A prediction method for the mechanical properties of a woven composite material structure

Abstract

The present invention relates to a method for predicting the mechanical properties of a woven composite material structure, including obtaining a microscopic CT image of the woven composite material structure; performing image preprocessing on the CT image; segmenting fiber bundles in the preprocessed image to obtain segmented fiber bundle images; extracting the centerlines of the fiber bundle orientations from the segmented fiber bundle images; establishing a macroscopic geometric model of the woven composite material structure and performing element mesh division on the geometric model; establishing a connection between the centerlines of the fiber bundle orientations and the elements of the geometric model; calculating the mechanical property parameters of each element one by one; and calculating the macroscopic mechanical response of the composite material structure. The method for predicting the mechanical properties of the woven composite material structure of the present invention uses mesh division independent of microscopic structural features, can improve the modeling efficiency, reduce the modeling difficulty, and is easy to implement.

Description

Technical Field

[0001] The present invention belongs to the technical field of prediction methods for the mechanical properties of composite material structures, and particularly relates to a method for predicting the mechanical properties of woven composite materials. Background Art

[0002] In the engineering field, understanding the mechanical properties of materials is of great significance for the design and application of materials. Woven composite materials are a type of composite material composed of fibers and a matrix, where the fibers are intertwined in a woven manner and then cured through the matrix material. Woven composite materials have the advantages of high strength, high stiffness, low specific gravity, corrosion resistance, etc., and are widely used in the fields of aerospace, automotive, shipbuilding, construction, sports goods, etc. Accurately predicting the mechanical properties of woven composite materials is crucial for the performance evaluation and structural design of materials. Woven composite materials are typical anisotropic materials, and the mesoscopic structure of woven composite materials is complex, with the existence of irregular and non-uniform pores, as well as problems such as the interlacing, extrusion, and interference between fiber bundles, which will all affect the mechanical properties of woven composite materials. The accurate prediction of the properties of woven composite materials is complex and difficult.

[0003] Existing methods for predicting the mechanical properties of woven composite materials generally rely on the equivalent averaging method, and obtain the overall mechanical properties of the material by studying the mechanical properties of the Representative Volume Element (RVE for short). However, the above methods generally obtain the mechanical properties of woven composite materials based on the periodicity of the mesoscopic structure of woven composite materials. In fact, woven composite materials have a complex mesoscopic structure, and differences are easily generated during the forming process of woven composite materials, resulting in significant non-uniformity in woven composite materials. These non-uniformities are difficult to accurately consider in the mechanical property prediction method based on RVE, and bring many difficulties to model establishment, resulting in errors in the prediction of the mechanical properties of woven composite materials. Summary of the Invention

[0004] The purpose of the present invention is to propose a method for predicting the mechanical properties of woven composite materials based on the center line of the fiber bundle orientation, which improves the modeling efficiency and reduces the modeling difficulty.

[0005] The technical solution proposed by the present invention is a method for predicting the mechanical properties of woven composite materials, including:

[0006] Step 101, obtaining the CT image of the woven composite material;

[0007] Step 102, performing image preprocessing on the CT image;

[0008] Step 103, segmenting the fiber bundles in the preprocessed image to obtain the segmented fiber bundle image;

[0009] Step 104: Extract the centerline of the fiber bundle orientation from the segmented fiber bundle image;

[0010] Step 105: Establish a macroscopic geometric model of the woven composite material structure and perform element mesh division on the geometric model;

[0011] Step 106: Establish the connection between the centerline of the fiber bundle orientation and the elements of the geometric model;

[0012] Step 107: Calculate the mechanical property parameters of each element one by one;

[0013] Step 108: Calculate the macroscopic mechanical response of the composite material structure.

[0014] Furthermore, in step 101, obtaining the CT image of the woven composite material further includes: the resolution of the CT image is at least sub-millimeter level.

[0015] Furthermore, in step 102, performing image preprocessing on the CT image further includes: the image preprocessing method includes noise removal algorithm, contrast enhancement algorithm, etc.

