Method and system for generating two-dimensional finite element model of composite materials based on XCT images

Abstract

The invention discloses a method and system for generating a two-dimensional finite element model of a composite material based on an XCT image, and belongs to the technical field of finite element analysis. The method includes: collecting an initial state XCT image of a composite material and preprocessing it; performing statistical analysis on two-dimensional microscopic structural parameters; generating a virtual XCT image based on a Pix2PixHD style transfer network; realizing accurate segmentation of each component region of the virtual XCT image based on a Swin‑Tranformer network; completing contour feature extraction and smoothing of the segmented image based on the mapping relationship between pixels and nodes; constructing a spatial coordinate system and performing discrete derivation of all yarn center points of each segmented image to construct a feature database; constructing feature details and grid division based on the position information of the contour features of each component; inserting a cohesive force unit; customizing material properties and allocating local material directions and boundary condition settings.

Description

Technical Field

[0001] The invention belongs to the technical field of finite element analysis, and in particular relates to a method and system for generating a two-dimensional finite element model of a composite material based on XCT images. Background Art

[0002] Basalt fiber reinforced composites (BFRP) are composite products made of basalt fiber as reinforcement, usually using resin transfer molding (RTM) processing technology. Basalt fiber reinforced composites have excellent functional structural characteristics such as light weight and designable performance. They have gradually replaced some metal and alloy materials and have been widely used in transportation, aerospace, military and national defense. They are of great significance to the upgrading of my country's manufacturing industry and the transformation of industrial lightweighting.

[0003] XCT (X-ray Computed Tomography) scanning is a non-destructive testing technology based on X-rays, which can provide high-resolution information about the internal structure of materials. Through XCT scanning, we can obtain the microstructure of basalt fiber composite materials. XCT scanning technology can be used to obtain the internal microstructure characterization of composite materials under non-destructive conditions. This provides an important means of image data acquisition for further characterizing the internal microstructure of composite materials.

[0004] However, due to the obvious anisotropy and inhomogeneity of composite materials, their macroscopic mechanical properties testing often requires more experimental samples, complex testing equipment, high testing costs and long test cycles. Therefore, for the deformation and damage of composite materials under external loads, traditional test methods only focus on characterizing the average mechanical properties at the macroscopic level, ignoring the mechanical properties and spatial distribution characteristics of each component inside the composite material, and failing to establish a cross-scale connection between the micro-macroscopic mechanical properties.

[0005] Therefore, in recent years, finite element numerical simulation methods have become a common means of studying the mechanical properties of composite materials. By qualitatively and quantitatively characterizing the microstructural parameters, a more realistic finite element model is constructed, material properties and boundary conditions are set, and the deformation and damage evolution of materials under real experimental conditions are simulated, so that the surface and internal fracture mechanisms of the material can be observed at a low cost. However, most finite element models of basalt fiber reinforced composite materials tend to consider the three-dimensional whole, ignoring the intrinsic connection between the two-dimensional finite element model and the three-dimensional finite element model. It is difficult to accurately express the mechanical response characteristics by only considering the three-dimensional finite element model. Summary of the invention

[0006] In view of the shortcomings of the prior art, the present invention proposes a method and system for generating a two-dimensional finite element model of composite materials based on XCT images, which can be used to quickly and batch establish two-dimensional finite element models of basalt composite materials based on XCT images, and study their microstructural damage evolution process and fracture state.

[0007] To achieve the above object, the present invention provides the following solutions:

[0008] The method for generating a two-dimensional finite element model of a composite material based on XCT images includes the following steps:

[0009] S1: Acquire XCT images of the composite material of basalt fiber in the initial state and perform preprocessing;

[0010] S2: performing parametric statistical analysis on the pre-processed XCT image of the composite material, quantifying the microscopic structural parameters of the yarn inside the material and performing two-dimensional parametric random modeling to obtain a parametric random modeling image;

[0011] S3: learning the parameterized random modeling image based on the adversarial learning style transfer network Pix2PixHD to generate a virtual XCT image;

[0012] S4: Accurate segmentation of each component region of virtual XCT images based on Swin-Tranformer network;

[0013] S5: Based on the segmented virtual XCT image, the microscopic features of the contour are extracted and the local contour is smoothed to build a yarn and pore feature database;

[0014] S6: Based on the yarn and pore feature database, extract the contour feature of each yarn in each segmented image, calculate the coordinate value of the center of the yarn contour, establish a spatial coordinate system, and based on the spatial coordinate system, perform discrete derivation on the center of the yarn contour in each segmented image to construct a feature database;

[0015] S7: Based on the feature database, extract feature data information of a single segmented image, classify each component area, and divide it into grids;

[0016] S8: Based on the divided grid, insert cohesive force units and define material properties for each component area feature, and assign local material directions;

[0017] S9: Setting boundary conditions and numerical simulation based on the assigned local material direction;

[0018] S10: Repeat S7-S9 to realize the rapid generation and numerical simulation of a two-dimensional finite element model based on XCT digital images of composite materials.

[0019] Preferably, in S1, the method of collecting XCT images of composite materials of basalt fibers in an initial state and performing preprocessing includes:

[0020] S11: Acquire complete XCT scan source image data, and perform cropping and filtering on the XCT scan source image data;

[0021] S12: Based on the cropped and filtered XCT scan source image data, a quantitative number of segmented images are extracted at equal intervals, and each component in each segmented image is manually labeled;

[0022] S13: Perform feature extraction on the manually annotated segmented image to obtain a true label set.

