Crystal performance prediction method based on four-field iteration and dynamic combination
Through the method of four-domain iteration and dynamic merging, the problems of information loss and operational complexity in establishing crystal models in the existing technology are solved, and efficient crystal performance prediction and modeling are achieved, especially the prediction of anisotropy, strain hardening and fatigue damage of metal materials.
Patent Information
- Application Number
- CN202510859549.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-25
- Publication Date
- 2025-10-17
AI Technical Summary
Existing methods for establishing multi-scale crystal models have problems such as information loss caused by data conversion, complex operation procedures, and high learning costs, making it difficult to achieve efficient crystal performance prediction.
A four-domain iterative and dynamic merging approach is adopted to construct a unified system for crystal data computation, processing, and modeling. This includes preprocessing of crystal structure materials, image acquisition, crystal model initialization, meshing and phase parameter calculation using the four-domain iterative method, and dynamic merging modeling. The model is built using Abaqus, Matlab, and Python modules.
It enables rapid identification of crystal data, avoids information loss, simplifies the modeling process, improves modeling efficiency, and can predict the anisotropy, strain hardening, fatigue damage and other properties of metallic materials.
Smart Images
Figure CN120808995A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of crystal performance prediction, and particularly relates to a crystal performance prediction method based on four-field iteration and dynamic merging. BACKGROUND
[0002] Crystal plasticity theory, as a bridge connecting material microstructure and macroscopic mechanical behavior, has become an important paradigm in modern materials science research. This theory quantitatively describes the micro-mechanism of dislocation slip, twinning deformation, etc. in the crystal, and builds a cross-scale correlation model from atomic scale dislocation dynamics to continuous medium plasticity. Based on this theoretical framework, researchers have successfully revealed the nature of macroscopic mechanical phenomena such as anisotropy, strain hardening, and fatigue damage of metal materials, providing theoretical support for the research and development of advanced materials in the fields of aerospace, biomedicine, etc.
[0003] Existing multi-scale crystal model building methods, such as Coupled phase field damage and crystal plasticity analysis of intragranular fracture: The role of crystallographic orientation and voids and Crystal plasticity modeling of ductile fracture locus in advanced high-strength steel, require cross-platform operation, and the multi-software collaboration mode has three key defects: data conversion leads to information loss, complex operation process, and high learning cost.
[0004] Therefore, how to realize a multi-scale crystal model building method that reduces the loss of grain boundary information, is efficient in operation process, and conforms to the real performance of materials is a technical problem to be solved at present. SUMMARY
[0005] In view of the shortcomings of the existing method, the present application constructs a four-field iteration method for multi-material cubic crystal dynamic merging modeling. The unified crystal data calculation processing and modeling method overcomes the problems of complex modeling operation process, information loss caused by data conversion, and low modeling efficiency of the existing modeling method in processing cubic crystal materials.
[0006] The technical scheme adopted by the present application is: a crystal performance prediction method based on four-field iteration and dynamic merging includes the following steps:
[0007] Step one, obtain a crystal structure material and perform pretreatment;
[0008] As a preferred embodiment of the present application, the pre-treatment comprises: rough polishing, fine polishing and electrolytic polishing on the surface of the crystal structure material.
[0009] Step two, image acquisition and crystal model initialization are performed on the pre-treated crystal structure material;
[0010] As a preferred embodiment of the present application, the image acquisition is performed by EBSD.
[0011] As a preferred embodiment of the present application, the crystal model initialization comprises:
[0012] Starting the crystal model plug-in, a responsive interactive interface of the data processing and modeling plug-in is constructed;
[0013] Selecting the electron backscattering diffraction data position, data storage position and crystal plasticity simulation subroutine position;
[0014] Setting the material parameters and load size.
[0015] As a preferred embodiment of the present application, the data of the crystal model initialization comprises: crystal image size, pixel point X and Y direction step length, pixel point coordinates, pixel point crystal orientation and pixel point phase parameter.
[0016] Step three, phase parameter and orientation angle difference calculation and judgment are performed on the grided crystal by using the four-field iteration method, and the grain group division is performed on the adjacent grid crystal;
[0017] As a preferred embodiment of the present application, step three specifically comprises:
[0018] Step 31, the crystal image is divided into a plurality of sub-grids, and the crystal edge coordinate points of the sub-grids are extracted;
[0019] Step 32, taking a sub-grid as a starting point, the phase parameter and the orientation angle difference of the adjacent sub-grids are calculated;
[0020] As a preferred embodiment of the present application, the formula of the orientation angle difference is:
[0021]
[0022] Wherein, G ω is the orientation difference matrix of two crystals; trace() is the trace function, and err is the orientation difference threshold.
