A method and system for constructing an alloy precipitate phase model based on bit orientation relationship

Abstract

The application provides a method and system for constructing an alloy precipitate phase model based on a positional relationship. The method comprises: obtaining a parameter set of a positional relationship between an alloy precipitate phase and a matrix; preprocessing and screening the positional relationship and crystal face of the precipitate phase to determine an optimal model construction positional relationship; performing rotation and combination operations on the crystal face, and then performing optimization on interface atoms to obtain an alloy-precipitate phase atomic model; and by adjusting parameters in the parameter set, a series of alloy-precipitate phase models with different shapes, sizes and orientations are obtained in a high-throughput manner. The application can significantly improve the construction efficiency of the alloy-precipitate phase model, and the gradient and systematic model generation mode enables users to quickly obtain a model library covering different precipitate phase characteristic parameters, thereby providing an efficient, reproducible and thermodynamically consistent atomic model construction scheme for multi-scale simulation and performance prediction of alloy materials, and significantly shortening the new material research and development cycle.

Description

Technical Field

[0001] This invention belongs to the interdisciplinary field of materials science and computer intelligent model construction, specifically relating to a method and system for high-throughput generation of precipitate-matrix composite atomic models for materials computational simulations by utilizing the orientational relationship between precipitates and the matrix in an alloy. Specifically, it relates to a method for constructing atomic-scale models based on crystal orientational relationships to simulate the strengthening mechanism, interface behavior, and service performance of precipitates in alloys; a method for providing initial atomic configurations for molecular dynamics simulations, particularly for alloy systems containing precipitates; a method for batch generation of a series of models with different precipitate characteristic parameters through parametric scanning and automated processes, serving materials database construction and multi-scale performance prediction; and a method for obtaining, verifying, and converting the crystal plane parallelism and crystal orientation parallelism between precipitates and the matrix into three-dimensional modeling parameters. Background Technology

[0002] Precipitated phases are important microstructural components of metals and their alloys. Their type, size, distribution density, and orientation relationship with the matrix play a crucial role in the mechanical and corrosion properties of materials. The orientation relationship determines the degree of coherence between the precipitated phase and the matrix interface, typically categorized into coherent, semi-coherent, and incoherent interfaces. Coherent interfaces have the lowest interfacial energy and significantly hinder dislocation movement; while incoherent interfaces may become channels for preferential corrosion or element segregation. Therefore, accurately obtaining and describing the orientation relationship between the precipitated phase and the matrix is ​​a core prerequisite for understanding the evolution mechanism of alloy properties.

[0003] However, traditional experimental methods have significant limitations in identifying the specific roles of precipitates in mechanical and corrosion aspects. While existing quantitative assessment methods based on empirical formulas can estimate the contribution of precipitates to material properties to some extent, the calculation results are heavily simplified. For example, empirical formulas often neglect the anisotropy of precipitates and their elastic mismatch effect with the matrix, fail to fully consider the differentiated influence of orientation relationships on dislocation slip and crack initiation, and struggle to describe microscopic features such as atomic arrangement, dislocation networks, and elemental segregation at interfaces. Therefore, existing methods often cannot intuitively and accurately reveal the true contribution of precipitates to material properties.

[0004] Existing technologies for constructing alloy-precipitate atomic models suffer from the following shortcomings: First, modeling efficiency is low, manual operation is cumbersome and prone to errors; second, orientation relationships are not accurately handled, requiring manual calculation of rotation matrices, which demands a high level of crystallographic knowledge; third, there is a lack of high-throughput generation capability, making it difficult to generate a series of models with different parameter combinations in batches; and fourth, interface optimization is insufficient, with models containing non-physical high-energy configurations that affect simulation stability. Therefore, there is an urgent need for a solution that can utilize alloy precipitation structure information to automatically generate precipitate-matrix composite atomic models in a high-throughput manner to address the aforementioned technical challenges.

[0005] In addition, while existing open-source modeling tools (such as the lattice command in ASE, Atomsk, and LAMMPS) can construct basic crystal supercells, they have the following key limitations: (1) Orientation relation processing relies on manual calculation of the rotation matrix by the user and only supports input of single crystal plane parallel relations, which cannot simultaneously satisfy the joint constraints of three crystal direction families; (2) It cannot automatically search for the best matching crystal plane combination from experimental orientation relation data (such as transmission electron microscopy diffraction spectra), which requires manual trial and error; (3) High-throughput modeling can only traverse the geometric parameters such as phase size and shape, but cannot incorporate orientation relation types into automated scanning variables; (4) The removal of overlapping atoms in the interface region is mainly based on simple distance threshold judgment, lacking the ability to intelligently identify and correct non-physical high-energy configurations. Therefore, there is an urgent need for a fully automated solution that can integrate automatic analysis of crystal orientation relations, multi-crystal plane joint constraint optimization, high-throughput parameterized scanning, and intelligent pre-optimization of interface geometry. Summary of the Invention

[0006] This invention addresses the problems of low modeling efficiency, inaccurate handling of orientation relationships, lack of high-throughput generation capability, and insufficient optimization in existing technologies for constructing atomic models of alloys containing precipitates. It provides a solution that can automatically generate precipitate-matrix composite atomic models for material simulation calculations in high throughput using alloy precipitate structure information.

[0007] To solve the above-mentioned technical problems, the present invention provides the following specific technical solutions: On one hand, the present invention provides a method for constructing an alloy precipitate model based on orientation relationships, the method comprising: S1. Input or obtain the crystal structure parameters of the precipitated phase, the crystal structure parameters of the matrix, the orientation relationship, and the geometric parameters of the precipitated phase in the target alloy system to form the parameter set Parameters; S2. Based on the matrix crystal structure parameters, construct a matrix supercell containing a predetermined number of atoms, and calculate the matrix supercell lattice basis vector matrix M. m and matrix atomic coordinates; S3. Construct the lattice basis matrix M of the precipitated phase based on the crystal structure parameters and orientation relationships of the precipitated phase. p The supercell of the precipitated phase and the atomic coordinates of the precipitated phase; S4, Based on the matrix supercell basis vector matrix M m , Precipitation phase lattice basis matrix M p Based on the aforementioned orientation relationship, three candidate crystal plane families are constructed using a three-level hierarchical screening method. F 1. F 2. F 3; wherein the directions of the matrix supercell basis vectors and the directions of the precipitated phase supercell basis vectors satisfy the aforementioned orientation relationship; S5. For the three types of candidate crystal plane families constructed in step S4 F 1. F 2. F 3. Based on the constraints of three aspects—crystal plane normal deviation factor, crystal orientation orthogonality preservation factor, and interface coherence factor—the optimal model for constructing orientation relationships is determined. S6. Based on the optimal model determined in step S5, construct the orientation relationship, generate the matrix crystal planes and precipitated phase crystal planes, and calculate the rotation matrix through the joint constraint of the three crystal planes. ; S7. The precipitated phase supercell is rotated according to the aforementioned rotation matrix. Rotate to a crystal plane that aligns with the crystal plane corresponding to the matrix supercell, record the coordinates of the precipitated phase atoms, and thus obtain the precipitated phase supercell; S8. Based on the geometric parameters of the precipitated phase, calculate the Boolean vector b in the matrix supercell to determine the embedding region; embed the rotated precipitated phase supercell into the matrix supercell according to the embedding region to obtain the initial composite atomic model; perform interface optimization on the interface atoms of the initial composite atomic model, and eliminate the non-physical high-energy model in the initial composite atomic model through interface region geometric pre-optimization to obtain the alloy-precipitated phase atomic model.