[0023] Preferably, in S2, the method of performing parametric statistical analysis on the pre-processed XCT image of the composite material, quantifying the microscopic structural parameters of the yarn inside the material and performing two-dimensional parametric random modeling includes:

[0024] S21: Extract the mesoscopic structural parameters of each yarn in each segmented image based on the manually annotated mesoscopic structural features of the dataset;

[0025] S22: Based on the mesoscopic structural parameters of each yarn, the mesoscopic structural characteristics are qualitatively and quantitatively expressed, and statistical methods are used for parametric statistical analysis;

[0026] S23: Based on the results of the parametric statistical analysis, the data set is expanded by parametric random modeling to generate a virtual data set, namely a parametric random modeling image.

[0027] Preferably, in S3, the method of learning the parameterized random modeling image based on the adversarial learning style transfer network Pix2PixHD to generate a virtual XCT image includes:

[0028] S31: generating a warp yarn random model based on warp yarn parameterized statistical analysis;

[0029] S32: Generate a random model of weft yarns according to the parameterized statistical analysis of weft yarns;

[0030] S33: generating an artificial synthetic label image based on the warp yarn random model and the weft yarn random model;

[0031] S34: Based on the artificially synthesized label image, an improved Pix2PixHD network is applied to generate a basalt fiber composite image with the same microstructure as the real one, and a hybrid data set, i.e., a virtual XCT image, is constructed.

[0032] Preferably, in S4, the method for realizing accurate segmentation of each component region of a virtual XCT image based on a Swin-Tranformer network comprises:

[0033] S41: Input the mixed data set to the Swin-Tranformer network to obtain training images;

[0034] S42: Divide the training image into different small blocks by pixels through a block division module, and then convert it into a one-dimensional feature vector image;

[0035] S43: Accurately segment each component region in the one-dimensional feature vector image to obtain a semantic segmentation result image;

[0036] S44: Based on the semantic segmentation result map, a watershed segmentation algorithm is used to complete region segmentation for each yarn.

[0037] Preferably, in S5, the method of extracting contour microscopic features and smoothing local contours based on the segmented virtual XCT image includes:

[0038] S51: Selecting a resolution that meets preset requirements to downsample the segmented virtual XCT image;

[0039] S52: performing feature screening on the virtual XCT image after downsampling;

[0040] S53: extracting image contour features of pores and yarns based on the virtual XCT image after feature screening, and converting coordinate values ​​according to the positional relationship between pixels and nodes to obtain contour coordinate values;

[0041] S54: setting a convolution kernel that meets preset requirements, and using the operation between the convolution kernel and the contour coordinate value to achieve contour smoothing;

[0042] S55: Process all segmented images according to S51-S54, build a yarn and pore feature database, and store the contour coordinate values ​​of each component after the coordinate system is transformed.

[0043] Preferably, in S6, based on the yarn and pore feature database, the contour feature of each yarn in each segmented image is extracted, the coordinate value of the center of the yarn contour is calculated, a spatial coordinate system is established, and based on the spatial coordinate system, the method for performing discrete derivation of the center of the contour of each yarn in each segmented image includes:

[0044] S61: extracting the coordinate value of the center point of each yarn contour in each segmented image from the yarn and pore feature database, and establishing a spatial coordinate system in the order of the segmented images;

[0045] S62: Based on the spatial coordinate system, use the first-order intermediate differential law to perform discrete derivative on the center point of each yarn contour in each segmented image to obtain the tangent direction of the center point, that is, the 1 direction of the yarn;

[0046] S63: Obtain direction 2 from direction 1 obtained by calculation.

[0047] Preferably, in S7, based on the feature database, extracting feature data information of a single segmented image, classifying each component region, and dividing the grid comprises:

[0048] S71: extracting all the yarn contour and pore contour coordinate value data of a single segmented image from the feature database;

[0049] S72: Call the underlying functions of Abaqus 2021 software in sequence based on the Python language;

[0050] S73: Based on the coordinate value data of all yarn profiles and pore profiles and the underlying functional function, the component regions are divided and restored according to the internal microstructure position characteristics of the material, and the grid size is set for grid division.

[0051] Preferably, in S8, based on the divided grid, the method of inserting the cohesion unit and defining the material properties of each component region feature and allocating the local material direction includes:

[0052] S81: inserting cohesive force units at the contact interfaces between yarns and yarns and between yarns and matrix from the divided grid;

[0053] S82: gradually allocating the calculated local material direction vector to each unit of each yarn;

[0054] S83: Define material properties and mechanical response characteristics for each component region.

