Heat-resistant steel random grain stacking modeling and thermal shock simulation method
By constructing a random grain stacking model and assigning it anisotropic properties, combined with thermo-mechanical coupling equations, the thermal shock process of heat-resistant steel is accurately simulated. This solves the problem that the influence of microstructure is not considered in existing technologies, and achieves high-precision stress distribution and damage law revelation, guiding the optimized design of heat-resistant steel.
Patent Information
- Application Number
- CN202511398750.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-28
- Publication Date
- 2026-01-16
AI Technical Summary
Existing material simulation methods cannot accurately describe the microstructure of heat-resistant steel. They ignore grain size, orientation and grain boundary characteristics, fail to quantify the contribution of slip systems, and do not consider thermo-mechanical coupling effects, resulting in large deviations between thermal shock simulation results and experimental results.
A random grain stacking model was constructed using the Voronoi mosaic algorithm, which was then given anisotropic properties. Combined with the thermo-mechanical coupling equation and the crystal plastic constitutive relation, thermal shock simulation of heat-resistant steel was performed. The stress distribution and slip system contribution were accurately captured through mesh generation.
The simulation error of stress distribution was less than 8%, which was consistent with the experimental results. It revealed the stress concentration at grain boundaries and the origin of damage, guided the optimization of the microstructure of heat-resistant steel, reduced the stress concentration at grain boundaries by 25%, and reduced the rate of non-stop accidents of the unit.
Smart Images
Figure CN121354751A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of thermal shock material simulation technology, and specifically discloses a method for modeling random grain stacking of heat-resistant steel and simulating thermal shock. Background Technology
[0002] Driven by the "dual carbon" target, thermal power units need to frequently participate in deep peak shaving. The heat-resistant steel of the boiler heating surface is subjected to severe alternating thermal shocks of "heating-cooling", which leads to complex micro-damage inside the material (such as grain boundary slip, dislocation proliferation, and structure degradation), and ultimately causes macroscopic failure.
[0003] Traditional materials simulation methods have the following limitations:
[0004] Macroscopic homogenization models treat materials as continuous homogeneous bodies, ignoring the effects of grain size, orientation, and grain boundary characteristics, and therefore cannot explain the origin of microscopic damage. For example, SA-213TP347H austenitic steel and 12Cr1MoVG pearlitic steel have significantly different thermal shock resistance due to their different crystal structures, but macroscopic models cannot quantify this difference.
[0005] Simplified grain model: Existing microscopic simulations mostly use simplified grain models with regular shapes such as cubes or random orientations, without considering the irregular morphology and spatial distribution characteristics of grains in real materials, resulting in large deviations between the stress distribution simulation results and experiments.
[0006] Single-physics simulations only consider the individual effects of mechanical loads or temperature fields, neglecting thermo-mechanical coupling effects. However, thermal shock damage is essentially the result of the interaction between thermal stress induced by temperature gradients and the material's microstructure; single-physics simulations cannot reflect the true evolution of damage.
[0007] Insufficient quantification of slip system contribution: The contribution law of slip on basal plane, cylindrical plane, and conical plane to shear stress is not clearly defined, making it difficult to reveal the microscopic mechanism of material fatigue damage. For example, in the martensitic lath structure of SA-213T91, cylindrical slip is the main source of plastic deformation.
[0008] While existing technologies involve simulation of polycrystalline materials, they do not optimize models for the thermal shock characteristics of heat-resistant steel, nor do they establish the correlation between grain anisotropy and macroscopic properties. Therefore, there is an urgent need for a simulation method that can accurately describe the microstructure, couple the thermo-mechanical field, and quantify the contribution of the slip system, to provide support for the thermal shock resistance design of heat-resistant steel. Summary of the Invention
[0009] To address the shortcomings and problems of current material simulation methods, this invention provides a method for modeling random grain stacking and simulating thermal shock in heat-resistant steel. This method mainly includes the following steps:
[0010] S1. Obtain the microstructure parameters of the target heat-resistant steel, including grain size distribution, crystal orientation distribution, and grain boundary characteristics;
[0011] S2. Based on the Voronoi mosaic algorithm, a random grain stacking model was constructed and optimized using Neper software. The spatial distribution of grains in the model conforms to a log-normal distribution.
