, also known as the original fatigue crack propagation threshold. Specifically, the workpiece to be predicted corresponds to G=46GPa, ∆K th =6.462 E=109.9GPa, σ r =513MPa, ∆σ r =525MPa.

[0069] like Figure 2 As shown, in this embodiment, by predicting the average proportion of the wear area A, the average proportion of the wear perimeter L, the average aspect ratio R, and the average fractal dimension D of the wear surface in the workpiece, a comprehensive feature that can represent the morphology of the wear surface is fitted. Furthermore, by using the measured cross-sectional depth h and the measured cross-sectional radius of curvature ρ0 corresponding to the wear cross-section in the workpiece to be predicted, the stress concentration state of the wear track can be calculated. Simultaneously, the stress concentration state is corrected using residual stress. Since the comprehensive features of the wear surface morphology, the stress concentration state, and the residual stress all affect the fatigue life of the workpiece to be predicted formed from TC4 titanium alloy in this embodiment, this embodiment, based on the above three aspects of influence, modifies the N... i By refining the basic model, a more accurate N can be obtained. i .

[0070] Furthermore, fatigue life is composed of the sum of fatigue initiation life and fatigue propagation life. In the above embodiment, the fatigue initiation life N... i The acquisition of a0 and K was explained. Furthermore, the acquisition of a0 and K... t ’ The method then includes:

[0071] S500: Generate the initial crack length a based on a0 and h. ’ 0. a ’ 0 satisfies the following condition: a ’ 0 = a0 + h.

[0072] The grinding marks left by belt grinding can alter the micro-geometry of titanium alloy surfaces to some extent, thus affecting the crack propagation rate and path. Therefore, in calculating fatigue propagation life, existing techniques, such as Giga-fatigue life prediction of FV520B-I with surface roughness, rely on publicly available fatigue propagation life models based on fracture mechanics theory. Where 'a' represents the initial fatigue crack size, which can generally be considered the initial size of the wear mark. Since the depth of the groove corresponding to the wear mark is also a type of initial crack, the initial size of the wear mark in this embodiment is related not only to the size of the wear mark on the workpiece surface but also to the depth of the wear mark in the workpiece cross-sectional direction. Therefore, in this embodiment, the two sizes are directly superimposed. For example... Figure 3 As shown, in this embodiment, a0 combines multiple feature values ​​of the wear surface, thus representing the morphological differences of the wear marks on the workpiece surface, including size differences. Therefore, in this embodiment, a is directly replaced with a0. ’ 0, a ’ 0 = a0 + h.

[0073] Furthermore, the stress concentration effect at the wear mark also affects the fatigue crack propagation life. Specifically, the greater the stress concentration effect at the wear mark, the shorter the fatigue crack propagation life. Therefore, in this embodiment, the existing N... p The computational model introduces K to represent the stress concentration effect. t ’ Since stress concentration effect is negatively correlated with fatigue crack propagation life, it is added to the denominator. Meanwhile, considering the N disclosed in the prior art... p The workpiece processing conditions corresponding to the basic model differ somewhat from those in this embodiment. Therefore, all instances of 'n' in the original formula are replaced with 'm', and the exponent of π is changed from 'n / 2' to 'm'. After these adjustments, a more accurate fatigue crack propagation life can be obtained. Specifically, the corrected N... p The calculation formula is shown in S600.

[0074] S600: According to a ’ 0 and K t ’ Generate the fatigue crack propagation life N of the workpiece to be predicted. p N p The following conditions must be met:

[0075] .

[0076] Where β1 is a geometric constant, β1 = 1.772. C and m are material property parameters of TC4 titanium alloy, C = 4.58e-12, m = 3.9.

[0077] S700: According to N i and N p Generate the fatigue life N of the workpiece to be predicted. f The following condition must be met: N f =N i +N p .

[0078] N f It can also be expressed as .

[0079] To verify the accuracy of the fatigue life prediction model proposed in this invention, the following titanium alloy belt grinding and tensile / compressive fatigue experiments were conducted. The material used in the experiments was TC4 titanium alloy, and its main mechanical properties and chemical composition are shown in Tables 2 and 3, respectively.