[0016] Furthermore, in step 103, segmenting the fiber bundles in the preprocessed image to obtain the segmented fiber bundle image further includes: one or more of the segmentation methods such as threshold segmentation, region growing, edge detection, and deep learning-based segmentation methods can be applied to segment the fiber bundles.

[0017] Furthermore, in step 104, extracting the centerline of the fiber bundle orientation from the segmented fiber bundle image further includes: one or more of the algorithms such as skeleton extraction algorithm, thinning algorithm, distance transformation algorithm, and medial axis transformation algorithm can be applied to extract the centerline of the fiber bundle orientation.

[0018] Furthermore, in step 106, establishing the connection between the centerline of the fiber bundle orientation and the elements of the geometric model further includes: setting the range of calculating the distance; determining the centerline of the fiber bundle orientation included within the center point range of the geometric model element; calculating the distance from the center point of the geometric model element to the centerline of the fiber bundle orientation included within the range.

[0019] Furthermore, in step 107, calculate the mechanical property parameters of each element one by one. The mechanical properties of the composite material usually include properties such as the stiffness, strength, toughness, and thermal expansibility of the material. Taking the stiffness property of the material as an example, the calculation idea of this patent is introduced, and the calculation formula is:

[0020]

[0021] [C m is the stiffness matrix of the geometric model element of the woven composite material;

[0022] i = 1 - n represents the centerline of the fiber bundle orientation within the range of the center point of the geometric model element;

[0023] f i represents the weighting factor of the centerline of the fiber bundle orientation. Its value depends on the distance from the centerline of the fiber bundle orientation to the center point of the geometric model element and is used to determine the influence of each centerline of the fiber bundle orientation on the mechanical properties of the geometric model element;

[0024] [C i is the stiffness matrix of the woven composite material fiber bundle in its local coordinate system;

[0025] [T i is the coordinate transformation matrix used to transform the stiffness matrix of the fiber bundle from the local coordinate system to the global coordinate system;

[0026] [T i -1 is the inverse matrix of the coordinate transformation matrix, used to transform the stiffness matrix of the fiber bundle from the local coordinate system to the global coordinate system.

[0027] Other mechanical property parameters can refer to the calculation method of stiffness parameters. For example, the coefficient of thermal expansion of the material:

[0028]

[0029] α m is the coefficient of thermal expansion of the geometric model element of the woven composite material;

[0030] i = 1 - n represents the centerline of the fiber bundle orientation within the range of the center point of the geometric model element;

[0031] f i represents the weighting factor of the centerline of the fiber bundle orientation. Its value depends on the distance from the centerline of the fiber bundle orientation to the center point of the geometric model element and is used to determine the influence of each centerline of the fiber bundle orientation on the mechanical properties of the geometric model element. It should be noted that the weighting coefficients of different mechanical property parameters are not the same;

[0032] α f is the coefficient of thermal expansion of the fiber bundle of the woven composite material, usually determined by experiments;

[0033] Furthermore, in step 108, calculate the macroscopic mechanical response of the composite material structure. The macroscopic mechanical response includes the stress-strain field, overall deformation, etc. of the composite material structure, which can be directly calculated by using the finite element method based on the unit mechanical property parameters obtained in 107.

[0034] ​The beneficial effects of the present invention are as follows: By establishing the connection between the geometric model unit of the woven composite material and the center line of the fiber bundle orientation, the mechanical properties of the woven composite material are calculated. Using a regular grid division independent of the mesoscopic structural characteristics improves the modeling efficiency of the structural mechanics analysis of the woven composite material, reduces the modeling difficulty, facilitates the preform process design during the initial configuration design of the woven composite material structure, and the method is easy to implement. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 is the CT image of the woven composite material specimen;

[0036] Figure 2 is the CT image of the initial woven composite material specimen;

[0037] Figure 3 is the CT image of the preprocessed woven composite material specimen;

