A full-automatic finite element modeling and performance analysis method for two-phase microstructure images

Abstract

This invention provides a fully automated finite element modeling and performance analysis method for two-phase microstructure images, relating to the field of image processing technology. First, the microstructure photograph of the alloy material is preprocessed to extract the boundary contour of the second phase from the preprocessed image. Then, a two-dimensional planar deformable body component containing a periodically expanding region is automatically constructed in finite element software, generating different face sets. The two-dimensional planar deformable body component is then instantiated to generate an assembly. Material properties are assigned to the face sets, and an elastoplastic constitutive relation is applied. The assembly is then automatically meshed to obtain a mesh model. Periodic boundary conditions and loads are applied to the boundary of the central rectangular region. The finite element analysis job is automatically created, configured, and submitted to complete the mechanical calculations. The mechanical calculation results are extracted and post-processed, and mechanical performance data files and contour plots are output.

Description

Technical Field

[0001] This invention relates to the field of image processing technology, and in particular to a fully automated finite element modeling and performance analysis method for two-phase microstructure images. Background Technology

[0002] The macroscopic mechanical properties of alloy materials are fundamentally determined by the characteristics of their microstructure. Establishing a quantitative structure-property relationship between microstructure and macroscopic properties is a core scientific problem for achieving rational material design and performance optimization. Finite element simulation based on representative volume elements (RVEs) predicts macroscopic behavior by numerically reproducing the mechanical response of materials under load at the microscale. It has become a key bridge connecting the multi-scale properties of materials and has irreplaceable value in materials science research and engineering applications.

[0003] Currently, the technical workflow for constructing and performing finite element analysis based on microstructure images typically includes: first, manually cropping, enhancing, and segmenting the image in image processing software to extract the microstructure contour; then, manually reconstructing the model, components, and assemblies in finite element preprocessing software, generating meshes, assigning material properties (distinguishing between the matrix and the second phase), and setting boundary conditions and loads; then submitting the calculation to the solver; and finally, manually extracting and organizing the data in the post-processing environment. This is the mainstream technical approach connecting the simulation of material microstructure and macroscopic properties.

[0004] Existing technical processes suffer from systemic bottlenecks characterized by heavy reliance on manual labor, low efficiency, and poor consistency. These bottlenecks are mainly reflected in:

[0005] 1. Fragmented and cumbersome workflow: From real image processing to finite element pre- and post-processing, multiple independent software programs are involved, including commonly used ones like Photoshop, FreeCAD, and Abaqus. Each step requires extensive manual operation. Photoshop requires adjusting image size, FreeCAD requires manual modeling based on the real image and conversion to a file format readable by Abaqus, and Abaqus requires modeling the image, adding periodic boundary conditions, and submitting the job sequentially. Pre-processing of a single model often takes 1-2 hours, which cannot meet the needs of high-throughput analysis.

[0006] 2. The results are heavily dependent on subjective experience: From the selection of image binarization threshold, the cleaning and simplification of geometric contours, and the determination of grid size, to the setting of periodic boundary conditions to meet the requirements of representative volume elements (which requires accurate identification and pairing of massive boundary nodes), each step is heavily dependent on the operator's experience and judgment, resulting in poor reproducibility of modeling results and low comparability between different studies.

[0007] 3. Low and fragmented level of automation: Although some studies have attempted to develop dedicated scripts to improve specific steps (such as automatic mesh generation), these efforts are isolated and fragmented, failing to solve the problem of end-to-end automation integration from "raw image" to "final mechanical property analysis report", especially in the areas of handling complex irregular morphologies, automatically applying accurate periodic boundary conditions, and integrated post-processing. Summary of the Invention

[0008] To address the shortcomings of existing technologies, this invention proposes a fully automated finite element modeling and performance analysis method for two-phase microstructure images. This method can automatically complete the entire process from microstructure image preprocessing, finite element model construction, material property allocation, mesh generation, boundary condition setting to finite element analysis and post-processing, and supports batch processing.

[0009] On the one hand, this invention provides a fully automated finite element modeling and performance analysis method for two-phase microstructure images, comprising the following steps:

[0010] Step 1: Pre-processed microstructure photographs of alloy materials;

[0011] The SEM images of the microstructure of the alloy material were cropped using image processing software to remove scale bars and areas that did not need to be processed; contrast enhancement technology was used to preprocess the images to enhance their visual effect.