[0008] Preferably, after S8, the method further includes: S9. Repeat S1 to S8 to generate composite atomic models with different parameter combinations in batches by changing one or more parameters in the parameter set Parameters.

[0009] Preferably, the modifiable parameters include one or more of the following in the parameter set: matrix structure, precipitate structure, precipitate size, shape, volume fraction, orientation relationship type, and spatial distribution density.

[0010] Preferably, in step S1, the parameter set Parameters specifically includes:

[0011] in,S matrixs and S phases These represent the sets of unit cell structures of the matrix and the precipitated phase, respectively. R Indicates the orientation relationship between the matrix and the precipitated phase; size, num, shape, and position. core The dimensions, number, shape, and nucleus coordinates of the precipitates in the matrix are represented sequentially; each set of unit cell structures may contain one or more unit cell structures s containing precipitates.

[0012] Preferably, the cell structure s of the precipitated phase includes the following parameters:

[0013] Where length: a, b, c and angle: α, β, γ represent lattice constants; atom_type represents the atom type; and atom_positions represents the atom coordinates.

[0014] Preferably, the orientation relationship R is in the form of:

[0015]

[0016]

[0017] Where h, k, and l represent the crystal plane indices of the crystal in different basis directions, and the subscripts are respectively. m Subscript represents the matrix. p This represents the precipitated phase.

[0018] Preferably, in S2 and S3, the basis vector matrix M of the matrix supercell lattice m and the lattice basis matrix M of the precipitated phase p The unified calculation method is based on the lattice constant of each unit cell structure, and the calculation is performed as follows:

[0019] Where a, b, and c represent the lengths of the three edges in the unit cell structure, α, β, and γ represent the included angles between the edges of the unit cell structure; M represents the lattice basis vector matrix, which here represents the general calculation formula (or unified calculation method) of the matrix in this invention.

[0020] Preferably, in step S4, the three-level cascaded screening method includes: S41. Select potential crystal planes parallel to the matrix crystal planes from the precipitated phase crystal planes, and calculate the interplanar spacing of each potential crystal plane and the matrix crystal plane respectively. Then, use the interplanar spacing matching tolerance factor. or dBy screening potential crystal planes, a family of better candidate crystal planes corresponding to the crystal planes of the matrix is ​​obtained; S42. Calculate the angle between the normal vector of each crystal plane in the preferred candidate crystal plane family and the corresponding matrix crystal plane, and combine it with the crystal plane angle tolerance. d Crystal planes from the superior candidate crystal plane family are screened by their interplanar angles, retaining those within the tolerance range of interplanar angles. d Crystal planes within the range; S43. For the crystal planes retained after screening in S42, generate all their equivalent crystal planes according to crystal symmetry, and group these equivalent crystal planes into three orthogonal candidate crystal plane families corresponding to the matrix crystal planes. F 1. F 2. F 3.

[0021] Preferably, in S41, the interplanar spacing d hkl The calculation method is as follows:

[0022]

[0023]

[0024] Where g represents the direction of the normal vector of the crystal plane, and x, y, z represent the vectors in the lattice basis matrix M.

[0025] Preferably, in the calculation of interplanar spacing, the maximum value of the crystal plane index is 6.

[0026] Preferably, in S41, the screening method for potential crystal planes is as follows: Potential crystal planes satisfying the following formula are retained:

[0027] in, Indicates the interplanar spacing of the matrix. This indicates the interplanar spacing of each potential precipitate crystal plane.

[0028] Preferably, the interplanar spacing matches the tolerance factor. or d The value range is 0.03 to 0.10.

[0029] Preferably, in S42, the tolerance of the crystal plane angle is... d The calculation method is as follows:

[0030] Where k represents the adaptive factor, This represents the initial value of the tolerance for the crystal plane angle.

[0031] Preferably, S5 further includes: S51, Calculate candidate crystal plane families F 1. F 2. F The weighted composite function value Φ for each crystal plane combination in section 3; S52, Screening candidate crystal plane families F 1. F 2. F 3. The crystal plane combination formed by the crystal plane with the smallest angle to the matrix crystal plane is selected, and it is determined whether the Φ value of the crystal plane combination is the smallest. If it is, it is determined as the optimal model to construct the orientation relationship. If not, the crystal plane combination is deleted, and the remaining possible crystal plane combinations are traversed and the judgment in this step is performed.

[0032] Preferably, the weighted composite function value Φ is calculated as follows:

[0033] in, w 1. w 2. w 3 is the weighting coefficient; f 1 represents the deviation factor of the crystal plane normal vector. f 2 is the crystal orientation orthogonality preservation factor. f 3 represents the interface coherence factor;

[0034]

[0035]

[0036] Where v represents the crystal plane vector, i、j The numbers represent the three crystal plane families, with the subscript p indicating the precipitated phase and the subscript m indicating the matrix.

[0037] Preferably, if multiple combinations have similar Φ values, the combination of crystal planes with the smallest sum of absolute values ​​of crystal plane indices is selected as the optimal model for constructing orientation relationships.

[0038] Preferably, in S6, the rotation matrix The construction method is as follows: S61. Construct the orientation relationship based on the optimal model, and convert the crystal plane indices of the corresponding matrix crystal plane and precipitated phase crystal plane into real space normal vectors. S62. Correct the real space normal vectors obtained in S61 and construct them into orthogonal normalized basis vectors e1, e2, e3; S63. Assemble the orthogonal normalized basis vectors e1, e2, and e3 into the basis orthogonal basis matrix B. m and the extruded orthogonal basis matrix B p ; S64. Calculate the rotation matrix : .

[0039] Preferably, in step S8, the interface region geometry pre-optimization specifically includes: S81. Perform Voronoi polyhedral volume analysis on each atom in the interface region. V i Calculate, if the atoms V i If the volume is lower than the reference volume of the same type of atom in the precipitated phase or matrix, then the atom is removed; S82. For the atoms retained after processing in S81, calculate the radial distribution function of the atoms in the remaining interface region, and compare it with the peak distribution of the radial distribution functions of the precipitated phase and the matrix to determine the position of the abnormal peak. Further calculate the rigid displacement required by the atoms on both sides of the interface region. Based on the rigid displacement, adjust the position of the atoms in the interface region along the interface normal direction. S83. Calculate the effective coordination number of atoms in the interface region after processing by S82. If the effective coordination number of an atom deviates from the reference value of its block by more than a preset threshold, then delete the atom.

