Multi-scale simulation method for thermal compression deformation behavior of TC4 titanium alloy

Abstract

The invention discloses a multi-scale simulation method for a thermal compression deformation behavior of TC4 titanium alloy, and belongs to the field of metal material processing simulation. Comprising the following steps: 1, carrying out thermal compression experiments at different temperatures to obtain a deformed sample; 2, analyzing the grain orientation, orientation difference angle and texture evolution of the sample by using an electron back scattering diffraction technology; 3, constructing a crystal plasticity finite element model, and simulating a polycrystalline deformation process; 4, constructing a molecular dynamics model, and simulating phase change and dislocation evolution in thermal compression and cooling processes; and 5, based on a multi-scale simulation result, hot working parameters are optimized, dynamic recrystallization and a beta-to-alpha phase change path are regulated and controlled, and grain refinement and performance homogenization are achieved. The invention discloses a correlation mechanism of the microstructure and the mechanical property in TC4 titanium alloy thermal compression deformation, and provides a theoretical basis for optimization of a thermal processing technology of a high-performance titanium alloy component in the aerospace field.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of metal material processing simulation, in particular to a multi-scale simulation method for studying the hot compression deformation behavior of TC4 titanium alloy by combining hot compression experiment, electron backscatter diffraction technology, crystal plasticity finite element simulation and molecular dynamics simulation. BACKGROUND

[0002] TC4 titanium alloy, as an important structural material, has excellent mechanical properties and high-temperature resistance, and is widely used in the fields of aerospace, medical treatment and the like. However, its room-temperature formability is poor, and the microstructure and performance need to be regulated through hot working. However, the existing technology has the following technical problems. Due to the significant anisotropy, the cooperative deformation mechanism of the hexagonal close-packed (HCP) alpha phase and the body-centered cubic (BCC) beta phase is complex, resulting in uneven grain size after deformation. In addition, due to the difficulty in controlling the phase transformation path, the variant selection in the beta to alpha phase transformation process is affected by the coupling of multiple factors such as temperature and strain rate, and it is difficult to accurately regulate the orientation and distribution of acicular alpha phase through traditional process. During the hot compression deformation process, the deformation behavior and phase transformation mechanism of the material are relatively complex, involving the interaction of multiple factors such as grain orientation, twinning activation and dislocation movement. The texture evolution mechanism is unknown, and the existing model cannot effectively capture the interaction of dynamic recrystallization and twinning activation at high temperature, resulting in high cost and low efficiency of the process optimization relying on trial and error method. At present, the research on the hot compression deformation behavior of TC4 titanium alloy mainly focuses on single-scale analysis. Experimental research is difficult to capture the micro details of texture evolution and phase transformation mechanism, and single simulation method is also difficult to fully reveal the complex deformation mechanism. At the experimental level, the microstructure evolution is analyzed through metallographic observation and hardness test, but the atomic scale phase transformation mechanism cannot be revealed. At the simulation level, the finite element model ignores the grain orientation difference, and the molecular dynamics simulation lacks connection with the macro process. Therefore, a multi-scale simulation method is needed to systematically study the hot compression deformation behavior of TC4 titanium alloy from macro experiment to micro simulation and then to atomic scale analysis. SUMMARY

[0003] The purpose of the present application is to provide a multi-scale simulation method for the hot compression deformation behavior of TC4 titanium alloy, which combines hot compression experiment, electron backscatter diffraction technology, crystal plasticity finite element simulation and molecular dynamics simulation, reveals the synergistic effect of slip, phase transformation and texture evolution through multi-scale simulation, solves the problems of grain coarsening and uneven performance in traditional process, and systematically reveals the high-temperature deformation behavior and phase transformation mechanism of TC4 titanium alloy, thereby providing a theoretical basis for the hot working process optimization of high-performance TC4 titanium alloy components in the field of aerospace.

[0004] In order to achieve the above purpose, the technical scheme adopted by the present application is as follows:

[0005] A multi-scale simulation method for the hot compression deformation behavior of TC4 titanium alloy includes the following steps:

[0006] Step 1: Conduct hot compression tests: Using a Gleeble-3800 instrument, the TC4 titanium alloy sample was subjected to hot compression tests at 800℃, 900℃, and 1000℃, with a strain rate of 0.1 s⁻¹. -1 The deformation was 70%. After the hot compression test, the sample was water-quenched and cooled to room temperature to obtain the deformed sample.

