Method for constructing nuclear power steel hot manufacturing full-process grain evolution model based on high-throughput characterization and data driving

Abstract

According to the nuclear power steel hot manufacturing full-process grain evolution model construction method based on high-flux characterization and data driving, a sample of high-flux grain configuration is designed, global high-flux microscopic structure characterization is carried out, crystallographic information of mass micro-area structures is obtained, and parameterized expression suitable for describing each grain configuration is given. Through cross-scale multi-physical field modeling of a rigid viscoplastic finite element method and a crystal plasticity finite element method, the distribution rule and quantitative description of deformation energy storage of each micro-area of the large-size nuclear power austenitic steel under different thermal deformation process conditions are explored. And a data driving model of the thermal deformation-deformation energy storage-solid solution grain configuration relation of the austenitic steel for nuclear power is constructed, and an active regulation and control strategy and method for controlling coarse grains / mixed grains of the austenitic steel for nuclear power are formed. The method can provide theoretical and method support for accurate quantitative prediction of nuclear power key part thermal manufacturing full-process grain evolution and active control of abnormal grains.

Description

Technical Field

[0001] The present invention relates to the field of material science and engineering technology, and in particular to a method for constructing a grain evolution model for the entire process of thermal manufacturing of nuclear power steel based on high-throughput characterization and data-driven. Background Art

[0002] In the field of materials science, especially in the manufacturing process of key materials such as austenitic steel for nuclear power, grain size and uniformity are key factors that determine the final service performance of the product. Hot deformation (such as forging and rolling) and heat treatment have a significant impact on the grain configuration of the material. Although domestic and foreign scholars have conducted a lot of research on grain evolution, these studies often focus on the grain changes of specific materials at specific process stages (such as only in hot deformation or heat treatment processes), lacking a comprehensive analysis covering the entire thermal manufacturing process.

[0003] At present, the quantitative description of grain evolution mainly relies on phenomenological models and physical models. These models fail to fully consider the coupling effect of thermal deformation and heat treatment process, and are also significantly insufficient in coupling grain topological structure parameters and micro-area energy parameters. In addition, due to the limitations of the number of experimental samples and the combination of process parameters, it is difficult to fully reflect the complexity of grain evolution in actual production. In addition, the accuracy of physical models is limited by the accuracy of basic assumptions and the rationality of parameter selection, which often leads to large calculation errors.

[0004] Therefore, how to comprehensively consider the interactions between multiple processes, explore the intrinsic mechanisms from the energy perspective, and establish a data-driven full-process grain evolution model has become an urgent problem to be solved. Summary of the invention

[0005] In order to overcome the problem that the existing prediction models in the field of materials cannot consider the influence between multiple processes and cannot couple the grain topological structure parameters and micro-area energy parameters, in view of this, the present invention proposes a method for constructing a grain evolution model for the entire process of hot manufacturing of nuclear power steel based on high-throughput characterization and data-driven. By designing a sample with high-throughput grain configuration and performing global high-throughput microstructure characterization, a large amount of crystallographic information of micro-area organization is obtained, and a parametric expression suitable for describing each grain configuration is given to construct a mapping database of sample micro-area coordinates-thermal deformation grains-solid solution grains. Through cross-scale multi-physics field modeling of rigid-viscous-plastic finite element method and crystal plasticity finite element method, the distribution law and quantitative description of deformation energy storage in each micro-area of ​​large-size nuclear power austenitic steel under different hot deformation process conditions are explored. A data-driven model of the relationship between thermal deformation-deformation energy storage-solid solution grain configuration of nuclear power austenitic steel is constructed to form an active regulation strategy and method for controlling the coarse grain / mixed grain of nuclear power austenitic steel. The present invention can provide theoretical and methodological support for the accurate quantitative prediction of grain evolution and active control of abnormal grains in the entire process of thermal manufacturing of key nuclear power components, and is expected to break through the process control bottleneck of grain size and uniformity of nuclear power main pipeline forgings made of large austenitic steel.

[0006] To this end, the present invention provides the following technical solutions:

[0007] The present invention discloses a method for constructing a grain evolution model of the entire process of nuclear power steel thermal manufacturing based on high-throughput characterization and data-driven, comprising:

[0008] Several sets of hot deformation and heat treatment experiments were conducted on large-size gradient nuclear austenitic steel samples with a large strain range. High-throughput characterization was performed on the samples after hot deformation and heat treatment to obtain a large amount of grain crystallographic data under hot deformation and solid solution conditions.

