Automatic modeling method and system of macro-micro multi-scale model based on real microstructure

Abstract

The application belongs to the field of material science and engineering, and discloses a macro-micro multi-scale model full-automatic modeling method and system based on real microstructure; fine EBSD data processing technology is adopted, so that the orientation, grain boundary and size information of the grain can be accurately obtained, and the overall quality of the data is improved through effective filtering of the smallest grains; through accurate data input, it is ensured that the microstructure can be fully reflected in the macro model; through effective integration of the EBSD data and ABAQUS finite element simulation, the close correlation between the macro and micro scales is successfully established; on this basis, the change of the microstructure can be tracked in real time, and the change is transmitted to the macro stress and strain response model, so that the stress distribution and plastic behavior of the material under different load conditions can be more accurately described.

Description

Technical Field

[0001] The present invention belongs to, but is not limited to, the technical field of materials science and engineering, and in particular relates to a fully automatic modeling method and system for a macro-micro multi-scale model based on real microstructures. Background Art

[0002] In materials science and engineering, grain structure plays a crucial role in determining a material's macroscopic mechanical properties. In metals in particular, grain size, morphology, and orientation directly influence the material's strength, ductility, hardness, and fatigue resistance. Therefore, studying the evolution of grains under varying conditions, particularly orientation changes under complex stress fields and their impact on the material's macroscopic properties, is a crucial research topic in materials science.

[0003] Currently, electron backscatter diffraction (EBSD) technology has been widely used in the characterization of the microstructure of materials. EBSD can provide detailed microstructural information such as grain orientation, grain boundaries, and phase information, allowing researchers to deeply understand the relationship between the microstructure and macroscopic mechanical properties of materials. However, EBSD technology usually only provides static grain information and cannot directly capture the evolution of grains during dynamic loading. To make up for this deficiency, the finite element method (FEM) is widely used to simulate and predict the deformation behavior of materials during loading. By combining EBSD experimental data with finite element simulation, researchers can more accurately describe the grain rotation and deformation behavior of materials under complex stress states.

[0004] Although the method of combining EBSD data with finite element simulation has made some progress in materials research, existing technologies still have certain limitations. In macro-micro multi-scale analysis, existing technologies have difficulty in effectively establishing an accurate correlation between the two. Traditional finite element methods often focus on the macro-scale material response and ignore the evolution of the microscopic grain structure under stress. While EBSD technology can characterize the microstructure in detail, it lacks the ability to dynamically update in real time when correlating it with macroscopic properties. This results in the influence of microstructure on the macroscopic mechanical properties of materials under complex loading conditions not being fully considered, limiting the accuracy and applicability of the simulation results.

[0005] In summary, establishing accurate multiscale models is of paramount importance in materials science and engineering. Combining EBSD data with finite element simulation not only enables in-depth analysis of structural features such as grain orientation and grain boundary evolution at the microscopic level, but also allows this microscopic information to be transferred to the macroscopic scale for more accurate predictions of the material's overall mechanical properties. Therefore, developing dynamic models that effectively integrate the macro and microscales is crucial for improving the accuracy and reliability of material behavior predictions. This not only helps advance materials science but also provides a solid theoretical foundation for optimizing material properties in engineering applications. Summary of the Invention

[0006] In response to the problems existing in the prior art, the present invention provides a fully automatic modeling method and system for a macro-micro multi-scale model based on real microstructures.

[0007] The present invention is achieved by a fully automatic modeling method of a macro-micro multi-scale model based on real microstructures, the method comprising:

[0008] S1: Call basic controls such as uifigure and uilabel through MATLAB, use SelectionChangedFcn and SizeChangedFcn to achieve responsive interaction of the multi-scale modeling system, and build the user interface of the macro-micro multi-scale modeling system;

[0009] S2: Select the corresponding macro model and its corresponding boundary conditions in the interface of S1, and use MATLAB to write the py file to create the ABAQUS initial model entity; at the same time, complete the creation of the equipment body, establishment of analysis steps, creation of reference points, establishment of reference point coupling constraints, addition of fixed constraints, addition of load constraints, and the setting of corresponding boundary conditions;

[0010] S3: Use MATLAB to read the real microstructure data obtained by EBSD, automatically obtain the length, width and scanning step of the EBSD experimental scanning area, and divide the area of ​​the initial macro model where the micro model needs to be added into a surface domain with the same length and width as the EBSD scanning area, and record the initial vertex coordinates of the surface domain; then, perform coarse meshing on the initial macro model, mesh the micro model area according to the size of the EBSD experimental scanning step, and generate an INP file;

[0011] S4: Use MATLAB to call the METX toolbox to import EBSD data and create an EBSD dataset in the MATLAB workspace;

[0012] S5: Use methods such as grain segmentation, confidence index filtering, and small grain filtering to create grain sets for EBSD data, delete extremely small grains, reduce noise in EBSD data, and improve the quality of EBSD data;

[0013] S6: extract the denoised EBSD data, draw the denoised grain distribution map, and compare it with the grain distribution map of the EBSD data;

[0014] S7: Read each node data and unit data of the initial model through MATLAB; generate a data file of the node and unit set of each grain in ABAQUS based on the position of each grain (X, Y coordinates) and the pixel value occupied by each grain; then align the generated data file with the initial vertex coordinates of the surface domain recorded in S3 and write it into the INP file of the initial model to generate an INP file of the multi-scale crystal plasticity finite element model containing both the microscopic grain structure model and the macroscopic model;

[0015] S8: Convert the grain Euler angle into radian value, and use MATLAB to output the PYTHON file that assigns the material properties of each grain and the material properties of the macro model;

[0016] S9: Call ABAQUS through MATLAB to execute the PYTHON file of material property assignment, assign material properties to each grain and macro model at the same time, and generate the final macro-micro multi-scale crystal plasticity finite element model INP file;

[0017] S10: Call ABAQUS through MATLAB to run the final generated INP file, call the subroutine of the crystal plasticity finite element simulation, output the orientation information of each grain in the final generated crystal plasticity finite element model, and call METX to draw the pole figure of each grain in the final output; at the same time, call the METX toolbox through MATLAB to draw the pole figures of the original EBSD data and the pole figures of the grain set after noise reduction, and compare the consistency of the pole figures finally output by the model with the initial pole figures and the pole figures after noise reduction;

[0018] S11: Write a Python file to extract the stress-strain curve of the crystal plasticity finite element simulation through MATLAB, call ABAQUS through MATLAB to execute the Python file, obtain the stress-strain curve of the macro-micro multi-scale crystal plasticity finite element simulation and compare it with the experimental results to verify the accuracy of the model.

