A woven composite strength prediction method based on real microstructure

Abstract

The application provides a woven composite material strength prediction method based on a real microstructure, and belongs to the technical field of composite material mechanics. The method obtains and reconstructs a three-dimensional microstructure of a material through X-ray CT, selects a representative volume element for yarn, matrix and pore defect segmentation, calculates component volume fractions and extracts yarn orientation characteristic parameters, establishes an analytical prediction model in combination with component strength parameters, and quickly obtains tensile, compression and shear strength prediction values of the woven composite material. The application takes real structure information as input, does not need full-cell finite elements and damage evolution iteration, has high calculation efficiency and clear parameters, and is suitable for engineering rapid evaluation and scheme comparison.

Description

Technical Field

[0001] This invention relates to the field of composite material mechanics, and more specifically to a method for predicting the strength of woven composite materials based on real microstructure. Background Technology

[0002] Woven composite materials, due to their interlaced fiber bundles in space, excellent structural integrity, and high specific strength, specific stiffness, and good damage tolerance, have been widely used in aerospace and other engineering fields. During service, woven composite components are typically subjected to high-temperature and complex load environments, and their tensile, compressive, and shear strength indicators directly affect structural design margins, safety assessments, and service reliability. To shorten the design iteration cycle of these components and reduce testing costs, establishing a strength prediction method that balances structural characteristic characterization with computational efficiency is of great significance in the engineering application phase.

[0003] Currently, several patented technologies have proposed different implementation paths for predicting the strength properties of woven composite materials. For example, patent CN119047231A predicts the strength of woven composite materials under multiaxial loads by establishing a micro-finite element model and introducing damage criteria. This method relies on complex numerical modeling and calculation processes, resulting in high computational costs and long cycles, making it difficult to meet the needs of rapid engineering evaluation. Patent CN105956347 proposes a method for predicting the mechanical properties of composite materials based on microstructure modeling, achieving material strength analysis through multi-scale model coupling. However, its implementation involves multi-level modeling and parameter settings, leading to high complexity in engineering applications and poor scalability. Another patent, CN120577133A, addresses the interlaminar properties of ceramic matrix composites by introducing pore type to evaluate shear strength. This method can reflect the influence of defect factors, but it mainly focuses on single structural factors and cannot comprehensively characterize the spatial distribution of multiple yarn types in woven composite materials and their combined effect on directional strength. Therefore, existing technologies either focus on micro-scale finite element analysis and progressive damage analysis, making it difficult to meet the needs of rapid engineering evaluation, or they focus on local defects or single indicators, making it difficult to comprehensively characterize the contribution of woven microstructure to strength.

[0004] The strength performance of woven composite materials depends not only on the mechanical properties of the components themselves, but also on the spatial distribution and orientation of the yarns. Furthermore, defects such as pores directly affect their load-bearing capacity. Existing strength prediction methods either fail to fully incorporate the true microstructure of woven composite materials, neglecting key influencing factors such as yarn orientation angles and pore distribution, leading to significant deviations between predicted results and actual service performance; or their computational processes are complex and inefficient, failing to meet the practical needs of rapid engineering assessment. Therefore, there is an urgent need for a method that, based on true microstructural information, can rapidly predict the strength of woven composite materials while ensuring physical plausibility, thus balancing the dual requirements of true microstructural characterization and rapid engineering prediction. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method for predicting the strength of woven composite materials based on the real microstructure. This method can make rapid and accurate predictions of tensile, compressive and shear interlaminar strength under different loading directions by fully considering the comprehensive effects of yarn spatial orientation, component contribution and pore defects based on the real microstructure information of the material, thereby meeting the dual requirements of efficiency and accuracy in engineering design.

[0006] To achieve the above-mentioned technical objectives, the technical solution adopted by the present invention is as follows:

[0007] A method for predicting the strength of woven composite materials based on real microstructure, the method comprising the following steps:

[0008] S1, obtain the real microstructure of the woven composite material and perform three-dimensional reconstruction to obtain a three-dimensional digital microstructure model containing yarn, matrix and pore defects;

[0009] S2, in the three-dimensional digital microstructure model, representative volume units are selected and components are divided to divide the material into three components: yarn phase, matrix phase and pore phase.

[0010] S3, Statistically calculate the volume fraction of yarn phase, matrix phase, and pore phase within a representative volume unit;

[0011] S4. The skeleton of the yarn phase is extracted to obtain discrete points of the yarn centerline, construct yarn orientation characterization parameters, and calculate the orientation efficiency coefficient under the interlayer loading direction.