[0040] On the other hand, the present invention also provides a system for constructing an alloy precipitate model based on orientation relationships. This system is used to perform the method described above, and the system specifically includes: The information acquisition module is used to acquire the crystal structure parameters of the precipitated phase, the crystal structure parameters of the matrix, the orientation relationship, and the geometric parameters of the precipitated phase in the target alloy system, forming a parameter set. The supercell construction module is used to construct a matrix supercell containing a preset number of atoms based on the matrix crystal structure parameters, and to calculate the matrix supercell lattice basis vector matrix M. m The matrix atomic coordinates; and the lattice basis matrix M of the precipitated phase constructed based on the crystal structure parameters and orientation relationships of the precipitated phase. p The supercell of the precipitated phase and the atomic coordinates of the precipitated phase; The optimal orientation relationship determination module is used to determine the basis vector matrix M of the matrix supercell lattice. m , Precipitation phase lattice basis matrix M p Based on the aforementioned orientation relationship, three candidate crystal plane families are constructed using a three-level hierarchical screening method. F 1. F 2. F 3; and for three types of candidate crystal plane families F 1. F 2. F 3. Based on the constraints of three aspects—crystal plane normal deviation factor, crystal orientation orthogonality preservation factor, and interface coherence factor—the optimal model for constructing orientation relationships is determined. The orientation alignment and embedding module is used to construct the orientation relationship according to the optimal model, generate the matrix crystal plane and the precipitated phase crystal plane, calculate the rotation matrix through the joint constraint of the three crystal planes, and rotate and align the precipitated phase supercell according to the rotation matrix and embed it into the matrix supercell to obtain the initial composite atom model. The interface optimization module is used to optimize the interface of the initial composite atomic model to obtain the alloy-precipitate atomic model. The high-throughput generation module is used to generate a series of alloy-precipitate atomic models in batches by changing parameters.

[0041] Compared with existing technologies, this solution has at least the following beneficial effects: (1) Automatic Orientation Relationship Resolution and Polycrystalline Facet Joint Optimization. Unlike existing tools (such as ASE and Atomsk) that rely on users to manually calculate and input rotation matrices, this invention introduces an automatic orientation search framework of "three-level layered screening + multi-objective weighted optimization": the candidate space is compressed to 20%~40% of its original size through pre-screening of crystal plane spacing, then mismatched pairs are eliminated using adaptive tolerance, and finally global optimization is performed using the three-plane normal vector deviation, orthogonality preservation, and interface coherence as joint objective functions. This method can automatically start from experimental orientation relationship data and output the optimal rotation matrix without artificial crystallographic derivation, significantly reducing the requirements for users' professional knowledge. Moreover, the joint constraint of the three crystal planes ensures the complete determination of the three-dimensional orientation and avoids orientation ambiguity that may be caused by single crystal plane matching.

[0042] (2) Interface geometry intelligent pre-optimization. This invention innovatively proposes a pure geometry three-stage pre-optimization algorithm (Voronoi volume determination, abnormal peak detection, and coordination number anomaly removal) to be executed before energy relaxation, which can eliminate more than 95% of non-physical high-energy configurations without relying on any empirical potential function. This method not only fundamentally solves the problem of the incompatibility of the traditional "distance threshold deduplication" method with complex interfaces, but also reduces the average number of convergence iterations of subsequent energy relaxation by 40% to 60%, effectively avoiding simulation divergence or false convergence caused by the "inherent defects" of the initial model.

[0043] (3) Orientation relation is included in high-throughput scanning variables. Unlike existing high-throughput modeling methods that only traverse geometric parameters such as precipitate size and shape, this invention also includes orientation relation types (including coherent, semi-coherent and incoherent orientation variants) as scannable variables, enabling users to systematically evaluate the quantitative impact of different orientation relations on interfacial energy, strengthening effect and corrosion behavior, and providing complete parameter space coverage for the directional optimization of precipitate microstructure.

[0044] (4) Calculation method of rotation matrix of three crystal plane joint orthogonal basis. The covariance matrix eigendecomposition hybrid orthogonal basis construction strategy proposed in this invention maintains the orientation information of the normal vectors of the three crystal planes in the least squares sense in the orientation alignment of the precipitates and the matrix in non-cubic crystal system (hexagonal, monoclinic, triclinic) crystal system, and ensures the mathematical rigor of the obtained rotation matrix by forced orthogonalization, thus solving the problem of accuracy degradation of the traditional two-vector cross product method when dealing with non-strictly orthogonal crystal plane groups. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0046] Figure 1 This is a schematic diagram of the overall process of the present invention; Figure 2 The candidate crystal orientation family face angle difference calculated based on orientation in Embodiment 1 of the present invention; Figure 3 In Embodiment 1 of the present invention, the rotational precipitate is determined based on the orientation relationship obtained from experiments; Figure 4 Different shapes and sizes of precipitates were constructed for Embodiment 1 of the present invention; Figure 5 In Embodiment 1 of the present invention, the precipitated phase precipitates according to the orientation relationship between the precipitate and the matrix; Figure 6 This is an atomic model diagram of the alloy-precipitate distribution formed in Example 1 of the present invention; Figure 7 In Embodiment 2 of the present invention, multiple alloy-precipitate atomic models with different rotation angles, sizes, and structures are constructed in the matrix for high throughput. Detailed Implementation

[0047] The technical solutions described below in conjunction with the embodiments will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0048] The technical solution of the present invention will be further described below with reference to specific embodiments.

[0049] Example 1 In this embodiment, the MgZn precipitate phase in aluminum alloy is used as an example to illustrate the main implementation method of this scheme. Figure 1 As shown, the method of this scheme includes the following main steps: Step S1, Obtain the set of precipitation structure parameters: Obtain the crystal structure parameters, orientation relationship, and geometric parameters of the precipitates and matrix in the target alloy system; all parameter sets form a parameter set Parameters, the format of which is as follows:

[0050] Crystal structure parameters of the matrix S matrix and S phase It can include multiple structures, and their common form is as follows:

[0051]

[0052] Each unit cell structure set S It may contain n Individual cell structure s , n Greater than or equal to 1; unit cell structure s In this context, length (a, b, c) and angle (α, β, γ) represent lattice constants; atom_type represents the atom type; and atom_positions represents the atom coordinates. Lattice constants include the lengths a, b, and c of the three edges in the unit cell structure, as well as the angles α, β, and γ between the edges in the unit cell structure.

[0053] In this embodiment, we will take a structure containing one unit cell as an example for illustration, that is, the crystal structure parameters of the matrix. S matrix Includes a structure s 1. Among the crystal structure parameters of the matrix S phase Includes a structure s 1.

[0054] Crystal structure parameters of the matrix S matrix Includes a structure s 1. Its lattice constants are: length: a, b, c, where a=b=c=161.97, angle: α, β, γ, where α=β=γ=90°, atom_type is [Al,Al, ..., Al] and atom_positions are Cartesian atomic coordinates. The crystal structure parameters are represented as follows:

[0055] Crystal structure parameters of the matrix S phase Includes a structure s 1. Its lattice constants are: length: a, b, c, where a=b=5.25, c=8.44; angle: α, β, γ, where α=β=90°, γ=120°; atom type: [Mg, Mg, ..., Zn]; and Cartesian atomic coordinates: atom_positions are briefly as follows:

[0056] The orientation parameter R represents one or more orientation relationships between the matrix and the precipitated phase. Each orientation relationship contains two parallel relationships between the matrix and the precipitated phase crystal planes: crystal plane parallelism R1 and crystal orientation parallelism R2. The orientation parameter R takes the form of:

[0057]

[0058]

[0059] in, h , k , l These represent the crystal plane indices of a crystal in different basis directions, as discussed in the fundamentals of materials science. m Represents the matrix. p This represents the precipitated phase.

