A near-field dynamics prediction method for hydrogen-induced fatigue crack growth

Abstract

The application discloses a hydrogen-induced fatigue crack propagation near-field dynamics prediction method, comprising the following steps: meshless discretization is performed on a component, and a conventional state near-field dynamics model is constructed; the cyclic deformation response of a material is solved, the stress and strain of a material point and the cumulative plastic strain are obtained, and the yield strength of the material is corrected according to the local total hydrogen concentration; a hydrogen transport model coupled with multiple mechanisms is established, and the concentrations of lattice hydrogen, trapped hydrogen and desorbed hydrogen are solved; the critical energy density is adjusted downward under the local total hydrogen concentration, the crack evolution is characterized in combination with the near-field bond irreversible fracture, and the deformation field, the hydrogen concentration field, the material degradation state and the near-field bond state are iteratively updated by using an interlaced coupling strategy, so that the fatigue crack propagation prediction result is finally obtained. The application can realize the prediction of the fatigue crack propagation behavior of a hydrogen-containing metal material under different cyclic loads and temperature conditions.

Description

Technical Field

[0001] This invention relates to the fields of computational solid mechanics, fracture mechanics, hydrogen embrittlement damage and fatigue life prediction in materials, and specifically to a near-field dynamics prediction method for hydrogen-induced fatigue crack propagation. Background Technology

[0002] Metal components in hydrogen-powered equipment and related power systems typically endure cyclic loads during service and may be exposed to complex environments such as high temperatures and hydrogen-containing media. When hydrogen enters metallic materials, it migrates, traps, and accumulates under the combined effects of concentration gradients, hydrostatic stress gradients at crack tips, and localized plastic deformation. This reduces the fatigue resistance of the crack tip region, accelerates fatigue crack propagation, and shortens the safe service life of the components. For nickel-based superalloys and other hydrogen-sensitive metallic materials, hydrogen-induced fatigue crack propagation also exhibits a significant range of stress intensity factors and temperature dependence. Under high-temperature conditions, lattice hydrogen can migrate to and desorb from the material's free surface, causing competition between hydrogen accumulation and escape at the crack tip, increasing the complexity of crack propagation prediction.

[0003] Because hydrogen-induced fatigue testing is time-consuming and costly, and it is difficult to quantitatively characterize the hydrogen migration and enrichment processes at the crack tip, numerical simulation methods are usually required for research. Currently, commonly used methods include continuum mechanics methods such as the finite element method, extended finite element method, cohesive zone model, and phase field method. These methods can couple the solution of the material deformation field, hydrogen concentration field, and damage field. However, when dealing with the discontinuity of the displacement field caused by crack initiation and continuous propagation, it is usually necessary to pre-define the crack path, insert cohesive interfaces, introduce enrichment functions, or reconstruct the mesh, increasing the complexity of model construction and numerical solution, and limiting the automatic prediction of crack propagation paths.

[0004] To address the aforementioned limitations, near-field dynamics employs nonlocal integral equations to describe the interactions between material points within a certain neighborhood. It does not require the displacement field to remain continuous at the crack location, and crack initiation, propagation, and bifurcation can be naturally characterized through the degradation of interactions between material points and fracture. It eliminates the need for pre-setting crack paths or repeated mesh reconstruction, making it suitable for simulating complex fracture processes. Existing near-field dynamics research has covered hydrogen-induced fracture and crack propagation in hydrogen-containing materials; however, hydrogen transport is often simplified to concentration gradient-driven lattice diffusion, failing to adequately consider stress-assisted diffusion, hydrogen trapping induced by plastic deformation, and hydrogen migration carried by moving dislocations. Furthermore, it does not adequately consider high-temperature hydrogen desorption processes; and a unified coupling prediction method is still lacking between cyclic deformation, hydrogen redistribution, fatigue performance degradation, and crack propagation.

[0005] Therefore, it is necessary to propose a near-field dynamic prediction method for hydrogen-induced fatigue crack propagation, which can uniformly describe the cyclic deformation, multi-mechanism hydrogen transport, high-temperature hydrogen desorption, hydrogen-induced fatigue performance degradation and crack propagation process, and realize the prediction of fatigue crack propagation behavior of hydrogen-containing metal structures under different loads and temperatures, so as to provide technical support for the life assessment and safety design of related components. Summary of the Invention

[0006] To address the shortcomings of existing near-field dynamics methods in predicting hydrogen-induced fatigue crack propagation, such as simplification of hydrogen transport mechanisms, insufficient consideration of high-temperature hydrogen desorption, and lack of unified coupling between cyclic deformation, hydrogen redistribution, fatigue performance degradation, and crack propagation processes, this invention aims to provide a near-field dynamics prediction method for hydrogen-induced fatigue crack propagation. This method, within the framework of conventional near-field dynamics, uniformly describes the processes of cyclic deformation, hydrogen migration from moving dislocations, high-temperature hydrogen desorption, and hydrogen-induced fatigue performance degradation, enabling the prediction of fatigue crack propagation behavior in hydrogen-containing metallic materials under different cyclic loading and temperature conditions.

[0007] To achieve the above-mentioned technical objectives, the technical solution adopted by the present invention is as follows:

[0008] A near-field dynamic prediction method for hydrogen-induced fatigue crack propagation, the method comprising the following steps:

[0009] Step 1: The computational domain of the component to be predicted is uniformly discretized without mesh, material points and their near-field neighborhoods are established, a normal near-field dynamic model is constructed, and material parameters, initial cracks, boundary conditions, cyclic loads and initial hydrogen concentration are set.