[0012] S5, determine the single-component strength parameters of the yarn and matrix under interlayer tension, interlayer compression, and interlayer shear modes;

[0013] S6. Under the interlayer loading direction, the strength parameters of each component are weighted and superimposed according to the volume fraction and the orientation efficiency coefficient, and a porosity reduction term is introduced to obtain the predicted value of the interlayer strength of the woven composite material.

[0014] Step S1 further includes:

[0015] X-ray computed tomography was performed on woven composite material specimens to obtain a sequence of microstructure images.

[0016] The microstructure image sequence is reconstructed to obtain a three-dimensional digital microstructure model containing yarn, matrix and pore defects, resulting in three-dimensional gray volume data I(x,y,z) and voxel size data.

[0017] Step S2 further includes:

[0018] In the three-dimensional digital microstructure model, a representative volume element RVE that reflects the periodicity or statistical uniformity of the material structure is selected. The thickness dimension of the representative volume element RVE is consistent with the thickness of the actual specimen.

[0019] The CT volume data within the RVE range of the representative volume unit are preprocessed to remove noise, and the material is divided into yarn phase, matrix phase and pore phase based on grayscale threshold or histogram segmentation.

[0020] Further, in step S3, the volume fractions of yarn phase, matrix phase, and porous phase within a representative volume unit are calculated using the following formula:

[0021]

[0022] Where j is the component type identifier, j=f represents the yarn phase, j=p represents the porous phase, and j=m represents the matrix phase. Let j be the number of voxels in the j-th phase. The total prime number of the representative volume unit RVE; and satisfying the normalization condition:

[0023]

[0024] in, It is the yarn volume fraction. This is the volume fraction of the matrix. It represents the pore volume fraction.

[0025] Step S4 further includes:

[0026] The yarn phase is subjected to skeleton extraction to obtain discrete points of the yarn centerline. The yarn centerline is represented as a parametric curve. :

[0027]

[0028] In the formula, This represents the coordinate components of the r-th discrete point in the three-dimensional coordinate system. This is the arc length parameter of the yarn centerline. These are the coordinate components of the centerline in the three-dimensional coordinate system;

[0029] Perform tangential normalization on the local orientation unit vector of the yarn:

[0030]

[0031] In the formula, Let be the tangential guide vector of the yarn centerline. Let be the Euclidean norm of the vector. The unit tangent vector of the yarn centerline;

[0032] Statistical analysis is performed on all orientation vectors of the k-th type of yarn within the RVE, and a second-order orientation tensor is constructed:

[0033]

[0034] In the formula, k is the yarn type identifier. Let be the total number of orientation vectors for the k-th type of yarn. Let i be the unit tangent vector of the k-th type of yarn, with superscript... Indicates transpose;

[0035] Introducing the orientation efficiency coefficient :

[0036]

[0037] in Load a unit vector of direction for the target. Indicates the k-th type of yarn in the loading direction The directional orientation efficiency coefficient.

[0038] Further, in step S5, the strength parameters of the yarn phase and the matrix phase are obtained through material standard tests or supplier data; wherein, for the yarn phase, the set of strength parameters for the k-th type of yarn is represented as:

[0039]

[0040] in The axial tensile strength of the k-th type of yarn The axial compressive strength of the k-th type of yarn, The yarn shear strength of the k-th type of yarn;

[0041] The matrix strength parameters are:

[0042]

[0043] in The tensile strength of the matrix. The compressive strength of the matrix, The shear strength of the matrix.

[0044] For similar material systems, macroscopic mechanical tests are used to verify and correct the consistency of strength parameters.

[0045] Further, in step S6, the predicted value of the interlaminar tensile or interlaminar compressive strength of the woven composite material is calculated using the following formula:

[0046]

[0047] superscript The porosity sensitivity index under tensile or compressive conditions. Indicates the target loading direction The predicted interlaminar strength values ​​are shown below, with the superscript + indicating tensile conditions and - indicating compressive conditions. This represents the summation over all yarn types within a representative volume cell. Used to identify yarn type; This represents the volume fraction of the k-th type of yarn; Indicates the k-th type of yarn in the loading direction The orientation efficiency coefficient is below. The axial tensile strength of the k-th type of yarn The axial compressive strength of the k-th type of yarn; Indicates the volume fraction of the matrix phase; The tensile strength of the matrix. The compressive strength of the matrix

[0048] The predicted interlaminar shear strength of woven composite materials is calculated using the following formula:

[0049]

[0050] Where q is the porosity sensitivity index under shear conditions. Load direction for target Predicted interlaminar shear strength values ​​below Shear strength of type k yarn The shear strength of the matrix.