[0060] In this embodiment, the orientation relationship R is taken as an example, which includes one orientation relationship, namely the orientation relationship. R Includes one ( R 1, R 2) Including the crystal plane indices of the MgZn phase and the Al matrix, as well as the crystal plane parallelism R1 and crystal orientation parallelism R2 between the precipitated phase and the matrix:

[0061] The geometric parameters of the precipitated phase include its shape, size, volume fraction (num), and core coordinates (position). core These geometric parameters can each include one or more parameters. In this embodiment, each of the above parameters is assumed to have only one unique value, which are size = 1, num = 1, shape = hexagonal prism, and position = 1. core The coordinates are (220, 145, 20).

[0062] At this point, the extraction structure information has been summarized and collected, forming a parameter set.

[0063] In this embodiment, the orientation relationship R can be obtained in the following ways: experimental determination, including transmission electron diffraction, Kikuchi line analysis or electron backscattering diffraction; literature data or crystallographic database; calculation derivation based on lattice matching theory.

[0064] More preferably, the obtained orientation relationship can be verified, including: determining whether the orientation relationship satisfies the geometric constraint conditions of parallel crystal planes and parallel crystal directions; determining whether the interface corresponding to the orientation relationship satisfies the low-index crystal plane matching principle; and outputting a prompt message if the orientation relationship is not valid.

[0065] Step S2, Constructing the matrix supercell: Based on the matrix crystal structure parameters, construct a matrix supercell containing a preset number of atoms, and record the lattice basis vector matrix M of the matrix supercell. m and all atomic coordinates.

[0066] In this embodiment, based on the crystal structure parameters of the matrix S matrix In s 1. Construct a matrix supercell with a=b=c=4.05, α=β=γ=90°, and record its lattice basis matrix M. m With a preset number of atoms of 250,000, the supercell size is set to 40×40×40 units, which means the supercell side length is approximately 162 Å × 162 Å × 162 Å, containing approximately 256,000 Al atoms.

[0067] In this invention, the lattice basis vector matrix M of the crystal is calculated as follows:

[0068] Here, each parameter represents the basic crystal structure information input in step S1. In specific implementations, the lattice basis matrix includes the matrix supercell lattice basis matrix M. m and the lattice basis matrix M of the precipitated phase p Etc. a, b, and c represent the lengths of three edges in the unit cell structure, and α, β, and γ represent the included angles between the edges in the unit cell structure. M in the above formula represents the lattice basis vector matrix, which here represents the general calculation formula (or unified calculation method) for this matrix in this embodiment.

[0069] Continuing with this embodiment, based on the crystal structure parameters, the corresponding matrix basis vectors of the matrix supercell lattice can be calculated. as follows: .

[0070] Step S3, Construction of precipitate structure parameters: Based on the crystal structure parameters and orientation relationships of the precipitate, construct the precipitate lattice basis matrix M. p Supercell and precipitated phase atomic coordinates. Precipitated phase atomic coordinates can be directly recorded based on user selections, such as the coordinates selected directly when the user selects one or more precipitated phases.

[0071] Continuing with the above embodiments, a supercell of the MgZn precipitate phase is constructed, with a preset atomic number of 10,000. Based on the crystal structure parameters of the MgZn precipitate phase... S phase In s 1. A supercell for the precipitated phase was constructed with structural parameters a=b=5.25 Å, c=8.44 Å, α=β=90°, and γ=120°. Considering subsequent embedding operations, the supercell size was slightly larger than the target precipitated phase size, set to 10×10×10 units, containing approximately 11,000 atoms. The lattice basis vector matrix M of the precipitated phase was recorded. p and the coordinates of all atoms. The lattice basis vector matrix M of the precipitated phase. p The calculation results are as follows: .

[0072] Step S4, constructing candidate crystal plane families based on pre-screening of interplanar spacing and adaptive tolerance: based on the lattice basis vector matrix M of the matrix and precipitates. m and M p Based on the aforementioned orientation relationship, a candidate crystal plane family is constructed using a three-level hierarchical screening method. F .

[0073] Furthermore, the three-level layered screening method includes pre-screening of interplanar spacing, pre-screening of interplanar angles and adaptive tolerance, equivalent plane expansion and plane family merging. Based on the unit cell matrix parameter M... m and M p Candidate crystal plane families are constructed by calculating the orientation relationship, and the lattice basis vector directions of the precipitated phase supercell and the lattice basis vector directions of the matrix supercell satisfy the orientation relationship R.

[0074] In this embodiment, this step preferably includes the following sub-steps: Sub-step S4-1, Pre-screening of interplanar spacing: Calculate the interplanar spacing of all crystal plane indices (hkl). d hkl The maximum value of the exponent is 5. In this embodiment, the exponent crystal plane used for calculation refers to the potential crystal plane in the precipitated phase that is parallel to the matrix crystal plane. That is, the three-digit number of the crystal plane index, where each digit can be any integer from -5 to 5, resulting in a total of 1330 combinations. Considering the simplification of the three digits in the crystal plane index, there are a total of 1154 combinations. The interplanar spacing of each potential crystal plane and the matrix crystal plane is calculated separately, and then the tolerance factor is matched using the interplanar spacing. or d Potential crystal planes are screened, and candidate crystal planes that are obviously mismatched are quickly eliminated to obtain a family of better candidate crystal planes for the precipitates corresponding to the matrix crystal planes. This step can effectively reduce the amount of computation by 60%-80%.

[0075] In a preferred embodiment, the interplanar spacing is matched with the tolerance factor. or d The settings can be determined according to specific screening requirements, thereby calculating the matching between the precipitated phase and the matrix crystal plane using a screening threshold method:

[0076] Then pass the result through factorization. or d This serves as a threshold for filtering. Generally, the tolerance factor for the interplanar spacing is a specific value. The higher the value, the greater the degree of interplanar matching, but the more difficult it is to find a suitable interplanar surface, and the more candidate interplanar surfaces are filtered out.

[0077] Furthermore, the interplanar spacing is calculated as follows:

[0078]

[0079]

[0080]

[0081] Where g is the direction of the normal vector of the crystal plane (hkl), and x, y, z represent the vectors in the unit cell structure basis matrix M. Indicates the interplanar spacing of the matrix. This represents the interplanar spacing of each potential crystal plane of the precipitated phase.

[0082] In a more preferred embodiment, the interplanar spacing is matched with a tolerance factor. or d The value range is 0.03 to 0.10.

[0083] Continuing with the example of this embodiment, a pre-screening of interplanar spacing is performed: searching for potential interplanar phases in the precipitates that are intersecting with the matrix (100). Al For crystal planes parallel to each other, calculate the matrix (100). Al The interplanar spacing is as follows:

[0084]

[0085] In this example, the maximum value of the hkl crystal plane index is set to 5, the minimum value to -5, and the step size to 1. The traversal forms a (100) matrix. Al candidate crystal plane families of precipitated phases F :

[0086] Continuing with the examples above, let's look at candidate crystal plane families. F Taking the precipitated phase (110) crystal plane as an example, the interplanar spacing can be calculated as follows:

[0087]

[0088] The interplanar spacing matching degree between the precipitated phase and the matrix is ​​evaluated, and the interplanar spacing matching tolerance factor η is used. d Setting it to 0.1 revealed a poor match between the crystal plane of the precipitated phase (110) and the crystal plane of the Al matrix (100), i.e.:

[0089] Therefore, the precipitated phase (110) crystal plane is removed from the candidate crystal plane family. F Removed from the middle.