[0012] S3. Based on the microstructure parameters of heat-resistant steel, temperature-related anisotropic property parameters are assigned to different grains, including thermophysical properties, mechanical properties and slip system parameters;
[0013] S4. Import the model into the simulation software and define the full-field crystal plastic constitutive relation, which includes dislocation slip, hardening model and thermo-mechanical coupling equation;
[0014] S5. Set thermal shock boundary conditions, including shock temperature range, heating / cooling rate and mechanical constraints;
[0015] S6. Run the simulation and output the distribution of intergranular thermal stress, accumulation of plastic strain, and contribution of slip system.
[0016] The above-mentioned method for modeling random grain stacking and simulating thermal shock in heat-resistant steel includes SA-213T91, SA-213TP347H or 12Cr1MoVG type heat-resistant steel.
[0017] The above-mentioned method for modeling random grain stacking and simulating thermal shock in heat-resistant steel, specifically includes the following steps in step S2:
[0018] (1) Voronoi polycrystalline structure generation: Using Neper software, based on the grain size distribution obtained step by step, crystal nuclei are randomly generated through the Poisson point process, and the three-dimensional polycrystalline structure is constructed using the Voronoi mosaic algorithm.
[0019] (2) Grain morphology optimization: The sharp edges in the initial model are iteratively optimized by Lloyd relaxation algorithm. The objective function is to minimize the ratio of grain surface area to volume. The grain boundary position is adjusted to make the grain volume more uniform and to eliminate sharp edges.
[0020] (3) Mesh generation: Import the model with optimized grain morphology into the simulation software for mesh processing.
[0021] The above-mentioned method for modeling and simulating the random grain stacking of heat-resistant steel is characterized in that: the total number of mesh elements in step (3) is 1 million to 5 million.
[0022] The above-mentioned modeling and thermal shock simulation method for random grain stacking of heat-resistant steel uses C3D8T thermo-mechanical coupling elements for mesh generation, implements local refinement of the element size ≤1μm in the grain boundary region, and sets the element size inside the grain to 5μm; performs mesh quality checks to ensure that the Jacobian determinant is ≥0.7 and the distortion is ≤15°, so as to avoid non-convergence of calculation due to mesh distortion.
[0023] The above-mentioned method for modeling random grain stacking and simulating thermal shock in heat-resistant steel includes the following thermophysical properties in step S3: thermal conductivity, specific heat capacity, and coefficient of thermal expansion; mechanical properties: elastic modulus and Poisson's ratio; and slip system parameters: critical shear stress of the base plane, cylindrical plane, and conical plane.
[0024] The above-mentioned method for modeling random grain stacking and simulating thermal shock in heat-resistant steel uses thermophysical property parameters based on experimental data of each target heat-resistant steel.
[0025] In the above-mentioned method for modeling random grain stacking and simulating thermal shock in heat-resistant steel, step S4 defines the crystal plastic constitutive relation through a user subroutine. The crystal plastic constitutive relation includes kinematic equations, flow rules, and hardening models.
[0026] The above-mentioned method for modeling random grain stacking of heat-resistant steel and simulating thermal shock includes the following thermal shock boundary conditions: temperature range, heating rate, cooling rate, mechanical constraints, and cyclic conditions.
[0027] The above-mentioned random grain stacking modeling and thermal shock simulation method for heat-resistant steel has a thermal shock edge temperature range of 450℃-800℃ for SA-213T91 and SA-213TP347H type heat-resistant steels; and a thermal shock temperature range of 450℃-750℃ for 12Cr1MoVG type heat-resistant steels.
[0028] Compared with the prior art, the beneficial effects of the present invention are:
[0029] This invention simulates material thermal shock by random grain stacking and anisotropic assignment, achieving a stress distribution simulation error of ≤8%. Verification shows good consistency with EBSD experimental results, accurately capturing microscopic features such as grain boundary stress concentration. This method comprehensively considers the influence of temperature on material properties and the heat generated by plastic work, realistically reflecting damage evolution under thermal shock and overcoming the limitations of single-physics-field simulation. The method quantifies the contribution of different slip systems, clarifies the correlation between grain boundary stress concentration and damage origin, and provides a basis for microstructure optimization. By refining grains, grain boundary stress concentration can be reduced by 25%, providing guidance for the selection and operational optimization of heat-resistant steel. Furthermore, this method can reduce the rate of unit outages and decrease the rate of leakage accidents at heated surfaces. Attached Figure Description
[0030] Figure 1 This is a flowchart of the method for modeling random grain stacking and simulating thermal shock in heat-resistant steel according to the present invention.
