A finite element modeling method for fabric composite materials based on micro-geometry model

Abstract

The present invention relates to the field of woven composite material modeling, specifically a finite element modeling method for fabric composite materials based on a micro-geometry model, the method comprising establishing a representative volume unit micro-geometry model of composite material woven fabric, outputting a surface triangular patch model after calculation; periodically mapping the triangular patch model to a composite material domain, calculating the surface intersections of the triangular patch in the grid line domain; repairing the penetration and narrow gaps between different yarns along the grid line direction; adjusting the grid node coordinates to match the adjacent surface intersections and setting grid node labels; judging the surface splitting mode according to the four nodes on the surface of the hexahedral unit; splitting the hexahedral unit again into tetrahedron, pyramid and triangular prism units according to the surface splitting mode, setting a corresponding label set for the unit according to the label type of the unit, and saving the material direction corresponding to the unit at the same time. The present invention can establish a finite element model of a high-true woven composite material, and has a very broad application prospect.

Description

Technical Field

[0001] The invention relates to the field of woven composite material modeling, in particular to a finite element modeling method for fabric composite materials based on a micro-geometric model. Background Art

[0002] As a new type of high-performance composite material, woven composite materials have the characteristics of high specific modulus, high specific strength and good processability, and have been widely used in aerospace, military and other fields. The woven fabric reinforcement in woven composite materials is woven by warp yarn, weft yarn and binding yarn according to a specific topological structure. It has obvious multi-scale characteristics, and its microstructure and mesostructure directly affect the macroscopic properties of the composite material.

[0003] The mesoscopic unit cell of a woven composite material is composed of yarn and matrix, where the yarn is composed of fiber bundles and matrix infiltrated between the fibers. However, due to the mutual extrusion and deformation between the fiber bundles during the weaving process, the mesoscopic yarns of woven composite materials have complex geometric shapes. In previous modeling research work, certain ideal assumptions were made on the cross-section and path of the mesoscopic yarns, but the cross-section of the real mesoscopic yarn changes dynamically with the path, and the ideal assumptions often have a certain gap with the real structure. Therefore, in order to more realistically reflect the mesoscopic yarn structure of woven composite materials, it is necessary to fully consider the deformation of the fabric during modeling.

[0004] For the microscopic modeling of woven composite materials, the usual method is to make certain cross-section and path assumptions about the yarn based on the scanning results of scanning electron microscope, optical microscope and Micro-CT, so as to create an ideal microscopic structure model of the woven composite material in a parametric way. However, the ideal geometric modeling method is mostly used for woven composite materials with low fiber volume content. When the fiber volume content is high, the mutual extrusion and deformation between the yarns will cause the yarns to have different corrugations and the yarn cross-section will have certain asymmetry and torsion. The regular yarn morphology and trajectory assumed by the ideal geometric model are quite different from the real model, and even the yarn geometric model will have certain interference, which cannot meet the needs of mechanical property analysis.

[0005] Finite element-based numerical calculation method is an important and effective method for studying composite materials. This method requires the establishment of a finite element model of the material. Whether the model can effectively reflect the internal structure of the material directly affects the accuracy of subsequent prediction of various mechanical properties of the material; Chinese patent CN113987882 A provides a digital modeling method for the microscopic yarn structure of a woven composite material. This method realizes the geometric modeling of the composite reinforcement, but this method cannot realize the composite matrix modeling and overall meshing.

[0006] At present, the traditional meshing method is only applicable to the meshing of idealized (yarn cross-section is elliptical, track-shaped, lens-shaped, etc.) woven composite microstructures, and only meshes each yarn of the reinforcement and the matrix separately, and then combines the reinforcement and matrix meshes into a composite mesh, resulting in certain interference between the meshes at the yarn-yarn and yarn-matrix interfaces, which affects the subsequent finite element simulation results. Summary of the invention

[0007] Based on the problems existing in the prior art, the present invention aims to provide a finite element modeling method for fabric composite materials based on a micro-geometric model. The present invention can not only establish an accurate geometric model based on the micro-geometric model of woven fabrics, but also consider the interface relationship between the components of the woven composite material, eliminate the interference between the yarns and the narrow gaps, and generate an accurate finite element model of the woven composite material. Based on the micro-scale woven fabric unit cell model, a high-fidelity finite element model of the composite material is quickly established, the accuracy of finite element mechanical property prediction is improved, and it is suitable for the development of two-dimensional and three-dimensional woven composite materials of various geometric structures, which greatly improves the development efficiency of woven fabric composite materials and reduces the development cost, and has great application prospects.