[0012] Step 2: Extract the boundary contour of the second phase in the preprocessed image and generate boundary data file and image data file;

[0013] Specifically, the process involves: reading the image preprocessed in step 1, converting it to grayscale and scaling it; binarizing the image to distinguish different phases; extracting the boundary contours of the second phase using a contour detection algorithm; performing geometric cleanup on the extracted boundary contours, including removing duplicate points, simplifying polygons, and merging adjacent or overlapping contours; converting the coordinates of the cleaned boundary contours from pixel coordinates to actual physical coordinates; generating internal feature points for each independent second phase region; saving all boundary contour coordinates and feature point data in a text file, i.e., the boundary data file; and saving the actual length and width information of the image and the pixel-to-actual-size conversion coefficients in a text file, i.e., the image data file.

[0014] Step 3: Based on the boundary data file and image data file, automatically construct a two-dimensional planar deformable body component containing a periodically expanding region in the finite element software, generate different face sets, and instantiate the two-dimensional planar deformable body component to generate an assembly;

[0015] Specifically: Based on the image data file and boundary data file, a new model is created in the finite element software, generating a two-dimensional planar deformable body component; the sketch of the two-dimensional planar deformable body component is opened, and a basic rectangular plate is drawn. The size of the basic rectangular plate is set to 2n+1 times the size of the original image, and the basic rectangular plate is divided into (2n+1)×(2n+1) rectangles of the same size by grid lines; all boundary contour coordinates and corresponding feature point data are imported, the original contour is drawn in the innermost rectangle, and the contour close to the edge of the innermost rectangle is repeatedly drawn in the remaining rectangles excluding the innermost rectangle to generate a periodic contour pattern; during the drawing process, the contour point coordinates are detected in real time, and the points located on the four boundaries of the innermost rectangle are divided into... Record the boundary points to the corresponding list, and then use periodic contour patterns and grid lines to perform face segmentation on the entire base rectangular plate, dividing all faces into four types of regions: the second phase face containing feature points, the base phase face within the innermost rectangle without feature points, all faces within the innermost rectangle, and other faces excluding all faces within the innermost rectangle. On the 2D planar deformable body component, create different face sets for the first three types of regions, including the "all face set", i.e., all faces within the innermost rectangle, the "second phase face set", i.e., all faces in the second phase region, and the "base phase set", i.e., all faces in the base phase region. Finally, instantiate the 2D planar deformable body component to generate an assembly, and locate and create a set of corner points on the assembly, including the four corner points of the innermost rectangle.

[0016] Step 4: Based on the phase type, automatically assign material properties to the different surface sets created in Step 3 and apply elastoplastic constitutive relations;

[0017] Specifically: create a second phase material and define its elastic properties; create a matrix phase material and define its elastoplastic properties; create a homogeneous solid section and assign it to the second phase and matrix phase materials; assign the material sections to different face sets created in step 3.

[0018] Step 5: Perform automatic mesh generation on the assembly generated in Step 3 to obtain a mesh model;

[0019] Specifically: Set the global mesh seed size, which is 10 times the conversion factor from pixels to actual size by default; select the mixed cell type of CPE4R and CPE3, call the mesh generation function of the finite element software, and automatically generate an unstructured finite element mesh on the assembly to obtain the mesh model, that is, the assembly with mesh generation.

[0020] Step 6: Post-process the mesh model by deleting nodes and cells in all regions except the central rectangular region, while retaining the mesh in the central rectangular region.

[0021] Specifically, the steps are: determine the range of the central rectangular area; filter and mark external nodes that do not belong to the central rectangular area; and delete external nodes and associated units.

[0022] Step 7: Create a static generic analysis step and configure historical output requests;

[0023] Step 8: Apply periodic boundary conditions and loads to the boundary of the central rectangular region;

[0024] Specifically, this involves: identifying and pairing nodes on the boundary of the central rectangular region; creating and applying periodic constraint equations; applying fixed constraints to eliminate rigid body motion; and calculating and applying loads.

[0025] Step 9: Automatically create, configure, and submit the finite element analysis job to complete the mechanical calculations;

[0026] Specifically, the process involves: creating an analysis job and configuring computational resources; creating an inp input file from the existing Model and performing a consistency check; submitting the job and monitoring the computational status of the finite element solution process; and generating a finite element calculation result file in .dat format after the solution is completed.

[0027] Step 10: Automatically extract and post-process the mechanical calculation results, and output the mechanical property data file and contour plot;

[0028] Specifically, the process involves: reading the finite element calculation result file output in step 9; extracting the calculated stress and strain components and performing normalization preprocessing; then visualizing the preprocessed data in the form of a heatmap, saving the stress cloud map, and storing the data as an npy format file.