[0031] Figure 2 This is a flowchart illustrating the construction process of the random grain stacking model of this invention.
[0032] Figure 3 This is a flowchart illustrating the design process for the thermal shock simulation conditions of this invention. Detailed Implementation
[0033] To address the shortcomings of existing simulation methods, such as neglecting the influence of microstructure, insufficient thermo-mechanical coupling, and inadequate understanding of damage mechanisms, this invention provides a method for modeling random grain stacking and simulating thermal shock in heat-resistant steel. This method achieves accurate simulation of the microscopic damage behavior of materials under thermal shock, revealing the stress distribution between grains, the plastic slip mechanism, and fatigue damage patterns. It provides a theoretical basis for optimizing the microstructure and assessing the lifespan of heat-resistant steel. The invention will be further described below with reference to the accompanying drawings and specific embodiments.
[0034] Example 1: This example provides a method for modeling random grain stacking and simulating thermal shock in heat-resistant steel, such as... Figure 1 As shown, the method mainly includes the following contents.
[0035] Step 1: Obtain the microstructure parameters of the target heat-resistant steel
[0036] The microstructure of the target heat-resistant steel (e.g., SA-213T91, SA-213TP347H, or 12Cr1MoVG) was characterized by electron backscatter diffraction or scanning electron microscopy to obtain key parameters, including:
[0037] Grain size distribution: The diameters of at least 500 grains were statistically analyzed to obtain log-normal distribution curves; the grain diameters of 12Cr1MoVG are mainly distributed in the range of 12-18μm, the grain diameters of SA-213TP347H are concentrated in the range of 8-12μm, and the grain diameters of SA-213T91 are 10-15μm.
[0038] Crystal orientation: The orientation distribution of grains was analyzed by EBSD spectrum; the austenite grains of SA-213TP347H are mostly randomly oriented, while the ferrite matrix of 12Cr1MoVG may have slight texture due to the rolling process.
[0039] Grain boundary characteristics: Measure the grain boundary angle. Generally, large-angle grain boundaries account for ≥60% and the distribution of precipitated phases at grain boundaries; the grain boundary carbide size of SA-213T91 is about 0.5-1μm, and the pearlite lamellars of 12Cr1MoVG are prone to coarsening at grain boundaries.
[0040] Step 2: Construction and Optimization of Random Grain Stacking Model
[0041] Voronoi polycrystalline structure generation: Using Neper software, based on the grain size distribution in step S1, crystal nuclei are randomly generated through a Poisson point process, and a three-dimensional polycrystalline structure is constructed using the Voronoi mosaic algorithm. The core of the Voronoi algorithm is to divide the space through the perpendicular bisectors between crystal nuclei. Each grain is a convex polyhedron surrounding a crystal nucleus, mathematically expressed as: for the i-th grain, its geometric domain V i It consists of all the points closest to the crystal nucleus Si.
[0042]
[0043] This algorithm can generate irregular polyhedra that closely resemble the grain morphology of real materials.
[0044] Grain morphology optimization: The initial model was iteratively optimized using the Lloyd relaxation algorithm to adjust the grain boundary positions, making the grain volume more uniform and eliminating sharp edges. The edge radius was ≥2μm to avoid stress concentration distortion caused by geometric abrupt changes in the simulation. The optimized grain morphology is closer to the actual metallurgical observation results. The austenite grains of SA-213TP347H are nearly equiaxed, while the grains of 12Cr1MoVG are slightly irregular due to the distribution of pearlite lamellars.
[0045] Mesh generation: The model is imported into simulation software for mesh discretization using C3D8T. Local refinement is performed in grain boundary regions, with element sizes ≤1μm, to accurately capture stress and strain changes at grain boundaries. Mesh generation needs to balance computational accuracy and efficiency; the total number of elements is typically controlled between 1 million and 5 million.
[0046] Step 3: Assigning values to anisotropic properties
[0047] Temperature-dependent anisotropic parameters were assigned to different grains using a Python script, with parameter values based on experimental data.