[0008] In order to achieve the above object, the present invention provides the following technical solutions:

[0009] A finite element modeling method for a fabric composite material based on a micro-geometric model comprises the following steps:

[0010] (1) Establish a representative volume unit micro-geometric model of woven fabrics;

[0011] (2) Calculate the yarn axis direction based on the representative volume unit micro-geometry model and output the surface triangle patch model of the woven fabric yarn;

[0012] (3) Establishing a composite material domain, adding matrix materials and discretizing the composite material domain into hexahedral units;

[0013] (4) mapping the surface triangular patch model of the yarn in step (2) to the composite material domain, calculating the intersection points of the triangular patch and the grid lines in the composite material domain, and saving the topological relationship between the grid line surface intersection points;

[0014] (5) Repair the inter-yarn penetration and narrow gaps caused by scale conversion between different yarns along the grid line direction;

[0015] (6) Adjust the mesh nodes of the composite material domain to the intersection points of adjacent mesh lines and surfaces, and set the topological labels of the mesh nodes;

[0016] (7) Traversing the surface of the hexahedral unit, judging the splitting mode of the corresponding six surfaces according to the four mesh nodes of each surface, and determining the splitting line;

[0017] (8) According to the splitting pattern of the six surfaces in the hexahedral unit, the hexahedral unit is split into tetrahedral, pyramidal and triangular prism units again, and the corresponding component unit sets are stored according to the labels of each unit, and the material direction corresponding to each unit is obtained.

[0018] Compared with the prior art, the present invention has the following advantages:

[0019] The present invention can develop woven composite products in a completely digital manner without the need for real fabrics. The product development cycle is greatly reduced. At the same time, the fidelity of the finite element model reconstructed based on the micro-geometric structure can be effectively guaranteed, and the penetration and narrow gaps between different yarns can be identified and repaired based on the overall grid division, and the yarn-yarn and yarn-matrix interfaces can be matched to make all components mesh perfectly together. The present invention can improve the accuracy of the prediction of the mechanical properties of complex woven composite materials, and well guide the development and verification of woven composite products. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A flow chart of a finite element modeling method for a fabric composite material based on a micro-geometric model according to an embodiment of the present invention;

[0021] Figure 2 A schematic diagram of a representative volume unit topological structure of a three-dimensional orthogonal fabric according to an embodiment of the present invention;

[0022] Figure 3 A representative volume unit micro-geometric model of a three-dimensional orthogonal fabric according to an embodiment of the present invention;

[0023] Figure 4 A schematic diagram of fiber nodes of a weft yarn cross-section slice of a three-dimensional orthogonal fabric according to an embodiment of the present invention;

[0024] Figure 5 A schematic diagram of boundary nodes of a weft yarn cross section of a three-dimensional orthogonal fabric according to an embodiment of the present invention;

[0025] Figure 6 A three-dimensional orthogonal fabric surface triangular patch model according to an embodiment of the present invention;

[0026] Figure 7 A schematic diagram of mapping a triangular facet to a composite material domain according to an embodiment of the present invention;

[0027] Figure 8 A schematic diagram of grid line surface intersection calculation according to an embodiment of the present invention;

[0028] Fig. 9 A schematic diagram of inter-yarn penetration and narrow gap adjustment according to an embodiment of the present invention;

[0029] Fig.10A schematic diagram of adjusting a grid node according to an embodiment of the present invention;

[0030] Fig.11 A schematic diagram of a hexahedral unit surface splitting mode according to an embodiment of the present invention;

[0031] Fig.12 A schematic diagram of the decomposition of a hexahedral unit according to an embodiment of the present invention;

[0032] Fig.13 This is a finite element mesh model of a three-dimensional orthogonal fabric composite material according to an embodiment of the present invention. DETAILED DESCRIPTION

[0033] 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.

[0034] Among them, the drawings are only used for illustrative explanations, and they only represent schematic diagrams rather than actual pictures, and should not be understood as limitations on the present invention. In order to better illustrate the embodiments of the present invention, some parts of the drawings may be omitted, enlarged or reduced, and do not represent the size of actual products. For those skilled in the art, it is understandable that some well-known structures and their descriptions in the drawings may be omitted.