[0010] Step 2: Solve the cyclic deformation response of the material based on the cyclic plastic or cyclic viscoplastic constitutive relation to obtain the stress and strain of the material points and the cumulative plastic strain, and correct the yield strength of the material according to the local total hydrogen concentration.

[0011] Step 3: Establish a hydrogen transport model that includes concentration gradient diffusion, stress-assisted diffusion, plastic deformation-induced hydrogen trapping, hydrogen migration carried by moving dislocations, and high-temperature hydrogen desorption. Use the near-field dynamic differential operator to convert the hydrogen transport equation into a nonlocal integral form and solve for the concentrations of lattice hydrogen, trapped hydrogen, and desorbed hydrogen.

[0012] Step 4: Establish fatigue bond breaking criteria based on the cumulative elastic strain energy density and cumulative plastic strain energy density of near-field bonds, reduce the critical energy density according to the local total hydrogen concentration, and characterize crack initiation and propagation through irreversible fracture of near-field bonds.

[0013] Step 5: The cyclic deformation field, hydrogen concentration field, hydrogen-induced performance degradation state and near-field bond state are updated sequentially using an interleaved coupling method until the preset number of cycles or crack length is reached, and the predicted results such as fatigue crack propagation curve and crack propagation rate are output.

[0014] Step 1, the process of constructing the normal state near-field dynamics model includes the following sub-steps:

[0015] The computational domain of the component to be predicted is discretized into a series of near-field dynamic material points with volume and mass properties; for each material point, its corresponding near-field neighborhood is established according to the preset near-field scale parameters.

[0016] Define the reference bond vector and relative displacement vector between a material point and its neighboring material points in the neighborhood, solve for the elongation scalar state of the near-field bond, and further decompose the elongation scalar state into a spherical part and a biased part to characterize the volumetric deformation effect and the shear deformation effect, respectively.

[0017] Based on the material elastic parameters, a normal state near-field dynamic scalar force state is constructed, and the force vector state of the interaction between material points is obtained from the scalar force state.

[0018] Establish the normal state peri-field dynamics control equations, and set boundary conditions and cyclic loads according to the geometry of the component to be predicted; for components with initial cracks, set the peri-field bonds intersecting with the initial crack surface as the initial failure state.

[0019] Step 2 further includes:

[0020] The bias portion of the elongation scalar state of the near-field bond is decomposed into elastic and plastic components;

[0021] A yield function incorporating kinematic hardening and isotropic hardening effects is constructed, and an expression for the yield strength, modulated by the local total hydrogen concentration, is introduced into the yield function:

[0022]

[0023] in, Let be the yield function. For equivalent von Mises stress, Back stress in scalar form To accumulate plastic strain, An isotropic hardening function characterizing the change in the yield surface; The yield strength in the hydrogen-containing state is expressed as:

[0024]

[0025] in, This is the reference yield strength in the hydrogen-free state. The hydrogen-induced hardening coefficient, This represents the local total hydrogen concentration. For reference hydrogen concentration; cumulative plastic strain is expressed as:

[0026]

[0027] In the formula, Let be the near-field influence function. The correlation coefficient is dimensional; for cyclic viscoplastic conditions, the cumulative plastic strain rate satisfies the following evolution form:

[0028]

[0029] in, For viscosity parameters, It is a strain rate sensitive parameter. For Macaulay brackets; It is a viscoplastic flow function;

[0030] The yield strength is dynamically corrected based on the real-time updated local total hydrogen concentration. The elastic or viscoplastic state of the material point is determined by elastic trial calculation and return mapping algorithm. In the elastic state, the plastic-related variables remain unchanged. In the viscoplastic state, the stress, plastic strain, cumulative plastic strain, plastic strain rate and hardening variables are updated.

[0031] Step 3 further includes:

[0032] S3.1, Based on cumulative plastic strain Update dislocation density And calculate the dislocation trap density. :

[0033]

[0034] in, For dislocation trap coefficients, is the lattice constant. It is Avogadro's constant;

[0035] Based on the updated dislocation trap density With lattice hydrogen concentration The calculated trap hydrogen concentration is expressed as:

[0036]

[0037] in, For lattice site density, The number of hydrogen atoms that a single trap can hold. This is the equilibrium constant for the hydrogen trap;

[0038] Combining lattice hydrogen and trapped hydrogen, the local total hydrogen concentration is calculated and expressed as: ;

[0039] S3.2 Construct two types of hydrogen transport fluxes: lattice hydrogen flux and moving dislocation hydrogen-carrying flux. The lattice hydrogen flux is driven by both the concentration gradient and the hydrostatic stress gradient.