[0038] Figure 4 is the segmented fiber bundle image;

[0039] Figure 5 is the schematic diagram of the center line of the fiber bundle orientation of the woven composite material specimen;

[0040] Figure 6 is the grid diagram of the geometric model unit of the woven composite material specimen;

[0041] Figure 7 is the schematic diagram of the projection distance from the center point of the geometric model unit to the center line of the fiber bundle orientation;

[0042] Figure 8 is the flow analysis diagram of the mechanical property prediction method. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0043] The drawings are only for illustrative purposes and should not be construed as a limitation of this patent; for a better illustration of this embodiment, some components in the drawings will be omitted, enlarged or reduced, and do not represent the actual size of the product; for those skilled in the art, it is understandable that some well-known structures and their descriptions in the drawings may be omitted.

[0044] In the description of the present invention, it should be noted that unless otherwise clearly defined and limited, the terms "installation" and "connection" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the internal connection of two components. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances. The technical solutions of the present invention will be further described below with reference to the drawings and embodiments.

[0045] The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. As Figure 8 shown, the present invention provides a method for predicting the mechanical properties of a woven composite material structure, and the method includes the following steps:

[0046] Step 101, obtaining a CT image of the woven composite material structure;

[0047] Step 102, performing image preprocessing on the CT image;

[0048] Step 103, segmenting the fiber bundles in the preprocessed image to obtain a segmented fiber bundle image;

[0049] Step 104, extracting the center line of the fiber bundle orientation from the segmented fiber bundle image;

[0050] Step 105, establishing a geometric model of the woven composite material and performing element mesh division on the geometric model;

[0051] Step 106, establishing the connection between the center line of the fiber bundle orientation and the elements of the geometric model;

[0052] Step 107, calculating the mechanical property parameters of each element one by one;

[0053] Step 108, calculating the macroscopic mechanical response of the composite material structure.

[0054] In this embodiment, a woven composite material specimen is selected as the research object, and the woven composite material specimen is subjected to CT scanning to obtain a high-resolution CT image, as Figure 1 shown.

[0055] In step 101, the resolution of the CT image of the woven composite material structure needs to reach at least the sub-millimeter level to clearly display the mesoscopic features such as the fiber arrangement of the woven composite material, so as to perform analysis. At the same time, it is also necessary to determine the most suitable resolution level according to the specific application situation. The initial CT image of the woven composite material specimen is as Figure 2 shown. The woven composite material structure in this embodiment is a long strip test piece of a woven ceramic matrix composite material.

[0056] Step 102, performing image preprocessing on the CT image.

[0057] In this embodiment, the Avizo software is used to perform image preprocessing on the CT image obtained in step 101. The methods of image preprocessing include removing noise by adjusting parameters such as the size of Gaussian filtering and median filtering, where the Gaussian kernel size and neighborhood size are usually selected between 3x3 and 7x7. The methods of enhancing contrast include histogram equalization and contrast stretching. Through step-by-step attempts and evaluations based on the noise level and contrast change degree of the image after preprocessing, the effect is achieved that the fiber bundles are clearly visible, the fiber bundles are clearly distinguishable from the background, and the details and shapes of the fiber bundles are retained. The preprocessed image is as Figure 3 shown.

[0058] Step 103: Segment the fiber bundles in the preprocessed image to obtain the segmented fiber bundle image.

[0059] In this embodiment, the image processed in step 102 is imported into the Avizo software, and the threshold segmentation technique is used to convert the image into a binary image. Based on the difference between the gray level features of the fiber bundles and the background gray level, automatic threshold segmentation techniques such as Otsu and Triangle in the Avizo software are used to automatically select an appropriate threshold according to the statistical characteristics of the image, and then the threshold is visually adjusted to observe the effect of the binary image, and multiple tests and adjustments are carried out to achieve the best segmentation effect. The pixels higher than the threshold are marked as fiber bundles, and the pixels lower than the threshold are marked as the background, so as to segment the fiber bundles from the image and ensure that the fiber bundles are separated from other materials as accurately as possible. The segmented fiber bundle image is as Figure 4 shown.