[0051] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0052] First, the strength prediction method for woven composite materials based on real microstructure of the present invention takes the real microstructure of woven composite materials as the core input, relies on CT scanning and three-dimensional reconstruction technology to accurately extract the spatial distribution information of yarn, matrix and pores inside the material, and performs component segmentation and quantitative statistics on CT three-dimensional reconstruction data through professional image segmentation algorithm. It can not only directly obtain the accurate volume fraction of each component (yarn, matrix and pores), but also accurately capture the fine structural features such as spatial undulation and deflection angle of yarn, and completely restore the real structural state inside the material. It breaks the limitation of traditional prediction methods that only rely on idealized models and have large deviations from the actual material structure, and ensures the authenticity and accuracy of prediction results from the structural characterization level.

[0053] Secondly, the strength prediction method for woven composite materials based on real microstructure of the present invention abandons the complex process of relying on whole-cell micro-finite element modeling and damage evolution iteration calculation in the prior art. It adopts an analytical strength prediction model based on component volume fraction and yarn orientation parameters, organically integrating yarn orientation efficiency, component performance parameters and the influence of pore defects. Through scientific weighted calculation, it directly outputs the tensile strength, compressive strength and shear strength results of the material under different loading directions. The entire process does not require complex numerical simulation and boundary condition settings, which greatly reduces the calculation cost and modeling difficulty, and takes into account both prediction accuracy and engineering application efficiency.

[0054] Third, the strength prediction method for woven composite materials based on real microstructure of the present invention not only ensures the consistency between the prediction results and the actual service performance of the material, but also meets the needs of rapid evaluation and scheme comparison in the engineering design stage. It effectively solves the technical pain points of existing methods, such as complex modeling, long calculation cycle, high cost and poor adaptability. It provides efficient and accurate technical support for the engineering design and safety assessment of woven composite materials, and is both practical and economical. Attached Figure Description

[0055] Figure 1 This is a flowchart illustrating the strength prediction method for woven composite materials based on real microstructure according to the present invention.

[0056] Figure 2 This is a schematic diagram of XCT slices of woven composite materials and the selection of representative volumetric elements (RVEs);

[0057] Figure 3 This is a schematic diagram of the RVE model component segmentation results and yarn centerline extraction;

[0058] Figure 4 This is a comparison chart of the prediction results of this invention and the reference intensity data. Detailed Implementation

[0059] The embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0060] This invention discloses a method for predicting the strength of woven composite materials based on real microstructure, the method comprising the following steps:

[0061] S1, obtain the real microstructure of the woven composite material and perform three-dimensional reconstruction to obtain a three-dimensional digital microstructure model containing yarn, matrix and pore defects;

[0062] S2, in the three-dimensional digital microstructure model, representative volume units are selected and components are divided to divide the material into three components: yarn phase, matrix phase and pore phase.

[0063] S3, Statistically calculate the volume fraction of yarn phase, matrix phase, and pore phase within a representative volume unit;

[0064] S4. The skeleton of the yarn phase is extracted to obtain discrete points of the yarn centerline, construct yarn orientation characterization parameters, and calculate the orientation efficiency coefficient under the interlayer loading direction.

[0065] S5, determine the single-component strength parameters of the yarn and matrix under interlayer tension, interlayer compression, and interlayer shear modes;

[0066] S6. Under the interlayer loading direction, the strength parameters of each component are weighted and superimposed according to the volume fraction and the orientation efficiency coefficient, and a porosity reduction term is introduced to obtain the predicted value of the interlayer strength of the woven composite material.

[0067] Specifically, the strength prediction method for woven composite materials based on real microstructure of the present invention includes the following steps:

[0068] Step 1: Acquisition of realistic detailed structure and 3D reconstruction.

[0069] like Figure 2 As shown, a woven composite material sample (in this embodiment, a 2.5D woven C / SiC composite material sample) was selected. X-ray computed tomography (XCT) was used to obtain a sequence of slices showing the internal microstructure of the sample, resulting in three-dimensional grayscale data I(x,y,z). Figure 2 (a) in the image represents a typical grayscale slice. To ensure the identification of fine features such as yarn gaps and pores, a voxel size of 15 μm was selected. The output is three-dimensional grayscale volume data and spatial resolution information, which are used for subsequent segmentation and statistics.

[0070] Step 2: Selection of representative volumetric units and component segmentation.

[0071] like Figure 2 As shown in (b), a representative volumetric unit (RVE) is selected in the three-dimensional grayscale data, which covers one or more weaving cycles in the warp and weft directions, and is consistent with the effective thickness of the sample or reaches statistical stability in the thickness direction; the RVE grayscale data is preprocessed to remove noise, resulting in a denoised grayscale value. Subsequently, based on grayscale features, threshold segmentation was used to divide the material into yarn phase, matrix phase, and porosity / defect phase. After segmentation, connected component filtering and correction were performed on the results to remove isolated noise points and ensure boundary continuity; the output is a set of yarn voxels. matrix voxel set Pore ​​defect voxel set .