[0019] Furthermore, the initial model entities in S2 include macroscopic standard models commonly used in experimental tests, such as tensile models, crack extension models, and impact toughness models; the data set in S4 contains important information such as grain type, grain ID, grain orientation, phase sequence, and grain rotation.

[0020] Furthermore, the formulas for grain segmentation, confidence index filtering, and small grain filtering in S5 are:

[0021]

[0022] Wherein, the grain rotation is represented as R1 and R2, and Δθ is the orientation angle difference between the two points. If Δθ>angle, the two points are considered to belong to different grains;

[0023]

[0024] For each measurement point i, its confidence index is CI i ;

[0025]

[0026] Where g is the size of each grain and size(g) is the size of each grain.

[0027] Furthermore, the EBSD data in S6 includes data such as the grain number, equivalent diameter, grain orientation Euler angle, grain pixel coordinates, and grain set of each grain.

[0028] Furthermore, the macro model material property assignment code in S8 includes four parts: automatically reading the radian value of the grain Euler angle, converting the radian value into the Miller index, assigning the grain size effect, and converting the macro model engineering stress and strain into real stress and strain.

[0029] Furthermore, the formulas for converting the radian value into the Miller index, the grain size effect, and the engineering stress and strain into the true stress and strain are:

[0030]

[0031] τ i =τ0+Kd -0.5 (i=1,2,3…)

[0032] σ t =σ e (1+ε e )

[0033] ε t =ln(1+ε e )

[0034] Where θ is the nutation angle, ψ is the precession angle, φ is the rotation angle; u, v, w are the crystal direction indices; h, k, l are the crystal plane indices; i is the grain number, τ0 is the initial yield strength of the grain, K is the influence coefficient of the reaction grain boundary on the deformation, d is the equivalent diameter of the grain, σ e is the engineering stress, σ tis the true stress, ε e is the engineering strain, ε t For true resilience.

[0035] Another object of the present invention is to provide a fully automatic modeling system for a macro-micro multi-scale model based on a real microstructure based on the fully automatic modeling method for a macro-micro multi-scale model based on a real microstructure, the system specifically comprising:

[0036] The interface construction module calls basic controls such as uifigure and uilabel through MATLAB, uses SelectionChangedFcn and SizeChangedFcn to achieve responsive interaction of the multi-scale modeling system, and builds the user interface of the macro-micro multi-scale modeling system;

[0037] The boundary condition setting module is connected to the interface construction module. In the user interface, the corresponding macro model and its corresponding boundary conditions are selected, and MATLAB is used to write the py file to create the ABAQUS initial model entity. At the same time, the module completes the creation of the equipment body, establishment of the analysis step, creation of the reference point, establishment of the reference point coupling constraint, addition of the fixed constraint, addition of the load constraint, and the setting of the corresponding boundary conditions.

[0038] The INP file generation module is connected to the boundary condition setting module. It reads the real microstructure data obtained by EBSD through MATLAB, automatically obtains the length, width and scanning step of the EBSD experimental scanning area, and divides the area of ​​the initial macro model where the micro model needs to be added into a surface domain with the same length and width as the EBSD scanning area, and records the initial vertex coordinates of the surface domain; then, the initial macro model is coarsely meshed, and the micro model area is meshed according to the size of the EBSD experimental scanning step, and the INP file is generated;

[0039] The dataset creation module is connected to the INP file generation module, uses MATLAB to call the METX toolbox to import EBSD data, and creates an EBSD dataset in the MATLAB workspace;

[0040] The grain establishment module is connected to the data set creation module. It uses methods such as grain segmentation, confidence index filtering, and small grain filtering to establish grain sets for EBSD data, delete extremely small grains, and reduce noise in EBSD data to improve the quality of EBSD data.

[0041] The distribution map drawing module is connected to the grain establishment module to extract the denoised EBSD data, draw the denoised grain distribution map, and compare it with the grain distribution map of the EBSD data;

[0042] The INP file writing module is connected to the distribution map drawing module and the INP file generation module, and reads each node data and unit data of the initial model through MATLAB; based on the X, Y coordinates of the position of each grain and the pixel value occupied by each grain, a data file of the node and unit set of each grain in ABAQUS is generated; then, the generated data file is aligned with the initial vertex coordinates of the surface domain recorded by the INP file generation module and written into the INP file of the initial model, thereby generating an INP file of a multi-scale crystal plasticity finite element model containing both the microscopic grain structure model and the macroscopic model;

[0043] The property assignment module converts the grain Euler angle into radian value, and uses MATLAB to output the Python file that assigns the material properties of each grain and the material properties of the macro model;

[0044] The final INP file generation module is connected to the INP file writing module and the property assignment module. It calls ABAQUS through MATLAB to execute the Python file for material property assignment, assigns material properties to each grain and macro model at the same time, and generates the final macro-micro multi-scale crystal plasticity finite element model INP file;

[0045] The pole figure drawing module is connected to the final INP file generation module. MATLAB calls ABAQUS to run the final INP file and calls the crystal plasticity finite element simulation subroutine to output the orientation information of each grain in the final crystal plasticity finite element model. It calls METX to draw the pole figures of each grain in the final output. At the same time, MATLAB calls the METX toolbox to draw the pole figures of the original EBSD data and the pole figures of the grain set after noise reduction. The consistency of the pole figures finally output by the model is compared with the initial pole figures and the pole figures after noise reduction.