[0035] Figure 1 FIG. 1 is a flow chart of a finite element modeling method for a fabric composite material based on a micro-geometric model according to an embodiment of the present invention; Figure 1 As shown, the finite element modeling method of a fabric composite material based on a micro-geometric model of the present invention comprises the following steps:

[0036] (1) Establish a representative volume unit micro-geometric model of woven fabrics, as follows:

[0037] (1.1) Define the parameters of each component yarn of the woven fabric, the size of the representative volume unit and the key nodes of its organization, and establish the topological structure of the representative volume unit of the woven fabric;

[0038] In some embodiments of the present invention, the step (1.1) may also specifically include: setting parameters such as the initial cross-sectional area and modulus of the yarn, setting the warp (length direction) and weft (width direction) dimensions of the fabric, defining the coordinates of key points of the woven fabric and the initial path of the yarn, and establishing a representative volume unit topological structure of the fabric with a circular yarn cross-sectional shape and a folded line yarn path, such as Figure 2 As shown, the three-dimensional orthogonal fabric topology structure includes three types of yarns: warp yarns, weft yarns and binding yarns.

[0039] (1.2) The boundary conditions of the topological structure unit are set and the yarn tension is applied. The yarn is discretized into micro-geometric units through yarn fiber discretization and fiber digital unit discretization. The discretized yarn is numerically calculated to obtain the micro-geometric unit cell model.

[0040] In some embodiments of the present invention, the step (1.2) may also specifically include: setting periodic boundary conditions in the warp and weft directions, and dividing the warp yarn, weft yarn and binding yarn into 64, 64 and 16 virtual fibers respectively while ensuring that the cross-sectional area of ​​the yarn remains unchanged, and discretizing the virtual fibers into digital units (rod units and fiber nodes), and the length of the digital unit is 0.5 times the virtual fiber diameter. Applying constant tension at both ends of the yarn causes relative movement between the fibers, realizing changes in the yarn path and cross-sectional shape, until it approaches the real fabric structure. Figure 3 Shown is a representative volume unit micro-geometry model of a three-dimensional orthogonal fabric, and the final size of the representative volume unit is 13.64*5.976*5.9753mm (length*width*thickness).

[0041] (2) Calculate the yarn axis direction based on the representative volume unit micro-geometry model and output the triangular patch model of the woven fabric yarn surface, as follows:

[0042] (2.1) Obtain the three-dimensional coordinate data of the fiber nodes of the model in step (1) and record the corresponding topological relationship;

[0043] It can be understood that, in the embodiment of the present invention, virtual fibers belonging to the same yarn have the same number of fiber nodes, and the fiber nodes of the same virtual fiber are arranged in sequence according to the direction of the yarn path. According to the above relationship, the topological relationship of the fiber nodes can be constructed.

[0044] (2.2) Calculate the yarn mid-axis path nodes through the node three-dimensional coordinates, slice the yarn cross section evenly along the path direction of each yarn, and obtain the interpolation point of each fiber on the yarn cross section;

[0045] In some embodiments of the present invention, the step (2.2) may further specifically include: calculating the mean of the fiber nodes of the same yarn and the same serial number as the yarn path node, and sequentially connecting the yarn path nodes as the yarn path direction. Calculating the normal vector of the cutting plane along the yarn path direction, the distance between two adjacent cutting planes is about 1-5 times the length of the virtual fiber rod unit, and obtaining the interpolation point of the virtual fiber rod unit on the cutting plane, such as Figure 4 Shown is a schematic diagram of the interpolation points of the weft yarn along the yarn path direction.

[0046] (2.3) Adopting the Delaunay triangulation algorithm and the Alpha-shape algorithm to adaptively obtain the boundary of the yarn cross-section interpolation point in step (2.2), and obtaining the polygonal boundary contour of the yarn cross-section;

[0047] In some embodiments of the present invention, the step (2.3) may further specifically include: traversing the interpolation points of the cutting plane in step (2.2), taking the interpolation point as the center of the circle, adding an average of 8-20 interpolation points on the cutting plane with the corresponding virtual fiber radius as the radius, and using the Delaunay triangulation algorithm to calculate the Delaunay triangulation of all interpolation points on the same cutting plane. Delete any triangle with an edge greater than Alpha in the Delaunay triangulation to obtain the Alpha-shape of the triangulation, where Alpha is 5-10 times the virtual fiber diameter. Extract the boundary of the remaining Delaunay triangulation, which is the polygonal boundary contour of the yarn cross section, and perform homogenization interpolation along the contour to obtain the cross-section boundary nodes, such as Figure 5 Schematic diagram of the boundary nodes of a weft yarn section.