[0048] Thermophysical properties: including thermal conductivity, specific heat capacity, and coefficient of thermal expansion, are all functions of temperature; the thermal conductivity of SA-213T91 increases with increasing temperature, being 30 W / (m·℃) at 20℃ and 39 W / (m·℃) at 750℃; the coefficient of thermal expansion of SA-213TP347H is significantly higher than that of martensitic steel, at 16.8 × 10⁻⁶ at 20℃. -6 / ℃, 18.5×10 at 750℃ -6 / ℃, which is one of the reasons for its large thermal stress at high temperatures.
[0049] Mechanical properties: Elastic modulus and Poisson's ratio change with temperature. The elastic modulus of SA-213T91 at 750℃ is only 48% of that at room temperature, decreasing from 210GPa to 101GPa; the elastic modulus of 12Cr1MoVG decreases to 143GPa at 650℃, reflecting the high-temperature softening characteristics of pearlitic structures.
[0050] Slip system parameters: Define the critical shear stress for different slip systems, including the base surface, cylindrical surface, and conical surface. Base surface of SA-213T91. The critical shear stress of slip is 80-100 MPa at room temperature, and drops to 60 MPa at 750℃. The critical shear stress of cylindrical slip of SA-213TP347H is relatively low, so it is preferentially activated at high temperature and becomes the main contributor to plastic deformation.
[0051] Step 4: Definition of the full-field crystal plastic constitutive relation
[0052] In simulation software, crystal plastic constitutive relations are defined through user subroutines, considering thermo-mechanical coupling effects. The main aspects considered are as follows:
[0053] Kinematic equations: The total deformation gradient is decomposed into elastic and plastic components. Elastic deformation consists of thermal expansion and elastic strain, while plastic deformation is contributed by dislocation slip, reflecting the plastic flow characteristics within the grains.
[0054] Flow law: The shear strain rate of a slip system generally follows a power law model, which is as follows:
[0055] γ (α) =γ0|τ (a) / g (a) | (1 / m) sgn(τ (a) )
[0056] In the formula: γ0=1s -1 As the reference strain rate, τ (a) To decompose shear stress, g (α) The value represents the slip resistance, and m = 0.2 is the strain rate sensitivity coefficient.
[0057] Hardening model: The evolution of slip resistance is described using Voce's hardening law.
[0058] g (α) =Σβh a β|γ(β)|,
[0059] In the formula, h a β is the hardening parameter, which includes self-hardening and latent hardening; h0 is the initial hardening modulus; g is the saturated slip resistance; a is the hardening exponent; and q is the latent hardening coefficient.
[0060] Thermo-mechanical coupling equation: Temperature change affects strain through the coefficient of thermal expansion, while plastic work is converted into heat, and the heat generation rate Q = τ (α) γ (α) Through the heat conduction equation Update the temperature field to achieve bidirectional coupling of heat and force.
[0061] Step 5: Setting Thermal Shock Boundary Conditions
[0062] Thermal boundary: Convection-radiation combined heat transfer is applied to the outer surface to simulate high-temperature flue gas impact, with a heat transfer coefficient h = 50-100 W / (m²). 2 ·℃), ambient temperature T∞=800℃;
[0063] The inner surface is in contact with steam, and the temperature is kept constant at T = 450℃; the initial temperature is 450℃, simulating the stable operating state before peak shaving.
[0064] Mechanical boundaries: The bottom is fixedly constrained (ux = uy = uz = 0) to simulate the connection between the pipe and the header; the rest of the surface is free, allowing thermal expansion and deformation, and avoiding stress distortion caused by excessive constraints.
[0065] Thermal shock conditions: The temperature range covers the critical failure temperature of the material, 450℃-800℃ for SA-213T91 and SA-213TP347H, and 450℃-750℃ for 12Cr1MoVG.
[0066] The cycle mode includes a heating time of 5-15 seconds and a cooling time of 5-15 seconds to simulate the rapid heating-cooling cycle during peak shaving; the cycle number is 300 times to capture the cumulative effect of fatigue damage.
[0067] Step Six: Simulation Result Output and Analysis
[0068] Intergranular stress distribution: There are significant stress concentration zones at grain boundaries, with stress values generally 1.5-2 times that inside the grain, and the stress concentration is more pronounced at large-angle grain boundaries; under impact at 725℃, the stress at large-angle grain boundaries of 12Cr1MoVG reaches 294MPa, while that inside the grain is about 150MPa, which is related to the strength weakening caused by the coarsening of pearlite lamellae at the grain boundaries.