[0080] Table 2

[0081]

[0082] Table 3

[0083]

[0084] Before belt grinding, titanium alloy sheets were machined into standard "bone-like" fatigue specimens using wire EDM. Subsequently, the "neck" section of the specimens was ground using 80# alumina belts for a length of 30 mm. Specific process parameters are shown in Table 4. All grinding experiments were conducted on a 2MGY5540 CNC belt grinder with a rated power of 7.5 kW, a maximum machining length of 1000 mm, a machining accuracy of ±0.8 µm, and a positioning accuracy of 1 µm. The surface morphology of the ground specimens was examined using a white light interferometer (MFT-5000, Rtec Instruments, Inc. America). Figure 4 As shown in the figure, the objective lens used was a ×10 magnification lens with a resolution of 1024×1024 and a field of view of 1.11 mm × 0.89 mm. All experiments were conducted at room temperature (temperature range of 18°C ​​to 30°C, relative humidity range of 45% to 70%).

[0085] Table 4

[0086]

[0087] Surface defects and hardened layers may be introduced into the sides of the workpiece after wire EDM. To reduce their impact on fatigue life test results, the sides of the test piece were successively polished with 400, 800, 1000, and 2000 grit sandpaper until sufficiently smooth (Ra < 0.2 µm). Simultaneously, the workpiece was ultrasonically cleaned to thoroughly remove any residual cutting fluid, grinding dust, and other contaminants, ensuring a clean surface and avoiding interference from external factors in the fatigue test results. After treatment, the dimensions of each standard specimen, as well as the minimum width and thickness of the "neckback" region, were measured using vernier calipers, and the corresponding load was calculated.

[0088] This study employed an axial / torsion testing system (MTS-809) for tensile-compressive fatigue experiments. The axial load capacity was + / -100 kN, and the measurement accuracy was 0.5 grade. Based on the fatigue properties of titanium alloys, the maximum stress was set to 450 MPa, the load input waveform was a sine wave with a frequency of 10 Hz, and the stress ratio was 0. Other experimental parameters remained consistent with those described previously. Specific fatigue test parameter settings are shown in Table 5.

[0089] To reduce the impact of randomness in fatigue testing on fatigue life assessment, this study conducted three sets of repeated experiments on fatigue specimens under the same working condition, and took the average value as the fatigue life value of the workpiece.

[0090] Table 5

[0091]

[0092] Substituting the mechanical property parameters of TC4 titanium alloy (Table 5) into the various calculation formulas provided in this invention, the key parameters for fatigue life prediction are obtained as shown in Table 6, including fatigue crack initiation life (N). i ), fatigue crack propagation life (N) p ), predicting total fatigue life (N) f ) and actual fatigue life N ’ f The results are shown in Table 7. The table shows that the fatigue crack initiation life is much longer than the fatigue crack propagation life, indicating that fatigue crack initiation plays a dominant role in fatigue failure, which is consistent with the fatigue development patterns in existing technologies. Furthermore, the fatigue life range is found to be between 30,000 and 60,000 cycles, which is also largely consistent with current practical results.

[0093] Table 6

[0094]

[0095] Table 7

[0096]

[0097] To verify the accuracy of the fatigue life prediction model proposed in this study, experimental and theoretical fatigue life calculations were compared. Calculations of the above data showed that the maximum error was less than 33.3%, and the minimum error was greater than 9.1%. The error between the theoretically calculated fatigue life and the experimentally obtained fatigue life was approximately 20%, indicating that the theoretical model has high prediction accuracy. This is because the model comprehensively considers the influence of surface morphology and residual stress.

[0098] To comprehensively evaluate the performance of the model proposed in this study, the performance of the model proposed in this application was compared and analyzed with that of other existing models of various types. The results are shown in Table 8.

[0099] The first type is the traditional numerical prediction model based on theories such as fracture mechanics. Most of these models are based on theoretical and simulation calculations to predict fatigue life, which has good interpretability and can well reflect the physical meaning between various parameters and fatigue life. However, these models are often for specific materials or parameters, and the generalization ability of the models is relatively weak. In addition, since many parameters are idealized in the calculation process, the prediction accuracy is also low. Specifically, the prediction results of this type of model in this invention are from existing literature (hereinafter referred to as literature [1]) Wang, JL, Zhang, YL, Sun, QC, Liu, SJ, Shi, BW, Lu, HT, 2016. Giga-fatigue lifeprediction of FV520B-I with surfaceroughness. Mater. Des. 89, 1028–1034.