[0090] Based on the above example, the orientation relation module ultimately calculates the preferred candidate crystal plane family F1 for the precipitated phase, as follows: Figure 2 As shown, that is: .

[0091] Sub-step S4-2, Crystal plane angle screening and adaptive tolerance: For each candidate crystal plane, i.e., the better candidate crystal plane family in the above step, calculate the angle between its normal vector and the normal vector of the target crystal plane (i.e., the matrix crystal plane). i Crystal plane angle tolerance d It is not a fixed value, but rather a tolerance factor matching the interplanar spacing in sub-step S4-1. or d Adaptive determination. The closer the interplanar spacing, the greater the allowable angular deviation; the adaptive factor... k The value ranges from 0 to 10 because the coherence of the matrix-precipitated phase interface is mainly determined by the interplanar spacing matching. This adaptive crystal plane angle tolerance... d Candidate crystal planes outside the range were eliminated.

[0092] Crystal plane angle i The calculation method is as follows:

[0093] Crystal plane angle tolerance d The calculation method is as follows:

[0094] Where k represents the adaptive factor, This represents the initial value of the tolerance for the interplanar angle. Represents the normal vector of the target crystal plane. This represents the normal vector of the preferred candidate crystal plane to be screened.

[0095] In a more preferred embodiment, adaptive tolerance d Within the range of -6° ≤ i If the angle is ≤ 6°, it is considered a superior crystal plane that meets the screening requirements and is retained.

[0096] Continuing with the example in this embodiment, the preferred candidate crystal plane family for the precipitate is calculated. F 1 in ( Crystal plane and matrix (100) Al The angle difference of the crystal planes is as follows:

[0097]

[0098]

[0099] Superior crystal plane family F The candidate crystal planes in 1 and (100) Al The angular differences between the crystal planes are 1.14°, -1.13°, 1.14°, -1.13°, 1.1°, -1.15°, 0.9°, 2.23°, 1.34°, -2.45°, -1.44°, 3.19°, -4.83°, -5.89°, 2.95°, -3.29°, 5.22°, -5.78°, -3.42°, 5.83°, -4.65°, 5.1°, and 4.38°. The crystal plane differences are within the tolerance range. d -6° ≤ i If the angle is ≤ 6°, then the candidate crystal planes that meet the requirements of this range will be retained.

[0100] Sub-step S4-3, Equivalent crystal plane expansion and crystal plane family merging: For each candidate crystal plane obtained after screening in sub-step S4-2, equivalent crystal planes are reduced or expanded according to crystal symmetry, and finally merged into candidate crystal plane families. F 1. F 2. F 3. These correspond to three families of orthorhombic crystal planes of the matrix. Each family of crystal planes... F i The number of equivalent crystal planes depends on the degree of symmetry of the crystal system.

[0101] In this embodiment, an equivalent crystal plane refers to a crystal plane that can be transformed into another crystal plane through point group symmetry operations. For example, (001), (100), and (010) in a cubic crystal system are equivalent crystal planes with the same interplanar spacing. The worse the symmetry of the crystal, the fewer its equivalent crystal planes, and the greater the computational load.

[0102] Continuing with the example above, we obtain the crystal plane that satisfies sub-steps S4-1 and S4-2, which is perpendicular to the matrix (100). Al (010) Al (001) Al Approximately parallel, forming a family of crystal planes. F 1. F 2. F 3.

[0103] Step S5: Construct a multi-objective constrained optimization problem, and determine the optimal model and orientation relationship based on constraints in three aspects: minimizing the deviation of the crystal plane normal vector, maintaining the orthogonality of the crystal orientation, and the interface coherence factor.

[0104] In crystal family F 1. F 2. F In the optimization process, this embodiment constructs a multi-objective constrained optimization problem and sets a weighted synthesis function Φ. For all candidate crystal plane families output by S4, the value of Φ is calculated to intelligently determine the optimal crystal plane combination as the best model for constructing orientation relationships. If multiple combinations have similar Φ values, the crystal plane combination with the smallest sum of absolute values ​​of crystal plane indices is selected as the best model to ensure the simplicity of subsequent modeling. The weighted synthesis function Φ is calculated as follows:

[0105] in, w 1. w 2. w 3 represents the weighting coefficient, which can be set to 1 / 3 by default. f 1 represents the deviation factor of the crystal plane normal vector. f 2 is the crystal orientation orthogonality preservation factor. f 3 represents the interface coherence factor.

[0106] Crystal plane normal vector deviation factor f 1. The crystal plane angle between the precipitated phase crystal plane and the target matrix crystal plane was evaluated; the smaller the angle, the more accurate the result. The crystal orientation orthogonality retention factor was also assessed. f 2. The orthogonality of the candidate crystal planes is determined; the closer the value is to 0, the more orthogonal the three crystal planes are. f 3 is the interface coherence factor, which measures the degree of coherence between the precipitated phase and the matrix corresponding to the candidate crystal plane combination. The calculation methods for the above constraint factors are as follows:

[0107]

[0108]

[0109] Where v is the vector represented by the (hkl) crystal plane, i、j These are three crystal plane family numbers, with the subscript p indicating the precipitated phase and the subscript m indicating the matrix.

[0110] In one implementation, if more attention is paid to orientation accuracy during implementation, then the [measurement] is increased. w 1. Specific gravity; if the three candidate crystal planes are orthogonal, then f 2 is 0; if more emphasis is placed on interface commonality during implementation, then increase it. w 3. Specific gravity.

[0111] Continuing with the example above, after step S4, all three matrix crystal orientation families (100) are traversed. Al (010) Al (001) Al The three crystal plane families obtained later F 1, F 2, F 3. Determine the MgZn precipitate phase ( ), ( (354) and (355) are the crystal planes with the smallest included angles to the three matrix crystal planes mentioned above, with angle differences of 0.9°, -0.79° and 0.79°, respectively. However, the weighted composite function value of this combination is relatively high, so the weighted composite function of other crystal plane combinations is calculated.

[0112] Among all candidate crystal planes that satisfy the orientation relationship, calculate the Φ value of all other crystal plane combinations, and select the crystal plane combination with the smallest Φ value as the optimal orientation relationship. For example, crystal plane combination ( ), ( ), (354) and crystal plane combination ( (403), ( All of them are crystal plane combinations that meet the requirements of step S4-2.

[0113] But crystal plane combination ( ), ( The Φ values ​​of (354) are the lowest, at only 1.126, and are the three crystal orientations with the smallest orientation difference from the matrix. Therefore, they are determined as the optimal model orientation relationships. The crystal plane angle errors of this combination are 1.61°, 0.81° and 0.81°, respectively, and the sum of the absolute values ​​of the crystal plane angles is the lowest, at only 3.23. f 1), and the three crystal orientations are perpendicular to each other ( f 2) Satisfying the orientation relationship of the precipitated phase distribution. The calculation results of each constraint factor are as follows:

[0114]

[0115]

[0116] .

[0117] Step S6, Calculation of rotation matrix based on three-plane joint constraint: Construct the orientation relationship according to the optimal model determined in step S5, generate matrix crystal planes and precipitated phase crystal planes, and use the three-plane joint constraint orthogonal basis construction method to ensure the rotation matrix. Simultaneously satisfying the orientation consistency of the three crystal planes.