[0048] It should be noted that in order to ensure the correct generation of subsequent triangular patches, it is necessary to ensure that the number of boundary nodes of each cross section of the same yarn is the same.

[0049] (2.4) Perform homogenization interpolation on the cross-sectional polygons in step (2.3) to obtain the vertices of the triangular patch, and then generate the surface triangular patch model of the yarn in a specific order.

[0050] In some embodiments, Figure 6 As shown, the boundary nodes on two adjacent section planes are connected according to a specific boundary node sequence to obtain a surface triangle patch model of the yarn.

[0051] (3) Establishing a composite material domain, adding matrix materials and discretizing the composite material domain into hexahedral units, which may include the following:

[0052] (3.1) Read the size of the representative volume unit of the fabric and set the min and max key nodes of the composite material domain;

[0053] In some embodiments of the present invention, the step (3.1) may further specifically include: assuming that the midpoint of the composite material domain is P cmid (x cmid ,y cmid ,z cmid ), the midpoint of all fiber nodes in the representative volume unit microgeometric model of woven fabric is P tmid (x tmid ,y tmid ,z tmid ), then x cmid =x tmid +xoff ,y cmid =y tmid +y off , z cmid =z tmid , where x off and off The offset distance of the composite material domain midpoint along the x and y directions (longitude and latitude directions) (x direction is ±1 / 2*length, y direction is ±1 / 2*width). Define the minimum point P of the composite material domain cmin (x cmin ,y cmin ,z cmin ) and the maximum point P cmax (x cmax ,y cmax ,z cmax ), there is x cmin =x cmid -1 / 2*length,y cmin =y cmid -1 / 2*width,z cmin =z cmid -1 / 2*thickness*(1+z inc ), x cmax =x cmid +1 / 2*length,y cmax =y cmid +1 / 2*width,z cmax =z cmid +1 / 2*thickness*(1+z inc ), where z inc Increase the thickness percentage in the z direction (depending on the different composite material structures z inc is greater than or equal to 0). cmin and P cmax The rectangular area enclosed by the two points along the three positive directions of x, y and z is the composite material domain containing the woven fabric reinforcement and the matrix material.

[0054] (3.2) According to the fiber size and cross-sectional parameters of the fabric, set the appropriate hexahedral unit size, and evenly discretize the material domain in step (3.1) into hexahedral units of corresponding size.

[0055] In some embodiments of the present invention, the step (3.2) may further specifically include: setting the x, y and z dimensions of the hexahedral unit. The dimensions may be set according to the average diameter of all virtual fibers of the fabric. Too small a unit size may lead to a sharp increase in the number of meshes, increasing the consumption of computing resources. Too large a unit size may lead to the loss of some details of the fabric, affecting the subsequent simulation accuracy. The unit size is generally set to 1-5 times the average diameter of the virtual fibers. cmin and P cmax The composite material domain surrounded by the two points is uniformly discretized into hexahedral units, such as Figure 7 The rectangular area shown is the composite material domain after migration.

[0056] (4) Map the yarn surface triangular patch model in step (2) to the composite material domain, calculate the intersection points of the triangular patch and the grid lines in the composite material domain, and save the topological relationship of the grid line surface intersection points, as follows:

[0057] (4.1) Map the triangular patch of the yarn surface to the composite material domain, calculate the surface intersections of the triangular patch and the grid lines in the x, y and z directions, and record the topological information of the intersections;

[0058] In some embodiments of the present invention, the step (4.1) may further specifically include: Figure 7 As shown, the triangular facets on the yarn surface are mapped to the composite material domain. Since the composite material domain may be offset along the warp and weft directions in step (3), some of the triangular facets shown are not in the composite material domain. The triangular surface that is not in the composite material domain is periodically mapped to the other side of the representative volume unit, and the surface intersections of the triangular facets and the grid lines in the x, y and z directions are calculated, and the topological information of the intersections is recorded as follows: Figure 8 As shown in the schematic diagram of intersection calculation, triangle abc is a triangular patch on the yarn surface, and triangles a'b'c', a"b"c" and a"'b"'c"' are the vertical projections of triangle abc on the xoy, xoz and zoy planes respectively. Taking the xoy plane as an example, first determine whether there is a grid node inside triangle a'b'c' on the xoy plane. If there is a grid node, calculate the corresponding node coordinates and reversely project them onto the triangular patch abc to calculate the projection point coordinates; if not, no calculation is performed, and the projection point is the surface intersection of the yarn and the grid line. Determine the vertical projections of triangle abc on the xoz and zoy planes in turn. Points 1, 2, and 3 are the intersections of triangle abc on the yarn surface with the grid lines in the x, y, and z directions.