[0009] Based on the massive crystallographic data of grains under the conditions of thermal deformation and solid solution, the grain configuration shape characteristics are digitally expressed to form a micro-area process parameter and grain configuration data set;

[0010] Based on the massive crystallographic data of grains under the conditions of thermal deformation and solid solution, the micro-area energy of large-scale samples under full-range thermal deformation is calculated to obtain the deformation energy storage corresponding to the grain configuration of each micro-area in the full range;

[0011] The micro-region process parameters, grain configuration data set and deformation energy storage are formed into an austenitic steel data set;

[0012] The austenitic steel data set is used to construct a data-driven model of thermal deformation-deformation energy storage-solid solution grain configuration of austenitic steel for nuclear power; the input of the model is micro-region process parameters and grain configuration data set, and the output of the model is the strain range generated by coarse grains / mixed grains at each deformation temperature; the model uses a clustering algorithm to identify the local domain of the austenitic steel data, uses Gaussian process regression to establish a corresponding local prediction model for each local domain, integrates each local model through a finite mixing mechanism, and uses ensemble learning to combine and weight each local model to obtain the final prediction result.

[0013] Furthermore, high-throughput experiments were conducted on large-scale gradient nuclear austenitic steel specimens with a large strain range, and experimental results of high-throughput characterization were obtained, including:

[0014] Take samples from forged samples of austenitic steel for nuclear power and use thermal / mechanical simulation test equipment to obtain the material's high-temperature stress-strain curve and thermophysical property parameters; construct the constitutive equation of austenitic steel for nuclear power and use it for finite element simulation of temperature and strain fields after thermal deformation;

[0015] A double-wedge specimen was designed, and thermal deformation experiments were carried out in the range of 950℃ to 1200℃ to obtain specimens with gradient strain in the long axis direction. Samples were cut along the long axis section of the thermally deformed specimens and then electron backscatter diffraction (EBSD) sample preparation was performed. The large-area stitching module was used to perform rapid and high-throughput acquisition of the samples to obtain grain information over the entire domain.

[0016] Furthermore, it also includes:

[0017] Based on the experimental results of high-throughput characterization, the grain evolution law of the whole process of thermal manufacturing of austenitic steel for nuclear power is analyzed.

[0018] Furthermore, based on the experimental results of high-throughput characterization, the grain evolution law of the whole process of thermal manufacturing of austenitic steel for nuclear power is analyzed, including:

[0019] Finite element simulation was carried out on the temperature field and strain field of the sample after thermal deformation to obtain the strain and temperature distribution along the long axis direction, and the strain and temperature contour diagrams in the entire range after thermal deformation were drawn to study the evolution law of the microstructure in the continuous strain range after thermal deformation and solid solution.

[0020] Furthermore, based on the experimental results, the grain configuration shape characteristics are digitally expressed to form a grain configuration data set, including:

[0021] According to the shape characteristics of the grain configuration after thermal deformation in the experimental results, each grain configuration is expressed parameterized;

[0022] Combining the parametric expression of each grain configuration and the sample micro-area coordinates, digitizing the two-dimensional topological structure of the grain configuration after thermal deformation, and forming a sample micro-area coordinate-thermal deformation grain configuration mapping database;

[0023] According to the characteristics of equiaxed grains after solid solution in the experimental results, the configuration and shape characteristics of the grains after solid solution are determined;

[0024] The sample micro-area coordinate-thermal deformation grain configuration mapping database and the grain configuration shape characteristics after solid solution are combined to form a mapping database of micro-area coordinates-thermal deformation grains-solid solution grains.

[0025] Furthermore, based on the experimental results, the micro-area energy of the large-scale sample under the full-range thermal deformation is calculated to obtain the deformation storage energy corresponding to the grain configuration of each micro-area in the full range, including:

[0026] The temperature and strain fields of large-size wedge-shaped austenitic steel for nuclear power under macroscopic thermal load conditions are calculated using the rigid-viscoplastic finite element method.

[0027] According to the high-throughput EBSD characterization results, a large-scale, cross-scale microstructure mean field model based on dislocation density is constructed. The thermal stress component, free parameters and average activation energy are introduced to calculate the energy barrier parameters that control the thermal motion of dislocations. The control equation of dislocation motion at high temperature is established, and then a large-scale, cross-scale crystal plasticity model based on dislocation density under high-temperature deformation conditions is constructed.