[0007] Step 2, Electron Backscattering Diffraction (EBSD) Analysis: Electron backscattering diffraction analysis was performed on the deformed sample obtained in Step 1 to characterize grain orientation, orientation difference angle, and texture evolution.

[0008] Step 3: Crystal plasticity finite element simulation: Based on the analysis results of Step 2, a crystal plasticity constitutive model is established, a polycrystalline model is constructed using the Voronoi method, and polycrystalline deformation simulation is performed using the ABAQUS user material subroutine to analyze stress-strain distribution, plastic slip, and grain deformation deflection behavior.

[0009] Step 4: Molecular dynamics simulation: Construct an α-Ti model of compression along the

[0001] , [11-20] and [10-10] crystal orientations, and use LAMMPS software to simulate the thermal compression and cooling processes, tracking atomic displacement, twin formation and dislocation evolution trajectories;

[0010] Step 5: Based on the multi-scale simulation results, optimize the hot working parameters, control the dynamic recrystallization and β→α phase transformation path, and achieve grain refinement and performance homogenization.

[0011] Furthermore, in step one, before the hot compression test, the sample is processed into a cylinder with a diameter of φ6mm×9mm. The heating rate for the hot compression test is 10℃ / s, the holding time is 3min, and the sample is cooled by water quenching after deformation.

[0012] Furthermore, in step two, before electron backscatter diffraction analysis, the sample is electropolished using perchloric acid ethanol solution as the electrolyte. Spectra are acquired at an accelerating voltage of 20 kV, and the distribution of α and β phase variants, as well as the distribution and intensity of pole figure and inverse pole figure textures, are quantitatively analyzed using the EDAX-TSL system.

[0013] Furthermore, in step three, the crystal plastic constitutive model considers the slip system hardening criterion and the geometrically necessary dislocation evolution equation. The polycrystalline model is constructed using Neper software to generate the Voronoi polycrystalline model. In the crystal plastic finite element model, the α-phase slip system includes the basal plane {0001}<11-20>, the cylindrical plane {10-10}<11-20>, the conical plane {10-11}<11-20>, and {10-11}<11-23>, and the β-phase slip system is {110}. <111> The initial critical shear stress (CRSS) decreases with increasing temperature.

[0014] Furthermore, in step four, the DeePMD interatomic potential function model of titanium is used in the molecular dynamics simulation. The simulation process includes equilibration in the NPT ensemble for 500 ps and a compressive strain rate of 10. -9 / s, cooling rate 300℃ / ps, heating, compression and quenching process.

[0015] Furthermore, the TC4 titanium alloy of the present invention comprises at least 6.12% Al, 4.2% V, and 0.24% Fe by mass, with the balance being Ti.

[0016] Furthermore, the processing of EBSD data in step two of this invention includes:

[0017] Use OIM software to export the RGB values ​​and pixel coordinates corresponding to the Euler angles of the grains, and generate the grain_euler_angle_pixel_data.csv file;

[0018] The CSV file is parsed using a Python script, and the RGB values ​​are converted inversely to Rodrigues vectors to construct a Neper-recognizable grain geometry model tesr file.

[0019] The Neper software is used to generate a polycrystalline mesh model, and the output is an ABAQUS-compatible INP format file.

[0020] The present invention also provides a multi-scale simulation system for the control method described above, the system comprising:

[0021] Experimental modules: Gleeble-3800 hot compression testing machine, EBSD microanalyzer;

[0022] Simulation modules: Neper polycrystalline modeling software, ABAQUS finite element platform, LAMMPS molecular dynamics simulation program;

[0023] Data interaction module: Enables parameter mapping between experimental data and simulation models, and supports automated analysis using Python scripts.