[0040]

[0041] in, The lattice hydrogen diffusion coefficient is The partial molar volume of hydrogen. The gas constant is Absolute temperature For hydrostatic stress; the hydrogen-carrying flux of moving dislocations is expressed as:

[0042]

[0043] in, For Burgers' vector modulus, The unit vector representing the direction of hydrogen-carrying migration of moving dislocations;

[0044] S3.3 Integrating lattice hydrogen flux and moving dislocation hydrogen-carrying flux, a hydrogen transport equation is established that includes diffusion, trapping, dislocation hydrogen-carrying migration, and high-temperature desorption:

[0045]

[0046] in, The hydrogen trapping capacity factor is . This represents the evolution of trap density caused by plastic deformation and the exchange between lattice hydrogen and trapped hydrogen, where This is the high-temperature hydrogen desorption term;

[0047] The high-temperature hydrogen desorption term is represented as:

[0048]

[0049] in, This is the equivalent hydrogen desorption coefficient;

[0050] The cumulative desorbed hydrogen concentration satisfies:

[0051]

[0052] in, This represents the cumulative desorbed hydrogen concentration.

[0053] S3.4. The gradient, second derivative and divergence in the hydrogen transport equation are converted into nonlocal integrals in the neighborhood of the material point using the peri-field dynamic differential operator.

[0054] S3.5. Solve the hydrogen transport equation using a nonlocal differential operator to update the lattice hydrogen concentration at each material point, and then update the trapped hydrogen concentration, total hydrogen concentration, and cumulative desorbed hydrogen concentration in sequence. The local total hydrogen concentration obtained from the solution is used to correct the yield strength of the material in real time.

[0055] Step 4 further includes:

[0056] S4.1 Calculate the cumulative elastic strain energy density of the material points based on the elastic and plastic deformation of the near-field bonds during cyclic loading. and cumulative plastic strain energy density ;

[0057] S4.2. Define the hydrogen-induced degradation coefficient based on the total hydrogen concentration at the substance point:

[0058]

[0059] in, It is a sensitive parameter for hydrogen-induced degradation;

[0060] The hydrogen concentration-dependent critical energy density is expressed as:

[0061]

[0062] in, This is the reference critical energy density in the hydrogen-free state;

[0063] S4.3. The historical minimum value operator is used to update the hydrogen concentration-related fatigue critical energy density to maintain the irreversibility of hydrogen-induced fatigue performance degradation:

[0064] ;

[0065] In the formula, Indicates the first A point of matter, and They represent the first The calculation step and the first One calculation step; and For the first Step and the first Step matter point The critical energy density; For the first Step matter point The local total hydrogen concentration;

[0066] S4.4 Establish near-field bond fatigue fracture criteria:

[0067]

[0068] in, This indicates that the near-field bond is intact. This indicates that the near-field bond is invalid; the invalid near-field bond remains invalid in subsequent calculations. The hydrogen concentration-dependent fatigue critical energy density; Let be the relative position vector of the material points at both ends of the near-field bond. This is the current calculation time;

[0069] S4.5 Introduce the near-field bond state into the near-field influence function so that the failed near-field bonds no longer participate in the subsequent internal force calculation, and update the hydrogen transport neighborhood according to the near-field bond state; finally, define the local damage variable by the proportion of failed near-field bonds in the neighborhood of the material point.

[0070] Step 5 further includes:

[0071] S5.1 Apply the current cyclic load increment, solve the near-field dynamic motion control equations, and update the material point stress, strain, cumulative plastic strain, and plastic strain rate;

[0072] S5.2. Based on the updated cumulative plastic strain and plastic strain rate, update the dislocation density, trap density, and hydrogen-carrying flux of moving dislocations, and solve the nonlocal hydrogen transport equation.

[0073] S5.3 Update the lattice hydrogen concentration, trapped hydrogen concentration, total hydrogen concentration and cumulative desorbed hydrogen concentration, and correct the material yield strength and critical energy density based on the total hydrogen concentration;

[0074] S5.4 Calculate the cumulative elastic strain energy density and cumulative plastic strain energy density, determine whether the near-field bond is broken, and update the local damage variables and crack location;

[0075] S5.5 Determine whether the current cycle number, crack length, or damage state has reached the termination condition; if not, return to S5.1 to calculate the load increment for the next cycle; if so, output the crack length-cycle number curve, crack propagation rate-stress intensity factor range curve, hydrogen concentration distribution, desorbed hydrogen distribution, critical energy density distribution, local damage distribution, and fatigue life.

[0076] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0077] First, the near-field dynamics prediction method for hydrogen-induced fatigue crack propagation of the present invention couples cyclic deformation, multi-mechanism hydrogen transport, hydrogen-induced fatigue performance degradation and crack propagation process under a unified conventional near-field dynamics framework, and uses a unified material point discretization system for solution, reducing mesh mapping and field variable transfer between different numerical methods.

[0078] Second, the near-field dynamic prediction method for hydrogen-induced fatigue crack propagation of the present invention comprehensively considers stress-assisted diffusion, plastic deformation-induced hydrogen trapping, hydrogen migration carried by moving dislocations, and high-temperature hydrogen desorption, and uses the total hydrogen concentration to correct the material yield strength and critical fatigue energy density, which can describe the coupling effect between hydrogen redistribution, local fatigue resistance degradation and crack propagation.

[0079] Third, the near-field dynamics prediction method for hydrogen-induced fatigue crack propagation of the present invention naturally simulates crack initiation and propagation through irreversible fracture of near-field bonds. It does not require pre-setting crack paths or repeated mesh reconstruction, and can predict crack propagation rates and fatigue life under different cyclic loads and temperatures. It also obtains physical quantities such as hydrogen concentration at the crack tip, hydrogen desorption, critical energy density, and local damage, providing a basis for the design and life assessment of hydrogen-containing components. Attached Figure Description

[0080] Figure 1 This is a schematic diagram of a normal near-field dynamics model;

[0081] Figure 2 Flowchart for calculating hydrogen-induced fatigue crack propagation;

[0082] Figure 3 This is a schematic diagram showing the geometry of the compact tensile specimen and the near-field dynamics of the material points discretized in the example.