[0028] For each local micro-region uniformly selected in the large-size wedge block, the temperature and strain data of the local micro-region are extracted as the input of the crystal plasticity model to obtain the dislocation density of the local micro-region;

[0029] according to Calculate the deformation energy storage of the local micro-region, where ρ is the dislocation density, μ is the shear modulus, and b s is the Burgers vector; μ, b s is a constant;

[0030] The governing equation for dislocation motion under high temperature conditions is given by:

[0031]

[0032] The dislocation densities of each term in formula (1) correspond to the generation of monopole dislocation density and the reduction of dislocation density caused by dipole formation, respectively; where: is the time derivative of dislocation density, which indicates the rate of change of dislocation density with time, i.e., the evolution of dislocation density; di is the grain size; is the shear strain rate; v d is the diffusion coefficient associated with the dislocation; s is the mean free path; d is the dislocation slip distance;

[0033] The dislocation densities of each term in equation (2) correspond to the formation of dislocation dipoles, the spontaneous annihilation of dipoles, and the annihilation of dipoles due to dislocation climbing, respectively; where: and are the maximum and minimum values ​​of dislocation slip distance;

[0034] High temperature thermal stress and high temperature shear rate are obtained from equations (3) and (4):

[0035]

[0036]

[0037] in, represents the shear strain rate in the α dislocation system; ρ α represents the dislocation density in the dislocation system; v0 is the base velocity of the dislocation; ω0 is the number of times the dislocation attempts to cross the energy barrier per second during thermal activation; q is a constant related to the dislocation motion;

[0038] F0 is the activation free energy constant of dislocation motion; is an Arrhenius factor, which represents the probability that the system can overcome the energy barrier ΔF at a given temperature T; k B is the Boltzmann constant. τ α represents the total shear stress in the α-dislocation system; represents the shear stress determined by the dislocation density and microstructure of the material; represents the thermally activated stress term; represents the reference stress; T represents the temperature; p represents the empirical index of stress and thermal activation;

[0039] Based on the crystal plasticity constitutive equation of the crystal plasticity model, the deformation energy storage between adjacent micro-regions is obtained by linear interpolation, and then the deformation energy storage corresponding to the grain configuration of each micro-region in the entire domain is obtained.

[0040] Furthermore, the micro-region process parameters, grain configuration data set and deformation energy storage are formed into an austenitic steel data set, including:

[0041] The micro-area process parameters, grain configuration data set and deformation energy storage are integrated, and a multi-modal perturbation mechanism is used to randomly sample the data, add noise or artificially change it to generate different data modes.

[0042] Furthermore, it also includes: constructing an adaptive update module to incrementally update the prediction model, so that the model can be updated online according to newly obtained austenitic steel data.

[0043] Further, it also includes: model validation;

[0044] The model validation includes:

[0045] In the heat deformation temperature range of 950℃~1200℃ and the strain range of 0~1.39, heat deformation experiments and solid solution experiments with different process parameter combinations were carried out to obtain the grain size of the sample after solid solution. The actual value was compared with the model predicted value to verify the accuracy of the model and calibrate the model.

[0046] Furthermore, it also includes:

[0047] The model is used to determine the strain range generated by coarse grains / mixed crystals at each deformation temperature, draw a three-dimensional process range diagram of abnormal grain growth, clarify the relationship between abnormal grain growth and hot deformation process parameters and deformation energy storage, explore the correlation between the physical mechanism of coarse grain / mixed crystal formation and process parameters, establish a safe processing range to avoid abnormally grown grains, and regulate the abnormal grains in the entire process of hot manufacturing of austenitic steel for nuclear power based on the safe processing area.

[0048] Advantages and positive effects of the present invention:

[0049] (1) In the present invention, by designing high-throughput samples and performing global high-throughput microstructural characterization, a large amount of crystallographic information of microstructures is obtained, and these data provide a solid foundation for subsequent modeling and analysis; further, through a data-driven fusion modeling method, a data-driven model of the relationship between thermal deformation-deformation energy storage-solid solution grain configuration of austenitic steel for nuclear power is constructed, which can comprehensively consider the influence of multiple manufacturing processes on the grain evolution of the material, effectively describe and predict the grain evolution of the material under different process conditions, and improve the accuracy and practicality of the model prediction.

[0050] (2) In the present invention, a systematic data management method is established by constructing a mapping database of sample micro-area coordinates-thermal deformation grains-solid solution grains, which provides strong support for material property prediction and process optimization.