[0118] In this step, the joint constraint of the three crystal planes refers to calculating the rotation matrix based on the three sets of matrix crystal planes and the three sets of precipitated phase crystal planes determined in S4 and S5. In this embodiment, the method for calculating the rotation matrix specifically includes: Sub-step S6-1: Convert the crystal planes into real space normal vectors: To address the potential mismatch between the hexagonal, tetragonal, and other crystal plane indices of the precipitated phase and the matrix, the reciprocal basis vectors are used to convert the crystal plane indices of the three sets of matrix crystal planes and the three sets of precipitated phase crystal planes into real space normal vectors g.

[0119] Sub-step S6-2, construction of orthogonal normalized basis vector set: The three candidate crystal plane normal vectors (including three sets of matrix crystal plane normal vectors and three sets of precipitate crystal plane normal vectors) are usually not strictly orthogonal in non-cubic crystal systems. Therefore, for the universality of the method, in this embodiment, orthogonal normalized basis vectors e1, e2, e3 are constructed by orthogonalization combined with covariance matrix correction.

[0120] In a more preferred embodiment, during the orthogonalization process in this step, v1, v2, and v3 represent the crystal plane normal vectors of the matrix (subscript m) or the precipitated phase (subscript p), respectively. The three vectors are combined and subjected to Gram-Schmidt orthogonalization, i.e., the projection of the subsequent vector onto the preceding vectors (one or two) is removed, thus initially obtaining the preliminary orthogonal basis vectors u1, u2, and u3. The covariance matrix C is constructed, and eigenvalue decomposition is performed on the preliminary orthogonal matrix to obtain the orthogonal normalized basis vector U, i.e., U = [e1, e2, e3].

[0121] .

[0122] Sub-step S6-3, assembly of orthogonal basis matrices: Assemble the corrected orthogonal normalized basis vectors e1, e2, e3 into the basis orthogonal basis matrix B. m and the extruded orthogonal basis matrix B p .

[0123]

[0124]

[0125] In this context, the subscript m represents the matrix, and the subscript p represents the precipitated phase.

[0126] Sub-step S6-4: Calculate the rotation matrix based on the orthogonal basis matrices of the matrix and precipitated phases. : .

[0127] Continuing with the example above, construct the orientation relationship based on the determined optimal model, and calculate the rotation matrix between the precipitated phase and the matrix. : For precipitated phase ( ), ( The crystal plane indices of (354) and the matrix (100), (010), and (001) are used to construct the matrix orthogonal basis matrix B by vector calculation. M and the extruded orthogonal basis matrix B P :

[0128]

[0129] Calculate the rotation matrix based on the two orthogonal basis matrices The details are as follows:

[0130] .

[0131] In addition, in one embodiment, this step can also be used to verify the model, that is, to compare and verify the generated composite atom model with the results of transmission electron microscopy experiments to ensure that the crystallographic orientation and interface structure of the model are consistent with the experiment.

[0132] Step S7, precipitate structure rotation: The precipitate supercell is rotated according to the rotation matrix obtained in step S6. Rotate to a crystal plane aligned with the corresponding crystal plane of the matrix supercell to obtain the precipitate parameter coordinates. Furthermore, the precipitate can be segmented into different specified precipitate shapes based on the precipitate parameter coordinates.

[0133] Continuing with the example above, based on the rotation matrix calculated in step S6... The precipitated phase is treated as follows Figure 3 Rotate as shown. Use parameter r to represent the original precipitated phase. sThe atom_positions parameter (i.e., atomic coordinates) is 1. The parameter r' represents the atom_positions parameter (i.e., atomic coordinates) after the precipitate is rotated. The specific calculation process is as follows:

[0134]

[0135] Subsequently, based on the atomic coordinates r' of the rotated precipitate parameters, the MgZn precipitate is divided into specified shapes. These specified shapes can be, for example, plate-like, rod-like, spherical, cubic, parallelepiped, hexagonal prism, or other precipitate shapes. Figure 4 As shown.

[0136] Step S8, Intelligent pre-optimization of the precipitated phase interface geometry: Based on the geometric parameters of the precipitated phase, calculate the Boolean vector b in the matrix supercell to determine the embedding region; embed the rotated precipitated phase supercell into the matrix supercell according to the embedding region to obtain the initial composite atom model; perform interface optimization on the interface atoms of the initial composite atom model, and use a geometry-driven three-stage intelligent pre-optimization method to eliminate non-physical high-energy models in the initial composite atom model.

[0137] In this embodiment, the method for determining the embedding region is as follows: according to the shape and size of the precipitated phase, a corresponding spatial region is delineated in the matrix supercell, and this spatial region is the embedding region; the shape includes spherical, ellipsoidal, cubic, lath-shaped or needle-shaped.

[0138] Furthermore, the three stages of intelligent pre-optimization of interface region geometry include: overlapping atom removal based on Voronoi volume determination, interface layer spacing correction based on atomic distribution function, and interface reconstruction atom removal based on coordination number anomalies. The specific steps are as follows: Sub-step S8-1, the removal of overlapping atoms in the Voronoi volume determination, involves performing Voronoi polyhedral volume calculation on each atom in the interface region. V i calculate:

[0139] in, f Represents all faces of the Voronoi polyhedron. A f It is noodles f area, d f It is a seed point to a surface f The perpendicular distance. The seed point is the body center of the polyhedron, which is generally the atomic coordinate of the body center.

[0140] When a certain atom Vi If the volume of an atom is lower than the reference volume of the same type of atom in the precipitated phase or matrix, it is determined that the atom is in a non-physical proximity state and is removed.

[0141] Sub-step S8-2, the interface layer spacing correction based on the atomic distribution function, involves statistically analyzing the radial distribution function g(r) of atoms in the interface region between the matrix and the precipitated phase, comparing it with the peak distribution of g(r) of the precipitated phase and g(r) of the matrix, identifying the positions of abnormal peaks, and calculating the required rigid displacement Δ of the atomic layers on both sides of the interface region. d The atomic coordinates are automatically adjusted along the interface normal direction to restore the nearest neighbor distance of atoms in the interface region to a physically reasonable range. It can be understood that the interface region in this step refers to the boundary region between the precipitated phase and the matrix. In this embodiment, the radial distribution function is calculated as follows:

[0142] Where, r x It is the distance between the atomic pairs, which is the independent variable. Indicates the first i The and the first j The atomic distance between atoms; δ is a statistical function whose result represents the quantity, i.e., the statistical quantity obtained; ρ is the average atomic density.

[0143] Sub-step S8-3, the interface reconstruction atom removal based on coordination number anomalies, involves calculating the effective coordination number of atoms in the interface region. This value can also be calculated using the radial distribution function g(r). If the coordination number of an atom deviates from the reference value of its bulk, the atom is marked as an "interface reconstruction atom" and removed. The reference value is a custom preset threshold, for example, a value between 4 and 12.