[0055] The present invention also provides a system for generating a two-dimensional finite element model of a composite material based on XCT images, comprising: an acquisition module, a modeling module, a learning module, a precise segmentation module, a first database construction module, a second database construction module, a partitioning module, an allocation module, a calculation module and a model generation module;

[0056] The acquisition module is used to acquire the composite XCT image of the basalt fiber in the initial state and perform preprocessing;

[0057] The modeling module is used to perform parametric statistical analysis on the pre-processed XCT image of the composite material, quantify the microscopic structural parameters of the yarn inside the material and perform two-dimensional parametric random modeling to obtain a parametric random modeling image;

[0058] The learning module is used to learn the parameterized random modeling image based on the adversarial learning style transfer network Pix2PixHD to generate a virtual XCT image;

[0059] The precise segmentation module is used to realize precise segmentation of each component area of ​​the virtual XCT image based on the Swin-Tranformer network;

[0060] The first database construction module is used to extract contour microscopic features and smooth local contours based on the segmented virtual XCT image to construct a yarn and pore feature database;

[0061] The second database construction module is used to extract the contour features of each yarn in each segmented image based on the yarn and pore feature database, calculate the coordinate value of the center of the yarn contour, establish a spatial coordinate system, and perform discrete derivative of the center of the yarn contour in each segmented image based on the spatial coordinate system to construct a feature database;

[0062] The segmentation module is used to extract the feature data information of a single segmented image based on the feature database, classify each component area, and divide it into grids;

[0063] The allocation module is used to insert cohesion units based on the divided grids and define material properties for each component region feature, and allocate local material directions;

[0064] The calculation module is used to set boundary conditions and perform numerical simulation calculations based on the assigned local material directions;

[0065] The model generation module is used for repeatedly dividing the module-calculation module to realize the rapid generation and numerical simulation of a two-dimensional finite element model based on XCT digital images of composite materials.

[0066] Compared with the prior art, the present invention has the following beneficial effects:

[0067] The present invention discloses a method and system for generating a two-dimensional finite element model of a composite material based on an XCT image, collects an initial state of the composite material XCT image and preprocesses it, performs statistical analysis on two-dimensional micro-structure parameters, generates a virtual XCT image-virtual label image pair based on a Pix2PixHD style transfer network to achieve data enhancement, accurately segments each component inside the composite material based on a Swin Transformer network, extracts and smoothes the contour features of the segmented image based on the mapping relationship between pixels and nodes, constructs a spatial coordinate system and discretely derives all yarn center points of each segmented image, establishes a feature database, constructs a feature region based on the feature coordinate information of the contour of each component and divides the grid, inserts a cohesive force unit, customizes material properties, allocates local material directions, sets boundary conditions, and completes numerical simulation calculations. The present invention can be used to quickly and batch establish a two-dimensional finite element model of a composite material based on an XCT image, study its micro-structure damage evolution process and rupture state, and help reveal the intrinsic connection between the two-dimensional finite element model and the three-dimensional whole. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0069] Figure 1 This is a schematic diagram of a flow chart of a method for generating a two-dimensional finite element model of a composite material based on XCT images according to an embodiment of the present invention;

[0070] Figure 2 This is a schematic diagram of XCT image segmentation results based on deep learning according to an embodiment of the present invention;

[0071] Figure 3 This is a schematic diagram of preprocessing based on segmented XCT images according to an embodiment of the present invention; wherein: Figure 3 (a) in the figure is the segmented XCT image. Figure 3 (b) is the XCT image after downsampling and filtering of non-interest areas;

[0072] Figure 4 Schematic diagram of the local material direction assignment algorithm according to an embodiment of the present invention; wherein, Figure 4 (a) shows the distribution state of a single yarn feature in space and the extraction of the center line. Figure 4 (b) is the discrete derivative result of a single yarn at the center point on the XCT image (only two points are selected as examples). Figure 4 (c) in the calculation is the distance from the centroid of the unit to the center of all XCT images. Figure 4 (d) is the result of discrete derivation based on the shortest distance assignment unit;

[0073] Figure 5 A schematic diagram of generating a two-dimensional finite element model according to an embodiment of the present invention; wherein: Figure 5 (a) is a finite element model with cohesive elements. Figure 5 (b) in the figure is the boundary condition setting of the finite element model;

[0074] Figure 6 It is a schematic diagram of simulation results of a two-dimensional finite element model of an embodiment of the present invention. DETAILED DESCRIPTION

[0075] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0076] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0077] Embodiment 1

[0078] like Figure 1 , Figure 2 , Figure 3 , Figure 4 , Figure 5 and Figure 6 As shown, this embodiment provides a schematic flow chart of a method for generating a two-dimensional finite element model of a composite material based on XCT images, comprising the following steps:

[0079] S1, collecting an initial state XCT image of the composite material, and manually segmenting the image according to the internal component categories of the material based on the XCT image;

[0080] Specifically, the steps include:

[0081] Further preferably, S1 includes:

[0082] S11, acquiring complete XCT scan source image data, obtaining 1450 source images, appropriately cropping and filtering the source image data, and using a cropping tool to remove segmented images with blurred edge details;

[0083] S12, extracting one segmented image every 10 segmented images, manually marking each component in each extracted segmented image, and completely segmenting to obtain four components: resin base, warp yarn, weft yarn, and pores;

[0084] S13. Perform feature extraction on the annotated image set to obtain a true label set, which has 139 true label images in total.

[0085] S2. Perform parametric statistical analysis on the completed manually segmented images, quantify the microstructural parameters of the yarn inside the material and perform two-dimensional parametric random modeling;

[0086] Further preferably, S2 includes:

[0087] S21, extracting the meso-structural parameters such as yarn length and average width of each yarn in each segmented image based on the meso-structural features of the manually annotated dataset;

[0088] S22, qualitatively and quantitatively express the mesostructure characteristics, and use statistical methods to analyze the statistical results;

[0089] S23, based on the results of the parametric statistical analysis, the data set is expanded by parametric random modeling to generate a virtual data set;

[0090] The statistical analysis includes: statistical analysis of warp yarn geometric parameters and statistical analysis of weft yarn geometric parameters.