[0023] Step 33, if the phase parameters of two adjacent sub-grids are consistent and the orientation angle difference value is within the preset error range, the two sub-grids are classified into the same grain group;
[0024] Step 34, continuously circulating, when a certain sub-grid does not meet the condition of step 33, complete the division of the grain group;
[0025] Step 35, repeat steps 31-34 to complete the division of other grain groups.
[0026] Step 36, extracting grain modeling data for all divided grain groups.
[0027] Step four, using dynamic merging modeling method to establish crystal plastic model based on grain modeling data; merging sub-grids in different grain groups to obtain merged grids;
[0028] As a preferred embodiment of the application, step four specifically comprises:
[0029] Step 41, first, the sub-grid in a certain grain group is initially merged according to the preset merging threshold to obtain a plurality of first merged grids; second, a plurality of first merged grids are merged two by two to obtain second merged grids; third, a plurality of second merged grids are merged two by two to obtain third merged grids, and so on; finally, all sub-grids in the grain group are merged to obtain a first grid;
[0030] Step 42, merging sub-grids of other grain groups according to step 41.
[0031] As a preferred embodiment of the application, the crystal performance prediction system based on four-field iteration and dynamic merging comprises: a memory for storing instructions executable by a processor; and the processor is configured to execute the instructions to implement the crystal performance prediction method based on four-field iteration and dynamic merging.
[0032] As a preferred embodiment of the application, the computer readable medium storing computer program code, the computer program code implements the crystal performance prediction method based on four-field iteration and dynamic merging when executed by the processor.
[0033] The beneficial effects of the application are as follows:
[0034] 1. The application constructs a four-field iteration method to realize rapid identification of crystal data and obtain comprehensive crystal information, avoiding information loss.
[0035] 2. The application realizes complex structure model establishment by dynamic merging modeling method, reduces repeated merging operations in the modeling process to improve modeling efficiency.
[0036] 3. The model of the application can be used to predict anisotropy, strain hardening and fatigue damage of metal materials. BRIEF DESCRIPTION OF DRAWINGS
[0037] Figure 1A logic block diagram of a crystal performance prediction method based on four-field iteration and dynamic merging of the present application;
[0038] Figure 2 A schematic diagram of a four-field iteration method of the present application;
[0039] Figure 3 A schematic diagram of a dynamic merging method of the present application;
[0040] Figure 4 An EBSD structure of a material crystal used in the present application;
[0041] Figure 5 A crystal model established in an embodiment of the present application. DETAILED DESCRIPTION
[0042] The present application will be further described below in conjunction with the accompanying drawings and embodiments, which are simplified schematic diagrams and only schematically show the basic structure of the present application, and thus only show the components related to the present application.
[0043] As shown in Figure 1 , a crystal performance prediction method based on four-field iteration and dynamic merging includes the following steps:
[0044] The present application develops a secondary plug-in in Abaqus simulation software, calls three modules of Abaqus RSG, Matlab and Python to build an Abaqus plug-in to realize the establishment of a model.
[0045] Step one, obtain a crystal structure material and perform pretreatment;
[0046] The crystal structure material can be laser melting formed 316L stainless steel; it can also be other metal materials;
[0047] The pretreatment includes: first, 800#-2000# rough polishing is performed on the surface of 316L; second, 1.5 μm-0.25 μm emery polishing agent is used to cooperate with silk velvet cloth for fine polishing; finally, high-chloric acid: acetic acid solution is used for electrolytic polishing of 316L to make the surface meet the EBSD shooting requirements;
[0048] Step two, image acquisition and crystal model initialization are performed on the pretreated crystal structure material;
[0049] EBSD shooting is performed on 316L, and the hit rate should be ensured to be above 80% during shooting, the crystal structure data of 316L is obtained, and the post-processing software is used to post-process the EBSD data, save the data to obtain the 316L-EBSD IPF diagram as shown in Figure 4
[0050] Initializing the crystal model involves: opening the Abaqus software and launching the crystal model building plug-in. The plug-in calls the AbaqusRSG module to build a responsive interactive interface for the data processing and modeling plug-in. In the interface, the location of the electron backscatter diffraction data to be modeled, the location where the processed data is saved, and the location of the crystal plasticity simulation subroutine are selected; the material parameters and load size required by the subroutine are set; the material parameters such as elastic stiffness modulus and the load size such as displacement are set;
[0051] After the crystal model is initialized, initial crystal data is obtained; the initial crystal data includes: crystal image size, pixel X and Y direction step size, pixel coordinates, pixel crystal orientation and pixel phase parameters, etc.
[0052] Using the Matlab engine for Python function in the Python module to call the Matlab module to process the initial crystal data;
[0053] Use the METX toolkit to import EBSD data and enter the Matlab graphical interface to select the position and size of the crystal structure to be modeled; create the required coordinates of all crystal edge points based on the pixel point X and Y direction steps.