[0083] Figure 4 This is a comparison chart of the crack length-cycle number and crack propagation rate-stress intensity factor range results obtained at room temperature and the experimental results, as shown in the example.

[0084] Figure 5 The diagram shows the normalized total hydrogen concentration distribution at room temperature and different crack lengths, as well as the distribution of lattice hydrogen and trapped hydrogen at a selected material point at the crack tip, as shown in the example.

[0085] Figure 6 This is a comparison chart of the crack length-cycle number and crack propagation rate-stress intensity factor ranges obtained at 650 °C in the example with the experimental results;

[0086] Figure 7 The figure shows the results of hydrogen-assisted fatigue crack propagation acceleration ratio as a function of stress intensity factor at room temperature and 650 °C in the example.

[0087] Figure 8 The normalized total hydrogen concentration and cumulative desorbed hydrogen concentration distribution are shown in the example at 650 °C for different crack lengths. Detailed Implementation

[0088] The embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0089] This invention discloses a near-field dynamics prediction method for hydrogen-induced fatigue crack propagation, the overall process of which is as follows: Figure 2 As shown, the method includes the following steps:

[0090] A near-field dynamic prediction method for hydrogen-induced fatigue crack propagation includes the following steps:

[0091] Step 1: Uniformly discretize the computational domain of the component to be predicted without mesh, establish the material points and their near-field neighborhoods, construct a normal-state near-field dynamic model, and set material parameters, initial crack, boundary conditions, cyclic load, and initial hydrogen concentration. The near-field dynamic model is as follows: Figure 1 As shown;

[0092] Step 2: Solve the cyclic deformation response of the material based on the cyclic plastic or cyclic viscoplastic constitutive relation to obtain the stress and strain of the material points and the cumulative plastic strain, and correct the yield strength of the material according to the local total hydrogen concentration.

[0093] Step 3: Establish a hydrogen transport model that includes concentration gradient diffusion, stress-assisted diffusion, plastic deformation-induced hydrogen trapping, hydrogen migration carried by moving dislocations, and high-temperature hydrogen desorption. Use the near-field dynamic differential operator to convert the hydrogen transport equation into a nonlocal integral form and solve for the concentrations of lattice hydrogen, trapped hydrogen, and desorbed hydrogen.

[0094] Step 4: Establish fatigue bond breaking criteria based on the cumulative elastic strain energy density and cumulative plastic strain energy density of near-field bonds, reduce the critical energy density according to the local total hydrogen concentration, and characterize crack initiation and propagation through irreversible fracture of near-field bonds.

[0095] Step 5: The cyclic deformation field, hydrogen concentration field, hydrogen-induced performance degradation state and near-field bond state are updated sequentially using an interleaved coupling method until the preset number of cycles or crack length is reached, and the predicted results such as fatigue crack propagation curve and crack propagation rate are output.

[0096] Further, in step 1, the establishment of the normal state near-field dynamics model includes:

[0097] S1.1 Discretize the computational domain of the component to be predicted into a series of near-field dynamic material points with volume and mass properties; for any material point In near-field scale parameters Establish its near-field neighborhood within the range:

[0098]

[0099] in, For matter points Adjacent material points within the neighborhood, This is the region for near-field dynamics calculations.

[0100] S1.2, Define the material point With adjacent material points Reference key vector between and relative displacement vector :

[0101]

[0102]

[0103] in, For the displacement of a material point, Let be time. Correspondingly, the elongation scalar state of the near-field bond is represented as:

[0104]

[0105] Decompose it into spherical mass component and biased part :

[0106]

[0107] S1.3 Establish the near-field dynamic scalar force state based on the material's elastic parameters. :

[0108]

[0109] in, Equivalent bulk modulus Shear modulus This is the volume expansion amount. For the influence function, This is the weighted volume.

[0110] Force vector state Represented as:

[0111]

[0112] in:

[0113]

[0114] S1.4 Establish the control equations for normal near-field dynamics:

[0115]

[0116] in, For material density, It is the acceleration vector. It represents the density of body force.

[0117] S1.5 Set boundary conditions and cyclic loads according to the geometry of the component to be predicted; for components with initial cracks, set the near-field bonds intersecting with the initial crack surface as the initial failure state.

[0118] Further, in step 2, the solution for the cyclic deformation response of the material includes:

[0119] S2.1. Decompose the partial component of the bond elongation scalar state into elastic and plastic components. For a two-dimensional problem, this can be expressed as:

[0120]

[0121] in, and These represent the elastic and plastic components of the scalar elongation deviator, respectively.

[0122] S2.2. Define the hybrid hardening yield function applicable to normal near-field dynamics, expressed as:

[0123]

[0124] in, Back stress in scalar form To accumulate plastic strain, An isotropic hardening function characterizing the change in yield surface, This represents the yield strength in the hydrogen-containing state.

[0125] S2.3. Define cumulative plastic strain. For a two-dimensional problem, cumulative plastic strain is expressed as:

[0126]

[0127] For cyclic viscoplastic conditions, the evolution is as follows:

[0128]

[0129] in, For viscosity parameters, It is a strain rate sensitive parameter. The brackets are for Macaulay.