[0091] The method for statistical analysis of warp yarn geometric parameters is: to count the average width, width standard deviation, length and other information of the marked warp yarn area, calculate the centroid of each warp yarn, and read the distance between the centroids of each warp yarn to generate multiple warp yarns.

[0092] The method for statistical analysis of weft yarn geometric parameters is as follows: the ellipse is fitted by the fitellipse function in the OpenCV library, the weft yarn inclination angle is calculated, the farthest distance of each weft yarn in the image is manually marked and used as the long axis, the two intersection points of the straight line perpendicular to the long axis are calculated to calculate the length of the short axis, and the ratio of the long axis to the short axis is obtained. The collected data of each weft yarn is substituted into the ellipse representation equation with the same center, and the ratio of the actual radius of the pixel at each angle to the ideal radius is calculated to achieve the irregularity of the ellipse.

[0093] Statistical analysis of warp yarn geometric parameters and weft yarn geometric parameters: Use the fitter library to automatically search for the distribution supported by the scipy library for fitting, and automatically fit the mean width, standard deviation, length, etc. of the warp yarn, the major diameter, aspect ratio, inclination angle and other data of the weft yarn.

[0094] S3, the adversarial learning style transfer network Pix2PixHD learns the parameterized random modeling image to generate a virtual XCT image;

[0095] Further preferably, S3 includes:

[0096] S31, generating a warp yarn random model according to warp yarn parameterized statistical analysis;

[0097] S32, generating a weft yarn random model according to the weft yarn parameterized statistical analysis;

[0098] S33, generating an artificial synthetic label image based on the warp random model and the weft random model;

[0099] S34. Apply the improved Pix2PixHD network to generate basalt fiber composite images with the same real mesoscopic structure and construct a hybrid dataset.

[0100] The specific implementation method of S31 is:

[0101] First, a straight line is generated as its center line, and then the width of the center line is filled. The center line is divided equally by selecting 4 to 6 nodes in a straight line, including the upper and lower endpoints, and the nodes in the middle part are randomly offset. The maximum offset is selected by experience, and the generated points are fitted into a curve by spline interpolation, which is used as the center line. According to the length of the generated center line, a corresponding number of widths are generated, and 1-3 nodes are randomly selected from them. The generated widths are sorted according to the nodes, and the shape of the uneven width of the real warp yarn is approached in an ascending and descending staggered manner. At the same time, the tip part required by other categories is generated in this way. The generated curved irregular rectangle is subjected to random elastic deformation. Random elastic deformation randomly generates two matrices of the same size as the image, representing the horizontal and vertical position offsets of the coordinates of each pixel point, and smoothed by Gaussian filtering to make the matrix continuous, forming a coordinate offset matrix in two directions. The original coordinates of the image are applied with offsets in two different directions, and the grayscale values ​​of the original image coordinates are obtained by interpolation of the offset pixels, and an image is reconstructed to achieve the effect of random elastic deformation. In addition, the median filter method is used to remove the burrs generated during the width sorting process. Different types of warp yarns are arranged at a certain interval.

[0102] The specific implementation method of S32 is:

[0103] A basic elliptical shape is generated with the fitted minor axis length and major-minor axis ratio. Nodes are set with equal lengths on the circumference of the ellipse, and then offsets are generated for all nodes according to the fitted weft offsets, which are applied to the lengths of all nodes. Here, 720 nodes are collected during generation, and a special smoothing method is used to take the midpoints of adjacent nodes of the ellipse. After multiple smoothings, the entire ellipse is median filtered to eliminate burrs, and finally the ellipse is tilted according to the inclination angle generated by the fitting.

[0104] The specific implementation method of S33 is:

[0105] The original image size is used as the generated image size, and the upper and lower boundaries of the resin area are artificially limited. The warp yarn is generated first, and the warp yarn is arranged in position according to the generated spacing until the remaining width in the image cannot accommodate the next generated spacing and the warp yarn width. After that, the two ends of the warp yarn are selected as the lateral boundaries of the resin area. Then the weft yarn is generated, and the weft yarn is randomly selected for placement. The placement area is expanded upward and downward, and in this way, the shape of the truncated weft yarn on the boundary of the original image is generated. Then the overlap rate with the already placed fibers is calculated to decide whether to place the weft yarn. Finally, the placement of the weft yarn stops when the placement fails a certain number of times or the total fiber area of ​​the image reaches a certain level.

[0106] The specific implementation method of S34 is:

[0107] The Pix2PixHD network is a conditional generative adversarial network (cGAN). Through supervised learning, the generated image is more accurately similar to the input target image in grayscale distribution. Its main purpose is to train a generator G, pass the original image x through the generator network G to obtain a fake image G(x), and train a discriminator D at the same time. The generated fake image G(x) and the real image are input into the discriminator D. Through training, the features of the original image are similar to those of the generated image, so that the discriminator D cannot distinguish whether the image is forged. By changing the generation part of the Pix2PixHD network, the generation part can simultaneously train two generator models at different scales, and fuse the features of the two generator models at different scales. Based on the generator model after feature fusion, a fiber composite image with the same mesoscopic structure as the real one is generated. 120 pictures with good synthesis effects were manually selected from the 200 synthetic data sets to form a synthetic data set, and together with the original data set consisting of 139 pictures, they constitute the final data set. We randomly selected 15 images from the 139 original datasets as the validation set, and used the remaining 124 original datasets and 120 synthetic datasets to form the training set.