[0046] The verification module is connected to the pole figure drawing module. It writes a Python file to extract the stress-strain curve of the crystal plasticity finite element simulation through MATLAB, calls ABAQUS through MATLAB to execute the Python file, obtains the stress-strain curve of the macro-micro multi-scale crystal plasticity finite element simulation and compares it with the experimental results to verify the accuracy of the model.

[0047] Another object of the present invention is to provide a computer device, which includes a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the fully automatic modeling method of the macro-micro multi-scale model based on real microstructures.

[0048] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, enables the processor to execute the steps of the fully automatic modeling method of the macro-micro multi-scale model based on real microstructures.

[0049] Another object of the present invention is to provide an information data processing terminal, which is used to implement the fully automatic modeling system of the macro-micro multi-scale model based on real microstructures.

[0050] In combination with the above technical solutions and the technical problems solved, the advantages and positive effects of the technical solutions to be protected by the present invention are as follows:

[0051] First, the present invention addresses the shortcomings of existing technologies in modeling multi-scale grain plastic behavior by combining EBSD experimental data with finite element simulation technology to propose an improved multi-scale modeling method. First, the present invention uses sophisticated EBSD data processing technology, which not only accurately obtains information such as grain orientation, grain boundaries, and size, but also improves the overall quality of the data by effectively filtering out extremely small grains. Traditional methods often ignore the complexity of microscopic grains when processing these details, resulting in insufficient modeling accuracy. However, the present invention ensures that the microstructure can be fully reflected in the macroscopic model through precise data input.

[0052] Secondly, the present invention successfully established a close relationship between macro and micro scales by effectively integrating EBSD data with ABAQUS finite element simulation. On this basis, the changes in microstructure can be tracked in real time and transferred to the macro stress-strain response model to ensure that the stress distribution and plastic behavior of the material under different load conditions are more accurately described. This process does not require tedious post-processing steps, greatly improves the efficiency and reliability of the simulation, and reduces manpower and time costs. In addition, the implementation of this method relies on common software platforms, such as MATLAB and ABAQUS, and the system structure is simple and easy to operate, with wide applicability and promotion value.

[0053] In summary, this invention improves existing EBSD data processing and finite element simulation methods, not only enhancing the accuracy of multiscale analysis of dynamic grain behavior but also effectively capturing the evolution of microstructures under dynamic stress conditions, thereby improving the accuracy and applicability of macroscopic simulation results. This technology offers the advantages of simple structure, excellent performance, ease of implementation, and low cost, and holds broad application prospects in materials science research and industrial applications.

[0054] Second, as auxiliary evidence for the inventiveness of the claims of the present invention, it is also reflected in the following important aspects:

[0055] (1) The expected benefits and commercial value of the technical solution of the present invention after transformation are:

[0056] This invention significantly improves research efficiency in the field of materials science by combining automated modeling technology with EBSD experimental data and finite element simulation. This automated, multi-scale modeling system has important application value in aspects such as material property prediction, structural optimization, and failure analysis. It can provide more accurate simulation and prediction for the design and manufacturing process of materials, especially in fields such as aerospace, automobile manufacturing, shipbuilding, and high-end equipment manufacturing, where it has significant market demand and commercial value. By optimizing the mechanical properties of materials, the system can shorten R&D cycles, reduce experimental costs, improve product quality, and thus create considerable economic benefits for enterprises.

[0057] (2) The technical solution of the present invention fills the technical gap in the industry at home and abroad:

[0058] Currently, the industry lacks an automated modeling system that can effectively link microstructural information with macroscopic mechanical properties. This invention fills this gap, achieving for the first time the fully automated construction of a macro-micro multiscale model based on realistic microstructures. By integrating EBSD technology and finite element simulation, this invention dynamically captures the evolution of grain structure under complex stress states, thereby more accurately describing the impact of microstructure on the material's macroscopic properties. This innovation transcends the limitations of existing technologies and brings a completely new research approach to the field of materials science.

[0059] (3) The technical solution of the present invention solves the technical problems that people have been eager to solve but have never been able to solve successfully:

[0060] The grain rotation and deformation behavior of materials under complex stress fields has always been an important research problem in materials science. Traditional experimental methods are unable to capture the dynamic evolution of grains during loading in real time, and existing simulation methods also have difficulty effectively linking micro and macro structures. The present invention successfully solves this technical problem by developing a multi-scale automatic modeling system based on real microstructures, realizing the simulation of the dynamic evolution of grain structure under complex stress states and accurately transmitting this microscopic information to the macro scale. This technological breakthrough provides a reliable tool for studying the deformation behavior of materials under complex stress fields.

[0061] (4) The technical solution of the present invention overcomes technical prejudice:.

[0062] Existing technologies typically focus on predicting material properties at the macroscale, ignoring the impact of microstructural evolution under stress on overall performance. This methodological approach overcomes this bias by integrating microscopic grain structure with macroscopic material response, achieving a dynamic correlation between microstructure and macroscopic properties through a multiscale automated modeling system. This innovative perspective not only expands the application scope of existing materials simulation techniques but also provides a more comprehensive and accurate theoretical foundation and practical support for predicting material behavior.

[0063] Third, the technical solution of the present invention solves many problems in the prior art in industrial applications and achieves significant technological progress, which is specifically reflected in the following aspects:

[0064] 1. Existing technical problems solved

[0065] (1) Manual modeling is inefficient and difficult to handle complex multi-scale problems

[0066] In existing technologies, the construction of macro-micro multiscale models often relies on manual operations, especially when processing real-world microstructural data. This leads to high levels of human involvement, low efficiency, and the potential for errors. Especially in complex materials, the integration of microstructural processing and macroscopic models places extremely high demands on manual intervention, resulting in a lengthy and cumbersome modeling process, which in turn affects efficiency in industrial applications.