[0060] In step 103, using the segmentation algorithm in the three-dimensional data visualization and analysis software Avizo to separate the fiber bundles from the image is just a common method. According to the actual situation, including but not limited to segmentation methods such as threshold segmentation, region growing, edge detection, and deep learning-based segmentation methods, the fiber bundles can be separated from the image.

[0061] Step 104: Extract the centerline of the fiber bundle orientation from the segmented fiber bundle image, as Figure 5 shown.

[0062] The fiber bundle direction center line includes the warp yarn direction center line X1 and the weft yarn direction center line Y1. In this embodiment, the "skeletonization" tool provided by the Avizo software is used to perform "Thinning Algorithm" processing on the image data obtained in step 103. Select appropriate parameter settings, including parameters such as the number of iterations, threshold, and domain size. Regarding the setting of the number of iterations parameter, the number of iterations is set within a range of 10 to 100, and the changes in the extraction results are observed and the most appropriate number of iterations is selected. Regarding the setting of the threshold parameter, the threshold parameter is based on the threshold obtained in step 103, and then continuously debugged according to the changes in the brightness and contrast of the image to obtain a suitable threshold. Regarding the setting of the neighborhood size parameter, the neighborhood size is set within a range of 1 to 10 pixels, and the changes in the extraction results are observed to select a suitable neighborhood size. Through the iterative process, the boundary of the fiber bundle is gradually "eroded", and finally the fiber bundle direction center line information is obtained, and the extraction results are optimized and adjusted, including removing possible noise and interference to obtain more accurate fiber bundle direction center line information.

[0063] The fiber bundle centerline information in step 104 refers to the extracted center axis information of the fiber bundle in the form of a series of spatial coordinate values ​​of points. Through these spatial coordinate values, various geometric parameters of the fiber bundle can be obtained, including length, radius of curvature, and curvature.

[0064] In step 104, using the "skeletonization" tool in the three-dimensional data visualization and analysis software Avizo to obtain the centerline information of the fiber bundle of the woven composite material specimen is only a common method. According to the actual situation, one or more algorithms including but not limited to "Thinning Algorithm", "Distance Transform Algorithm" and "Medial Axis Transform Algorithm" can be used to extract the centerline information of the fiber bundle of the woven composite material. At the same time, this embodiment can also be used to predict the mechanical properties of self-designed woven composite materials.

[0065] Step 105, establishing a geometric model of the woven composite material, and performing unit mesh division on the geometric model; Figure 6 shown.

[0066] Using ABAQUS finite element analysis software, according to the macroscopic geometric characteristics of the woven composite specimens, the corresponding geometric model was established, and the appropriate grid density was selected for unit grid division.

[0067] In this embodiment, the size of the geometric model is 39 * 4.5 * 9.5, the shape of the unit mesh of the geometric model is a hexahedral structured mesh, the approximate global size is 1, and the unit is mm.

[0068] Mesh generation is performed. For example, if the woven composite specimen is a cuboid test piece, then a cuboid geometric model with corresponding dimensions is established, and then the cuboid geometric model is imported into the ABAQUS finite element analysis software for mesh generation. The geometric model element is m1.

[0069] Step 106: Establish the connection between the fiber bundle centerline and the geometric model element.

[0070] The connection in Step 106 means determining the fiber bundle centerlines included within the distance range of each geometric model element and determining the distance from the center point of the geometric model element to these fiber bundle centerlines.

[0071] In this embodiment, based on the fiber bundle centerline information obtained in Step 104 and the geometric model of the woven composite specimen established in Step 105, the Visual Studio programming software is used to establish the connection between the geometric model elements of the woven composite specimen and the fiber bundle centerlines of the woven composite specimen. Specifically, the parameter information of the geometric model element is the coordinates of the element center point. A suitable distance range is selected, such as Figure 7 shown by the middle circle. The size of the distance range is between half of the fiber bundle width and one fiber bundle width. According to the distance range, determine the fiber bundle centerlines included by the center point of the geometric model element within this range, and calculate the distance from the center point of the geometric model element to all the fiber bundle centerlines included within the range. The calculation method can adopt the mapping method.