[0144] Continuing with the example above, based on the extracted parameters (size=1, num=1, shape=hexagonal prism, position) in the parameter set Parameters... core Calculate the Boolean vector b in the matrix supercell for (220, 145, 20) to determine the embedding region; embed the rotated precipitated phase supercell into the matrix supercell and remove overlapping atoms to obtain the initial composite atomic model; perform interface region atomic relaxation on the initial composite atomic model and adjust the position in the parameter set Parameters. core By optimizing the parameters and eliminating non-physical high-energy configurations, a precipitate-matrix composite atom model for materials computational simulation is obtained:

[0145] in Let x, y, z represent Boolean vectors, where x, y, z represent the precipitated phases. iThe Cartesian coordinates of local atoms in three-dimensional space, where min and max are the upper and lower boundaries of the shape in three-dimensional space, and local atoms refer to atoms within the matrix. The specific calculation process is as follows: Sub-step S8-1, Removal of overlapping atoms in Voronoi volume determination: Select interface atoms and calculate V. i Value, such as Figure 5 As shown. V of the interface atoms i It must be significantly smaller than that of similar atoms, therefore, these V i Related atoms smaller than those of the same type are removed.

[0146] Sub-step S8-2, interlayer spacing correction based on atomic distribution function: Calculation of g(r) shows that the peak position corresponding to AlZn atoms at the interface is 2.76, and the peak position corresponding to AlMg atoms is 3.0. Both are within the normal atomic distance, so there is no need to adjust the position of the precipitated phase. core Adjust the parameters.

[0147] Sub-step S8-3, interface reconstruction atom removal based on coordination number anomalies: Calculations using g(r) show that the main coordination numbers of interface atoms are 8, 9, 10, 11, 12, 13, 14, 15, and 16. The proportions of each coordination number are 0.005%, 0.042%, 0.206%, 0.586%, 93.903%, 1.907%, 1.622%, 1.368%, and 0.361%, respectively. The vast majority of atoms have a coordination number around 12. Atoms with extremely high coordination (16 coordination) represent a serious non-physical region, and atoms containing this coordination number are removed accordingly.

[0148] Continuing with the example above, the atom_positions of the rotated precipitate phase and its filling atoms are merged with the atom_positions of the aluminum alloy matrix. The atoms at the interface are intelligently optimized, thus forming an alloy-precipitate atom model of the precipitate distribution, as shown below. Figure 6 As shown.

[0149] At this point, the composite atom model can be established.

[0150] Example 2 The precipitate distribution construction method in this embodiment is the same as that in Embodiment 1. The main difference lies in the parameter set. S phase R, size, num, position core The parameters include one or more parameters. Variations in these parameters allow for high-throughput generation of alloy-precipitate distribution models within a short time. That is, after step S8, the following is added: Step S9, High-throughput Generation: Repeat steps S1 to S8, generating composite atomic models with different parameter combinations in batches by changing one or more parameters in the precipitated phase and matrix structure information. The parameters that can be changed include one or more of the following: precipitated phase size, shape, volume fraction, orientation relationship type, and spatial distribution density, thereby simultaneously generating models such as... Figure 7 The distribution models of multiple precipitated phases are shown.

[0151] Continuing with the example above, the parameter set is as follows:

[0152]

[0153]

[0154] num = 4

[0155] In this embodiment, the embedding region is determined by delineating a corresponding spatial region in the matrix supercell according to the shape and size of the precipitated phase; the shape includes spherical, ellipsoidal, cubic, lath-shaped or needle-shaped.

[0156] The size distribution of the precipitated phase includes sizes such as 1 nm, 5 nm, 10 nm, 15 nm, 20 nm, 50 nm, and 100 nm, and includes intermediate values ​​of the above parameters. An average size can also be specified to randomly generate size distributions or size gradients, thereby generating a variety of random precipitated phase sizes that more closely approximate physical rules.

[0157] The volume fraction or number density of the precipitated phase (num) includes 0.0001 / nm. 3 0.001 / nm 3 0.01 / nm 3 0.1 / nm 3 1 / nm 3 2 / nm 3 4 / nm 3 5 / nm 3 10 / nm 3 20 / nm 3 50 / nm 3 It has the same density and includes the intermediate values ​​of the above parameters.

[0158] Example 3 Furthermore, the solution of the present invention can also be implemented in a systematic way, namely, a system for constructing an alloy precipitate model based on orientation relationships, comprising: The information acquisition module is used to acquire the crystal structure parameters of the precipitated phase, the crystal structure parameters of the matrix, the orientation relationship, and the geometric parameters of the precipitated phase in the target alloy system, forming a parameter set. The supercell construction module is used to construct a matrix supercell containing a preset number of atoms based on the matrix crystal structure parameters, and to calculate the matrix supercell lattice basis vector matrix M. m The matrix atomic coordinates; and the lattice basis matrix M of the precipitated phase constructed based on the crystal structure parameters and orientation relationships of the precipitated phase. p The supercell of the precipitated phase and the atomic coordinates of the precipitated phase; The optimal orientation relationship determination module is used to determine the basis vector matrix M of the matrix supercell lattice. m , Precipitation phase lattice basis matrix M p Based on the aforementioned orientation relationship, three candidate crystal plane families are constructed using a three-level hierarchical screening method. F 1. F 2. F 3; and for three types of candidate crystal plane families F 1. F 2. F 3. Based on the constraints of three aspects—crystal plane normal deviation factor, crystal orientation orthogonality preservation factor, and interface coherence factor—the optimal model for constructing orientation relationships is determined. The orientation alignment and embedding module is used to construct the orientation relationship according to the optimal model, generate the matrix crystal plane and the precipitated phase crystal plane, calculate the rotation matrix through the joint constraint of the three crystal planes, and rotate and align the precipitated phase supercell according to the rotation matrix and embed it into the matrix supercell to obtain the initial composite atom model. The interface optimization module is used to optimize the interface of the initial composite atomic model to obtain the alloy-precipitate atomic model. The high-throughput generation module is used to generate a series of alloy-precipitate atomic models in batches by changing parameters.

[0159] In addition, this solution can also be implemented by means of a device or a computer-readable storage medium. When the electronic device is running, and when the storage medium is read and executed by a computer device, the instructions therein can be executed to implement the method of constructing an alloy precipitate model based on orientation relationship as described in Example 1.

[0160] The above description is merely a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural transformations made using the contents of the present invention under the inventive concept of the present invention, or direct / indirect applications in other related technical fields, are included within the patent protection scope of the present invention.

Claims

1. A method for constructing an alloy precipitate model based on orientation relationships, the method comprising: S1. Input or obtain the crystal structure parameters of the precipitated phase, the crystal structure parameters of the matrix, the orientation relationship, and the geometric parameters of the precipitated phase in the target alloy system to form the parameter set Parameters; S2. Based on the matrix crystal structure parameters, construct a matrix supercell containing a predetermined number of atoms, and calculate the matrix supercell lattice basis vector matrix M. m and matrix atomic coordinates; S3. Construct the lattice basis matrix M of the precipitated phase based on the crystal structure parameters and orientation relationships of the precipitated phase. p The supercell of the precipitated phase and the atomic coordinates of the precipitated phase; S4, Based on the matrix supercell basis vector matrix M m , Precipitation phase lattice basis matrix M p Based on the aforementioned orientation relationship, three candidate crystal plane families are constructed using a three-level hierarchical screening method. F 1. F 2. F 3; wherein the directions of the matrix supercell basis vectors and the directions of the precipitated phase supercell basis vectors satisfy the aforementioned orientation relationship; S5. For the three types of candidate crystal plane families constructed in step S4 F 1. F 2. F 3. Based on the constraints of three aspects—crystal plane normal deviation factor, crystal orientation orthogonality preservation factor, and interface coherence factor—the optimal model for constructing orientation relationships is determined. S6. Based on the optimal model determined in step S5, construct the orientation relationship, generate the matrix crystal planes and precipitated phase crystal planes, and calculate the rotation matrix through the joint constraint of the three crystal planes. ; S7. The precipitated phase supercell is rotated according to the aforementioned rotation matrix. Rotate to a crystal plane that aligns with the crystal plane corresponding to the matrix supercell, record the coordinates of the precipitated phase atoms, and thus obtain the precipitated phase supercell; S8. Based on the geometric parameters of the precipitated phase, calculate the Boolean vector b in the matrix supercell to determine the embedding region; embed the rotated precipitated phase supercell into the matrix supercell according to the embedding region to obtain the initial composite atomic model; perform interface optimization on the interface atoms of the initial composite atomic model, and eliminate the non-physical high-energy model in the initial composite atomic model through interface region geometric pre-optimization to obtain the alloy-precipitated phase atomic model.