[0067] (2) Microscopic data processing is complex, and data noise reduction and grain feature extraction are difficult

[0068] In existing technologies, processing real-world microstructural data (such as EBSD data) often requires highly specialized software and complex workflows. Steps like noise reduction and grain feature extraction are cumbersome and susceptible to the effects of noise and small grains, resulting in inaccurate microfeature extraction and ultimately affecting simulation and analysis results.

[0069] (3) Inaccurate assignment of material properties to multi-scale models

[0070] Traditional modeling processes lack effective automated solutions for assigning material properties to multi-scale models. This is particularly true when integrating microscopic grain structures with macroscopic models, where the assignment of material properties is complex. Existing technologies often rely on manual operations or imperfect automated processes, often resulting in inaccurate material property assignments and impacting the reliability of subsequent simulation analysis.

[0071] 2. Significant technological advancement of the present invention

[0072] (1) Fully automatic modeling to improve modeling efficiency

[0073] This paper addresses the inefficiency of existing manual operations by introducing an automated modeling system. The system uses MATLAB to call multiple modules, integrating the entire process of data acquisition, processing, and model generation. This system enables fully automated operations, from interface design and boundary condition setting to INP file generation. This significantly reduces manual effort and improves the efficiency and accuracy of model building, making it particularly suitable for complex materials and multi-scale modeling requirements in industry.

[0074] (2) Integrated EBSD data processing to improve microstructure processing accuracy

[0075] The technical solution of this invention integrates MATLAB with the METX toolbox to automate EBSD data processing, including dataset creation, grain segmentation, confidence index filtering, and noise reduction. By introducing grain noise reduction and filtering, the interference of small grains and noise on model construction is eliminated, significantly improving the accuracy of microstructure feature extraction and ensuring the accuracy of microscopic components in multiscale models.

[0076] (3) Automatically assign material properties to ensure the consistency of model material properties

[0077] This method assigns material properties through an automated process. Using Python scripts generated by MATLAB and calling ABAQUS, material properties are assigned simultaneously at different scales during model construction. Whether it's the material properties of microscopic grains or the material properties of the macroscopic model, these properties can be accurately assigned through automation. This not only reduces the errors associated with manual material property assignment but also improves the accuracy of model simulation calculations.

[0078] (4) Multi-module collaboration to achieve data verification and model optimization

[0079] The technical solution of the present invention enables the coordinated operation of multiple modules at various stages of modeling, such as through INP file generation, pole figure drawing, and stress-strain curve analysis, to ensure model accuracy. In particular, the pole figure drawing module verifies the consistency of the denoised data with the simulation results by generating and comparing grain pole figures. The verification module further ensures the reliability of the simulation results by comparing them with experimental data, making them suitable for comprehensive performance evaluation in engineering applications.

[0080] 3. Significant progress in industrial applications

[0081] The technical solution of this invention achieves a fully automated process from data acquisition to model generation in industrial applications, not only improving the efficiency of model building, but also significantly enhancing modeling accuracy. The system can be widely used in a variety of fields, including materials research and development, industrial manufacturing, aerospace, and more. It is particularly suitable for building multi-scale models of complex materials, providing an efficient and reliable solution. Furthermore, the system integrates data verification and material performance analysis modules, which can help companies conduct product development and structural optimization more quickly, reducing development costs and improving work efficiency.

[0082] Therefore, the present invention solves many pain points in the existing technology through multi-module integration and a fully automated modeling system, and demonstrates significant technological progress in industrial applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] Figure 1 This is a flow chart of a fully automatic modeling method for a macro-micro multi-scale model based on real microstructures provided by an embodiment of the present invention;

[0084] Figure 2 This is an interface diagram of a multi-scale model building system provided by an embodiment of the present invention;

[0085] Figure 3 is an initial model and its boundary condition diagram provided by an embodiment of the present invention;

[0086] Figure 4 is an initial model grid division diagram provided by an embodiment of the present invention;

[0087] Figure 5 This is a comparison diagram of the original EBSD grain distribution and the grain distribution after noise reduction provided by an embodiment of the present invention;

[0088] Figure 6 is a macro-micro multi-scale crystal plasticity model diagram provided by an embodiment of the present invention;

[0089] Figure 7 This is a polar figure comparison diagram provided by an embodiment of the present invention;

[0090] Figure 8 This is a comparison diagram between simulation and experiment provided by an embodiment of the present invention;

[0091] Figure 9 This is a graph showing the change in dislocation density during a multi-scale model fatigue simulation process according to an embodiment of the present invention;

[0092] Figure 10 This is a life prediction curve diagram based on multi-scale model fatigue simulation according to an embodiment of the present invention;

[0093] Figure 11 This is a structural diagram of a fully automatic modeling system for a macro-micro multi-scale model based on real microstructures provided by an embodiment of the present invention;

[0094] In the figure: 1. Interface construction module; 2. Boundary condition setting module; 3. INP file generation module; 4. Data set creation module; 5. Grain establishment module; 6. Distribution map drawing module; 7. INP file writing module; 8. Attribute assignment module; 9. Final INP file generation module; 10. Pole figure drawing module; 11. Verification module. DETAILED DESCRIPTION

[0095] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0096] like Figure 1 As shown, an embodiment of the present invention provides a fully automatic modeling method of a macro-micro multi-scale model based on real microstructures, the method comprising:

[0097] S1: Call basic controls such as uifigure and uilabel through MATLAB, use SelectionChangedFcn and SizeChangedFcn to realize responsive interaction of multi-scale modeling system, and build the user interface of macro-micro multi-scale modeling system, such as Figure 2 shown.