[0051] (3) In the present invention, a cross-scale multi-physics field modeling scheme integrating the rigid-viscoplastic finite element method and the crystal plasticity finite element method is proposed, which provides an effective means for studying the deformation energy storage corresponding to the grain configuration of each micro-region within the entire domain of a large-scale specimen. It can deeply explore the distribution law of deformation energy storage in each micro-region of large-scale austenitic steel for nuclear power under different thermal deformation process conditions and make a quantitative description.

[0052] (4) In the present invention, based on model analysis, an active control strategy and method for controlling the coarse grain / mixed grain of austenitic steel for nuclear power is proposed, which provides a scientific basis for improving material uniformity and performance. The present invention can provide theoretical and methodological support for the accurate quantitative prediction of grain evolution and active control of abnormal grains in the whole process of thermal manufacturing of key nuclear power components, and is expected to break through the process control bottleneck of grain size and uniformity of nuclear power main pipeline forgings made of large austenitic steel, and improve the safety and reliability of nuclear power equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0054] Figure 1 It is a flowchart of a method for constructing a full-process grain evolution model for thermal steel for nuclear power based on high-throughput characterization and data-driven in an embodiment of the present invention;

[0055] Figure 2 A flowchart for obtaining material parameters and high-throughput experimental characterization in embodiments of the present invention;

[0056] Figure 3 A flow chart for establishing a grain characteristic parameter extraction and grain configuration mapping database in an embodiment of the present invention;

[0057] Figure 4 A flow chart for calculating the energy of micro-zone under thermal deformation in the entire domain in an embodiment of the present invention;

[0058] Figure 5 It is a schematic diagram of the whole process grain evolution model of the thermal manufacturing of nuclear power steel used in the data-driven embodiment of the present invention;

[0059] Figure 6 This is a flowchart of abnormal grain regulation in an embodiment of the present invention. DETAILED DESCRIPTION

[0060] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0061] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0062] like Figure 1 As shown, in an embodiment of the present invention, a method for constructing a grain evolution model for the entire process of hot manufacturing of nuclear power steel based on high-throughput characterization and data-driven includes:

[0063] S1. Conduct several sets of hot deformation and heat treatment experiments on large-size gradient nuclear power austenitic steel samples with a large strain range, perform high-throughput characterization on the samples after hot deformation and heat treatment, and obtain massive grain crystallographic data under hot deformation and solid solution conditions;

[0064] In specific implementation, Figure 2 As shown, material parameter acquisition and high-throughput experimental characterization, including:

[0065] Sampling was done on forged specimens of austenitic steel for nuclear power, and the high-temperature stress-strain curve and thermophysical property parameters of the material were obtained using thermal / mechanical simulation test equipment. The constitutive equation of austenitic steel for nuclear power was constructed for finite element simulation of temperature and strain fields after thermal deformation. Double-wedge specimens were designed, and thermal deformation experiments were carried out in the range of 950°C to 1200°C to obtain specimens with gradient strain in the long-axis direction. Specimens were cut along the long-axis section of the thermally deformed specimens and then subjected to electron backscatter diffraction EBSD sample preparation. The large-area stitching module was used to quickly and high-throughput collect the specimens to obtain grain information over the entire domain.

[0066] Furthermore, based on the experimental results of high-throughput characterization, the grain evolution law of the whole process of hot manufacturing of austenitic steel for nuclear power can be analyzed. Specifically, the temperature field and strain field of the sample after hot deformation are simulated by finite element, the strain and temperature distribution along the long axis direction are obtained, and the strain and temperature contour map in the whole domain after hot deformation is drawn, and the microstructure evolution law in the continuous strain range after hot deformation and solid solution is studied (the dependence of the recrystallization mechanism on strain at different deformation temperatures; the quantitative expression of the critical conditions for the start of different recrystallization types; the distribution law of grain characteristics after solid solution; the correlation between hot-deformed grains and solid solution grains).

[0067] S2. Based on the massive crystallographic data of grains under thermal deformation and solid solution conditions, the grain configuration shape characteristics are digitally expressed to form micro-area process parameters and grain configuration data sets;

[0068] In specific implementation, Figure 3 As shown, a database for grain feature parameter extraction and grain configuration mapping is established, including:

[0069] S21. Based on the shape characteristics of the grain configuration after thermal deformation in the experimental results, each grain configuration is expressed parameterized;

[0070] S22, combining the parametric expression of each grain configuration and the sample micro-area coordinates, digitizing the two-dimensional topological structure of the grain configuration after thermal deformation, and forming a sample micro-area coordinate-thermal deformation grain configuration mapping database;

[0071] S23. According to the characteristics of equiaxed grains after solid solution in the experimental results, determine the configuration and shape characteristics of the grains after solid solution;

[0072] S24. Combining the sample micro-area coordinate-thermal deformation grain configuration mapping database and the grain configuration shape characteristics after solid solution, a mapping database of micro-area coordinate-thermal deformation grain-solid solution grain is formed.