[0059] (4.2) The grid line surface intersections are periodically mapped on the six surface boundaries of the composite material domain so that the grid line surface intersections satisfy the periodic boundary conditions of the representative volume unit.

[0060] It can be understood that by periodically mapping the triangular patch mesh to the grid line surface intersections on the six surface boundaries of the review material domain, the topological relationship of the grid line surface intersections can be reflected, so that the grid line surface intersections meet the periodic boundary conditions of the representative volume unit.

[0061] (5) Repair the inter-yarn penetration and narrow gaps caused by scale conversion between different yarns along the grid line direction, as follows:

[0062] (5.1) Extract the intersection points of the yarn on the grid line surface in the x, y and z directions respectively and sort them according to their x, y or z coordinate values; Fig. 9 As shown in (a)-(b), there are 2-4 unequal grid line surface intersections on the vertical grid lines;

[0063] (5.2) According to the x, y or z values ​​of the grid line intersection points and their topological relationships, the intersection points of two mutually penetrating different yarn surfaces are adjusted to the midpoint of the two points, and the topological labels of the two nodes are swapped. The intersection points of two different yarn surfaces whose distance is less than the set threshold are adjusted to the midpoint of the two points.

[0064] In some embodiments of the present invention, the step (5.2) may further specifically include: according to the x, y or z values ​​of the grid line surface intersections and their topological relationships, such as Fig. 9 In (a), Yarn1 and Yarn2 are two interlaced yarns, where i1 and i2 are the intersection points of the grid lines of Yarn1 and Yarn2, respectively. Infiltration occurs on the surface of Yarn1 and Yarn2. Similarly, Fig. 9 In (b), a narrow gap appears between Yarn 1 and Yarn 2 (the threshold for judging a narrow gap is less than 1 / 3 of the grid size). Adjust the intersection of the two grid lines where penetration or narrow gaps appear to the midpoint of the two points, such as Fig. 9 (a)-(b) The hollow nodes are the adjusted surface intersection points, and the dotted line is the material boundary between the two yarns.

[0065] (6) Adjust the mesh nodes of the composite material domain to the intersection points of adjacent mesh lines and surfaces, and set the topological labels of the mesh nodes as follows:

[0066] (6.1) Read the adjusted grid line surface intersection in step (5), determine the grid line surface intersection adjacent to the grid node and store the grid line surface intersection in the corresponding grid node adjacency linked list, such as Fig.10 As shown in (a), taking the plane as an example, there are three surface intersection points S1, S2 and S3 in the adjacency list of grid node N1.

[0067] (6.2) Traverse all grid nodes. If the grid node adjacency list is zero, the grid node does not need to be adjusted, and the grid node label is set to the corresponding single component label. Otherwise, the grid node coordinates are adjusted to the midpoint of all adjacent surface intersections, and the grid node label is set to the corresponding multi-component label.

[0068] In some embodiments of the present invention, the step (6.2) may further specifically include: traversing all grid nodes, such as Fig.10 In (a), N1 has three adjacent grid surface intersections. Adjust N1 to the midpoint of S1, S2, and S3, and set the label of N1 to a multi-component label; Fig.10 In (a), N2 has only one adjacent grid surface intersection point S4. Adjust the N2 node to S4 and set the N2 label to a multi-component label. Fig.10 In (a), the adjacency list of grid node N3 is zero, so there is no need to adjust the grid node, and the label of grid node N3 is set to the corresponding single component label. Fig.10 (b) is a schematic diagram of the adjusted grid nodes, where the unadjusted grid nodes are all single component labels.