[0098] S2: Select the corresponding macro model and its corresponding boundary conditions in the interface of step 1, and use MATLAB to write the py file to create the initial model entity of ABAQUS. The initial model entity includes macro standard models such as tensile model, crack growth model, impact toughness model, etc. commonly used in experimental tests. At the same time, complete the creation of equipment body, establishment of analysis steps, creation of reference points, establishment of reference point coupling constraints, addition of fixed constraints, addition of load constraints, etc., which are corresponding to the boundary conditions in step 1. Figure 3 shown.

[0099] S3: Use MATLAB to read the real microstructure data obtained by EBSD, automatically obtain the length, width and scanning step of the EBSD experimental scanning area, and divide the area of ​​the initial macro model where the micro model needs to be added into the area with the same length and width as the EBSD scanning area, and record the initial vertex coordinates of the area. Then, perform coarse meshing on the initial macro model, mesh the micro model area according to the size of the EBSD experimental scanning step, and generate an INP file, such as Figure 4 shown.

[0100] S4: Use MATLAB to call the METX toolbox to import EBSD data and create an EBSD dataset in the MATLAB workspace. The dataset contains important information such as grain type, grain ID, grain orientation, phase sequence, and grain rotation.

[0101] S5: Use methods such as grain segmentation, confidence index filtering, and small grain filtering to establish grain sets for EBSD data, delete extremely small grains, reduce noise in EBSD data, and improve the quality of EBSD data. The formulas for grain segmentation, confidence index filtering, and small grain filtering are:

[0102]

[0103] Here, the grain rotations are represented as R1 and R2, and Δθ is the orientation angle difference between the two points. If Δθ > angle, the two points are considered to belong to different grains.

[0104]

[0105] For each measurement point i, its confidence index is CI i .

[0106]

[0107] Where g is the size of each grain and size(g) is the size of each grain.

[0108] S6: Extract the EBSD data after noise reduction, including the grain number, equivalent diameter, grain orientation Euler angle, grain pixel coordinates, grain set and other data of each grain, draw the grain distribution map after noise reduction, and compare it with the grain distribution map of EBSD data, such as Figure 5 shown.

[0109] S7: Read each node data and unit data of the initial model through MATLAB. According to the position of each grain (X, Y coordinates) and the pixel value occupied by each grain, generate a data file of the node and unit set of each grain in ABAQUS. Then, align the generated data file with the initial vertex coordinates of the face domain recorded in step 3, write it into the INP file of the initial model, and generate an INP file of the multi-scale crystal plasticity finite element model containing both the microscopic grain structure model and the macroscopic model, such as Figure 6 shown.

[0110] S8: Convert the grain Euler angles to radians, and use MATLAB to output a Python file that assigns material properties to each grain and the macro model. The material property assignment code includes four parts: automatically reading the grain Euler angle radians, converting the radians to Miller indices, assigning grain size effects, and converting the macro model engineering stress and strain to true stress and strain. The formulas for converting radians to Miller indices, grain size effects, and engineering stress and strain to true stress and strain are:

[0111]

[0112] τ i =τ0+Kd -0.5 (i=1,2,3…)

[0113] σ t =σ e (1+ε e )

[0114] ε t =ln(1+ε e )

[0115] Where θ is the nutation angle, ψ is the precession angle, and φ is the rotation angle. u, v, and w are the crystal direction indices. h, k, and l are the crystal plane indices. i is the grain number, τ0 is the initial yield strength of the grain, K is the coefficient of influence of the reactive grain boundary on deformation, d is the equivalent diameter of the grain, and σ is the initial yield strength of the grain. e is the engineering stress, σ t is the true stress, ε e is the engineering strain, ε t For true resilience.

[0116] S9: Call ABAQUS through MATLAB to execute the PYTHON file for assigning material properties, assign material properties to each grain and macro model at the same time, and generate the final macro-micro multi-scale crystal plasticity finite element model INP file.

[0117] S10: MATLAB calls ABAQUS to run the final generated INP file, and calls the subroutine of the crystal plasticity finite element simulation to output the orientation information of each grain in the final generated crystal plasticity finite element model, and calls METX to draw the pole figure of each grain in the final output. At the same time, MATLAB calls the METX toolbox to draw the pole figures of the original EBSD data and the pole figures of the grain set after noise reduction, and compares the consistency of the pole figures finally output by the model with the initial pole figures and the pole figures after noise reduction, as shown in Figure 1. Figure 7 shown.

[0118] S11: Write a Python file to extract the stress-strain curve of the crystal plasticity finite element simulation through MATLAB, call ABAQUS through MATLAB to execute the Python file, obtain the stress-strain curve of the macro-micro multi-scale crystal plasticity finite element simulation and compare it with the experimental results to verify the accuracy of the model, such as Figure 8 At the same time, after assigning the fatigue crack growth boundary conditions, the multi-scale model can also output the dislocation density change after crack growth, fatigue life prediction and other results, as shown in the figure below. Figure 9 and Figure 10 shown.

[0119] In summary, MATLAB uses ABAQUS and Python for co-simulation programming, proposing a fully automated modeling method and system for macro-micro multiscale models based on the material's real microstructure. This method and system can fully automatically build macro-micro multiscale models from EBSD microstructure data. This fully automated multiscale modeling method and system are universally applicable and can be used to build macro-micro multiscale crystal plasticity finite element models for materials such as aluminum alloys, titanium alloys, magnesium alloys, and stainless steel.