[0072] Step 3: Volume fraction statistics.

[0073] The number of voxels in each phase within the RVE was statistically analyzed, and the volume fraction of each phase was as follows:

[0074] ;

[0075] Where j is the component type identifier, j=f represents the yarn phase, j=p represents the porous phase, and j=m represents the matrix phase. Let j be the number of voxels in the j-th phase. The total prime number of the representative volume unit RVE; and satisfying the normalization condition:

[0076] ;

[0077] in, It is the yarn volume fraction. This is the volume fraction of the matrix. It represents the pore volume fraction.

[0078] In this embodiment, the yarn volume fraction was statistically obtained. =0.3244, matrix volume fraction =0.6250, pore volume fraction .

[0079] Step 4: Yarn orientation extraction and orientation characterization parameter construction.

[0080] like Figure 3 As shown, the yarn phase skeleton is extracted to obtain discrete points of the yarn centerline { The centerline is represented as a parametric curve:

[0081] ;

[0082] Perform tangential normalization on the local orientation unit vector of the yarn:

[0083] ;

[0084] Statistical analysis is performed on all orientation vectors of the k-th type of yarn within the RVE, and a second-order orientation tensor is constructed:

[0085] ;

[0086] To facilitate the calculation of directional strength, a directional orientation efficiency coefficient is introduced:

[0087] .

[0088] Step 5: Determine the component strength parameters.

[0089] The set of component strength parameters is determined based on the material system and reference experimental data. For the yarn phase, the set of strength parameters for the k-th type of yarn is expressed as:

[0090]

[0091] in The axial tensile strength of the k-th type of yarn The axial compressive strength of the k-th type of yarn, The yarn shear strength of the k-th type of yarn;

[0092] The matrix strength parameters are:

[0093]

[0094] in The tensile strength of the matrix. The compressive strength of the matrix, This represents the shear strength of the matrix. This embodiment uses tensile strength prediction as an example, taking the equivalent tensile strength of the yarn as the reference. =570MPa, matrix tensile strength =200MPa. Table 1 shows the parameters of each component used in this embodiment.

[0095] Table 1. Component volume fraction and component strength parameters

[0096]

[0097] Step 6: Analyze the model for rapid prediction and output of results.

[0098] Given a loading direction, the strengths of each component are weighted and superimposed according to their volume fraction and directional efficiency coefficient to obtain the overall strength prediction value. When used for rapid engineering evaluation and the loading direction is consistent with the direction of the main load-bearing yarn, the main load-bearing yarn is... Setting it to 1 yields a simplified, fast prediction form:

[0099]

[0100] Substituting the data into Table 1, we get:

[0101]

[0102] Furthermore, when it is necessary to include porosity defect reduction, a porosity reduction term is introduced:

[0103]

[0104] Where w is the porosity sensitivity index, obtained through independent calibration. The calibration data comes from the tensile strength test results of several groups of samples with different pore volume fractions under the same material system. Calibration of w:

[0105]

[0106] This embodiment takes =0.213, substituting this value into the porosity reduction model, we get:

[0107] .

[0108] like Figure 4 The figure shows a comparison between the predicted values ​​of this invention and the reference strength data. The error between the analytical method's predicted values ​​and the experimental values ​​is 1.1%. This demonstrates that this invention, by utilizing the volume fraction and orientation characterization parameters obtained from the statistical analysis of the actual microstructure, combined with an analytical prediction model, can achieve rapid prediction of the strength of woven composite materials without requiring microscopic finite element iteration and damage evolution calculations.

[0109] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0110] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for predicting the strength of woven composite materials based on real microstructure, characterized in that, The method includes the following steps: S1, obtain the real microstructure of the woven composite material and perform three-dimensional reconstruction to obtain a three-dimensional digital microstructure model containing yarn, matrix and pore defects; S2, in the three-dimensional digital microstructure model, representative volume units are selected and components are divided to divide the material into three components: yarn phase, matrix phase and pore phase. S3, Statistically calculate the volume fraction of yarn phase, matrix phase, and pore phase within a representative volume unit; S4. The skeleton of the yarn phase is extracted to obtain discrete points of the yarn centerline, construct yarn orientation characterization parameters, and calculate the orientation efficiency coefficient under the interlayer loading direction. S5, determine the single-component strength parameters of the yarn and matrix under interlayer tension, interlayer compression, and interlayer shear modes; S6. Under the interlayer loading direction, the strength parameters of each component are weighted and superimposed according to the volume fraction and the orientation efficiency coefficient, and a porosity reduction term is introduced to obtain the predicted value of the interlayer strength of the woven composite material.