[0024] Compared with the prior art, the present invention has the following advantages:

[0025] By combining hot compression experiments with electron backscatter diffraction (ESD), microstructure and texture data of TC4 titanium alloy after deformation were obtained, providing an experimental basis for simulations. Crystal plasticity finite element simulations revealed the stress-strain distribution and grain deformation behavior during deformation, while molecular dynamics simulations elucidated the phase transformation and dislocation evolution mechanisms at the atomic scale. Multi-scale simulation methods comprehensively and systematically revealed the correlation mechanism between the microstructure and mechanical properties of TC4 titanium alloy during hot compression deformation, providing a reliable theoretical basis for optimizing the hot working process of high-performance titanium alloy components in the aerospace field. Process optimization strategies include a low-temperature strengthening mechanism (800℃) that preserves the subgrain structure and coordinates deformation through {10⁻¹²} compression twins to enhance work hardening capacity, suitable for components requiring high yield strength. A medium-temperature homogenization mechanism (900℃) controls the dynamic recrystallization ratio to >50%, promoting the precipitation of equiaxed α phase, combined with 10% cold deformation to refine the grain size. A high-temperature phase transition control mechanism (1000℃) utilizes deformation in the complete β-phase region to induce acicular α' phases, aligning them at 45° along the compression direction, thus enhancing fracture toughness. The gain effect is significant grain refinement; through 900℃ hot compression followed by cold deformation, the average grain size is reduced, with a grain refinement efficiency of 53%. The process cycle is shortened; multi-scale simulation replaces the traditional trial-and-error method, reducing the number of experiments by over 60% and production costs by 30%. A theoretical breakthrough reveals that the variant selection in the BCC→HCP phase transition follows the principle of "shortest..." <100> The principle of "prioritizing the shift in the β direction" provides new targets for texture design. Attached Figure Description

[0026] Figure 1 The diagram shows the changes in the sample before and after hot compression. (a) is the sample before compression, and (b) is the sample after compression.

[0027] Figure 2 The crystal orientation diagram of the α phase in the 1000℃ compressed sample is shown. The grain boundaries are colored according to three different orientation difference angle ranges. In the CDxTD plane coordinate system of the spectrum, the vertical axis CD represents the compression direction parallel to the

[100] direction, and the horizontal axis TD represents the transverse direction parallel to the

[010] direction. For each determined crystal orientation, the color intensity corresponds to the deviation angle, which ranges from a minimum of 0° to a maximum of 10°.

[0028] Figure 3 (a) Comparison of inverse pole figures of the original specimen and specimens compressed at (b) 800℃, (c) 900℃, and (d) 1000℃. CD indicates the compression direction;

[0029] Figure 4 The orientation difference angular distribution of the compressed specimens at 800℃ (black), 900℃ (blue), and 1000℃ (red) and the original specimen (green dashed line) is compared and shown.

[0030] Figure 5 (a) TC4 polycrystalline model of crystal plasticity finite element simulation, and equivalent stress distribution after 10% compression along the Y-axis at different temperatures: (b) 300℃, (c) 800℃, (d) 1000℃;

[0031] Figure 6 Comparison of stress-strain curves along the

[0001] , [10-10] and [11-20] crystal orientations for the α-phase single crystal model;

[0032] Figure 7 The average absolute value of stress along the X||

[100] , Y||

[010] and Z||

[001] axes during the cooling stage of the compression model in the

[0001] direction. The stress is plotted as a function of temperature. Specifically, the stress values ​​of BCC atoms (solid lines) and HCP atoms (dashed lines) are respectively represented by their atomic number N = n. BCC and N=n HCP The average is calculated, and the corresponding value is displayed on the left vertical axis. The right vertical axis displays n. BCC and n HCP The evolution curve over time;

[0033] Figure 8 For

[0001] directional compression system (a) and Pole diagram (b) of HCP grains in a directional compression system. {0001} and Polar figures are indicated by solid circles and hollow squares, respectively. The red dashed circle corresponds to... Figure 9 The position of the texture peak in the {0001} pole figure in the experimental results of electron backscatter diffraction technique shown;

[0034] Figure 9 The orientation distribution function and pole figure of the {0001} and {11-20} crystal planes of the 1000℃ compressed sample;

[0035] Figure 10 The distribution states of the α and β phases of the original sample in the EBSD experiment;

[0036] Figure 11 The left image shows the grain orientation distribution map derived from EBSD data, with RGB values ​​assigned by three Euler angles. This serves as a basis for... Figure 10 The input file for the code to construct a two-dimensional cross-section of the α-phase grain model from EBSD data; the right figure shows the code based on... Figure 10 The code for constructing a two-dimensional cross-section of the α-phase grain model from EBSD data generates the shape and numbering distribution of the two-dimensional cross-section of the α-phase grain model. Detailed Implementation

[0037] The present invention will be further described below with reference to specific embodiments. These embodiments are only used to more clearly illustrate the technical solutions of the present invention and should not be construed as limiting the scope of protection of the present invention. Anything not described in detail in the present invention patent application is considered to be common knowledge in the art.