[0029] This method also includes: automated control and batch processing systems;

[0030] Steps 1-10 are integrated into a linear automated workflow through the master script; all operating parameters of the system are managed uniformly using a centralized configuration file; multiple microscopic tissue SEM images under a specified folder are automatically traversed to perform batch modeling and analysis; status monitoring, anomaly capture, log recording and resource cleanup are performed during the execution process.

[0031] On the other hand, this application proposes a computer-readable storage medium storing executable instructions that, when executed, cause a processor to perform the fully automated finite element modeling and performance analysis method for two-phase microstructure images.

[0032] Thirdly, this application proposes a computer program product, including a computer program or instructions, which, when executed by a processor, implements the aforementioned fully automated finite element modeling and performance analysis method for two-phase microstructure images.

[0033] The beneficial effects of adopting the above technical solution are as follows:

[0034] This invention provides a fully automated finite element modeling and performance analysis method for two-phase microstructure images. This method aims to fully automate the entire process from microstructure image preprocessing, feature extraction, geometric modeling, material property assignment, intelligent mesh generation, automatic application of periodic boundary conditions, finite element solution, to integrated post-processing and visualization report generation. It also supports batch processing, completely solving the bottleneck problems of low efficiency, poor consistency, and reliance on manual labor in existing technologies. This achieves high-throughput, standardized, and repeatable predictive analysis of the mechanical properties of material microstructures. Compared with existing technologies, it specifically includes the following beneficial effects:

[0035] 1. Achieved a leapfrog improvement in analysis efficiency, enabling high-throughput analysis: This invention transforms the complex process that traditionally relied on hours of manual operation by experts into a fully automated, batch-processable "one-click" solution. The system can perform continuous, unattended modeling and calculation on massive amounts of microstructure images, making it possible to obtain statistically significant material mechanical property data in a very short time. The overall analysis efficiency is improved by one to two orders of magnitude compared to traditional methods.

[0036] 2. Fundamentally guarantees the objectivity, consistency, and strict repeatability of results: By embedding the entire process into an automated program driven by parameter files, this invention completely eliminates subjective judgment and operational errors caused by human factors in every step, from image processing to boundary condition setting. It ensures that, given the same input image and parameters, consistent modeling and analysis results can be obtained regardless of when or where the work is executed, significantly improving the reliability, rigor, and comparability of different studies. Attached Figure Description

[0037] Figure 1 The overall flowchart of the fully automated finite element modeling and performance analysis method according to an embodiment of the present invention;

[0038] Figure 2 A schematic diagram of the microscopic tissue image preprocessing and contour extraction results according to an embodiment of the present invention;

[0039] Wherein, (a) - example input image, (b) - boundary contour and feature point extraction result image;

[0040] Figure 3 A schematic diagram illustrating the application of periodic boundary conditions in an embodiment of the present invention;

[0041] Among them, (a) is a periodic modeling diagram, and (b) is a schematic diagram of additional periodic boundary conditions;

[0042] Figure 4Schematic diagram of stress cloud diagram results in an embodiment of the present invention;

[0043] Wherein, (a) - first example input image, (b) - second example input image, (c) - first example stress distribution cloud map in the tensile direction, and (d) - second example stress distribution cloud map in the tensile direction. Detailed Implementation

[0044] The specific implementation methods of this application will be further described in detail below with reference to the accompanying drawings and embodiments.

[0045] Example 1:

[0046] This embodiment provides a fully automated finite element modeling and performance analysis method for two-phase microstructure images, such as... Figure 1 As shown, it includes the following steps:

[0047] Step 1: Pre-processed microstructure photographs of alloy materials;

[0048] The SEM images of the microstructure of the alloy material were cropped using image processing software to remove scale bars and areas that did not need to be processed; contrast enhancement technology was used to preprocess the images to enhance their visual effect.

[0049] Step 2: Extract the boundary contour of the second phase in the preprocessed image and generate boundary data file and image data file;

[0050] Specifically, the process involves: reading the image preprocessed in step 1, converting it to grayscale and scaling it; binarizing the image to distinguish different phases; extracting the boundary contours of the second phase using a contour detection algorithm; performing geometric cleanup on the extracted boundary contours, including removing duplicate points, simplifying polygons, and merging adjacent or overlapping contours; converting the coordinates of the cleaned boundary contours from pixel coordinates to actual physical coordinates; generating internal feature points for each independent second phase region; saving all boundary contour coordinates and feature point data in a text file, i.e., the boundary data file; and saving the actual length and width information of the image and the pixel-to-actual-size conversion coefficients in a text file, i.e., the image data file.