[0100] The second type is the fatigue life intelligent prediction model represented by "machine learning" and "deep learning". This type of model is trained on experimental data. The data is relatively real and the source data already contains errors. The prediction accuracy of fatigue life is generally high. However, the interpretation of this type of prediction method is poor because the algorithm itself is a "black box" model. Specifically, the prediction results of the neural network type model in this invention are from existing literature (hereinafter referred to as literature [2]) Yu, XR, Zhang, GF, Jin, HH, Song, AX, 2023. A data driven model forestimating the fatigue life of 7075-T651 aluminum alloy based on the updatedBP model. J. Mater. Res. Technol. 24, 1252–1263.

[0101] The prediction results of the machine learning type model in this invention are derived from existing literature (hereinafter referred to as literature [3]) Liao,XX, Li, YD, Qiang, B., Wu, J., Yao, CR, Wei, X., 2022. An improved crack growth model of corrosion fatigue for steel inartificial seawater. Int.J. Fatigue 160, 106882.

[0102] The third type is a hybrid-driven prediction model that combines the advantages of the two models mentioned above. This type of model has good generalization ability and its accuracy is significantly improved compared to the traditional numerical prediction model. This is also a current research hotspot. However, since this type of model is not yet mature, its stability is relatively poor and its accuracy is not very satisfactory. Specifically, the prediction results of the hybrid-driven prediction model in this invention are from existing literature (hereinafter referred to as literature [4]) Wang, QY, Bathias, C., Kawagoishi, N., Chen, Q., 2002. Effect of inclusion on subsurface crack initiation and gigacycle fatiguestrength. Int. J. Fatigue24, 1269–1274.

[0103] Table 8

[0104]

[0105] Residual stress and disordered micro-grinds generated during titanium alloy belt grinding are the main areas for fatigue crack initiation. The disorder of the grinding marks and the coupling effect of morphology / residual stress result in low accuracy in predicting workpiece fatigue life. Therefore, this invention proposes a fatigue life prediction method. Combining surface / section characteristics of the surface morphology and residual stress, the YOLOv8 algorithm is introduced. Based on slip band dislocation theory, energy theory, and fracture mechanics theory, fatigue crack initiation and propagation life prediction models for titanium alloy belt grinding are established. Fatigue tensile and compressive tests are conducted, and the results show that the prediction accuracy of this model is relatively low (average error of 21.2%), but it has higher prediction accuracy compared to existing hybrid drive models.

[0106] Furthermore, although the steps of the method in this disclosure are described in a specific order in the accompanying drawings, this does not require or imply that the steps must be performed in that specific order, or that all the steps shown must be performed to achieve the desired result. Additional or alternative steps may be omitted, multiple steps may be combined into one step, and / or a step may be broken down into multiple steps.

[0107] From the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, mobile terminal, or network device, etc.) to execute the methods according to the embodiments of this disclosure.

[0108] In an exemplary embodiment of this disclosure, an electronic device capable of implementing the above-described method is also provided.

[0109] Those skilled in the art will understand that various aspects of the present invention can be implemented as systems, methods, or program products. Therefore, various aspects of the present invention can be specifically implemented in the following forms: entirely in hardware, entirely in software (including firmware, microcode, etc.), or in a combination of hardware and software, collectively referred to herein as “circuit,” “module,” or “system.”

[0110] An electronic device according to this embodiment of the invention. The electronic device is merely an example and should not be construed as limiting the functionality or scope of the embodiments of the invention.

[0111] Electronic devices are manifested in the form of general-purpose computing devices. Components of an electronic device may include, but are not limited to: at least one processor, at least one memory, and buses connecting different system components (including memory and processor).

[0112] The memory stores program code that can be executed by a processor, causing the processor to perform the steps described in the "Exemplary Methods" section above, according to various exemplary embodiments of the present invention.

[0113] The storage may include readable media in the form of volatile storage, such as random access memory (RAM) and / or cache memory, and may further include read-only memory (ROM).

[0114] The storage may also include programs / utilities having a set (at least one) of program modules, including but not limited to: an operating system, one or more applications, other program modules, and program data, each or some combination of these examples may include an implementation of a network environment.

[0115] A bus can represent one or more of several bus architectures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus that uses any of the various bus architectures.