[0108] S4, based on the Swin-Tranformer network, accurate segmentation of XCT images is achieved, and the contour features between each component are clear;

[0109] Further preferably, S4 includes:

[0110] S41, input the mixed data set to the Swin-Tranformer network;

[0111] S42, dividing the image into different small blocks by pixels through a block division module, and then converting them into a one-dimensional feature vector;

[0112] S43, network model training, segmenting each component region in the image;

[0113] S44, completing region segmentation for each yarn based on a watershed segmentation algorithm;

[0114] The specific implementation method of S41 is:

[0115] The image with the size of 1119×1431 is randomly cropped to 448×448, and the image is horizontally flipped, vertically flipped, and rotated for data enhancement. The image is randomly cropped because the original image size is large. In order to reduce the amount of calculation and adapt to the Swin Transformer, the training speed is accelerated. Effective data enhancement can enhance the robustness of the model and prevent overfitting. Finally, the obtained image is input into the network for training.

[0116] The specific implementation method of S42 is:

[0117] The training images in the mixed dataset are input into the network and divided into blocks of size 7×7 pixels, and then the images are divided into multiple small blocks. Then the pixels of the small blocks on all channels are stretched into one-dimensional feature vectors, and finally all one-dimensional feature vectors are combined into a feature vector map.

[0118] The specific implementation method of S43 is:

[0119] The feature vector graph is processed through four stages to obtain four feature vector graphs at different scales, where each stage consists of block merging and Swin-Transformer Block. Swin-Transformer Block includes layer normalization layer, window multi-head self-attention layer, and multi-layer perception layer. Finally, the normalized category prediction probability of length 2 is obtained through the softmax function, which corresponds to pores, resin base, weft yarn, and warp yarn in the four channels respectively. The channel with the highest prediction probability is determined as the corresponding category, thereby obtaining the semantic segmentation result graph of the entire input image, as shown in the figure below: Figure 1 shown.

[0120] The specific implementation method of S44 is:

[0121] Based on the semantic segmentation of the yarn by the ResUNet+ model, the XCT image segmentation results are extracted along the warp and weft directions respectively, and the yarn is binarized, and the binarization result is used as the labeled image; secondly, the distance transform operation is performed on the binarized image to obtain the distance transform map; finally, the watershed segmentation is performed based on the map, the maximum value of the distance transform is used as the water injection point, and the binary image of the yarn is used as a mask to limit the segmentation area. Considering that the watershed segmentation effect of the long strip yarn is not ideal, the binary image of the weft needs to be extracted and processed along the weft laying direction. After the segmentation is completed, a small number of misclassified areas are manually corrected to obtain the instance segmentation results.

[0122] S5, such as Figure 3 As shown, based on the segmented XCT image, Figure 3 As shown in (a), the contour microscopic features are extracted and the local contour is smoothed;

[0123] Further preferably, S5 includes:

[0124] S51, select a suitable resolution to downsample the image, retain the detailed features of the microstructure as much as possible, and reduce the cost and amount of computing resources;

[0125] S52, perform feature screening on the downsampled image, filter out a small part of the non-interesting area, and retain the necessary feature areas such as pores and yarns, such as Figure 3 As shown in (b);

[0126] S53, extracting the image contour features of the pores and the yarns, and establishing a coordinate value conversion based on the positional relationship between pixels and nodes.

[0127] S54, setting a suitable convolution kernel, and utilizing the operation between the convolution kernel and the extracted contour coordinate values ​​to achieve contour smoothing, thereby avoiding stress concentration and error interruption that may occur during finite element numerical calculations.

[0128] S55. Process all segmented images according to the steps of S51-S54, construct a yarn and pore feature database, and store the contour coordinate values ​​of each component after the coordinate system transformation.

[0129] The specific implementation method of S51 is:

[0130] Based on the image that has completed the task of segmenting each component region, a secondary cropping is performed to remove the surrounding non-interested areas, and an image with a length and width of 1020×1350 pixels and a resolution of 4.9819um is obtained. On the premise of retaining the detailed features of the microscopic structure as much as possible, the image is downsampled by taking one pixel every 10 pixels along the length and width directions, and an image of 102×135 pixels and a resolution of 49.819um is obtained by integration, thereby reducing the number of pixels that can be processed and reducing the cost of computing resources.

[0131] The specific implementation method of S52 is:

[0132] Due to the geometric characteristics of the yarn itself, it is inevitable that the yarn itself has a high degree of dispersion in the segmented image. In a single image, the characteristics of a single yarn are distributed in multiple areas. Therefore, the contourArea function in the OpenCV library is used to filter small areas composed of multiple fibers. By setting the screening threshold to 5, the area smaller than the fixed threshold will not be considered, and only the characteristics of the yarn area with a larger area will be retained, such as Figure 2 shown.