[0051] Step 3: Based on the boundary data file and image data file, automatically construct a two-dimensional planar deformable body component containing a periodically expanding region in the finite element software, generate different face sets, and instantiate the two-dimensional planar deformable body component to generate an assembly;

[0052] Specifically: Based on the image data file and boundary data file, a new model is created in the finite element software, generating a two-dimensional planar deformable body component; the sketch of the two-dimensional planar deformable body component is opened, and a basic rectangular plate is drawn. The size of the basic rectangular plate is set to 2n+1 times the size of the original image (n=1 in this embodiment), and the basic rectangular plate is divided into (2n+1)×(2n+1) rectangles of the same size by grid lines; all boundary contour coordinates and corresponding feature point data are imported, the original contour is drawn in the innermost rectangle, and the contour close to the edge of the innermost rectangle is repeatedly drawn in the remaining rectangle excluding the innermost rectangle to generate a periodic contour pattern; the contour point coordinates are detected in real time during the drawing process, and the coordinates of the contour points located on the four boundaries of the innermost rectangle are... Each point is recorded to its corresponding boundary point list. Then, the entire base rectangular plate is segmented using a periodic contour pattern and grid lines, dividing all faces into four types of regions: the second phase face containing feature points, the base phase face within the innermost rectangle without feature points, all faces within the innermost rectangle (i.e., the sum of the first two), and all other faces excluding those within the innermost rectangle. Different face sets are created on the two-dimensional planar deformable body component for the first three types of regions, including the "all face set" (all faces within the innermost rectangle), the "second phase face set" (all faces in the second phase region), and the "base phase set" (all faces in the base phase region). Finally, the two-dimensional planar deformable body component is instantiated to generate an assembly, and a set of corner points, including the four corner points of the innermost rectangle, is created on the assembly.

[0053] Step 4: Based on the phase type, automatically assign material properties to the different surface sets created in Step 3 and apply elastoplastic constitutive relations;

[0054] Specifically: create a second phase material and define its elastic properties; create a matrix phase material and define its elastoplastic properties; create a homogeneous solid section and assign it to the second phase and matrix phase materials; assign the material sections to different face sets created in step 3.

[0055] Step 5: Perform automatic mesh generation on the assembly generated in Step 3 to obtain a mesh model;

[0056] Specifically: Set the global mesh seed size, which is 10 times the conversion factor from pixels to actual size by default; select a mixed cell type of CPE4R (two-dimensional four-node bilinear quadrilateral, reduced integral) and CPE3 (two-dimensional three-node linear triangle); call the mesh generation function of the finite element software to automatically generate an unstructured finite element mesh on the assembly and obtain the mesh model, that is, the assembly with mesh generation.

[0057] Step 6: Post-process the mesh model by deleting nodes and cells in all regions except the central rectangular region, while retaining the mesh in the central rectangular region.

[0058] Specifically, the steps are: determine the range of the central rectangular area; filter and mark external nodes that do not belong to the central rectangular area; and delete external nodes and associated units.

[0059] Step 7: Create a static generic analysis step and configure historical output requests;

[0060] Step 8: Apply periodic boundary conditions and loads to the boundary of the central rectangular region;

[0061] Specifically, this involves: identifying and pairing nodes on the boundary of the central rectangular region; creating and applying periodic constraint equations; applying fixed constraints to eliminate rigid body motion; and calculating and applying loads.

[0062] Step 9: Automatically create, configure, and submit the finite element analysis job to complete the mechanical calculations;

[0063] Specifically, the process involves: creating an analysis job and configuring computational resources; creating an inp input file from the existing model (including two-dimensional planar deformable parts and assemblies) and performing a consistency check; submitting the job and monitoring the computational status of the finite element solution process; and generating a finite element calculation result file in .dat format after the solution is completed.

[0064] Step 10: Automatically extract and post-process the mechanical calculation results, and output the mechanical property data file and contour plot;

[0065] Specifically, the process involves: reading the finite element calculation result file output in step 9; extracting the calculated stress and strain components and performing normalization preprocessing; then visualizing the preprocessed data in the form of a heatmap, saving the stress cloud map, and storing the data as an npy format file.