2. The method according to claim 1, characterized in that, Following S8, the following also includes: S9. Repeat S1 to S8 to generate composite atomic models with different parameter combinations in batches by changing one or more parameters in the parameter set Parameters.

3. The method according to claim 1, characterized in that, In S1, the parameter set Parameters specifically includes: in, S matrixs and S phases These represent the sets of unit cell structures of the matrix and the precipitated phase, respectively. R Indicates the orientation relationship between the matrix and the precipitated phase; size, num, shape, and position. core The dimensions, number, shape, and nucleus coordinates of the precipitates in the matrix are represented sequentially; each set of unit cell structures may contain one or more unit cell structures s containing precipitates.

4. The method according to claim 1, characterized in that, In S2 and S3, the matrix supercell basis vector matrix M m and the lattice basis matrix M of the precipitated phase p The unified calculation method is based on the lattice constant of each unit cell structure, and the calculation is performed as follows: Where a, b, and c represent the lengths of the three edges in the unit cell structure, α, β, and γ represent the included angles between the edges of the unit cell structure, and M represents the lattice basis matrix.

5. The method according to claim 1, characterized in that, In S4, the three-level cascaded filtering method includes: S41. Select potential crystal planes parallel to the matrix crystal planes from the precipitated phase crystal planes, and calculate the interplanar spacing of each potential crystal plane and the matrix crystal plane respectively. Then, use the interplanar spacing matching tolerance factor. η d By screening potential crystal planes, a family of better candidate crystal planes corresponding to the crystal planes of the matrix is ​​obtained; S42. Calculate the angle between the normal vector of each crystal plane in the preferred candidate crystal plane family and the corresponding matrix crystal plane, and combine it with the crystal plane angle tolerance. δ Crystal planes from the superior candidate crystal plane family are screened by their interplanar angles, retaining those within the tolerance range of interplanar angles. δ Crystal planes within the range; S43. For the crystal planes retained after screening in S42, generate all their equivalent crystal planes according to crystal symmetry, and group these equivalent crystal planes into three orthogonal candidate crystal plane families corresponding to the matrix crystal planes. F 1. F 2. F 3.

6. The method according to claim 1, characterized in that, S5 further includes: S51, Calculate candidate crystal plane families F 1. F 2. F The weighted composite function value Φ for each crystal plane combination in section 3; S52, Screening candidate crystal plane families F 1. F 2. F 3. The crystal plane combination formed by the crystal plane with the smallest angle to the matrix crystal plane is selected, and it is determined whether the Φ value of the crystal plane combination is the smallest. If it is, it is determined as the optimal model to construct the orientation relationship. If not, the crystal plane combination is deleted, and the remaining possible crystal plane combinations are traversed and the judgment in this step is performed.

7. The method according to claim 6, characterized in that, The weighted composite function value Φ is calculated as follows: in, w 1. w 2. w 3 is the weighting coefficient; f 1 represents the deviation factor of the crystal plane normal vector. f 2 is the crystal orientation orthogonality preservation factor. f 3 represents the interface coherence factor; Where v represents the crystal plane vector, i, j The numbers represent the three crystal plane families, with the subscript p indicating the precipitated phase and the subscript m indicating the matrix.

8. The method according to claim 1, characterized in that, In S6, the rotation matrix The construction method is as follows: S61. Construct the orientation relationship based on the optimal model, and convert the crystal plane indices of the corresponding matrix crystal plane and precipitated phase crystal plane into real space normal vectors. S62. Correct the real space normal vectors obtained in S61 and construct them into orthogonal normalized basis vectors e1, e2, e3; S63. Assemble the orthogonal normalized basis vectors e1, e2, and e3 into the basis orthogonal basis matrix B. m and the extruded phase orthogonal basis matrix B p ; S64. Calculate the rotation matrix : .

9. The method according to claim 1, characterized in that, In step S8, the interface region geometry pre-optimization specifically includes: S81. Perform Voronoi polyhedral volume analysis on each atom in the interface region. V i Calculate, if the atoms V i If the volume is lower than the reference volume of the same type of atom in the precipitated phase or matrix, then remove the atom; S82. For the atoms retained after processing in S81, calculate the radial distribution function of the atoms in the remaining interface region, and compare it with the peak distribution of the radial distribution function of the precipitated phase and the matrix to determine the position of the abnormal peak. Further calculate the rigid displacement required by the atoms on both sides of the interface region. Based on the rigid displacement, adjust the position of the atoms on both sides of the interface region along the interface normal direction. S83. Calculate the effective coordination number of atoms in the interface region after processing by S82. If the effective coordination number of an atom deviates from the reference value of its block by more than a preset threshold, then delete the atom.

10. A system for constructing an alloy precipitate model based on orientation relationships, the system being used to perform the method as described in any one of claims 1-9, characterized in that, The system specifically includes: The information acquisition module is used to acquire the crystal structure parameters of the precipitated phase, the crystal structure parameters of the matrix, the orientation relationship, and the geometric parameters of the precipitated phase in the target alloy system, forming a parameter set. The supercell construction module is used to construct a matrix supercell containing a preset number of atoms based on the matrix crystal structure parameters, and to calculate the matrix supercell lattice basis vector matrix M. m The matrix atomic coordinates; and the lattice basis matrix M of the precipitated phase constructed based on the crystal structure parameters and orientation relationships of the precipitated phase. p The supercell of the precipitated phase and the atomic coordinates of the precipitated phase; The optimal orientation relationship determination module is used to determine the basis vector matrix M of the matrix supercell lattice. m , Precipitation phase lattice basis matrix M p Based on the aforementioned orientation relationship, three candidate crystal plane families are constructed using a three-level hierarchical screening method. F 1. F 2. F 3; and for three types of candidate crystal plane families F 1. F 2. F 3. Based on the constraints of three aspects—crystal plane normal deviation factor, crystal orientation orthogonality preservation factor, and interface coherence factor—the optimal model for constructing orientation relationships is determined. The orientation alignment and embedding module is used to construct the orientation relationship according to the optimal model, generate the matrix crystal plane and the precipitated phase crystal plane, calculate the rotation matrix through the joint constraint of the three crystal planes, and rotate and align the precipitated phase supercell according to the rotation matrix and embed it into the matrix supercell to obtain the initial composite atom model. The interface optimization module is used to optimize the interface of the initial composite atomic model to obtain the alloy-precipitate atomic model. The high-throughput generation module is used to generate a series of alloy-precipitate atomic models in batches by changing parameters.

Images