[0073] S3. Based on the massive crystallographic data of grains under thermal deformation and solid solution conditions, the micro-area energy of large-scale samples under full-range thermal deformation is calculated to obtain the deformation energy storage corresponding to the grain configuration of each micro-area in the full range;

[0074] In specific implementation, Figure 4 As shown in the figure, the micro-area energy of a large-scale specimen under full-range thermal deformation is calculated, including:

[0075] S31. Use rigid-viscoplastic finite element method to calculate the temperature and strain fields of large-size wedge-shaped austenitic steel for nuclear power under macroscopic thermal load conditions;

[0076] S32. Based on the high-throughput EBSD characterization results, a large-scale, cross-scale microstructure mean field model based on dislocation density is constructed. The thermal stress component, free parameters and average activation energy are introduced to calculate the energy barrier parameter △F that controls the thermal motion of dislocations, and the control equation of dislocation motion at high temperature is established. Then, a large-scale, cross-scale crystal plasticity model based on dislocation density under high-temperature deformation conditions is constructed;

[0077] Among them, the crystal plasticity model refers to a theoretical framework used to describe the plastic deformation behavior of materials at the microscopic scale (such as crystal structure, dislocation movement, etc.). The crystal plasticity model takes into account the microstructural characteristics and deformation mechanism of the material, and usually includes the description of slip systems, twins, phase transitions and other phenomena. It provides a basis for understanding the anisotropy, hardening behavior and deformation mechanism of materials.

[0078] S33, for each local micro-region uniformly selected in the large-size wedge block, extracting temperature and strain data of the local micro-region as input of the crystal plasticity model to obtain the dislocation density of the local micro-region;

[0079] S34, according to Calculate the deformation energy storage of the local micro-region, where ρ is the dislocation density, μ is the shear modulus, and b s is the Burgers vector; μ, b s is a constant;

[0080] The governing equation for dislocation motion under high temperature conditions is given by:

[0081]

[0082] The dislocation densities of each term in formula (1) correspond to the generation of monopole dislocation density and the reduction of dislocation density caused by dipole formation, respectively; where: is the time derivative of dislocation density, which indicates the rate of change of dislocation density with time, i.e., the evolution of dislocation density; di is the grain size; is the shear strain rate; b s is the Burgers vector; v d is the diffusion coefficient associated with the dislocation; s is the mean free path; d is the dislocation slip distance;

[0083] The dislocation densities of each term in equation (2) correspond to the formation of dislocation dipoles, the spontaneous annihilation of dipoles, and the annihilation of dipoles due to dislocation climbing, respectively; where: and are the maximum and minimum values ​​of dislocation slip distance;

[0084] High temperature thermal stress and high temperature shear rate are obtained from equations (3) and (4):

[0085]

[0086] in, represents the shear strain rate in the α dislocation system; ρ α represents the dislocation density in the dislocation system; v0 is the base velocity of the dislocation; ω0 is the number of times the dislocation attempts to cross the energy barrier per second during the thermal activation process; q is a constant related to the dislocation motion; F0 is the activation free energy constant of the dislocation motion; is an Arrhenius factor, which represents the probability that the system can overcome the energy barrier ΔF at a given temperature T; k B is the Boltzmann constant. τ α represents the total shear stress in the α-dislocation system; represents the shear stress determined by the dislocation density and microstructure of the material; represents the thermally activated stress term; represents the reference stress; T represents the temperature; and p represents the empirical index of stress and thermal activation.

[0087] S35. Based on the crystal plasticity constitutive equation of the crystal plasticity model, the deformation energy storage between adjacent micro-regions is obtained by linear interpolation, and then the deformation energy storage corresponding to the grain configuration of each micro-region in the entire domain is obtained, which solves the calculation problem of large-scale models.

[0088] Among them, the crystal plasticity constitutive equation is a mathematical expression of the crystal plasticity model, which specifies the stress-strain relationship of the material under external loading. It is usually given in the form of an equation, describing the flow behavior, hardening or softening characteristics of the material under different conditions.