[0069] PEEQ spatial distribution: The outer surface has a high value area at the geometric center, while the inner surface is concentrated at the corners. The PEEQ saturation value increases significantly under high temperature impact. The PEEQ of SA-213TP347H at 800℃ is 1.8 times that at 650℃, which reflects the accelerated plasticity accumulation characteristic of austenitic steel at supercritical temperature.
[0070] Slip system contribution law: basal plane Slip contributes more in the low-temperature region, cylindrical surface Slip becomes dominant in the high-temperature region, conical surface<c+a> Slip is activated only at high strains due to its high critical shear stress.
[0071] Failure mechanism characteristics: After the supercritical temperature, the weakening of grain boundaries leads to a surge in PEEQ, and the failure area expands from the center to the periphery, which is consistent with the center-originating failure observed in experiments.
[0072] Example 2: This example uses SA-213TP347H austenitic heat-resistant steel for the high-temperature superheater of a 600MW supercritical unit as the research object, and elaborates on the specific implementation steps of the present invention. This material has been in long-term service in an environment of 650-750℃ and needs to withstand the thermal shock load caused by frequent peak shaving. Its microstructure characteristics and thermal shock damage law are typical and representative. All experimental data and simulation parameters are derived from the actual unit test and laboratory verification results.
[0073] Step 1: Precise characterization of the microstructure parameters of SA-213TP347H
[0074] A 10mm × 10mm × 5mm sample was cut from the superheater tube, and after mounting, grinding, and polishing, the surface strain layer was removed by electrolytic polishing to ensure the accuracy of EBSD analysis. A 500μm × 500μm region was scanned using the EBSD detector of a FEIQuanta 650 scanning electron microscope at an accelerating voltage of 20kV and a step size of 0.2μm to obtain the data.
[0075] Grain size distribution: 1200 complete grains were statistically analyzed, with diameters ranging from 8 to 12 μm and an average of 10 μm, conforming to a log-normal distribution;
[0076] Crystal orientation: Austenite grains are randomly oriented, with an orientation difference angle between 15° and 60°;
[0077] Grain boundary characteristics: Large-angle grain boundaries account for 68%, small-angle grain boundaries account for 32%, and a small amount of Nb particles are precipitated at the grain boundaries.
[0078] Step 2: Construction and Optimization of a 3D Random Grain Stacking Model
[0079] Based on the microstructure parameters of the heat-resistant steel obtained in step one, a high-precision model is constructed using a three-step method: geometric modeling, morphological optimization, and mesh generation.
[0080] The number of crystal nuclei is set to 800, and the coordinates of the crystal nuclei are randomly generated using the Poisson point process, with a spatial distribution density of 16 nuclei / mm². 3 The initial polycrystalline structure was generated using the Voronoi mosaic algorithm, with the volume deviation of individual grains controlled within ±15%. For the sharp edges present in the initial model, the Lloyd relaxation algorithm was used for iterative optimization: the number of iterations was set to 50, and the objective function was to minimize the ratio of grain surface area to volume. After optimization, the grain edge radius was ≥2μm, and the average radius of curvature increased from 1.2μm to 3.5μm, consistent with the "near-equiaxed austenitic grain" morphology observed by SEM. The grain boundary energy distribution was calculated using a Python script to ensure that the deviation between the grain boundary energy of the optimized model and the experimental measurement was ≤5%.
[0081] The optimized model was imported into the simulation software and meshed as follows: C3D8T thermo-mechanical coupling elements were selected, local refinement was implemented in the grain boundary region, and the element size inside the grain was 5μm; mesh quality checks were performed to ensure that the Jacobian determinant was ≥0.7 and the distortion was ≤15° to avoid non-convergence of calculations due to mesh distortion; the final model contains 3.2 million elements and 4.8 million nodes, with an average of 4000 elements per grain, which can accurately capture the stress gradient at the grain boundary.
[0082] Step 3: Assigning values to anisotropic physical property parameters
[0083] Temperature-dependent anisotropic parameters are assigned to the model using Python scripts, with parameter values strictly following experimental measurements.