[0133] The specific implementation method of S53 is:

[0134] The findContours function in the OpenCV library is used to extract the pixel coordinate value information of the contour of each yarn and pore, and the information is converted according to the positional relationship between the pixel and the node. In addition, considering that the coordinate system in the OpenCV library is different from the coordinate system of Abaqus2021, the node coordinate values ​​converted from the pixel contour coordinate values ​​need to be converted into a coordinate system to ensure that the segmented image and the constructed two-dimensional finite element model are highly consistent in the microstructure.

[0135] The specific implementation method of S54 is:

[0136] When considering finite element numerical simulation, the junction of each component is prone to stress concentration and interruption of calculation. Therefore, it is necessary to perform secondary processing on the extracted contour feature information of each component, set a suitable convolution kernel, and create additional nodes to achieve contour smoothing when the local contour coordinate value changes greatly.

[0137] The specific implementation method of S55 is:

[0138] The contour coordinate value information of each yarn and pore in each segmented image is obtained along the X-axis and Y-axis directions respectively, and the S51-S54 operations are repeated. For each segmented image, a corresponding json file is created to store the information in a dictionary form in the python language to build a feature database.

[0139] S6, extracting contour features of each yarn in each segmented image, calculating the coordinate value of the center of the yarn contour, establishing a spatial coordinate system, and performing discrete derivation of the center of the contour of each yarn in each segmented image;

[0140] Further preferably, S6 includes:

[0141] S61. Extract the coordinate value of the center point of each yarn contour in each segmented image from the feature database, and establish a spatial coordinate system in the order of the segmented images.

[0142] S62. Use the first-order intermediate differential law to perform discrete differentiation on the center point of each yarn contour in each segmented image to obtain the tangent direction of the center point, that is, the 1 direction of the yarn.

[0143] S63. Direction 2 can be obtained from direction 1 obtained by calculation.

[0144] The specific implementation method of S61 is:

[0145] Considering that the segmented image is converted into a finite element model and a single yarn is transversely isotropic, a local material orientation allocation algorithm is proposed to consider the influence of fiber orientation on mechanical properties. The contour feature coordinate value information of each yarn is extracted from the feature database along the X-axis and Y-axis directions respectively, and the spatial coordinate system is established in the order of the segmented images. It is assumed that the distance between adjacent segmented images is i *4.9819um, all the yarn profile feature coordinate value information is mapped to the constructed space coordinate system, such as Figure 4 As shown in (a) in .

[0146] The specific implementation method of S62 is:

[0147] Statistical Computing i The center point of the yarn cross section ( ). The warp or weft center points of each cross-sectional image in different directions are statistically analyzed, and then the discrete derivative of each center point is solved to obtain the or direction, such as Figure 4 (b) Figure 4 As shown in (c) in .

[0148] (1)

[0149] (2)

[0150] (3)

[0151] (4)

[0152] (5)

[0153] (6)

[0154] in, i is the segmented image number, v represents the XCT image resolution, T yes Y Axis or Z The pixel length on the axis, yes t Pixels in the i The corresponding gray value on the yarn cross section, and yes t Pixels in the i The corresponding Y Axis and Z Axis coordinates, ( ) corresponds to ( ) on the yarn cross section, h is the corresponding solution step size.

[0155] The specific implementation method of S63 is:

[0156] Since the yarn is transversely anisotropic, the main direction is direction 1, and directions 2 and 3 can meet the requirements as long as they are perpendicular to direction 1, that is, the sum of the products of the corresponding components in each direction is equal to 0. Therefore, the component of direction 2 on the X-axis is defined as 0, and the component of direction 2 can be calculated. At the same time, the calculated discrete derivative value results are stored in the feature database, such as Figure 4 As shown in (d) in .

[0157] (7)

[0158] (8)

[0159] S7, extracting feature database data and constructing regional features of each component;

[0160] Further preferably, S7 includes:

[0161] S71. Extract all yarn contour and pore contour coordinate value data in a single segmented image from a feature database.

[0162] S72. Based on the Python language, call the underlying functional functions of Abaqus 2021 software in sequence.

[0163] S73. Divide and restore the component regions according to the internal microstructure and position characteristics of the material, and set the grid size for grid division.

[0164] The specific implementation method of S7 is:

[0165] Considering that the composite material only contains resin matrix, warp yarn, weft yarn, and pores, a rectangle with fixed length and width is considered as the initial model. The length and width of the rectangle can be obtained by multiplying the resolution by the length and width pixel length of the image respectively. Each region is segmented in the rectangular area, and the pore area is removed. The remaining area will be classified as the resin matrix area. Taking the 330th segmented image as an example, create the yarn contour coordinate value information of the image from the feature data, use the Line function of Abaqus 2021 to draw each yarn area and pore area, then use the PartitionFaceBySketch function to segment the area, and finally use the Cut function to remove the pore area. Set the unit types of the resin matrix and yarn to CPS3 and CPS4R respectively, call the seedPart function of Abaqus 2021 to reasonably set the mesh seed, and the size of the mesh density greatly affects the calculation cost and efficiency of numerical simulation.

[0166] S8, inserting cohesion units and defining material properties for each component region feature, and assigning local material directions;

[0167] Further preferably, S8 includes:

[0168] S81, in the mesh model divided by S7, insert cohesive units at the contact interfaces between yarns and yarns and between yarns and matrix, such as Figure 5 As shown in (a) in .

[0169] S82. The local material direction vector calculated by S6 is gradually distributed to each unit of each yarn.

[0170] S83. Define material properties and mechanical response characteristics for each component region.