[0066] This method also includes: automated control and batch processing systems;

[0067] Steps 1-10 are integrated into a linear automated workflow through the main control script (main.ipynb); all operating parameters of the system are managed uniformly using a centralized configuration file; multiple microstructure SEM images under a specified folder are automatically traversed to perform batch modeling and analysis; status monitoring, anomaly capture, log recording and resource cleanup are performed during the execution process.

[0068] Example 2:

[0069] In this embodiment, as shown Figure 2 As shown, where Figure 2 (a) is the example input image. Figure 2 (b) shows the boundary contour and feature point extraction results. The blue filled area in the figure represents all extracted second-phase particles, and the green dots represent feature points within the corresponding second-phase particle regions.

[0070] In this embodiment, as shown Figure 3 As shown, where Figure 3 (a) For a periodic modeling diagram, the precipitate boundary contours on the edge of the central rectangle are copied to the surrounding rectangles to construct a periodic structure; Figure 3 (b) is an additional schematic diagram for periodic boundary conditions, showing the specific ways to apply displacement constraints or equation coupling on the model boundary.

[0071] In this embodiment, as shown Figure 4 As shown, Figure 4 (a) is the first example input image. Figure 4 (b) is the input image for the second example. Figure 4 (c) is a contour map of the stress distribution in the tensile direction for the first example. Figure 4 (d) is the stress distribution contour map in the tensile direction for the second example. In the stress contour map, the color changes from blue to red to indicate that the stress value gradually increases.

[0072] Example 3:

[0073] This embodiment proposes a computer-readable storage medium that stores executable instructions. When these instructions are executed, if they are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium.

[0074] The computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the fully automated finite element modeling and performance analysis method for two-phase microstructure images described in the various embodiments of this application.

[0075] The aforementioned storage media include: flash memory, hard disk, multimedia card, card-type memory (e.g., SD (Secure Digital Memory Card) or DX (Memory Data Register, MDR) memory, random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic storage, disk, optical disk, server, APP (Application) application store, and other media capable of storing program verification codes. These media store computer programs, which, when executed by a processor, can implement the various steps of the aforementioned fully automated finite element modeling and performance analysis method for two-phase microstructure images.

[0076] This embodiment proposes a computer program product, including a computer program or instructions, which, when executed by a processor, implements the aforementioned fully automated finite element modeling and performance analysis method for two-phase microstructure images.

[0077] Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or part of the technical solution, can be embodied in the form of a computer program product.

[0078] The various embodiments in this application are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0079] The scope of protection of this application is not limited to the embodiments described above. Obviously, those skilled in the art can make various modifications and variations to this disclosure without departing from the scope and spirit of this disclosure. If such modifications and variations fall within the scope of the methods disclosed herein and their equivalents, then the intent of this disclosure also includes such modifications and variations.

Claims

1. A fully automated finite element modeling and performance analysis method for two-phase microstructure images, characterized in that, Includes the following steps: Step 1: Pre-processed microstructure photographs of alloy materials; Step 2: Extract the boundary contour of the second phase in the preprocessed image and generate boundary data file and image data file; Step 3: Based on the boundary data file and image data file, automatically construct a two-dimensional planar deformable body component containing a periodically expanding region in the finite element software, generate different face sets, and instantiate the two-dimensional planar deformable body component to generate an assembly; Step 4: Based on the phase type, automatically assign material properties to the different surface sets created in Step 3 and apply elastoplastic constitutive relations; Step 5: Perform automatic mesh generation on the assembly generated in Step 3 to obtain a mesh model; Step 6: Post-process the mesh model by deleting nodes and cells in all regions except the central rectangular region, while retaining the mesh in the central rectangular region. Specifically determine the range of the central rectangular region, filter and mark external nodes that do not belong to the central rectangular region, and delete the external nodes and their associated cells. Step 7: Create a static generic analysis step and configure historical output requests; Step 8: Apply periodic boundary conditions and loads to the boundary of the central rectangular region; Step 9: Automatically create, configure, and submit the finite element analysis job to complete the mechanical calculations; Step 10: Automatically extract and post-process the mechanical calculation results, and output mechanical property data files and contour maps.

2. The fully automated finite element modeling and performance analysis method for two-phase microstructure images according to claim 1, characterized in that, Step 1 specifically involves: cropping the SEM image of the microstructure of the alloy material using image processing software to remove the scale bars and areas that do not need to be processed; and using contrast enhancement technology to preprocess the image to enhance its visual effect.