[0038] This invention provides a multi-scale simulation method for the hot compression deformation behavior of TC4 titanium alloy, the steps of which are as follows:

[0039] Step 1: Conduct a hot compression test: Process TC4 titanium alloy bars into cylindrical specimens with a diameter of φ6mm × 9mm. The specimens must contain at least the following chemical composition (mass fraction): 6.12% Al, 4.2% V, 0.24% Fe, with the balance being Ti. A Gleeble-3800 instrument is used for the hot compression test. The test temperatures are set at 800℃ (dominantly dynamic recovery), 900℃ (early stage of dynamic recrystallization), and 1000℃ (complete β-phase deformation), with a holding time of 3 minutes and a strain rate of 0.1s. -1 The deformation was 70%. After the hot compression test, the sample was cooled to room temperature by water quenching to retain the high-temperature phase structure and obtain the deformed sample.

[0040] Step 2: Electron Backscattering Diffraction (EBSD) Analysis: The deformed sample was electropolished using a perchloric acid-ethanol solution as the electrolyte to obtain a smooth surface. Electron backscattering diffraction patterns were acquired using the EDAX-TSL system at an accelerating voltage of 20 kV. The pole figures, inverse pole figures, grain orientation, orientation difference angle, and texture evolution characteristics were analyzed using TSL-OIMAnalysis software, with a focus on the large deformation region at the center of the cross-section. (Appendix) Figure 1 A schematic diagram showing the changes in the sample before and after hot compression is provided, where (a) is the sample before compression and (b) is the sample after compression. (See attached diagram.) Figure 2 The crystal orientation diagram of the α phase in a 1000℃ compressed sample is shown. Grain boundaries are colored according to three different orientation difference angle ranges. In the CDxTD plane coordinate system of the diagram, the vertical axis CD represents the compression direction parallel to the

[100] direction, and the horizontal axis TD represents the transverse direction parallel to the

[010] direction. For each defined crystal orientation, the color intensity corresponds to the deviation angle, ranging from a minimum of 0° to a maximum of 10°. (Appendix) Figure 3 The diagram shows a comparison of inverse pole figures of the original sample (Figure a) and samples compressed at 800℃ (Figure b), 900℃ (Figure c), and 1000℃ (Figure d). CD in the diagram indicates the compression direction. (See attached diagram.) Figure 4 The orientation difference angular distribution of the compressed specimens at 800℃ (black), 900℃ (blue), and 1000℃ (red) and the original specimen (green dashed line) is compared and shown.

[0041] In some embodiments of the present invention, the processing of EBSD data in this step includes: using OIM software to export the RGB values ​​and pixel coordinates corresponding to the Euler angles of the grains, generating a grain_euler_angle_pixel_data.csv file; parsing the CSV file using a Python script, converting the RGB values ​​inversely into Rodrigues vectors, and constructing a Neper-recognizable grain geometry model (.tesr file); calling Neper software to generate a polycrystalline mesh model and outputting an ABAQUS-compatible INP format file.