[0069] (7) Traverse the hexahedral unit surface, determine the splitting mode of the corresponding six surfaces according to the four mesh nodes of each surface, and determine the splitting line, as follows:

[0070] Read the four vertex labels corresponding to the quadrilateral, and determine whether the quadrilateral needs to be divided into two triangles based on the number and type of labels. If division is required, add the adjacent mesh nodes of the mesh node, such as Fig. 9 Six surface splitting modes are shown, with m mesh nodes belonging to two or more materials and s mesh nodes belonging to a single material. Fig.11 In (a), there are 4 mesh nodes s mesh nodes, and no division lines need to be added to this surface; Fig.11 In (b)-(c), one node or two nodes on the same side are m-nodes, so the m-node and the s-node must contain the same label, and no dividing line needs to be added to the surface; Fig.11 If (d)-(f) contains 2 or more m-nodes and all the m-nodes contain the labels of s-nodes, then the surface is of one material and no dividing lines need to be added. Otherwise, dividing lines need to be added at the two m-nodes.

[0071] It should be noted that the regular quadrilateral and the order of m nodes and s nodes shown in Figure (11) are all schematic diagrams. A non-regular quadrilateral surface as shown in Figure (10) may appear. Different node distribution orders can be obtained by rotating the corresponding node distribution order in Figure (11).

[0072] (8) According to the splitting mode of the six surfaces in the hexahedral unit, the hexahedral unit is further split into tetrahedral, pyramidal and triangular prism units. The topological labels of the grid intersections of each unit are stored as corresponding component unit sets, and the material direction corresponding to each unit is obtained, as follows:

[0073] (8.1) Read the neighboring points of the hexahedral unit nodes, divide the hexahedral unit into one component, two components, or three components according to the node labels, and then split each component into tetrahedral, pyramidal, or triangular prism units. If necessary, add nodes inside the hexahedral unit for splitting. Fig.12 As shown in (a), there are two dividing lines between the two faces of the hexahedral unit, and the unit is split into two triangular prism units; Fig.12 As shown in (b), the hexahedral unit belongs to two or three components, and the unit is split into a tetrahedral unit, a pyramid unit and a triangular prism unit. Fig.13 (a)-(c) are the composite finite element mesh, composite reinforcement woven fabric mesh and matrix mesh, respectively.

[0074] (8.2) Determine whether each unit in the unit set corresponds to an adjacent yarn path node in step (2), and set the unit material direction to the path direction of the node.

[0075] In summary, the finite element modeling method of fabric composite materials provided by the present invention realizes full digital finite element modeling without the need for idealized assumptions and accurate real models; can complete the accurate conversion of fabric models from microscopic to mesoscopic scales; realizes inter-yarn penetration and narrow gap repair and grid node adjustment and attribute allocation; perfect meshing between yarn-yarn and yarn-matrix interfaces; obtains a composite material grid model with material property information, providing a basis for defining the isotropic properties of the material; at the same time, the grid model satisfies the periodic boundaries of representative volume units; greatly improves the development efficiency and simulation accuracy of fabric composite materials, and reduces the development cycle of fabric composite materials.

[0076] A person skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the relevant hardware through a program, and the program can be stored in a computer-readable storage medium, which can include: ROM, RAM, disk or CD, etc.

[0077] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A finite element modeling method for fabric composite materials based on a micro-geometric model, comprising the following steps: (1) Establish a representative volume unit micro-geometric model of woven fabrics; (2) Calculate the yarn axis direction based on the representative volume unit micro-geometry model and output the surface triangle patch model of the woven fabric yarn; (3) Establishing a composite material domain, adding matrix materials and discretizing the composite material domain into hexahedral units; (4) mapping the surface triangular patch model of the yarn in step (2) to the composite material domain, calculating the intersection points of the triangular patch and the grid lines in the composite material domain, and saving the topological relationship between the grid line surface intersection points; (5) Repair the inter-yarn penetration and narrow gaps caused by scale conversion between different yarns along the grid line direction; (6) Adjust the mesh nodes of the composite material domain to the intersection points of adjacent mesh lines and surfaces, and set the topological labels of the mesh nodes; (7) Traversing the surface of the hexahedral unit, judging the splitting mode of the corresponding six surfaces according to the four mesh nodes of each surface, and determining the splitting line; (8) According to the splitting pattern of the six surfaces in the hexahedral unit, the hexahedral unit is further split into tetrahedral, pyramidal and triangular prism units, which are stored as corresponding component unit sets according to the labels of each unit, and the material direction corresponding to each unit is obtained.