3. The fully automated finite element modeling and performance analysis method for two-phase microstructure images according to claim 1, characterized in that, Step 2 specifically involves: reading the image preprocessed in Step 1, performing grayscale conversion and scaling; binarizing the image to distinguish different phases; extracting the boundary contours of the second phase using a contour detection algorithm; performing geometric cleanup on the extracted boundary contours, including removing duplicate points, simplifying polygons, and merging adjacent or overlapping contours; converting the coordinates of the cleaned boundary contours from pixel coordinates to actual physical coordinates; generating internal feature points for each independent second phase region; saving all boundary contour coordinates and feature point data in a text file, i.e., a boundary data file; and saving the actual length and width information of the image and the pixel-to-actual-size conversion coefficients in a text file, i.e., an image data file.

4. The fully automated finite element modeling and performance analysis method for two-phase microstructure images according to claim 1, characterized in that, Step 3 specifically involves: based on the image data file and boundary data file, creating a new model in the finite element software and generating a two-dimensional planar deformable body component; opening the sketch of the two-dimensional planar deformable body component, drawing a basic rectangular plate, setting the size of the basic rectangular plate to 2n+1 times the size of the original image, and dividing the basic rectangular plate into (2n+1)×(2n+1) rectangles of the same size using grid lines; importing all boundary contour coordinates and corresponding feature point data, drawing the original contour within the innermost rectangle, and repeatedly drawing the contour close to the edge of the innermost rectangle within the remaining rectangle excluding the innermost rectangle to generate a periodic contour pattern; During the drawing process, the coordinates of the contour points are detected in real time. Points located on the four boundaries of the central rectangle are recorded in the corresponding boundary point list. Then, the entire base rectangular plate is segmented using a periodic contour pattern and grid lines, dividing all faces into four types of regions: the second phase face containing feature points, the base phase face within the central rectangle without feature points, all faces within the central rectangle, and other faces excluding all faces within the central rectangle. Different face sets are created on the two-dimensional planar deformable body component for the first three types of regions, including the "all face set" (all faces within the central rectangle), the "second phase face set" (all faces in the second phase region), and the "base phase set" (all faces in the base phase region). Finally, the two-dimensional planar deformable body component is instantiated to generate an assembly, and a set of corner points, including the four corner points of the central rectangle, is created on the assembly.

5. The fully automated finite element modeling and performance analysis method for two-phase microstructure images according to claim 1, characterized in that, Step 4 specifically involves: creating a second phase material and defining its elastic properties; creating a matrix phase material and defining its elastoplastic properties; creating a homogeneous solid cross section and assigning it to the second phase and matrix phase materials; and assigning the material cross section to the different face sets created in step 3.

6. The fully automated finite element modeling and performance analysis method for two-phase microstructure images according to claim 1, characterized in that, Step 5 specifically involves: setting the global mesh seed size, which is 10 times the conversion factor from pixels to actual size by default; selecting a hybrid cell type of CPE4R and CPE3; calling the mesh generation function of the finite element software; automatically generating an unstructured finite element mesh on the assembly; and obtaining a mesh model, i.e., an assembly with mesh generation.

7. The fully automated finite element modeling and performance analysis method for two-phase microstructure images according to claim 1, characterized in that, Step 8 specifically involves: identifying and pairing nodes on the boundary of the central rectangular region; creating and applying periodic constraint equations; Apply fixed constraints to eliminate rigid body motion; calculate and apply loads.

8. The fully automated finite element modeling and performance analysis method for two-phase microstructure images according to claim 1, characterized in that, Step 9 specifically involves: creating an analysis job and configuring computing resources; creating an inp input file from the existing Model and performing a consistency check; submitting the job and monitoring the computational status of the finite element solution process; and generating a finite element calculation result file in .dat format after the solution is completed.

9. The fully automated finite element modeling and performance analysis method for two-phase microstructure images according to claim 1, characterized in that, Step 10 specifically involves: reading the finite element calculation result file output in step 9; extracting the calculated stress and strain component data and performing normalization preprocessing; The preprocessed data is then visualized as a heatmap, and the stress cloud map is saved. The data is also stored as an npy file.

10. The fully automated finite element modeling and performance analysis method for two-phase microstructure images according to claim 1, characterized in that, Also includes: Automated control and batch processing system: Steps 1-10 are integrated into a linear automated workflow through the master control script; The system employs a centralized configuration file to manage all operational parameters; it automatically traverses multiple microscopic tissue SEM images in a specified folder to perform batch modeling and analysis; and it performs status monitoring, anomaly capture, log recording, and resource cleanup during execution.

Images