[0042] Step 3: Crystal Plasticity Finite Element Simulation: Based on data such as grain size and orientation obtained by electron backscatter diffraction, a crystal plasticity constitutive model is established. Considering the slip system hardening criterion and the geometrically necessary dislocation evolution equation, the Voronoi method is used to construct representative polycrystalline volume elements. A three-dimensional polycrystalline model is generated using Neper software and a hexahedral mesh is generated. Each grain is divided into 10-30 elements to ensure calculation accuracy at grain boundaries. An ABAQUS-readable INP file is output. The crystal plasticity constitutive model is embedded in the ABAQUS user material subroutine developed in Fortran. The elastic moduli of the α phase are C11 = 162.4 GPa and C44 = 49.7 GPa; the elastic moduli of the β phase are C11 = 135 GPa and C44 = 54.9 GPa. The reference shear rate is γ0 = 0.001 / s. Polycrystalline deformation simulation is performed to analyze stress-strain distribution, plastic slip, and grain deformation deflection behavior. A 10% displacement load is applied along the Y-axis in the compression direction, with a strain rate of 0.001 / s. Constraint settings: Surfaces X=0, Y=0, and Z=0 are fixed; surfaces X=L and Z=L are free. (See attached diagram) Figure 5 The TC4 polycrystalline model (a) of the finite element simulation of crystal plasticity is shown, as well as the equivalent stress distribution after 10% compression along the Y-axis at different temperatures: (b) 300℃, (c) 800℃, (d) 1000℃.

[0043] In some embodiments of the present invention, in the crystal plastic finite element model of this step, the α-phase slip system includes the basal plane {0001}<11-20>, the cylindrical plane {10-10}<11-20>, the conical plane {10-11}<11-20>, and {10-11}<11-23>, and the β-phase slip system is {110} <111> The initial critical shear stress (CRSS) decreases with increasing temperature.

[0044] Step 4: Molecular Dynamics Simulation: Based on LAMMPS software simulation, a 10×10×30 supercell model of α-Ti compressed along the

[0001] , [11-20], and [10-10] crystal orientations was constructed, using the DeePMD interatomic potential function model developed based on deep learning. The initial model was first equilibrated for 500 picoseconds at 300 K and 1 bar pressure in an isothermal-isobaric ensemble using a Nose-Hoover heat bath. After equilibration, the model was heated to 1273 K, and then a strain rate of 10⁻¹⁰ along the Z-axis was applied at the same temperature. -9 Uniaxial compression was applied at 1 / s until the compressive strain reached 0.2, with periodic boundary conditions applied only along the X and Y axes, and the pressure controlled at 1 bar using the NPT equations of motion. Finally, the model was quenched to 300 K at 1 nanosecond step sizes in the NPT ensemble. Using Ovito software, the crystal structure was identified using a polyhedral template matching method, tracing atomic shifts, twin formation, and dislocation evolution trajectories. During the cooling process, quenching to 300 K was performed at 2 fs / step sizes, tracing the BCC→HCP phase transformation path, and identifying the formation mechanism of the 60° ({10-11} twin) and 90° ({10-12} twin) orientation difference angles. (Appendix) Figure 6 A comparison of stress-strain curves along the

[0001] , [10-10], and [11-20] crystal orientations of the α-phase single-crystal model is shown. Figure 7 The figure shows the average absolute values ​​of stress along the X||

[100] , Y||

[010] and Z||

[001] axes during the cooling phase of the compression model in the

[0001] direction. The stress is plotted as a function of temperature. Specifically, the stress values ​​of BCC atoms (solid lines) and HCP atoms (dashed lines) are respectively represented by their atomic number N = n. BCC and N=n HCP The average value is calculated, and the corresponding value is displayed on the left vertical axis. Figure 7 The right vertical axis shows n BCC and n HCP The evolution curve over time. (Attached) Figure 8 For

[0001] directional compression system (a in the figure) and Pole diagram of HCP grains in the directional compression system (b in the figure). {0001} and Polar figures are indicated by solid circles and hollow squares, respectively. The red dashed ring corresponds to the attached... Figure 9 The position of the texture peak in the {0001} pole figure in the experimental results of electron backscatter diffraction technique shown. Figure 9 The orientation distribution functions and pole figures of the {0001} and {11-20} crystal planes of the 1000℃ compressed sample are shown.

[0045] Step 5: Based on the multi-scale simulation results, optimize the hot working parameters, control the dynamic recrystallization and β→α phase transformation path, and achieve grain refinement and performance homogenization.

[0046] Example 1

[0047] The sample in this embodiment is a Ti-6Al-4V rod with an average grain size of 2.89 μm. The chemical composition (mass fraction) is: 6.12% Al, 4.2% V, 0.24% Fe, 0.06% C, 0.0023% H, 0.16% O, 0.002% N, and the balance is Ti.