[0089] In an embodiment of the present invention, after establishing a crystal plasticity model, the deformation energy storage in a micro-region is calculated, and the fast Fourier transform equation method and the augmented Lagrangian algorithm are used to accurately and quickly solve the crystal plasticity constitutive equation, and then the constitutive equation is used to expand the deformation energy storage in the micro-region to the macroscopic level, forming a cross-scale modeling method.

[0090] S4, forming an austenitic steel data set by transforming micro-region process parameters, grain configuration data set and deformation energy storage;

[0091] Based on the above steps, the micro-area process parameters, grain configuration data set and deformation energy storage are obtained. The micro-area process parameters, deformation energy storage and grain configuration data set are integrated, and a multimodal perturbation mechanism is used to randomly sample the data, add noise or artificially change them to generate different data modes to increase the diversity of training data.

[0092] S5. Use austenitic steel dataset to build a data-driven model of thermal deformation-deformation energy storage-solid solution grain configuration of austenitic steel for nuclear power;

[0093] The input of the model is the micro-area process parameters and grain configuration data set, and the output of the model is the strain range generated by coarse grains / mixed grains at each deformation temperature; the model uses a clustering algorithm to identify the local domain of austenitic steel data, and uses Gaussian process regression to establish a corresponding local prediction model for each local domain. The local models are integrated through a finite mixing mechanism, and ensemble learning is used to combine and weight the local models to obtain the final prediction result.

[0094] Furthermore, an adaptive update module can be constructed to incrementally update the prediction model, so that the model can be updated online according to newly obtained austenitic steel data.

[0095] Furthermore, a verification experiment was designed to carry out hot deformation experiments and solutionizing experiments with different combinations of process parameters within the hot deformation temperature range of 950℃~1200℃ and the strain range of 0~1.39, to obtain the grain size of the sample after solutionizing, and to compare the actual value with the model predicted value to verify the accuracy of the hot deformation process parameters (temperature, strain)-deformation energy storage-grain configuration mapping model, and to calibrate the model to realize the prediction of grain evolution in the entire process of hot manufacturing of austenitic steel for nuclear power.

[0096] In the above embodiment, high-throughput global characterization technology and cross-scale modeling and simulation are combined to construct a data-driven model for grain prediction of austenitic steel for nuclear power in the hot deformation-solution process. The technical advantages are:

[0097] (1) Aiming at the actual hot deformation process characteristics of austenitic steel for nuclear power, we innovatively proposed high-throughput experiments and full-domain EBSD characterization to obtain the crystallographic feature information of grains corresponding to a large number of process parameter combinations, providing a big data foundation for the establishment of the mapping model.

[0098] (2) Considering the mutual influence among multiple processes, exploring the internal causes from the perspective of energy, and quantifying the relationship between hot deformation process parameters and grain structure and deformation energy storage, it is expected to form a new method to predict grain evolution under complex multi-physical fields and multiple processes, breaking through the bottleneck of related theoretical research.

[0099] (3) By coupling the grain topological structure parameters with the micro-area energy parameters, a data-driven thermal deformation-solid solution grain prediction model is constructed. The relationship between abnormal grain growth and thermal deformation process parameters such as deformation temperature and strain as well as deformation energy storage is clarified through the model.

[0100] In another embodiment, the data driven model constructed in the above embodiment can be used for abnormal grain control, specifically, as follows: Figure 5 As shown, the model is used to determine the strain range of coarse grains / mixed crystals at each deformation temperature, draw a three-dimensional process range diagram of abnormal grain growth, clarify the relationship between abnormal grain growth and hot deformation process parameters and deformation energy storage, explore the correlation between the physical mechanism of coarse grain / mixed crystal formation and process parameters, establish a safe processing range to avoid abnormally grown grains, and regulate the abnormal grains in the entire process of hot manufacturing of austenitic steel for nuclear power based on the safe processing area.

[0101] In the above embodiment, based on the constructed data-driven model for grain prediction of hot deformation-solid solution process of austenitic steel for nuclear power, an active control strategy for coarse-grained / mixed-grained microstructure defects in the full-process hot manufacturing process of austenitic steel for nuclear power is formed, providing theoretical and methodological support for the accurate quantitative prediction of grain evolution and abnormal grain growth control in the full-process hot manufacturing process of large forgings.