[0171] The specific implementation method of S8 is:

[0172] Insert cohesive units without thickness at the contact interface between yarns and yarns and between yarns and matrix. Accurately define material properties, such as density of each component material, material modulus, and Poisson's ratio, and select appropriate material damage constitutive model. Then, Figure 4 As shown, for each yarn in the model, the distance from the centroid of each unit to the discrete points on all segmented images is calculated from the corresponding discrete derivative results in the feature database, and the directional derivative component at the discrete points with the shortest distance is selected as the local material directional property of the unit, as shown in Figure 4 As shown in (c) in .

[0173] (9)

[0174] (10)

[0175] (11)

[0176] (12)

[0177] (13)

[0178] in, Corresponding to the i On the cross section of the yarn, j Unit k Node X axis, Y Axis and Z Axis coordinate values. Corresponding to i On the cross section of the yarn, j Unit centroid X axis, Y Axis and Z Axis coordinate values.

[0179] S9, set boundary conditions, numerical simulation calculation, such as Figure 5 As shown in (b) in .

[0180] The specific implementation method of S9 is:

[0181] Select two reference points, A and B, to be rigidly coupled with the upper movable platform and the lower fixed platform, respectively. Apply displacement conditions U1=0, U2=-1 and UR3=0 to reference point A to simulate the continuous loading process of the sample in the vertical direction. At the same time, set U1=0, U2=0 and UR3=0 on the lower fixed platform to ensure that the sample is subjected to compression effects, such as Figure 5 shown.

[0182] S10, repeating steps S7-S9, can realize the rapid generation of two-dimensional batch finite element models and numerical simulation, such as Figure 6 shown.

[0183] Embodiment 2

[0184] The present invention also provides a system for generating a two-dimensional finite element model of a composite material based on XCT images, comprising: an acquisition module, a modeling module, a learning module, a precise segmentation module, a first database construction module, a second database construction module, a partitioning module, an allocation module, a calculation module and a model generation module;

[0185] The acquisition module is used to acquire XCT images of the composite material of basalt fiber in the initial state and perform preprocessing;

[0186] The modeling module is used to perform parametric statistical analysis on the pre-processed XCT image of the composite material, quantify the microscopic structural parameters of the yarn inside the material and perform two-dimensional parametric random modeling to obtain a parametric random modeling image;

[0187] The learning module is used to learn the parameterized random modeling image based on the adversarial learning style transfer network Pix2PixHD to generate a virtual XCT image;

[0188] The precise segmentation module is used to achieve precise segmentation of each component area of ​​the virtual XCT image based on the Swin-Tranformer network;

[0189] The first database construction module is used to extract contour microscopic features and smooth local contours based on the segmented virtual XCT image, and to construct a yarn and pore feature database;

[0190] The second database construction module is used to extract the contour features of each yarn in each segmented image based on the yarn and pore feature database, calculate the coordinate value of the center of the yarn contour, establish a spatial coordinate system, and perform discrete derivative of the center of the yarn contour in each segmented image based on the spatial coordinate system to construct a feature database;

[0191] The segmentation module is used to extract the feature data information of a single segmented image based on the feature database, classify each component area, and divide the grid;

[0192] The allocation module is used to insert cohesive units based on the divided grid and define the material properties of each component area feature and allocate local material directions;

[0193] The calculation module is used to set boundary conditions and perform numerical simulation calculations based on the assigned local material directions;

[0194] The model generation module is used to repeatedly divide the module-calculation module to realize the rapid generation and numerical simulation of the two-dimensional finite element model based on the XCT digital image of the composite material.

[0195] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A method for generating a two-dimensional finite element model of composite materials based on XCT images, characterized in that: The following steps are involved: S1: Acquire XCT images of the composite material of basalt fiber in the initial state and perform preprocessing; S2: performing parametric statistical analysis on the pre-processed XCT image of the composite material, quantifying the microscopic structural parameters of the yarn inside the material and performing two-dimensional parametric random modeling to obtain a parametric random modeling image; S3: learning the parameterized random modeling image based on the adversarial learning style transfer network Pix2PixHD to generate a virtual XCT image; S4: Accurate segmentation of each component region of virtual XCT images based on Swin-Tranformer network; S5: Based on the segmented virtual XCT image, the microscopic features of the contour are extracted and the local contour is smoothed to build a yarn and pore feature database; In S5, based on the segmented virtual XCT image, the method of extracting contour micro features and smoothing the local contour includes: S51: Selecting a resolution that meets preset requirements to downsample the segmented virtual XCT image; S52: performing feature screening on the virtual XCT image after downsampling; S53: extracting image contour features of pores and yarns based on the virtual XCT image after feature screening, and converting coordinate values ​​according to the positional relationship between pixels and nodes to obtain contour coordinate values; S54: setting a convolution kernel that meets preset requirements, and using the operation between the convolution kernel and the contour coordinate value to achieve contour smoothing; S55: Process all segmented images according to S51-S54, build a yarn and pore feature database, and store the contour coordinate values ​​of each component after the coordinate system transformation; S6: Based on the yarn and pore feature database, extract the contour feature of each yarn in each segmented image, calculate the coordinate value of the center of the yarn contour, establish a spatial coordinate system, and based on the spatial coordinate system, perform discrete derivation on the center of the yarn contour in each segmented image to construct a feature database; In S6, based on the yarn and pore feature database, the contour feature of each yarn in each segmented image is extracted, the coordinate value of the center of the yarn contour is calculated, a spatial coordinate system is established, and based on the spatial coordinate system, the method for performing discrete derivation of the center of the contour of each yarn in each segmented image includes: S61: extracting the coordinate value of the center point of each yarn contour in each segmented image from the yarn and pore feature database, and establishing a spatial coordinate system in the order of the segmented images; The contour feature coordinate value information of each yarn is extracted from the feature database along the X-axis and Y-axis directions respectively, and the spatial coordinate system is established in the order of the segmented images. It is assumed that the distance between adjacent segmented images is i *4.9819um, mapping all the yarn profile feature coordinate value information to the constructed space coordinate system; S62: Based on the spatial coordinate system, use the first-order intermediate differential law to perform discrete derivative on the center point of each yarn contour in each segmented image to obtain the tangent direction of the center point, that is, the 1 direction of the yarn; S63: Obtain direction 2 from direction 1 obtained by calculation; Since the yarn is transversely anisotropic, the main direction is direction 1, and directions 2 and 3 meet the requirement as long as they are perpendicular to direction 1, that is, the sum of the products of the corresponding components in each direction is equal to 0; S7: Based on the feature database, extract feature data information of a single segmented image, classify each component area, and divide it into grids; S8: Based on the divided grid, insert cohesive force units and define material properties for each component area feature, and assign local material directions; S9: Setting boundary conditions and numerical simulation based on the assigned local material direction; S10: Repeat S7-S9 to realize the rapid generation and numerical simulation of a two-dimensional finite element model based on XCT digital images of composite materials.