[0048] Step 1: The sample is machined into a cylinder with a diameter of φ6mm × 9mm. A hot compression test is performed at 800℃ using a Gleeble-3800 instrument, with a holding time of 3 minutes and a strain rate of 0.1s. -1 The deformation was 70%, and the sample was water-quenched to room temperature.

[0049] Step 2: Electrolytic polishing was performed on the deformed sample using perchloric acid ethanol solution as the electrolyte. Electron backscatter diffraction patterns were collected, and analysis revealed that dynamic recovery was the main feature at low temperatures, with subcrystalline structures forming near the grain boundaries.

[0050] Step 3: Construct a finite element model of crystal plasticity. Simulation revealed that at 800℃, a few large grains bear most of the stress, and stress concentration occurs near the boundary.

[0051] Step 4: Construct an α-Ti model compressed along the

[0001] crystal orientation. Simulation shows that HCP grains with an orientation difference angle of about 60° are formed during the cooling process.

[0052] Example 2

[0053] The difference between this embodiment and Embodiment 1 is that the thermal compression test temperature is 900℃.

[0054] Step 1: Conduct a hot compression experiment at 900℃, with other conditions the same as in Example 1.

[0055] Step 2: Electron backscatter diffraction analysis showed that dynamic recrystallization and the activation of the β→α' phase transition at high temperature reduced the proportion of small-angle grain boundaries.

[0056] Step 3: Finite element simulation of crystal plasticity shows that at 900℃, more grains participate in load bearing, and the stress distribution is more uniform.

[0057] Step 4: Molecular dynamics simulations show the formation of twins with an orientation difference angle of approximately 90°.

[0058] Example 3

[0059] The difference between this embodiment and Embodiment 1 is that the thermal compression test temperature is 1000℃.

[0060] Step 1: Conduct a hot compression experiment at 1000℃, with other conditions the same as in Example 1.

[0061] Step 2: Electron backscattering diffraction analysis showed that the needle-like α' phases were arranged at an angle of 45° to the compression direction, indicating a significant phase transition enhancement effect.

[0062] Step 3: Finite element simulation of crystal plasticity shows that the degree of stress distribution non-uniformity decreases at 1000℃, and the crystal slip deformation is sufficient.

[0063] Step four: Molecular dynamics simulations verified the variant selection under the Burgers mechanism, resulting in an orientation difference angular distribution consistent with experimental results.

[0064] Appendix Figure 2 The crystal orientation diagram of the α phase in the 1000℃ compressed sample of this embodiment is shown. The grain boundaries are colored according to three different orientation difference angle ranges. In the CDxTD plane coordinate system of the spectrum, the vertical axis CD represents the compression direction parallel to the

[100] direction, and the horizontal axis TD represents the transverse direction parallel to the

[010] direction. For each determined crystal orientation, the color intensity corresponds to the deviation angle, which ranges from a minimum of 0° to a maximum of 10°.

[0065] Example 4

[0066] This example uses OIMAnalysis software to extract relevant data of α and β phase grains from EBSD data. A Python program then establishes a one-to-one correspondence between the grain orientation, Euler angles, RGB color values, and XY coordinates in the data to form an Excel table. Based on this table, the geometric and mesh models of the grains are generated from Neper.

[0067] Step 1: EBSD Data Processing and IPF Map Generation

[0068] The EBSD data was processed using OIM software to generate a distribution map of the Euler angle coloring for the α-phase grain orientation. In OIM software, EulerAngle output was selected, and the software generated corresponding RGB values ​​based on Bunge EulerAngles. The Euler Angle distribution map (`alpha.bmp`) was saved via `Map->EulerAngles RGB`, and a CSV file containing the coordinates and RGB values ​​of each pixel (`grain_euler_angle_pixel_data.csv`) was exported.

[0069] Step 2: Data Preparation

[0070] Ensure that the files `alpha.bmp` and `grain_euler_angle_pixel_data.csv` are in the same directory.

[0071] The `grain_euler_angle_pixel_data.csv` file contains the following:

[0072] Column 1: x-coordinate (pixel position).

[0073] Column 2: y-coordinate (pixel position).

[0074] Column 3: RGB values ​​(converted from EulerAngle).