[0102] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for constructing a grain evolution model for the entire process of nuclear power steel thermal manufacturing based on high-throughput characterization and data-driven, characterized in that: include: Several sets of hot deformation and heat treatment experiments were conducted on large-size gradient nuclear austenitic steel samples with a large strain range. High-throughput characterization was performed on the samples after hot deformation and heat treatment to obtain a large amount of grain crystallographic data under hot deformation and solid solution conditions. Based on the massive crystallographic data of grains under the conditions of thermal deformation and solid solution, the grain configuration shape characteristics are digitally expressed to form a micro-area process parameter and grain configuration data set; Based on the massive crystallographic data of grains under the conditions of thermal deformation and solid solution, the micro-area energy of large-scale samples under full-range thermal deformation is calculated to obtain the deformation energy storage corresponding to the grain configuration of each micro-area in the full range; The micro-region process parameters, grain configuration data set and deformation energy storage are formed into an austenitic steel data set; The austenitic steel data set is used to construct a data-driven model of thermal deformation-deformation energy storage-solid solution grain configuration of austenitic steel for nuclear power; the input of the model is micro-region process parameters and grain configuration data set, and the output of the model is the strain range generated by coarse grains / mixed grains at each deformation temperature; the model uses a clustering algorithm to identify the local domain of the austenitic steel data, uses Gaussian process regression to establish a corresponding local prediction model for each local domain, integrates each local model through a finite mixing mechanism, and uses ensemble learning to combine and weight each local model to obtain the final prediction result.

2. According to claim 1, a method for constructing a grain evolution model for the entire process of hot manufacturing of nuclear power steel based on high-throughput characterization and data-driven is characterized in that: High-throughput experiments were performed on large-scale gradient nuclear austenitic steel specimens with a large strain range, and experimental results of high-throughput characterization were obtained, including: Take samples from forged samples of austenitic steel for nuclear power and use thermal / mechanical simulation test equipment to obtain the material's high-temperature stress-strain curve and thermophysical property parameters; construct the constitutive equation of austenitic steel for nuclear power and use it for finite element simulation of temperature and strain fields after thermal deformation; A double-wedge specimen was designed, and thermal deformation experiments were carried out in the range of 950℃ to 1200℃ to obtain specimens with gradient strain in the long axis direction. Samples were cut along the long axis section of the thermally deformed specimens and then electron backscatter diffraction (EBSD) sample preparation was performed. The large-area stitching module was used to perform rapid and high-throughput acquisition of the samples to obtain grain information over the entire domain.

3. The method for constructing a grain evolution model for the entire process of nuclear power steel thermal manufacturing based on high-throughput characterization and data-driven according to claim 2 is characterized in that: Also includes: Based on the experimental results of high-throughput characterization, the grain evolution law of the whole process of thermal manufacturing of austenitic steel for nuclear power is analyzed.

4. The method for constructing a grain evolution model for the entire process of nuclear power steel thermal manufacturing based on high-throughput characterization and data-driven according to claim 3 is characterized in that: Based on the experimental results of high-throughput characterization, the grain evolution law of the whole process of thermal manufacturing of austenitic steel for nuclear power is analyzed, including: Finite element simulation was carried out on the temperature field and strain field of the sample after thermal deformation to obtain the strain and temperature distribution along the long axis direction, and the strain and temperature contour diagrams in the entire range after thermal deformation were drawn to study the evolution law of the microstructure in the continuous strain range after thermal deformation and solid solution.

5. The method for constructing a grain evolution model for the entire process of nuclear power steel thermal manufacturing based on high-throughput characterization and data-driven according to claim 2, characterized in that: Based on the experimental results, the grain configuration shape characteristics are digitally expressed to form a grain configuration data set, including: According to the shape characteristics of the grain configuration after thermal deformation in the experimental results, each grain configuration is expressed parameterized; Combining the parametric expression of each grain configuration and the sample micro-area coordinates, digitizing the two-dimensional topological structure of the grain configuration after thermal deformation, and forming a sample micro-area coordinate-thermal deformation grain configuration mapping database; According to the characteristics of equiaxed grains after solid solution in the experimental results, the configuration and shape characteristics of the grains after solid solution are determined; The sample micro-area coordinate-thermal deformation grain configuration mapping database and the grain configuration shape characteristics after solid solution are combined to form a mapping database of micro-area coordinates-thermal deformation grains-solid solution grains.