[0120] like Figure 11 As shown, an embodiment of the present invention provides a fully automatic modeling system for a macro-micro multi-scale model based on a real microstructure based on the fully automatic modeling method for a macro-micro multi-scale model based on a real microstructure, the system specifically comprising:

[0121] Interface construction module 1, through MATLAB calling uifigure, uilabel and other basic controls, using SelectionChangedFcn and SizeChangedFcn to achieve responsive interaction of the multi-scale modeling system, and build the user interface of the macro-micro multi-scale modeling system;

[0122] Boundary condition setting module 2 is connected to interface construction module 1. In the user interface, the corresponding macro model and its corresponding boundary conditions are selected, and MATLAB is used to write the py file for creating the ABAQUS initial model entity. At the same time, the creation of the equipment body, establishment of the analysis step, creation of the reference point, establishment of the reference point coupling constraint, addition of the fixed constraint, addition of the load constraint, and the setting of the corresponding boundary conditions are completed.

[0123] The INP file generation module 3 is connected to the boundary condition setting module 2, and reads the real microstructure data obtained by EBSD through MATLAB, automatically obtains the length, width and scanning step of the EBSD experimental scanning area, and divides the area of ​​the initial macro model where the micro model needs to be added into a surface domain with the same length and width as the EBSD scanning area, and records the initial vertex coordinates of the surface domain; then, the initial macro model is coarsely meshed, the micro model area is meshed according to the size of the EBSD experimental scanning step, and an INP file is generated;

[0124] The data set creation module 4 is connected to the INP file generation module 3, uses MATLAB to call the METX toolbox to import EBSD data, and creates an EBSD data set in the MATLAB workspace;

[0125] The grain establishment module 5 is connected to the data set creation module 4, and uses methods such as grain segmentation, confidence index filtering, and small grain filtering to establish a grain set for the EBSD data, delete extremely small grains, and reduce the noise of the EBSD data to improve the quality of the EBSD data;

[0126] The distribution map drawing module 6 is connected to the grain establishment module 5, extracts the EBSD data after noise reduction, draws the grain distribution map after noise reduction, and compares it with the grain distribution map of the EBSD data;

[0127] The INP file writing module 7 is connected to the distribution map drawing module 6 and the INP file generation module 3, and reads each node data and unit data of the initial model through MATLAB; based on the X, Y coordinates of the position of each grain and the pixel value occupied by each grain, a data file of the node and unit set of each grain in ABAQUS is generated; then, the generated data file is aligned with the initial vertex coordinates of the surface area recorded by the INP file generation module, and written into the INP file of the initial model, thereby generating an INP file of a multi-scale crystal plasticity finite element model containing both the microscopic grain structure model and the macroscopic model;

[0128] The property assignment module 8 converts the Euler angle of the grain into a radian value, and uses MATLAB to output a Python file that assigns the material properties of each grain and the material properties of the macro model;

[0129] The final INP file generation module 9 is connected to the INP file writing module 7 and the attribute assignment module 8. It calls ABAQUS through MATLAB to execute the Python file of material attribute assignment, assigns material attributes to each grain and macro model at the same time, and generates the final macro-micro multi-scale crystal plasticity finite element model INP file;

[0130] The pole figure drawing module 10 is connected to the final INP file generation module 9. MATLAB calls ABAQUS to run the final generated INP file, and calls the subroutine of the crystal plasticity finite element simulation to output the orientation information of each grain in the final generated crystal plasticity finite element model, and calls METX to draw the pole figure of each grain in the final output; at the same time, MATLAB calls the METX toolbox to draw the pole figures of the original EBSD data and the pole figures of the grain set after noise reduction, and compares the consistency of the pole figures finally output by the model with the initial pole figures and the pole figures after noise reduction;

[0131] The verification module 11 is connected to the pole figure drawing module 10, and writes a Python file for extracting the stress-strain curve of the crystal plasticity finite element simulation through MATLAB. The Python file is executed by calling ABAQUS through MATLAB to obtain the stress-strain curve of the macro-micro multi-scale crystal plasticity finite element simulation and compare it with the experimental results to verify the accuracy of the model.

[0132] An embodiment of the present invention provides a computer device, which includes a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the fully automatic modeling method of the macro-micro multi-scale model based on real microstructures.

[0133] An embodiment of the present invention provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the processor executes the steps of the fully automatic modeling method of the macro-micro multi-scale model based on real microstructures.

[0134] An embodiment of the present invention provides an information data processing terminal, which is used to implement the fully automatic modeling system of the macro-micro multi-scale model based on real microstructures.

[0135] The present invention's fully automated macro-micro multiscale modeling system, based on real microstructures, aims to achieve multiscale modeling of material microstructures and macroscale models through an automated process. The system integrates MATLAB and ABAQUS, with modules working collaboratively to achieve model generation, analysis, and verification. Its operating principle is specifically divided into the following six steps:

[0136] First, Interface Construction Module 1 provides the user interface, using MATLAB controls such as uifigure and uilabel. It also implements responsive interaction within the multiscale modeling system through the SelectionChangedFcn and SizeChangedFcn functions. The user selects the macromodel through this interface and sets the corresponding boundary conditions. Once the boundary conditions are set, the system generates a .py file for ABAQUS using MATLAB, ensuring the initial creation of the model and the correct setting of the boundary conditions. This module also includes the necessary operations for creating the rig, reference points, coupling constraints, fixed constraints, and load constraints.

[0137] Next, the Boundary Condition Setup Module 2 allows users to define the corresponding macromodel and its corresponding boundary conditions within the interface. This module, integrated with ABAQUS via MATLAB, automatically generates .py files related to the boundary conditions and completes the creation and analysis of the initial model entities. This module handles the initial definition of the model, including creating the equipment body, reference points, establishing coupling constraints, and adding fixed and load constraints. The generated files are used for the initial setup of the ABAQUS model.

[0138] Then, the INP file generation module 3 reads the EBSD experimental data through MATLAB, automatically obtains the scanning information of the real microstructure (such as the length, width, and step size of the scanning area), and generates a microstructure model with the same size within a specific area of ​​the macromodel. This module first performs a coarse mesh on the macromodel, then performs a finer mesh based on the step size of the microstructure, and generates the INP file used by ABAQUS, laying the foundation for the multiscale model.