[0130] S2.4. Correct the material yield strength based on the local total hydrogen concentration:

[0131]

[0132] in, This is the reference yield strength in the hydrogen-free state. The hydrogen-induced hardening coefficient, This represents the local total hydrogen concentration. For reference hydrogen concentration.

[0133] S2.5. Use elastic trial calculation and return mapping algorithm to determine the elastic or inelastic state of the material point, and update the stress, plastic strain, cumulative plastic strain, plastic strain rate and hardening variable.

[0134] Furthermore, in step 3, the establishment and solution of the multi-mechanism hydrogen transport model includes:

[0135] S3.1. The hydrogen in the material is divided into lattice hydrogen and trapped hydrogen, and the total hydrogen concentration is expressed as:

[0136]

[0137] in, The lattice hydrogen concentration, This represents the concentration of hydrogen trapped in the trap.

[0138] Based on the local equilibrium relationship, the hydrogen concentration in the trap is expressed as:

[0139]

[0140] in, For lattice site density, For the density of trap sites, The number of hydrogen atoms that a single trap can hold. This is the equilibrium constant for the hydrogen trap.

[0141] S3.2 Update dislocation density and dislocation trap density based on cumulative plastic strain:

[0142]

[0143] in, The dislocation density evolves with cumulative plastic strain. For dislocation trap coefficients, is the lattice constant. is Avogadro's constant.

[0144] S3.3, Lattice hydrogen flux is driven by both concentration gradient and hydrostatic stress gradient:

[0145]

[0146] in, The lattice hydrogen diffusion coefficient is The partial molar volume of hydrogen. The gas constant is Absolute temperature It is hydrostatic stress.

[0147] The hydrogen-carrying flux of moving dislocations is expressed as:

[0148]

[0149] in, For Burgers' vector modulus, This is the unit vector representing the direction of hydrogen migration carried by moving dislocations.

[0150] S3.4 Establish hydrogen transport equations that include diffusion, trapping, dislocation hydrogen-carrying migration, and high-temperature desorption:

[0151]

[0152] in, The hydrogen trapping capacity factor is . This represents the evolution of trap density caused by plastic deformation and the exchange between lattice hydrogen and trapped hydrogen. This is a high-temperature hydrogen desorption term.

[0153] The high-temperature hydrogen desorption term is represented as:

[0154]

[0155] in, is the equivalent hydrogen desorption coefficient.

[0156] The cumulative desorbed hydrogen concentration satisfies:

[0157]

[0158] in, This represents the cumulative desorbed hydrogen concentration.

[0159] S3.5. The gradient, second derivative, and divergence in the hydrogen transport equation are transformed into nonlocal integrals within the neighborhood of the matter point using near-field dynamical differential operators. For any scalar field... Its first and second nonlocal derivatives are expressed as:

[0160]

[0161]

[0162] The scalar field edge and The second-order nonlocal derivatives of the directions are expressed as follows:

[0163]

[0164]

[0165] Therefore, scalar field The Laplace operator can be expressed as:

[0166]

[0167] Furthermore, the lattice hydrogen concentration Laplace term and the hydrostatic stress Laplace term in the hydrogen transport equation are expressed as follows:

[0168]

[0169]

[0170] The inner product of the lattice hydrogen concentration gradient and the hydrostatic stress gradient is expressed as:

[0171]

[0172] S3.6 Solve the hydrogen transport equation in S3.4 using the nonlocal differential operator, update the lattice hydrogen concentration of each material point, and sequentially update the trapped hydrogen concentration, total hydrogen concentration, and cumulative desorbed hydrogen concentration.

[0173] Furthermore, in step 4, the prediction of hydrogen-induced fatigue damage and crack propagation includes:

[0174] S4.1 Calculate the cumulative elastic strain energy density of the material points based on the elastic and plastic deformation of the near-field bonds during cyclic loading. and cumulative plastic strain energy density :

[0175]

[0176]

[0177] S4.2. Define the hydrogen-induced degradation coefficient based on the total hydrogen concentration at the substance point:

[0178]

[0179] in, It is a sensitive parameter for hydrogen-induced degradation.

[0180] The hydrogen concentration-dependent critical energy density is expressed as:

[0181]

[0182] in, This is the reference critical energy density in the hydrogen-free state.

[0183] S4.3. Use historical minimum values ​​to maintain the irreversibility of hydrogen-induced fatigue performance degradation:

[0184]

[0185] S4.4 Establish near-field bond fatigue fracture criteria:

[0186]

[0187] in, This indicates that the near-field bond is intact. This indicates that the near-field bond is invalid; the invalid near-field bond remains invalid in subsequent calculations.

[0188] S4.5 Introducing the near-field bond state into the near-field influence function:

[0189]

[0190] The failed near-field bonds are no longer involved in subsequent internal force calculations, and the hydrogen transport neighborhood is updated based on the near-field bond state.

[0191] S4.6 Define the local damage variable by the proportion of failed near-field bonds in the neighborhood of the material point:

[0192]

[0193] in, The value range is from 0 to 1. The larger the value, the higher the degree of local damage at the material point. The hydrogen concentration-related critical energy density and the irreversible bond breaking criterion are used to establish the relationship between hydrogen enrichment, decreased fatigue resistance, and crack propagation.