6. The method for constructing a grain evolution model for the entire process of nuclear power steel thermal manufacturing based on high-throughput characterization and data-driven according to claim 2 is characterized in that: Based on the experimental results, the micro-area energy of the large-scale specimen under full-range thermal deformation is calculated, and the deformation storage energy corresponding to the grain configuration of each micro-area in the full range is obtained, including: The temperature and strain fields of large-size wedge-shaped austenitic steel for nuclear power under macroscopic thermal load conditions are calculated using the rigid-viscoplastic finite element method. According to the high-throughput EBSD characterization results, a large-scale, cross-scale microstructure mean field model based on dislocation density is constructed. The thermal stress component, free parameters and average activation energy are introduced to calculate the energy barrier parameters that control the thermal motion of dislocations. The control equation of dislocation motion at high temperature is established, and then a large-scale, cross-scale crystal plasticity model based on dislocation density under high-temperature deformation conditions is constructed. For each local micro-region uniformly selected in the large-size wedge block, the temperature and strain data of the local micro-region are extracted as the input of the crystal plasticity model to obtain the dislocation density of the local micro-region; according to Calculate the deformation energy storage of the local micro-region, where ρ is the dislocation density, μ is the shear modulus, and b s is the Burgers vector; μ, b s is a constant; The governing equation for dislocation motion under high temperature conditions is given by: The dislocation densities of each term in formula (1) correspond to the generation of monopole dislocation density and the reduction of dislocation density caused by dipole formation, respectively; where: is the time derivative of dislocation density, which indicates the rate of change of dislocation density with time, i.e., the evolution of dislocation density; di is the grain size; is the shear strain rate; v d is the diffusion coefficient associated with dislocation; Λ s is the mean free path; d is the dislocation slip distance; The dislocation densities of each term in equation (2) correspond to the formation of dislocation dipoles, the spontaneous annihilation of dipoles, and the annihilation of dipoles due to dislocation climbing, respectively; where: and are the maximum and minimum values ​​of dislocation slip distance; High temperature thermal stress and high temperature shear rate are obtained from equations (3) and (4): in, represents the shear strain rate in the α dislocation system; ρ α represents the dislocation density in the dislocation system; v0 is the base velocity of the dislocation; ω0 is the number of times the dislocation attempts to cross the energy barrier per second during thermal activation; q is a constant related to the dislocation motion; F0 is the activation free energy constant of dislocation motion; is an Arrhenius factor, which represents the probability that the system can overcome the energy barrier ΔF at a given temperature T; k B is the Boltzmann constant. τ α represents the total shear stress in the α-dislocation system; represents the shear stress determined by the dislocation density and microstructure of the material; represents the thermally activated stress term; represents the reference stress; T represents the temperature; p represents the empirical index of stress and thermal activation; Based on the crystal plasticity constitutive equation of the crystal plasticity model, the deformation energy storage between adjacent micro-regions is obtained by linear interpolation, and then the deformation energy storage corresponding to the grain configuration of each micro-region in the entire domain is obtained.

7. The method for constructing a grain evolution model for the entire process of nuclear power steel thermal manufacturing based on high-throughput characterization and data-driven according to claim 1 is characterized in that: The micro-region process parameters, grain configuration data set and deformation energy storage form an austenitic steel data set, including: The micro-area process parameters, grain configuration data set and deformation energy storage are integrated, and a multi-modal perturbation mechanism is used to randomly sample the data, add noise or artificially change it to generate different data modes.

8. The method for constructing a grain evolution model for the entire process of nuclear power steel thermal manufacturing based on high-throughput characterization and data-driven according to claim 1 is characterized in that: Also includes: An adaptive update module is constructed to incrementally update the prediction model, so that the model can be updated online according to the newly obtained austenitic steel data.

9. A method for constructing a grain evolution model for the entire process of hot manufacturing of nuclear power steel based on high-throughput characterization and data-driven according to any one of claims 1 to 8, characterized in that: Also included: model validation; The model validation includes: In the heat deformation temperature range of 950℃~1200℃ and the strain range of 0~1.39, heat deformation experiments and solid solution experiments with different process parameter combinations were carried out to obtain the grain size of the sample after solid solution. The actual value was compared with the model predicted value to verify the accuracy of the model and calibrate the model.

10. A method for constructing a grain evolution model for the entire process of hot manufacturing of nuclear power steel based on high-throughput characterization and data-driven according to any one of claims 1 to 8, characterized in that: Also includes: The model is used to determine the strain range generated by coarse grains / mixed crystals at each deformation temperature, draw a three-dimensional process range diagram of abnormal grain growth, clarify the relationship between abnormal grain growth and hot deformation process parameters and deformation energy storage, explore the correlation between the physical mechanism of coarse grain / mixed crystal formation and process parameters, establish a safe processing range to avoid abnormally grown grains, and regulate the abnormal grains in the entire process of hot manufacturing of austenitic steel for nuclear power based on the safe processing area.

Images