[0075] Step 3: Run the Python script

[0076] Execute the uploaded Python script (`24patent_neper.py`), and the script will perform the following operations:

[0077] 1. Read the files `alpha.bmp` and `grain_euler_angle_pixel_data.csv`.

[0078] 2. Reconstruct the grain distribution map based on the RGB values ​​in the CSV file and assign a unique ID to each grain.

[0079] 3. Calculate the centroid of each grain and label the grain ID on the graph.

[0080] 4. Convert EulerAngles to Rodrigues vectors to generate an input file (`0output.tesr`) suitable for Neper software.

[0081] 5. Use Neper software to process the generated `0output.tesr` file and output the visualization result of the two-dimensional model (`0img.png`).

[0082] Step 4: Output File

[0083] `saved_map.png`: The reconstructed grain distribution map, labeled with grain IDs.

[0084] `0output.tesr`: A two-dimensional model file that can be recognized by Neper software.

[0085] `0img.png`: Visualization result of a 2D model generated by Neper software.

[0086] Step 5: Verification and Adjustment

[0087] Check that the generated images and files meet expectations. If you need to adjust the grain distribution or model parameters, modify the relevant parameters in the Python script (such as `ymax` or `nuptolist`) and then rerun the script.

[0088] Step Six: Mesh Generation

[0089] The Neper command (neper-M 0output.tesr-cl0.02-format inp) is invoked to output the mesh model; the generated INP file is then imported into ABAQUS and coupled with the crystal plastic finite element model. Based on the generated and verified .tesr format file, the code 'neper-M0output.tesr-cl0.02-format inp' is input to divide the grain geometry model into a mesh model. The -cl option controls the mesh size and outputs an inp format file that can be imported into Abaqus.

[0090] Key technical points: EulerAngles to RGB conversion: OIM software automatically maps EulerAngles to RGB values ​​to ensure consistent grain color.

[0091] Rodrigues vector generation: The Python script reverses the RGB values ​​to EulerAngles, and then further converts them into Rodrigues vectors for use by the Neper software.

[0092] Automated processing: The script automates the entire process from data reading to model generation, improving efficiency and repeatability.

[0093] Through the steps described above, users can quickly convert EBSD data into two-dimensional model files required by Neper software, facilitating subsequent crystal structure analysis and visualization. The following is based on... Figure 10 The code for constructing a two-dimensional cross-section of the α-phase grain model from EBSD data generates a tesr format file suitable for Neper software. The full text of the code for this file is as follows:

[0094]

[0095]

[0096]

[0097]

[0098]

[0099]

[0100] Appendix Figure 11 The left image shows the grain orientation distribution map derived from EBSD data, with RGB values ​​assigned by three Euler angles. This serves as a basis for... Figure 10 The input file for the code to construct a two-dimensional cross-section of the α-phase grain model from EBSD data; the right figure shows the code based on... Figure 10 The code for constructing a two-dimensional cross-section of the α-phase grain model from EBSD data generates the shape and numbering distribution of the two-dimensional cross-section of the α-phase grain model.

[0101] Based on the above embodiments, the present invention provides a multi-scale simulation system for the above-described control method, the system comprising:

[0102] Experimental modules: Gleeble-3800 hot compression testing machine, EBSD microanalyzer;

[0103] Simulation modules: Neper polycrystalline modeling software, ABAQUS finite element platform, LAMMPS molecular dynamics simulation program;

[0104] Data interaction module: Enables parameter mapping between experimental data and simulation models, and supports automated analysis using Python scripts.

[0105] As demonstrated by the above embodiments, this invention systematically reveals the microscopic mechanism of hot deformation in TC4 titanium alloy through the deep integration of hot compression experiments and multi-scale simulations, and establishes a texture design method and grain refinement technology based on variant selection. This achievement not only breaks through the empirical dependence of traditional processes but also provides a replicable cross-scale research paradigm, opening up new paths for the precise manufacturing of high-performance metallic materials and providing effective guidance for the optimization of hot working processes.