2. The method for predicting the strength of woven composite materials based on real microstructure according to claim 1, characterized in that, Step S1 further includes: X-ray computed tomography was performed on woven composite material specimens to obtain a sequence of microstructure images. The microstructure image sequence is reconstructed to obtain a three-dimensional digital microstructure model containing yarn, matrix and pore defects, resulting in three-dimensional gray volume data I(x,y,z) and voxel size data.

3. The method for predicting the strength of woven composite materials based on real microstructure according to claim 1, characterized in that, Step S2 further includes: In the three-dimensional digital microstructure model, a representative volume element RVE that reflects the periodicity or statistical uniformity of the material structure is selected. The thickness dimension of the representative volume element RVE is consistent with the thickness of the actual specimen. The CT volume data within the RVE range of the representative volume unit are preprocessed to remove noise, and the material is divided into yarn phase, matrix phase and pore phase based on grayscale threshold or histogram segmentation.

4. The method for predicting the strength of woven composite materials based on real microstructure according to claim 1, characterized in that, In step S3, the volume fractions of yarn phase, matrix phase, and porous phase within a representative volume unit are calculated using the following formula: Where j is the component type identifier, j=f represents the yarn phase, j=p represents the porous phase, and j=m represents the matrix phase. Let j be the number of voxels in the j-th phase. The total prime number of the representative volume unit RVE; and satisfying the normalization condition: in, It is the yarn volume fraction. This is the volume fraction of the matrix. It represents the pore volume fraction.

5. The method for predicting the strength of woven composite materials based on real microstructure according to claim 1, characterized in that, Step S4 further includes: The yarn phase is subjected to skeleton extraction to obtain discrete points of the yarn centerline. The yarn centerline is represented as a parametric curve. : In the formula, This represents the coordinate components of the r-th discrete point in the three-dimensional coordinate system. This is the arc length parameter of the yarn centerline. These are the coordinate components of the centerline in the three-dimensional coordinate system; Perform tangential normalization on the local orientation unit vector of the yarn: In the formula, Let be the tangential guide vector of the yarn centerline. Let be the Euclidean norm of the vector. The unit tangent vector of the yarn centerline; Statistical analysis is performed on all orientation vectors of the k-th type of yarn within the RVE, and a second-order orientation tensor is constructed: In the formula, k is the yarn type identifier. Let be the total number of orientation vectors for the k-th type of yarn. Let i be the unit tangent vector of the k-th type of yarn, with superscript... Indicates transpose; Introducing the orientation efficiency coefficient : in Load a unit vector of direction for the target. Indicates the k-th type of yarn in the loading direction The directional orientation efficiency coefficient.

6. The method for predicting the strength of woven composite materials based on real microstructure according to claim 1, characterized in that, In step S5, the strength parameters of the yarn phase and the matrix phase are obtained through material standard tests or supplier data; wherein, for the yarn phase, the set of strength parameters for the k-th type of yarn is represented as: in The axial tensile strength of the k-th type of yarn The axial compressive strength of the k-th type of yarn, The yarn shear strength of the k-th type of yarn; The matrix strength parameters are: in The tensile strength of the matrix. The compressive strength of the matrix, The shear strength of the matrix. For similar material systems, macroscopic mechanical tests are used to verify and correct the consistency of strength parameters.

7. The method for predicting the strength of woven composite materials based on real microstructure according to claim 1, characterized in that, In step S6, the predicted interlaminar tensile or compressive strength of the woven composite material is calculated using the following formula: superscript The porosity sensitivity index under tensile or compressive conditions. Indicates the target loading direction The predicted interlaminar strength values ​​are shown below, with the superscript + indicating tensile conditions and - indicating compressive conditions. This represents the summation over all yarn types within a representative volume cell. Used to identify yarn type; This represents the volume fraction of the k-th type of yarn; Indicates the k-th type of yarn in the loading direction The orientation efficiency coefficient is below. The axial tensile strength of the k-th type of yarn The axial compressive strength of the k-th type of yarn; Indicates the volume fraction of the matrix phase; The tensile strength of the matrix. The compressive strength of the matrix The predicted interlaminar shear strength of woven composite materials is calculated using the following formula: Where q is the porosity sensitivity index under shear conditions. Load direction for target Predicted interlaminar shear strength values ​​below Shear strength of type k yarn The shear strength of the matrix.

Images