[0072] Establish the distance from the coordinates of the center point of the geometric model element to all the fiber bundle centerlines included within the range. The method for calculating the distance is as follows:

[0073] Calculate the distance from the coordinates of the center point of the geometric model element to the projection coordinates, and set this distance as the distance from the center point of the geometric model element to the fiber bundle centerline. The schematic diagram of the distance calculation method is as Figure 7 shown. Here, the fiber bundle centerline information refers to the distance from the center point of the geometric model element to the fiber bundle centerline, as well as the starting coordinates and ending coordinates of the fiber bundle centerlines included within the range of the geometric model element, etc.

[0074] Step 107: Calculate the mechanical property parameters of each unit one by one.

[0075] The mechanical properties of materials refer to the properties exhibited by materials when subjected to forces and deformations. Common mechanical properties of materials include properties such as the stiffness, strength, toughness, and thermal expansion of the materials. Among them, the stiffness property of a material is mainly represented by the stiffness matrix of the material. The stiffness matrix of a material is a tensor that describes the stiffness characteristics of the material under small strains. It reflects the rigid characteristics of the material or structure, that is, the degree of its deformation under external forces. The mechanical properties of materials are indicators for evaluating the mechanical behavior and properties of materials during the force application process, and can be calculated and evaluated through the material stiffness matrix. Based on the geometric model elements of the woven composite specimen in step 106 and the information on the centerline of the fiber bundle orientation associated therewith, using the formula the stiffness matrix of each geometric model element of the woven composite specimen can be calculated, thereby obtaining the overall stiffness property of the geometric model of the woven composite specimen. Common mechanical properties of materials include properties such as the stiffness, strength, toughness, and thermal expansion of the materials. In this embodiment, taking the stiffness property of the material as an example, a method for predicting the mechanical properties of woven composites is introduced to achieve the purpose of predicting the mechanical properties of woven composites.

[0076] [C m is the stiffness matrix of the geometric model element of the woven composite;

[0077] i = 1 - n represents the centerline of the fiber bundle orientation included within the center point range of the geometric model element;

[0078] f i represents the fiber bundle orientation centerline weighting factor, whose value depends on the distance from the centerline of the fiber bundle orientation to the center point of the geometric model element and is used to determine the influence of each centerline of the fiber bundle orientation on the mechanical properties of the geometric model element;

[0079] [C i is the stiffness matrix of the woven composite fiber bundle in its own local coordinate system, and this stiffness matrix is generally determined by experiments;

[0080] The stiffness matrix of the fiber bundle needs to be obtained through experiments. Mechanical experiments are conducted on the fiber bundle to obtain the stress-strain relationship of the fiber bundle, thereby obtaining stiffness parameters such as the elastic modulus, shear modulus, and Poisson's ratio of the fiber bundle. Through multiple experiments, the average value of the stiffness parameters of the fiber bundle is obtained as the representative parameter. Based on these stiffness parameters, the stiffness matrix of the fiber bundle can be calculated. Here, a simplified treatment is made, considering that the stiffness matrix of each fiber bundle is the same, which is convenient for subsequent calculation of the stiffness matrix of the geometric model element.