Claims

1. A multiscale simulation method of hot compression deformation behavior of TC4 titanium alloy, characterized in that, Comprising the following steps: Step one, heat compression experiment: using Gleeble-3800 equipment, TC4 titanium alloy sample is heat compressed at 800℃, 900℃, 1000℃, strain rate is 0.1s -1 , deformation is 70%, after heat compression experiment, water quenching cooling to room temperature, obtaining deformed sample; Step two, electron backscatter diffraction (EBSD) analysis: the deformed sample obtained in step one is analyzed by electron backscatter diffraction to characterize the grain orientation, orientation difference angle and texture evolution characteristics; Step three, crystal plasticity finite element simulation: based on the analysis results of step two, a crystal plasticity constitutive model is established, a polycrystal model is constructed using the Voronoi method, and a polycrystal deformation simulation is performed through the ABAQUS user material subroutine to analyze the stress-strain distribution, plastic slip and grain deformation deflection behavior; Step four, molecular dynamics simulation: α-Ti models are constructed along the [0001], [11-20] and [10-10] crystal directions for compression, and the LAMMPS software is used to simulate the hot compression process and the cooling process to track the atomic displacement, twinning formation and dislocation evolution trajectory; Step five, based on the results of multi-scale simulation, the hot working parameters are optimized to control the dynamic recrystallization and β→α phase transformation path, and grain refinement and performance uniformization are realized.

2. The multi-scale simulation method of hot compression deformation behavior of TC4 titanium alloy according to claim 1, characterized in that, In step one, before the hot compression experiment, the sample is processed into a φ6mm×9mm cylinder, the heating rate of the hot compression experiment is 10℃ / s, the holding time is 3min, and the deformed sample is water quenched.

3. The multi-scale modeling method of hot compression deformation behavior of TC4 titanium alloy according to claim 1, characterized in that, In step two, before the electron backscatter diffraction analysis, the sample is electrolytically polished, anhydrous hydrochloric acid ethanol solution is used as the electrolyte, the spectrum is collected at an acceleration voltage of 20kV, and the EDAX-TSL system is used to quantitatively analyze the distribution of α and β phase variants, as well as the pole figure and inverse pole figure texture distribution and intensity.

4. The multi-scale simulation method of hot compression deformation behavior of TC4 titanium alloy according to claim 1, characterized in that, In step three, the crystal plasticity constitutive model considers the slip system hardening criterion and the geometric necessary dislocation evolution equation, the Neper software is used to generate a Voronoi polycrystal model for the construction of the polycrystal model, in the crystal plasticity finite element model, the α phase slip system includes the basal plane {0001}<11-20>, the cylinder plane {10-10}<11-20>, the conical plane {10-11}<11-20> and {10-11}<11-23>, the β phase slip system is {110}<111>, and the initial critical shear stress (CRSS) decreases with increasing temperature.

5. The multi-scale modeling method of hot compression deformation behavior of TC4 titanium alloy according to claim 1, characterized in that, In step four, the DeePMD model of interatomic potential of titanium was used in the molecular dynamics simulation. The simulation process included 500 ps of equilibration in the NPT ensemble, a compression strain rate of 10 -9 / s, a cooling rate of 300 °C / ps, and the processes of heating, compression, and quenching.

6. The multi-scale modeling method of hot compression deformation behavior of TC4 titanium alloy according to claim 1, characterized in that, TC4 titanium alloy at least includes chemical components and mass fractions of 6.12% Al, 4.2% V, 0.24% Fe, and the balance of Ti.

7. The multi-scale modeling method of hot compression deformation behavior of TC4 titanium alloy according to claim 1, characterized in that, The processing of EBSD data in step two includes: Using OIM software to export the RGB values and pixel coordinates corresponding to the grain Euler angle to generate a grain_euler_angle_pixel_data.csv file; Using a Python script to parse the CSV file, convert the RGB values back to Rodrigues vectors, and construct a grain geometry model tesr file recognizable by Neper; Calling the Neper software to generate a polycrystal grid model and output an ABAQUS compatible INP format file.

8. A multiscale simulation system for the regulation method of claims 1-7, characterized by, Comprise: Experimental module: Gleeble-3800 hot compression testing machine, EBSD microanalyzer; Simulation module: Neper polycrystal modeling software, ABAQUS finite element platform, LAMMPS molecular dynamics simulation program; Data interaction module: Realize the parameter mapping between experimental data and simulation model, support Python script automatic analysis.

Images