[0116] The electronic device can also communicate with one or more external devices (e.g., keyboards, pointing devices, Bluetooth devices, etc.), one or more devices that enable a user to interact with the electronic device, and / or any device that enables the electronic device to communicate with one or more other computing devices (e.g., routers, modems, etc.). This communication can be performed via input / output (I / O) interfaces. Furthermore, the electronic device can communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter. The network adapter communicates with other modules of the electronic device via a bus. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with the electronic device, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0117] From the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, terminal device, or network device, etc.) to execute the methods according to the embodiments of this disclosure.

[0118] In exemplary embodiments of this disclosure, a computer-readable storage medium is also provided, on which a program product capable of implementing the methods described above is stored. In some possible embodiments, various aspects of the present invention may also be implemented as a program product comprising program code that, when the program product is run on a terminal device, causes the terminal device to perform the steps of the various exemplary embodiments of the present invention described in the "Exemplary Methods" section above.

[0119] The program product may employ any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: electrical connections having one or more wires, portable disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0120] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable signal medium may also be any readable medium other than a readable storage medium, capable of sending, propagating, or transmitting programs for use by or in conjunction with an instruction execution system, apparatus, or device.

[0121] The program code contained on the readable medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination thereof.

[0122] Program code for performing the operations of this invention can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can execute entirely on the user's computing device, partially on the user's device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0123] Furthermore, the above figures are merely illustrative of the processes included in the method according to exemplary embodiments of the present invention, and are not intended to be limiting. It is readily understood that the processes shown in the above figures do not indicate or limit the temporal order of these processes. Additionally, it is readily understood that these processes may be executed synchronously or asynchronously, for example, in multiple modules.

[0124] It should be noted that although several modules or units for the device used to perform actions have been mentioned in the detailed description above, this division is not mandatory. In fact, according to embodiments of this disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0125] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for predicting the grinding fatigue life of TC4 titanium alloy, characterized in that, The method includes the following steps: Based on the average proportion of the wear area (A), the average proportion of the wear perimeter (L), the average aspect ratio (R), and the average fractal dimension (D) of the wear surface in the workpiece to be predicted, a comprehensive defect feature value a0 of the wear marks on the workpiece surface is generated; a0 satisfies the following condition: a0 = 0.66A - 0.51L + 0.21R - 0.72D + 2.58; A reflects the quantity and size of the target wear marks on the surface to be tested; L reflects the quantity and size of the target wear marks on the surface to be tested, and whether the target wear marks are convex or concave functions, and can also reflect the superposition of the target wear marks on the surface to be tested; R reflects whether the shape of the target wear marks on the surface to be tested is slender or wide and short; D reflects the complexity of the boundary shape of the target wear marks on the surface to be tested; the target wear mark is a wear mark with a depth greater than a preset depth. Based on the measured cross-sectional depth h and cross-sectional radius of curvature ρ0 corresponding to the wear mark cross-section in the workpiece to be predicted, the initial stress concentration factor K of the wear mark on the surface of the workpiece to be predicted is generated. t ;K t The following conditions must be met: ; Based on the maximum tensile stress σ actually borne by the workpiece under the predicted operating conditions max And the residual stress Rs inside the workpiece after grinding, for K t The correction is performed to generate the target stress concentration factor K for the wear marks on the surface of the workpiece to be predicted. t ’ ;K t ’ The following conditions must be met: K t ’ =[(σ max +Rs)×K t ] / σ max ; Based on a0 and K t ’ Generate the fatigue crack initiation life N of the workpiece to be predicted. i N i The following conditions must be met: ； Where G is the shear modulus of the workpiece to be predicted, E is the elastic modulus of the workpiece to be predicted, and σ r Let Δσ be the fatigue strength of the workpiece to be predicted when the stress ratio is r. r Let ΔK be the tensile stress amplitude of the workpiece to be predicted when the stress ratio is r. th The fatigue crack propagation threshold; After obtaining a0 and K t ’ Subsequently, the method further includes: Based on a0 and h, the initial crack length a is generated. ’ 0; a ’ 0 satisfies the following condition: a ’ 0 = a0 + h; According to a ’ 0 and K t ’ Generate the fatigue crack propagation life N of the workpiece to be predicted. p N p The following conditions must be met: ； Wherein, β1 is a geometric constant, β1 = 1.772; C and m are material property parameters of TC4 titanium alloy, C = 4.58e-12; m = 3.