[0084] Thermal conductivity: 15 W / (m·℃) at 20℃, increasing to 25.5 W / (m·℃) at 750℃.
[0085] Specific heat capacity: 460 J / (kg·℃) at 20℃, 630 J / (kg·℃) at 750℃.
[0086] Coefficient of thermal expansion: 16.8 × 10⁻⁶ at 20℃ -6 / ℃, 18.5×10 at 750℃ -6 / ℃, and crystallographic anisotropy is taken into account.
[0087] Elastic modulus: 197 GPa at 20℃, 121 GPa at 750℃;
[0088] Poisson's ratio: 0.3-0.32;
[0089] Slip system parameters: 12 slip systems are defined, of which the critical shear stress of cylindrical slip is 90 MPa at room temperature and 70 MPa at 750℃; the critical shear stress of conical slip is 180 MPa at room temperature and 150 MPa at 750℃.
[0090] Step 4: Full-field crystal plastic constitutive relations and solution settings
[0091] In the simulation software, the crystal plastic constitutive model is defined through the user subroutine UMAT to achieve multi-physics coupling of thermo-mechanical-damage fields.
[0092] Kinematic equations: The total deformation gradient F is decomposed into the elastic part F e With plastic part F p Elastic strain includes contributions from thermal expansion;
[0093] Flow rule: Shear strain rate γ of slip system (α) =γ0|τ (α) / g (a) |^(1 / m)sgn(τ (a) ), where γ0=1s -1 m = 0.2;
[0094] Hardening model: using the improved Voce hardening law g (a) =g0+(g∞-g0)(1-exp(-h0γ) (α) / (g∞-g0))), where the initial slip resistance g0=90MPa, the saturation value g∞=450MPa, and the hardening modulus h0=2500MPa;
[0095] Potential hardening: The hardening coefficient q = 1.3 between different slip systems;
[0096] Define the heat generation rate Q = Στ (a) |γ (a) The thermal conductivity coefficient is taken as k(T) as defined in step three;
[0097] The time integration uses an implicit algorithm with time steps of 0.01s and 0.1s to ensure that the stress response at the moment of temperature change is captured;
[0098] The maximum number of iterations is set to 20, and the convergence tolerance is 1×10⁻⁶. -4 and 1×10 -3 Enable automatic time step adjustment; when the temperature change rate is >10℃ / s, automatically reduce the step size to 0.005s.
[0099] Step 5: Thermal Shock Boundary Conditions and Operating Case Design
[0100] Outer surface: Convection-radiation combined heat transfer is applied, with a heat transfer coefficient h = 80 W / (m²). 2 ·℃), ambient temperature T∞=800℃;
[0101] Inner surface: constant temperature 450℃, heat transfer coefficient h = 5000W / (m²) 2 ·℃);
[0102] Initial temperature: The model is uniformly 450℃.
[0103] Axial constraint: One end is fixed, while the other end is allowed to freely expand and contract axially, simulating the connection between a pipe and a header;
[0104] Radial constraint: The outer surface is unconstrained, allowing radial expansion;
[0105] Temperature range: 450℃→800℃→450℃;
[0106] Rate control: heating rate 30℃ / s, cooling rate 20℃ / s;
[0107] Cyclic mode: 30 seconds per cycle, 300 cycles in total;
[0108] Monitoring point setup: 30 monitoring points were set at the center of the outer surface of the model, the corners of the inner surface, and the dense grain boundary area to record the changes in temperature, stress, and PEEQ over time.
[0109] Step Six: Simulation Result Analysis and Engineering Verification
[0110] During thermal shock, the maximum stress occurs at large-angle grain boundaries, reaching 380 MPa at 800℃, which is 1.8 times the internal stress of the grains, verifying the theory that "grain boundaries are sensitive areas of stress concentration". The stress direction is distributed radially, and the stress on the outer surface is significantly higher than that on the inner surface, which is consistent with the actual stress state of the pipeline.
[0111] After 300 cycles, the maximum PEEQ value at the center of the outer surface is 0.0082, and at the corner of the inner surface it is 0.0065, showing a spatial distribution of "high on the outside and low on the inside".
[0112] The PEEQ growth rate at high temperature cycling is 2.3 times that at medium temperature cycling, indicating that supercritical temperature accelerates the accumulation of plastic damage.