[0139] On this basis, Dataset Creation Module 4 imports the EBSD data and uses the METX toolbox in MATLAB to generate the corresponding dataset in the MATLAB workspace. By processing the EBSD data, the system creates a collection of microscopic grains, ensuring accurate modeling of the material's microstructure. Furthermore, this module supports operations such as data denoising and grain segmentation, further improving data quality and modeling accuracy.

[0140] Next, the Grain Establishment Module 5 creates a grain set from the EBSD data through methods such as grain segmentation, confidence index filtering, and small grain filtering, removing extremely small grains. This module's core function is noise reduction, improving data quality and modeling accuracy. By extracting and optimizing the grain set, this module ensures that the grain data included in the micromodel truly reflects the material's microstructural properties.

[0141] Finally, the pole figure drawing module 10 verifies the resulting multi-scale crystal plasticity finite element model. MATLAB calls ABAQUS to simulate and draw the grain pole figures. The system then compares the original EBSD data with the noise-reduced data. The accuracy and consistency of the system are verified by repeatedly drawing and comparing the pole figures generated by the model. This module also verifies the mechanical properties of the model by extracting stress-strain curves and comparing them with experimental results to ensure the accuracy and reliability of the final modeling results.

[0142] In summary, the present invention automatically completes the entire process from data acquisition, model generation to simulation analysis through the efficient collaboration of various modules, greatly improving the efficiency and accuracy of multi-scale modeling based on real microstructures.

[0143] It should be noted that the embodiments of the present invention can be implemented by hardware, software, or a combination of software and hardware. The hardware portion can be implemented using dedicated logic; the software portion can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated design hardware. Those skilled in the art will appreciate that the above-mentioned devices and methods can be implemented using computer-executable instructions and / or contained in processor control code, for example, such as a carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The devices and modules of the present invention can be implemented by hardware circuits such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., can also be implemented by software executed by various types of processors, or can be implemented by a combination of the above-mentioned hardware circuits and software, such as firmware.

[0144] The above description is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions and improvements made by any technician familiar with this technical field within the technical scope disclosed by the present invention and within the spirit and principles of the present invention should be covered by the scope of protection of the present invention.

Claims

1. A fully automatic modeling method for macro-micro multi-scale models based on real microstructures, characterized by: The method includes: S1: Call the basic controls uifigure and uilabel through MATLAB, use SelectionChangedFcn and SizeChangedFcn to realize the responsive interaction of the multi-scale modeling system, and build the user interface of the macro-micro multi-scale modeling system; S2: Select the corresponding macro model and its corresponding boundary conditions in the interface of S1, and use MATLAB to write the py file to create the ABAQUS initial model entity; at the same time, complete the setting of the corresponding boundary conditions, including creating the equipment body, establishing the analysis step, creating the reference point, establishing the reference point coupling constraint, adding the fixed constraint, and adding the load constraint; S3: Use MATLAB to read the real microstructure data obtained by EBSD, automatically obtain the length, width and scanning step of the EBSD experimental scanning area, and divide the area of ​​the initial macro model where the micro model needs to be added into a surface domain with the same length and width as the EBSD scanning area, and record the initial vertex coordinates of the surface domain; then, perform coarse meshing on the initial macro model, mesh the micro model area according to the size of the EBSD experimental scanning step, and generate an INP file; S4: Use MATLAB to call the METX toolbox to import EBSD data and create an EBSD dataset in the MATLAB workspace; S5: Use grain segmentation, confidence index filtering, and small grain filtering methods to create grain sets for EBSD data, delete extremely small grains, reduce EBSD data noise, and improve EBSD data quality; S6: extract the denoised EBSD data, draw the denoised grain distribution map, and compare it with the grain distribution map of the EBSD data; S7: Read each node data and unit data of the initial model through MATLAB; generate a data file of the node and unit set of each grain in ABAQUS based on the position of each grain and the pixel value occupied by each grain; then align the generated data file with the initial vertex coordinates of the surface domain recorded in S3 and write it into the INP file of the initial model to generate an INP file of the multi-scale crystal plasticity finite element model containing both the microscopic grain structure model and the macroscopic model; S8: Convert the grain Euler angle into radian value, and use MATLAB to output the PYTHON file that assigns the material properties of each grain and the material properties of the macro model; S9: Call ABAQUS through MATLAB to execute the PYTHON file of material property assignment, assign material properties to each grain and macro model at the same time, and generate the final macro-micro multi-scale crystal plasticity finite element model INP file; S10: Call ABAQUS through MATLAB to run the final generated INP file, call the subroutine of the crystal plasticity finite element simulation, output the orientation information of each grain in the final generated crystal plasticity finite element model, and call METX to draw the pole figure of each grain in the final output; at the same time, call the METX toolbox through MATLAB to draw the pole figures of the original EBSD data and the pole figures of the grain set after noise reduction, and compare the consistency of the pole figures finally output by the model with the initial pole figures and the pole figures after noise reduction; S11: Write a Python file to extract the stress-strain curve of the crystal plasticity finite element simulation through MATLAB, call ABAQUS through MATLAB to execute the Python file, obtain the stress-strain curve of the macro-micro multi-scale crystal plasticity finite element simulation and compare it with the experimental results to verify the accuracy of the model.

2. The fully automatic modeling method of the macro-micro multi-scale model based on real microstructure according to claim 1 is characterized in that: The initial model entities in S2 include tensile models, crack extension models, and impact toughness models commonly used in experimental tests; the data set in S4 includes grain type, grain ID, grain orientation, phase sequence, and grain rotation.