[0194] Furthermore, in step 5, the interleaved coupling solution includes:

[0195] S5.1 Apply the current cyclic load increment, solve the near-field dynamic motion control equations, and update the material point stress, strain, cumulative plastic strain, and plastic strain rate;

[0196] S5.2. Based on the updated cumulative plastic strain and plastic strain rate, update the dislocation density, trap density, and hydrogen-carrying flux of moving dislocations, and solve the nonlocal hydrogen transport equation.

[0197] S5.3 Update the lattice hydrogen concentration, trapped hydrogen concentration, total hydrogen concentration and cumulative desorbed hydrogen concentration, and correct the material yield strength and critical energy density based on the total hydrogen concentration;

[0198] S5.4 Calculate the cumulative elastic strain energy density and cumulative plastic strain energy density, determine whether the near-field bond is broken, and update the local damage variables and crack location;

[0199] S5.5 Determine whether the current cycle number, crack length, or damage state has reached the termination condition; if not, return to S5.1 to calculate the load increment for the next cycle; if so, output the crack length-cycle number curve, crack propagation rate-stress intensity factor range curve, hydrogen concentration distribution, desorbed hydrogen distribution, critical energy density distribution, local damage distribution, and fatigue life.

[0200] Furthermore, to verify the applicability and prediction accuracy of the present invention in predicting hydrogen-induced fatigue crack propagation, the nickel-based superalloy FGH4096 was used as the research object, and its hydrogen-induced fatigue crack propagation behavior under room temperature and high temperature conditions was simulated and analyzed.

[0201] Example 1

[0202] like Figure 3 As shown in (a) of the example, this is a two-dimensional compact tensile specimen with a width W of 30 mm, a height of 18 mm for both the upper and lower halves, and a loading hole radius of 3.75 mm. The initial crack is set along the centerline of the specimen, and the initial crack length is 2.5 mm. Figure 3 As shown in (b), the sample region was discretized into two-dimensional material points, resulting in 8818 material points. The near-field bonds intersecting the initial crack surface were broken to form the initial crack. The loading hole neighborhood was set as the loading region, and cyclic loads in opposite directions were applied to the upper and lower loading regions. The cyclic load ratio R was set to 0.1, the frequency f to 1.0 Hz, and the maximum load P... max The material density is taken as 8200 kg / m³, Poisson's ratio as 0.32, room temperature elastic modulus as 213.2 GPa, and initial yield stress as 1010.8 MPa. A rate-independent cyclic plasticity model is used for room temperature cyclic deformation, and the correction factor for the total hydrogen concentration on the yield state is taken as 0.35. The initial total hydrogen concentration is taken as 1073.52 mol / m³; the room temperature lattice hydrogen diffusion coefficient is taken as 4.38 × 10⁻⁶. -14m² / s. The room temperature desorption coefficient was set to 0, so that the total hydrogen in the sample was spatially redistributed only through internal migration and lattice / trapped hydrogen partitioning. The hydrogen-free reference critical fatigue energy density was taken as 4.4 × 10¹² J / m³, and the hydrogen-induced degradation sensitivity index was taken as 5.83.

[0203] Through the above experimental simulation Figure 4 The crack length-cycle count (corresponding to) is given under room temperature hydrogen-free and pre-charged hydrogen conditions. Figure 4 (a) and the crack propagation rate-stress intensity factor range results (corresponding to) Figure 4 (b) shows that the simulation curves exhibit a consistent trend with the corresponding experimental data. Figure 5 The normalized total hydrogen concentration distribution is shown when the crack length 'a' is 6 mm, 12 mm, and 18 mm, where... Figure 5 (a) shows the evolution of hydrogen enrichment at the crack tip under different crack propagation lengths; as the crack continues to propagate forward, the stress-induced hydrogen enrichment effect at the crack tip is continuously enhanced, and the normalized total hydrogen concentration at the crack tip gradually increases. Figure 5 (b) Selects the crack tip tracing point when the crack length a is 18 mm, and compares the lattice hydrogen in the initial and final states. and trap hydrogen The proportion of hydrogen concentration changes. In the initial stage, only lattice hydrogen exists at the tracking point. After crack propagation, the total hydrogen concentration increases significantly, with lattice hydrogen accounting for 66% and trapped hydrogen accounting for 34%. The results show that the present invention can reliably predict the fatigue crack propagation behavior and the evolution of hydrogen concentration at the crack tip in hydrogen-free and pre-hydrogen-charged FGH4096 nickel-based superalloys at room temperature.

[0204] Example 2

[0205] The same specimen geometry, material point discretization, initial crack, and cyclic loading were used as in Example 1. The elastic modulus at 650 °C was taken as 198.3 GPa, the initial yield stress as 890.2 MPa, and the cyclic deformation was performed using a cyclic viscoplastic model; the lattice hydrogen diffusion coefficient was taken as 3.98 × 10⁻⁶. -10 m 2 / s, with the reference critical fatigue energy density without hydrogen taken as 3.3×10⁻⁶. 12 J / m 3 The hydrogen-induced degradation sensitivity index remains at 5.83. To reflect high-temperature hydrogen loss, the equivalent volumetric desorption coefficient is set to 2.4 × 10⁻⁶. -4 s -1 The desorption source term is initiated when the lattice hydrogen concentration at a material point is higher than the reference equilibrium lattice hydrogen concentration.