[0081] [T i is the coordinate transformation matrix used to transform the stiffness matrix of the fiber bundle from the local coordinate system to the global coordinate system;

[0082] The specific calculation steps of the coordinate transformation matrix are as follows:

[0083] 1. Determine the local coordinate system of the fiber bundle: Usually, the main axis direction of the fiber bundle is taken as the x-axis of the local coordinate system, and the normal direction of the fiber bundle cross-section is taken as the z-axis of the local coordinate system;

[0084] 2. Determine the position and direction of the fiber bundle in the global coordinate system: Determine the starting point coordinates and ending point coordinates of the fiber bundle in the global coordinate system, and calculate the direction vector of the fiber bundle by subtracting the starting point coordinates from the ending point coordinates;

[0085] 3. Calculate the rotation angle and coordinate transformation matrix of the fiber bundle: Calculate the rotation angle and coordinate transformation matrix of the fiber bundle according to the direction vector of the fiber bundle. The coordinate transformation matrix is a 6×6 matrix, which represents the rotation relationship of the local coordinate system of the fiber bundle relative to the global coordinate system.

[0086] [T i -1 is the inverse matrix of the coordinate transformation matrix, which is used to transform the stiffness matrix of the fiber bundle from the local coordinate system to the global coordinate system.

[0087] Step 108, calculate the macroscopic mechanical response of the composite structure.

[0088] The macroscopic mechanical response includes the stress-strain field and overall deformation of the composite structure, etc., which can be directly calculated by using the general finite element method based on the element mechanical property parameters obtained in Step 107.

[0089] As mentioned above, it is only the preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, makes equivalent substitutions or changes, and all should be covered within the protection scope of the present invention.​

Claims

1. A method for predicting mechanical properties of a woven composite material structure, characterized in that: Step 101, obtaining a CT image of the woven composite material; Step 102, performing image preprocessing on the CT image; Step 103, segmenting the fiber bundles in the preprocessed image to obtain a segmented fiber bundle image; Step 104, extracting the fiber bundle direction center line from the segmented fiber bundle image; Step 105, establishing a macroscopic geometric model of the woven composite material structure, and performing regular unit grid division on the geometric model; Step 106, establishing a connection between the fiber bundle centerline and the geometric model unit, including: Set the range of distance calculation; Establish the center line of the fiber bundle within the center point of the geometric model unit; Calculate the distance from the center point of the geometric model unit to the center line of the fiber bundle contained in the range; Step 107, calculating the mechanical performance parameters of each unit one by one, including: [C m ] is the stiffness matrix of the geometric model unit of the woven composite material; i = 1-n represents the center line of the fiber bundle within the center point of the geometric model unit; f i represents the weighting factor of the fiber bundle toward the center line; [C i ] is the stiffness matrix of the fiber bundle of the woven composite material in its own local coordinate system; [T i ] is the coordinate transformation matrix, which is used to transform the stiffness matrix of the fiber bundle from the local coordinate system to the global coordinate system; [T i ] -1 is the inverse matrix of the coordinate transformation matrix, which is used to transform the stiffness matrix of the fiber bundle from the local coordinate system to the global coordinate system; Step 108, calculating the macroscopic mechanical response of the composite material structure.

2. The mechanical property prediction method according to claim 1, characterized in that: Step 101, obtaining a CT image of a woven composite material specimen, further includes: The resolution of CT images is at least sub-millimeter.

3. The mechanical property prediction method according to claim 2, characterized in that: Step 102, performing image preprocessing on the CT image, further includes: Image preprocessing methods include noise removal algorithms and contrast enhancement algorithms.

4. The mechanical property prediction method according to claim 3, characterized in that: Step 103, segmenting the fiber bundles in the preprocessed image to obtain a segmented fiber bundle image, further comprising: The fiber bundle segmentation can apply one or more of threshold segmentation, region growing, edge detection and deep learning-based segmentation methods.

5. The mechanical property prediction method according to claim 4, characterized in that: Step 104, extracting the fiber bundle direction center line from the segmented fiber bundle image, also includes: To extract the center line of the fiber bundle, one or more of a skeleton extraction algorithm, a thinning algorithm, a distance transformation algorithm, and a medial axis transformation algorithm may be applied.

6. The mechanical property prediction method according to claim 5, characterized in that: Step 104, extracting the fiber bundle direction center line from the segmented fiber bundle image, also includes: The fiber bundle running center line includes the warp yarn running center line and the weft yarn running center line.

Images