[0054] Step 3: Use the four-domain iteration method to calculate and judge the phase parameters and orientation angle differences of the gridded crystals, divide the adjacent grid crystals into grain groups, and extract the modeling data of the grains;
[0055] The four-domain iteration method includes:
[0056] Step 31: Figure 2 As shown, Figure 4 The crystal image is divided into several sub-grids, and the four corners of each sub-grid are the crystal edge coordinate points;
[0057] Preferably the sub-grids are square.
[0058] by Figure 2 For example, the image is divided into 36 sub-grids, which are numbered from the 1st grid to the 36th grid from left to right and from top to bottom.
[0059] Step 32: Taking a subgrid as the starting point, calculate the phase parameters and orientation angle differences between it and its adjacent subgrids;
[0060] Adjacent directions are up, down, left and right;
[0061] like Figure 2 In the figure, the first subgrid is taken as the starting point, and the adjacent subgrids are the second subgrid (right) and the seventh subgrid (bottom);
[0062] The formula for orientation angle difference is:
[0063]
[0064] where G ω is the misorientation matrix between two crystals, trace() is the trace function.
[0065] atn ω = 1 means the two crystals are consistent, atn ω = 0 means the two crystals are inconsistent.
[0066] where f(θ1) is used to determine whether the input parameter is within the allowed misorientation threshold err, which can be expressed as:
[0067]
[0068] g(θ2) is used to handle the symmetry problem of cubic system, which can be expressed as:
[0069]
[0070] Step 33, if the phase parameters of two adjacent subgrids are consistent and the difference of orientation angle θ1 is within the preset error er, then the crystals of the two subgrids are classified into the same grain group;
[0071] Step 34, if the subgrids do not satisfy step 33, then the growth stops;
[0072] For example Figure 2 , 36 subgrids are divided into four grain groups according to color, taking dark gray as an example, serial numbers 1-10 are the growth direction of the subgrids; for example, since serial number 10 does not satisfy the condition of step 33, the growth stops;
[0073] Step 35, taking a subgrid in another grain group as the starting point (serial number 1), steps 31-33 are repeatedly executed until all subgrids are traversed; all grain groups are obtained;
[0074] Step 36, grain modeling data is extracted using all grain groups obtained by division;
[0075] The grain modeling data includes: first data, second data, third data, fourth data, and fifth data.
[0076] The first data includes: model size and grain number.
[0077] The second data is the crystal edge point coordinate point.
[0078] The third data is the connection order of the serial numbers of the crystal edge coordinate points.
[0079] The fourth data is all grain material parameters.
[0080] Grain material parameters such as orientation angle and phase parameters;
[0081] The fifth data is the crystal serial number corresponding to the grain.
[0082] Step 4: Use the dynamic merging modeling method to establish a crystal plasticity model based on the first to fifth data; use the characteristics of the cubic crystal system to complete the crystal modeling and simulate the model mesh division according to the four-point coordinates of the crystal edge through the stretching method; after the crystal model of the same grain group is completed, merge it into a grain model and assign it material properties to complete the establishment of a certain grain group; this can reduce the number of merging and save time:
[0083] The crystal plasticity model based on the first to fifth data includes:
[0084] Step 41: The program performs the merge set parameter establishment step: the total number of crystals included in the multi-scale model is calculated by reading the basic information text;
[0085] Step 42, enter the crystal model establishment link: the program will first read the grain material text, and create material properties in row order. After each row of material is created, the program reads the material allocation text and extracts the crystal numbers stored in the row in sequence. Every time a crystal number is extracted, the program will go to the crystal system edge coordinate connection sequence text to find the edge coordinate number connection sequence of the corresponding crystal. According to this sequence, the program reads the coordinate point data from the coordinate point data text and completes the establishment of the crystal two-dimensional graphics; finally, the crystal modeling is completed by stretching.
[0086] Step 43: Initially merge the subgrids within the first grain group according to a preset merging threshold to obtain a plurality of first merged grids, then sequentially merge the plurality of first merged grids in pairs to obtain a second merged grid, then sequentially merge the plurality of second merged grids in pairs to obtain a third merged grid, and so on, merging the subgrids within the first grain group into the first grid in this manner;
[0087] Step 44: Use step 43 to complete the merging of the inner subgrids of other grain groups to obtain the grid corresponding to the grain group;
[0088] like Figure 2 and 3 As shown, Figure 2 Take the medium-dark gray grain group as an example (there are 4 grain groups in total: dark gray, medium-dark gray, light gray, and white), which is set as the first grain group. The preset merging threshold is 2, and the preset merging threshold can be customized. The first merged grid is divided into 5, that is, the 1st and 2nd subgrids are merged into the first first merged grid, the 2nd and 3rd subgrids are merged into the second first merged grid, and so on. The 9th and 10th subgrids are merged into the fifth first merged grid.