3. The fully automatic modeling method of the macro-micro multi-scale model based on real microstructure according to claim 1, characterized in that: The formulas for grain segmentation, confidence index filtering, and small grain filtering in S5 are: Wherein, the grain rotation is represented by R1 and R2, Δθ is the orientation angle difference between the two points, and if Δθ>angle, the two points are considered to belong to different grains; For each measurement point i, its confidence index is CI i ; Where g is the size of each grain and size(g) is the size of each grain.

4. The fully automatic modeling method of the macro-micro multi-scale model based on real microstructure according to claim 1, characterized in that: The EBSD data in S6 includes the grain number, equivalent diameter, grain orientation Euler angle, grain pixel coordinates, and grain set of each grain.

5. The fully automatic modeling method of the macro-micro multi-scale model based on real microstructure according to claim 1, characterized in that: The macro model material property assignment code in S8 includes four parts: automatically reading the radian value of the grain Euler angle, converting the radian value into the Miller index, assigning the grain size effect, and converting the macro model engineering stress and strain into real stress and strain.

6. The fully automatic modeling method of the macro-micro multi-scale model based on real microstructure according to claim 5, characterized in that: The formulas for converting the radian value into the Miller index, grain size effect, and engineering stress and strain into true stress and strain are: t j =τ0+Kd -0.5 ,j=1,2,3... s t =s e (1+e e ) e t =ln(1+ε e ) Where θ is the nutation angle, ψ is the precession angle, φ is the rotation angle; u, v, w are the crystal direction indices; h, k, l are the crystal plane indices; j is the grain number, τ0 is the initial yield strength of the grain, K is the influence coefficient of the reaction grain boundary on the deformation, d is the equivalent diameter of the grain, σ e is the engineering stress, σ t is the true stress, ε e is the engineering strain, ε t For true resilience.

7. A fully automatic modeling system for a macro-micro multi-scale model based on a real microstructure based on the fully automatic modeling method for a macro-micro multi-scale model based on a real microstructure according to any one of claims 1 to 6, characterized in that: The system specifically includes: The interface construction module calls the basic controls uifigure and uilabe through MATLAB, uses SelectionChangedFcn and SizeChangedFcn to achieve responsive interaction of the multi-scale modeling system, and builds the user interface of the macro-micro multi-scale modeling system; The boundary condition setting module is connected to the interface construction module. In the user interface, the corresponding macro model and its corresponding boundary conditions are selected, and MATLAB is used to write the py file to create the ABAQUS initial model entity. At the same time, the corresponding boundary condition settings are completed, including creating the equipment body, establishing the analysis step, creating the reference point, establishing the reference point coupling constraint, adding the fixed constraint, and adding the load constraint. The INP file generation module is connected to the boundary condition setting module. It reads the real microstructure data obtained by EBSD through MATLAB, automatically obtains the length, width and scanning step of the EBSD experimental scanning area, and divides the area of ​​the initial macro model where the micro model needs to be added into a surface domain with the same length and width as the EBSD scanning area, and records the initial vertex coordinates of the surface domain; then, the initial macro model is coarsely meshed, and the micro model area is meshed according to the size of the EBSD experimental scanning step, and the INP file is generated; The dataset creation module is connected to the INP file generation module, uses MATLAB to call the METX toolbox to import EBSD data, and creates an EBSD dataset in the MATLAB workspace; The grain establishment module is connected to the data set creation module. It uses grain segmentation, confidence index filtering, and small grain filtering methods to establish grain sets for EBSD data, delete extremely small grains, and reduce noise in EBSD data to improve the quality of EBSD data. The distribution map drawing module is connected to the grain establishment module to extract the denoised EBSD data, draw the denoised grain distribution map, and compare it with the grain distribution map of the EBSD data; The INP file writing module is connected to the distribution map drawing module and the INP file generation module, and reads each node data and unit data of the initial model through MATLAB; based on the X, Y coordinates of the position of each grain and the pixel value occupied by each grain, a data file of the node and unit set of each grain in ABAQUS is generated; then, the generated data file is aligned with the initial vertex coordinates of the surface domain recorded by the INP file generation module and written into the INP file of the initial model, thereby generating an INP file of a multi-scale crystal plasticity finite element model containing both the microscopic grain structure model and the macroscopic model; The property assignment module converts the grain Euler angle into radian value, and uses MATLAB to output the Python file that assigns the material properties of each grain and the material properties of the macro model; The final INP file generation module is connected to the INP file writing module and the property assignment module. It calls ABAQUS through MATLAB to execute the Python file for material property assignment, assigns material properties to each grain and macro model at the same time, and generates the final macro-micro multi-scale crystal plasticity finite element model INP file; The pole figure drawing module is connected to the final INP file generation module. MATLAB calls ABAQUS to run the final INP file and calls the crystal plasticity finite element simulation subroutine to output the orientation information of each grain in the final crystal plasticity finite element model. It calls METX to draw the pole figures of each grain in the final output. At the same time, MATLAB calls the METX toolbox to draw the pole figures of the original EBSD data and the pole figures of the grain set after noise reduction. The consistency of the pole figures finally output by the model is compared with the initial pole figures and the pole figures after noise reduction. The verification module is connected to the pole figure drawing module. It writes a Python file to extract the stress-strain curve of the crystal plasticity finite element simulation through MATLAB, calls ABAQUS through MATLAB to execute the Python file, obtains the stress-strain curve of the macro-micro multi-scale crystal plasticity finite element simulation and compares it with the experimental results to verify the accuracy of the model.

8. A computer device, characterized in that: The computer device includes a memory and a processor, the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the fully automatic modeling method of the macro-micro multi-scale model based on real microstructures as described in any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the processor executes the steps of the fully automatic modeling method of the macro-micro multi-scale model based on real microstructures as described in any one of claims 1 to 6.

10. An information data processing terminal, characterized in that: The information data processing terminal is used to implement the fully automatic modeling system of the macro-micro multi-scale model based on real microstructure as described in claim 7.

Images