[0206] Through the above experimental simulation Figure 6 The crack length-cycle count (corresponding to) under hydrogen-free and pre-hydrogen-charged conditions at 650 °C is given. Figure 6 (a) and the crack propagation rate-stress intensity factor range results (corresponding to) Figure 6 (b) shows that the simulation curves exhibit a consistent trend with the corresponding experimental data. Figure 7 Further comparisons were made between the hydrogen-assisted fatigue crack propagation acceleration ratios at room temperature and 650 °C, and both simulations and experiments reflected the same temperature difference. Figure 8 The total hydrogen concentration and cumulative desorbed hydrogen concentration are given for crack lengths of 6 mm, 12 mm, and 18 mm, where... Figure 8 (a) in the figure represents the normalized total hydrogen concentration distribution. As the crack continues to propagate, the stress-induced hydrogen enrichment effect at the crack tip gradually weakens. Figure 8 In Figure (b), the normalized cumulative desorbed hydrogen concentration distribution shows that as the crack extends forward, the number of newly formed crack surfaces continuously increases, the hydrogen desorption region expands, and the cumulative desorbed hydrogen concentration along the crack path significantly increases. The results demonstrate that this invention can reliably predict the fatigue crack propagation behavior of hydrogen-free and pre-hydrogen-charged FGH4096 nickel-based superalloys at 650 °C, and characterize the evolution of hydrogen enrichment and desorption at high temperatures.

[0207] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0208] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A near-field dynamic prediction method for hydrogen-induced fatigue crack propagation, characterized in that, The method includes the following steps: Step 1: The computational domain of the component to be predicted is uniformly discretized without mesh, material points and their near-field neighborhoods are established, a normal near-field dynamic model is constructed, and material parameters, initial cracks, boundary conditions, cyclic loads and initial hydrogen concentration are set. Step 2: Solve the cyclic deformation response of the material based on the cyclic plastic or cyclic viscoplastic constitutive relation to obtain the stress and strain of the material points and the cumulative plastic strain, and correct the yield strength of the material according to the local total hydrogen concentration. Step 3: Establish a hydrogen transport model that includes concentration gradient diffusion, stress-assisted diffusion, plastic deformation-induced hydrogen trapping, hydrogen migration carried by moving dislocations, and high-temperature hydrogen desorption. Use the near-field dynamic differential operator to convert the hydrogen transport equation into a nonlocal integral form and solve for the concentrations of lattice hydrogen, trapped hydrogen, and desorbed hydrogen. Step 4: Establish fatigue bond breaking criteria based on the cumulative elastic strain energy density and cumulative plastic strain energy density of near-field bonds, reduce the critical energy density according to the local total hydrogen concentration, and characterize crack initiation and propagation through irreversible fracture of near-field bonds. Step 5: The cyclic deformation field, hydrogen concentration field, hydrogen-induced performance degradation state and near-field bond state are updated sequentially using an interleaved coupling method until the preset number of cycles or crack length is reached, and the predicted results such as fatigue crack propagation curve and crack propagation rate are output.

2. The near-field dynamic prediction method for hydrogen-induced fatigue crack propagation according to claim 1, characterized in that, Step 1, the process of constructing the normal state near-field dynamics model includes the following sub-steps: The computational domain of the component to be predicted is discretized into a series of near-field dynamic material points with volume and mass properties; for each material point, its corresponding near-field neighborhood is established according to the preset near-field scale parameters. Define the reference bond vector and relative displacement vector between a material point and its neighboring material points in the neighborhood, solve for the elongation scalar state of the near-field bond, and further decompose the elongation scalar state into a spherical part and a biased part to characterize the volumetric deformation effect and the shear deformation effect, respectively. Based on the material elastic parameters, a normal state near-field dynamic scalar force state is constructed, and the force vector state of the interaction between material points is obtained from the scalar force state. Establish the normal state peri-field dynamics control equations, and set boundary conditions and cyclic loads according to the geometry of the component to be predicted; for components with initial cracks, set the peri-field bonds intersecting with the initial crack surface as the initial failure state.

3. The near-field dynamic prediction method for hydrogen-induced fatigue crack propagation according to claim 1, characterized in that, Step 2 further includes: The bias portion of the elongation scalar state of the near-field bond is decomposed into elastic and plastic components; A yield function incorporating kinematic hardening and isotropic hardening effects is constructed, and an expression for the yield strength, modulated by the local total hydrogen concentration, is introduced into the yield function: in, Let be the yield function. For equivalent von Mises stress, Back stress in scalar form To accumulate plastic strain, An isotropic hardening function characterizing the change in the yield surface; The yield strength in the hydrogen-containing state is expressed as: in, This is the reference yield strength in the hydrogen-free state. The hydrogen-induced hardening coefficient, This represents the local total hydrogen concentration. For reference hydrogen concentration; cumulative plastic strain is expressed as: In the formula, Let m be the near-field influence function and m be the dimensional correlation coefficient. For cyclic viscoplastic conditions, the cumulative plastic strain rate satisfies the following evolution form: in, For viscosity parameters, It is a strain rate sensitive parameter. For Macaulay brackets; It is a viscoplastic flow function; The yield strength is dynamically corrected based on the real-time updated local total hydrogen concentration. The elastic or viscoplastic state of the material point is determined by elastic trial calculation and return mapping algorithm. In the elastic state, the plastic-related variables remain unchanged. In the viscoplastic state, the stress, plastic strain, cumulative plastic strain, plastic strain rate and hardening variables are updated.