9.

2. The method according to claim 1, characterized in that, After obtaining N i and N p Subsequently, the method further includes: According to N i and N p Generate the fatigue life N of the workpiece to be predicted. f The following condition must be met: N f =N i +N p .

3. The method according to claim 1, characterized in that, The workpiece to be predicted corresponds to G=46GPa, ΔK th =6.462 E=109.9GPa, σ r =513MPa, Δσ r =525MPa.

4. The method according to claim 1, characterized in that, The measured values ​​of the cross-sectional depth h and the cross-sectional radius of curvature ρ0 corresponding to the grinding marks in the workpiece to be predicted are obtained according to the following steps: Based on the topography information map representing the height information of the surface to be detected, the grayscale information of the image region corresponding to each target wear mark in the topography information map is obtained; the target wear mark is a wear mark with a depth greater than a preset depth; different heights in the topography information map are represented by different colors; the surface to be detected is the surface of the workpiece to be predicted after grinding; Based on the grayscale information corresponding to all target scratches, obtain the pixel depth feature value and pixel width feature value of the scratches on the surface to be detected; The pixel depth feature value is the mode of gray levels among all grayscale information; the pixel width feature value is the average pixel width of the image region corresponding to all target scratches. Based on the pixel depth feature value, generate h corresponding to the wear mark section in the workpiece to be predicted; Based on the pixel width feature value and h of the wear mark on the surface to be detected, ρ0 corresponding to the wear mark cross section in the workpiece to be predicted is generated.

5. The method according to claim 4, characterized in that, Based on the pixel width feature value and h of the wear mark on the surface to be detected, ρ0 corresponding to the wear mark cross-section in the workpiece to be predicted is generated, including: Based on the mapping relationship between the pixel width corresponding to the pixel width feature value and the actual width information, the cross-sectional width detection value w of the wear mark on the surface to be detected is generated; Based on h and w, generate ρ0 corresponding to the wear mark section in the workpiece to be predicted; ρ0 satisfies the following condition: 。 6. The method according to claim 1, characterized in that, The method further includes: Based on the topography information map representing the height information of the surface to be detected, the image region corresponding to each target wear mark in the topography information map is obtained; the target wear mark is a wear mark with a depth greater than a preset depth; the surface to be detected is the surface of the workpiece to be predicted after grinding; Based on the morphology information map and the pixel information of the image region corresponding to each target wear mark, the average percentage of wear mark area and the average percentage of wear mark perimeter are generated on the surface to be detected. Based on the pixel information of the image region corresponding to each target wear mark, the average aspect ratio and average fractal dimension of the wear mark on the surface to be detected are generated. The mean value D of the fractal dimension of the wear marks satisfies the following condition: ； Where n is the total number of target wear marks on the surface to be detected; i 2 Let i be the total number of boundary pixels in the image region corresponding to the i-th target wear mark; i 1 The total number of pixels remaining in the image region corresponding to the i-th target wear mark, excluding the boundary pixels; The average percentage of the wear area A, the average percentage of the wear perimeter L, and the average aspect ratio R of the wear surface to be tested shall satisfy the following conditions respectively: ； ； ； Among them, S i S represents the total number of pixels in the image region corresponding to the i-th target abrasion mark; z P represents the total number of pixels in the topographic image; i P represents the total number of boundary pixels in the image region corresponding to the i-th target wear mark; z Q represents the total number of boundary pixels in the topography image; iw Q represents the total number of pixels in the width direction of the bounding rectangle of the image region corresponding to the i-th target abrasion mark; ic Let be the total number of pixels in the length direction of the bounding rectangle of the image region corresponding to the i-th target abrasion mark.

7. The method according to claim 6, characterized in that, Based on the topography information map representing the height information of the surface to be detected, the image region corresponding to each target wear mark in the topography information map is obtained, including: The topography information map is input into the target Yolov8 algorithm model to generate the image region corresponding to each target wear mark in the topography information map; Gaussian filtering is used in the training samples of the target Yolov8 algorithm model to remove the small dot-like wear mark regions in the topography information map.

8. A non-transitory computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements a method for predicting the grinding fatigue life of TC4 titanium alloy as described in any one of claims 1 to 7.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements a method for predicting the grinding fatigue life of TC4 titanium alloy as described in any one of claims 1 to 7.

Images