2. The finite element modeling method for fabric composite materials based on a micro-geometric model according to claim 1, characterized in that: The step (1) is specifically as follows: (1.1) Define the parameters of each component yarn of the woven fabric, the size of the representative volume unit and the key nodes of its organization, and establish the topological structure of the representative volume unit of the woven fabric; (1.2) Topological boundary conditions are set and yarn tension is applied. The yarn is discretized into micro-geometric units through yarn fiber discretization and fiber digital unit discretization. The discretized yarn is numerically calculated to obtain a micro-geometric unit cell model.

3. The finite element modeling method for fabric composite materials based on a micro-geometric model according to claim 1, characterized in that: The step (2) is specifically as follows: (2.1) Obtain the three-dimensional coordinate data of the fiber nodes of the model in step (1) and record their corresponding topological relationships; (2.2) Calculate the yarn mid-axis path nodes through the node three-dimensional coordinates, slice the yarn cross section evenly along the path direction of each yarn, and obtain the interpolation point of each fiber on the yarn cross section; (2.3) Adopting the Delaunay triangulation algorithm and the Alpha-shape algorithm to adaptively obtain the boundary of the yarn cross-section interpolation point in step (2.2), and obtaining the polygonal boundary contour of the yarn cross-section; (2.4) Perform homogenization interpolation on the cross-sectional polygons in step (2.3) to obtain the vertices of the triangular patch, and then generate the surface triangular patch model of the yarn in a specific order.

4. The finite element modeling method for fabric composite materials based on a micro-geometric model according to claim 1, characterized in that: The step (3) is specifically as follows: (3.1) Read the size of the representative volume unit of the fabric and set the min and max key nodes of the composite material domain; (3.2) According to the fiber size and cross-sectional parameters of the fabric, set the appropriate hexahedral unit size, and evenly discretize the material domain in step (3.1) into hexahedral units of corresponding size.

5. The finite element modeling method for fabric composite materials based on a micro-geometric model according to claim 1, characterized in that: The step (4) is specifically as follows: (4.1) Map the triangular patch of the yarn surface to the composite material domain, calculate the surface intersections of the triangular patch and the grid lines in the x, y and z directions, and record the topological information of the intersections; (4.2) The grid line surface intersections are periodically mapped on the six surface boundaries of the composite material domain so that the grid line surface intersections satisfy the periodic boundary conditions of the representative volume unit.

6. The finite element modeling method for fabric composite materials based on a micro-geometric model according to claim 1, characterized in that: The step (5) is specifically as follows: (5.1) extracting the intersection points of the yarn on the grid line surface in the x, y and z directions respectively and sorting them according to their x, y or z coordinate values; (5.2) According to the x, y or z values ​​of the grid line intersection points and their topological relationships, the intersection points of two mutually penetrating yarn surfaces are adjusted to the midpoint of the two points, and the topological labels of the two nodes are swapped; the intersection points of two different yarn surfaces whose distance is less than the set threshold are adjusted to the midpoint of the two points.

7. The finite element modeling method for fabric composite materials based on a micro-geometric model according to claim 1, characterized in that: The step (6) is specifically as follows: (6.1) reading the grid line surface intersection point adjusted in step (5), determining the adjacent grid nodes of the grid line surface intersection point, and storing the grid line surface intersection point in the corresponding grid node adjacency linked list; (6.2) Traverse all grid nodes. If the grid node adjacency list is zero, the grid node does not need to be adjusted, and the grid node label is set to the corresponding single component label; otherwise, the grid node coordinates are adjusted to the midpoint of all adjacent surface intersections, and the grid node label is set to the corresponding multi-component label.

8. The finite element modeling method for fabric composite materials based on a micro-geometric model according to claim 1, characterized in that: The step (7) is specifically as follows: Read the four grid node labels corresponding to the quadrilateral on the surface of the hexahedral unit, and determine whether the quadrilateral needs to be divided into two triangles based on the number and type of labels. If division is required, add the adjacent nodes of the node.

9. The finite element modeling method for fabric composite materials based on a micro-geometric model according to claim 1, characterized in that: The step (8) is specifically as follows: (8.1) Read the neighboring points of the hexahedral unit nodes, divide the hexahedral unit into one component, two components or three components according to the node labels, and split the corresponding components into tetrahedral, pyramidal or triangular prism units; (8.2) Determine whether each unit in the unit set corresponds to an adjacent yarn path node in step (2), and set the unit material direction to the path direction of the node.

10. The finite element modeling method for fabric composite materials based on a micro-geometric model according to claim 9, characterized in that: In the step (8.1), the method of splitting the components includes adding nodes inside the hexahedral unit for splitting.

Images