4. The near-field dynamic prediction method for hydrogen-induced fatigue crack propagation according to claim 1, characterized in that, Step 3 further includes: S3.1, Based on cumulative plastic strain Update dislocation density And calculate the dislocation trap density. : in, For dislocation trap coefficients, is the lattice constant. It is Avogadro's constant; Based on the updated dislocation trap density With lattice hydrogen concentration The calculated trap hydrogen concentration is expressed as: in, For lattice site density, The number of hydrogen atoms that a single trap can hold. This is the equilibrium constant for the hydrogen trap; Combining lattice hydrogen and trapped hydrogen, the local total hydrogen concentration is calculated and expressed as: ; S3.2 Construct two types of hydrogen transport fluxes: lattice hydrogen flux and moving dislocation hydrogen-carrying flux. The lattice hydrogen flux is driven by both the concentration gradient and the hydrostatic stress gradient. in, The lattice hydrogen diffusion coefficient is The partial molar volume of hydrogen. The gas constant is Absolute temperature For hydrostatic stress; the hydrogen-carrying flux of moving dislocations is expressed as: in, For Burgers' vector modulus, The unit vector representing the direction of hydrogen-carrying migration of moving dislocations; S3.3 Integrating lattice hydrogen flux and moving dislocation hydrogen-carrying flux, a hydrogen transport equation is established that includes diffusion, trapping, dislocation hydrogen-carrying migration, and high-temperature desorption: in, The hydrogen trapping capacity factor is . This represents the evolution of trap density caused by plastic deformation and the exchange between lattice hydrogen and trapped hydrogen. This is a high-temperature dehydrogenation and desorption term; The high-temperature hydrogen desorption term is represented as follows: in, This is the equivalent hydrogen desorption coefficient; The cumulative desorbed hydrogen concentration satisfies: in, This represents the cumulative desorbed hydrogen concentration. S3.

4. The gradient, second derivative and divergence in the hydrogen transport equation are converted into nonlocal integrals in the neighborhood of the material point using the peri-field dynamic differential operator. S3.

5. Solve the hydrogen transport equation using a nonlocal differential operator to update the lattice hydrogen concentration at each material point, and then update the trapped hydrogen concentration, total hydrogen concentration, and cumulative desorbed hydrogen concentration in sequence. The local total hydrogen concentration obtained from the solution is used to correct the yield strength of the material in real time.

5. The near-field dynamic prediction method for hydrogen-induced fatigue crack propagation according to claim 1, characterized in that, Step 4 further includes: S4.1 Calculate the cumulative elastic strain energy density of the material points based on the elastic and plastic deformation of the near-field bonds during cyclic loading. and cumulative plastic strain energy density ; S4.

2. Define the hydrogen-induced degradation coefficient based on the total hydrogen concentration at the substance point: in, It is a sensitive parameter for hydrogen-induced degradation; The hydrogen concentration-dependent critical energy density is expressed as: in, This is the reference critical energy density in the hydrogen-free state; S4.

3. The historical minimum value operator is used to update the hydrogen concentration-related fatigue critical energy density to maintain the irreversibility of hydrogen-induced fatigue performance degradation: In the formula, Indicates the first A point of matter, and They represent the first The calculation step and the first One calculation step; and For the first Step and the first Step matter point The critical energy density; For the first Step matter point The local total hydrogen concentration; S4.4 Establish near-field bond fatigue fracture criteria: in, This indicates that the near-field bond is intact. This indicates that the near-field bond is invalid; the invalid near-field bond remains invalid in subsequent calculations. The hydrogen concentration-dependent fatigue critical energy density; Let be the relative position vector of the material points at both ends of the near-field bond. This is the current calculation time; S4.5 Introduce the near-field bond state into the near-field influence function so that the failed near-field bonds no longer participate in the subsequent internal force calculation, and update the hydrogen transport neighborhood according to the near-field bond state; finally, define the local damage variable by the proportion of failed near-field bonds in the neighborhood of the material point.

6. The near-field dynamic prediction method for hydrogen-induced fatigue crack propagation according to claim 1, characterized in that, Step 5 further includes: S5.1 Apply the current cyclic load increment, solve the near-field dynamic motion control equations, and update the material point stress, strain, cumulative plastic strain, and plastic strain rate; S5.

2. Based on the updated cumulative plastic strain and plastic strain rate, update the dislocation density, trap density, and hydrogen-carrying flux of moving dislocations, and solve the nonlocal hydrogen transport equation. S5.3 Update the lattice hydrogen concentration, trapped hydrogen concentration, total hydrogen concentration and cumulative desorbed hydrogen concentration, and correct the material yield strength and critical energy density based on the total hydrogen concentration; S5.4 Calculate the cumulative elastic strain energy density and cumulative plastic strain energy density, determine whether the near-field bond is broken, and update the local damage variables and crack location; S5.5 Determine whether the current cycle number, crack length, or damage state has reached the termination condition; if not, return to S5.1 to calculate the load increment for the next cycle; if so, output the crack length-cycle number curve, crack propagation rate-stress intensity factor range curve, hydrogen concentration distribution, desorbed hydrogen distribution, critical energy density distribution, local damage distribution, and fatigue life.

Images