[0089] Then, the first first merged grid and the second first merged grid are merged to obtain the first second merged grid; the third first merged grid and the fourth first merged grid are merged to obtain the second second merged grid; the fifth first merged grid is set as the third second merged grid;
[0090] Then merge the first and second second merged grids into the first third merged grid, and set the third second merged grid as the second third merged grid;
[0091] Finally, the first and second third merged grids are merged into the fourth merged grid, thus completing the merging of the first grain group;
[0092] And so on, complete Figure 2 The merging of the 2nd, 3rd and 4th grain groups.
[0093] The merging mesh operation of the dynamic merging modeling method finally results in an image without meshes;
[0094] like Figure 5 After the merging is completed, the parameters of the merged set are updated, and this cycle is repeated until all crystal models are completed; finally, the optimal merging of the crystal model grains is completed based on the number of grains included. After the merging is completed, operations such as meshing, load application, and analysis step setting are performed to complete the model establishment. Figure 5 The model shown, visible with Figure 4 The EBSD-IPF images shown are basically consistent.
[0095] With the above-described preferred embodiments of the present invention as a guide, and with reference to the above description, relevant personnel are fully capable of making various changes and modifications without departing from the technical scope of this invention. The technical scope of this invention is not limited to the contents of the specification and must be determined according to the scope of the claims.
Claims
1. A crystal performance prediction method based on four-domain iteration and dynamic merging, characterized in that: The following steps are involved: Step 1: Obtain crystal structure materials and perform pretreatment; Step 2: Capture images of crystal structure materials and initialize crystal models; Step 3: Calculate and judge the phase parameters and orientation angle difference of the gridded crystals using the four-domain iteration method, and divide the adjacent grid crystals into grain groups; Step 4: Use the dynamic merging modeling method to establish a crystal plasticity model based on the grain modeling data; merge the sub-grids in different grain groups to obtain a merged grid.
2. The crystal performance prediction method based on four-domain iteration and dynamic merging according to claim 1, characterized in that: Step three specifically includes: Step 31: Divide the crystal image into several sub-grids, and extract the crystal edge coordinate points of the sub-grids; Step 32: Taking a subgrid as a starting point, calculate the phase parameters and orientation angle differences of the subgrids adjacent to it; Step 33: If the phase parameters of two adjacent subgrids are consistent and the orientation angle difference is within a preset error range, the two subgrids are classified as the same grain group; Step 34: Repeat the process until a subgrid does not meet the conditions of step 33, and the division of the grain group is completed. Step 35: Repeat steps 31 to 34 to complete the division of other grain groups; Step 36: Extracting grain modeling data for all divided grain groups.
3. The crystal performance prediction method based on four-domain iteration and dynamic merging according to claim 2, characterized in that: The formula for orientation angle difference is: Among them, G ω is the orientation difference matrix of the two crystals; trace() is the trace function, and err is the orientation difference threshold.
4. The crystal performance prediction method based on four-domain iteration and dynamic merging according to claim 1, characterized in that: Step 4 specifically includes: Step 41: First, the subgrids in a certain grain group are initially merged according to a preset merging threshold to obtain a plurality of first merged grids; second, the plurality of first merged grids are sequentially merged in pairs to obtain a second merged grid; third, the plurality of second merged grids are sequentially merged in pairs to obtain a third merged grid, and so on; finally, all the subgrids in the grain group are merged to obtain a first grid; Step 42: Merge subgrids of other grain groups according to step 41.
5. The crystal performance prediction method based on four-domain iteration and dynamic merging according to claim 1, characterized in that: Preprocessing includes: The surface of crystal structure materials is subjected to rough polishing, fine polishing and electrolytic polishing.
6. The crystal performance prediction method based on four-domain iteration and dynamic merging according to claim 1, characterized in that: Images were acquired using EBSD for crystal structured materials.
7. The crystal performance prediction method based on four-domain iteration and dynamic merging according to claim 1, characterized in that: Crystal model initialization includes: Start the crystal model plug-in and build a responsive interactive interface for data processing and modeling plug-ins; Select the electron backscatter diffraction data location, data storage location and crystal plasticity simulation subroutine location; Set the material parameters and load magnitude.
8. The crystal performance prediction method based on four-domain iteration and dynamic merging according to claim 7, characterized in that: The data for crystal model initialization include: crystal image size, pixel X and Y direction step size, pixel coordinates, pixel crystal orientation and pixel phase parameters.
9. A crystal performance prediction system based on four-domain iteration and dynamic merging, characterized by: include: a memory for storing instructions executable by the processor; A processor, configured to execute instructions to implement the crystal performance prediction method based on four-domain iteration and dynamic merging as described in any one of claims 1 to 8.
10. A computer-readable medium storing computer program code, characterized in that When the computer program code is executed by a processor, the computer program code implements the crystal performance prediction method based on four-domain iteration and dynamic merging according to any one of claims 1 to 8.