[0113] Low temperature stage: Base surface and cylindrical surface slip work together, with cylindrical surface slip contributing 55%;
[0114] During the high-temperature stage: cylindrical slip becomes dominant, while conical slip is only activated in the stress concentration area, which is consistent with the phenomenon of "dense distribution of cylindrical slip bands at high temperature" observed by TEM.
[0115] The above description is only a preferred embodiment of the present invention and does not limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A method for modeling and thermal shock simulation of a random grain stack of heat resistant steel, characterized by: The method comprises the following steps: S1, obtaining microstructure parameters of target heat-resistant steel, the microstructure parameters including grain size distribution, crystal orientation distribution and grain boundary characteristics; S2, constructing a random grain stacking model and optimizing it by using Neper software based on a Voronoi tessellation algorithm, the spatial distribution of grains in the model conforming to a lognormal distribution; S3, giving different grains temperature-dependent anisotropic attribute parameters based on the microstructure parameters of the heat-resistant steel, the attributes including thermal physical properties, mechanical properties and slip system parameters; S4, importing the model into a simulation software and defining a full-field crystal plasticity constitutive relationship, the relationship including dislocation slip, hardening model and thermal-mechanical coupling equation; S5, setting thermal shock boundary conditions, including shock temperature range, heating / cooling rate and mechanical constraint; S6, running the simulation and outputting inter-grain thermal stress distribution, plastic strain accumulation and slip system contribution.
2. The heat-resistant steel random grain stack modeling and thermal shock simulation method according to claim 1, characterized in that: The target heat-resistant steel includes SA-213T91, SA-213TP347H or 12Cr1MoVG type heat-resistant steel.
3. The heat-resistant steel random grain stack modeling and thermal shock simulation method according to claim 1, characterized in that: Step S2 specifically comprises the following steps: (1) Voronoi polycrystalline structure generation: using Neper software, randomly generating crystal nuclei based on the grain size distribution obtained in step S1 by a Poisson point process, and constructing a three-dimensional polycrystalline structure by a Voronoi tessellation algorithm; (2) grain morphology optimization: iteratively optimizing the sharp corners existing in the initial model by a Lloyd relaxation algorithm, the objective function being to minimize the grain surface area to volume ratio, adjusting the grain boundary position to make the grain volume more uniform and eliminating sharp corners; (3) meshing: importing the grain morphology optimized model into a simulation software for meshing.
4. The heat-resistant steel random grain stack modeling and thermal shock simulation method according to claim 3, characterized in that: The total number of elements in step (3) is 1-5 million.
5. The heat-resistant steel random grain stack modeling and thermal shock simulation method according to claim 3, characterized in that: When meshing, C3D8T thermal-mechanical coupling elements are selected, the local encryption element size of the grain boundary area is ≤1 μm, the element size in the grain interior is set to 5 μm; mesh quality inspection is performed to ensure that the Jacobian determinant is ≥0.7 and the twist degree is ≤15°, avoiding calculation divergence caused by mesh distortion.
6. The heat-resistant steel random grain stack modeling and thermal shock simulation method according to claim 1, characterized in that: The thermal physical properties in step S3 include thermal conductivity, specific heat capacity and thermal expansion coefficient; the mechanical properties include elastic modulus and Poisson's ratio; and the slip system parameters include critical shear stress of basal plane, prism and pyramid.
7. The heat-resistant steel random grain stack modeling and thermal shock simulation method according to claim 6, characterized in that: The values of the thermal physical property parameters are based on experimental data of each target heat-resistant steel.
8. The heat-resistant steel random grain stack modeling and thermal shock simulation method according to claim 1, characterized in that: In step S4, the crystal plasticity constitutive relationship is defined by a user subroutine, the crystal plasticity constitutive relationship including kinematic equation, flow rule and hardening model.
9. The heat-resistant steel random grain stack modeling and thermal shock simulation method according to claim 1, characterized in that: The thermal shock boundary conditions include temperature range, heating rate, cooling rate, mechanical constraint and cycle working condition.
10. The heat-resistant steel random grain stack modeling and thermal shock simulation method according to claim 2, characterized in that: For SA-213T91 and SA-213TP347H type heat-resistant steel, the thermal shock temperature range is 450-800℃; and for 12Cr1MoVG type heat-resistant steel, the thermal shock temperature range is 450-750℃.