2. The method for generating a two-dimensional finite element model of composite materials based on XCT images according to claim 1, characterized in that: In S1, the method of collecting XCT images of composite materials of basalt fibers in an initial state and performing preprocessing includes: S11: Acquire complete XCT scan source image data, and perform cropping and filtering on the XCT scan source image data; S12: Based on the cropped and filtered XCT scan source image data, a quantitative number of segmented images are extracted at equal intervals, and each component in each segmented image is manually labeled; S13: Perform feature extraction on the manually annotated segmented image to obtain a true label set.

3. The method for generating a two-dimensional finite element model of composite materials based on XCT images according to claim 2, characterized in that: In S2, the method of performing parametric statistical analysis on the pre-processed XCT image of the composite material, quantifying the microscopic structural parameters of the yarn inside the material and performing two-dimensional parametric random modeling includes: S21: Extract the mesoscopic structural parameters of each yarn in each segmented image based on the manually annotated mesoscopic structural features of the dataset; S22: Based on the mesoscopic structural parameters of each yarn, the mesoscopic structural characteristics are qualitatively and quantitatively expressed, and statistical methods are used for parametric statistical analysis; S23: Based on the results of the parametric statistical analysis, the data set is expanded by parametric random modeling to generate a virtual data set, namely a parametric random modeling image.

4. The method for generating a two-dimensional finite element model of composite materials based on XCT images according to claim 3, characterized in that: In S3, the method of learning the parameterized random modeling image based on the adversarial learning style transfer network Pix2PixHD to generate a virtual XCT image includes: S31: generating a warp yarn random model based on warp yarn parameterized statistical analysis; S32: Generate a random model of weft yarns according to the parameterized statistical analysis of weft yarns; S33: generating an artificial synthetic label image based on the warp yarn random model and the weft yarn random model; S34: Based on the artificially synthesized label image, an improved Pix2PixHD network is applied to generate a basalt fiber composite image with the same microstructure as the real one, and a hybrid data set, i.e., a virtual XCT image, is constructed.

5. The method for generating a two-dimensional finite element model of composite materials based on XCT images according to claim 4, characterized in that: In S4, the method for realizing accurate segmentation of each component region of a virtual XCT image based on a Swin-Tranformer network includes: S41: Input the mixed data set to the Swin-Tranformer network to obtain training images; S42: Divide the training image into different small blocks by pixels through a block division module, and then convert it into a one-dimensional feature vector image; S43: Accurately segment each component region in the one-dimensional feature vector image to obtain a semantic segmentation result image; S44: Based on the semantic segmentation result map, a watershed segmentation algorithm is used to complete region segmentation for each yarn.

6. The method for generating a two-dimensional finite element model of composite materials based on XCT images according to claim 1, characterized in that: In S7, based on the feature database, the feature data information of a single segmented image is extracted, each component region is classified, and the method of dividing the grid includes: S71: extracting all the yarn contour and pore contour coordinate value data of a single segmented image from the feature database; S72: Call the underlying functions of Abaqus 2021 software in sequence based on the Python language; S73: Based on the coordinate value data of all yarn profiles and pore profiles and the underlying functional function, the component regions are divided and restored according to the internal microstructure position characteristics of the material, and the grid size is set for grid division.

7. The method for generating a two-dimensional finite element model of composite materials based on XCT images according to claim 6, characterized in that: In S8, based on the divided grid, the method of inserting the cohesion unit and defining the material properties of each component region feature and allocating the local material direction includes: S81: inserting cohesive force units at the contact interfaces between yarns and yarns and between yarns and matrix from the divided grid; S82: gradually allocating the calculated local material direction vector to each unit of each yarn; S83: Define material properties and mechanical response characteristics for each component region.

Images