CN-122016458-A - DVC calculation method and device for carbon fiber composite material

CN122016458ACN 122016458 ACN122016458 ACN 122016458ACN-122016458-A

Abstract

The invention relates to the technical field of mechanical property test and nondestructive detection of composite materials, in particular to a DVC calculation method and device of a carbon fiber composite material; and finally, performing DVC calculation on the unified coordinate data and the grid model to obtain three-dimensional displacement field and strain field distribution data inside the sample at each stage. According to the invention, the image matching is carried out on the data, the accurate grid is established according to the threshold segmentation, the influence of the tiny rigid body displacement on DVC calculation in the experimental process is eliminated, the calculation accuracy is improved, the analysis of the internal stress concentration rule of the carbon fiber composite material is facilitated, and the guidance is provided for the design of the carbon fiber composite material.

Inventors

JIANG LANXIN
Xiong Hongzi
WU JIANBO
WANG JIE

Assignees

四川大学

Dates

Publication Date: 20260512
Application Date: 20260123

Claims (10)

1.A method for calculating DVC of a carbon fiber composite material, the method comprising the steps of: s1, acquiring an X-ray tomographic image set of a carbon fiber composite material sample, wherein the X-ray tomographic image set comprises X-ray tomographic images of an experiment loading pre-stage, an experiment loading middle stage and an experiment loading post-stage of an in-situ experiment; s2, preprocessing the X-ray tomography image set to obtain a two-dimensional gray scale atlas of the carbon fiber composite material sample; based on the space parameters scanned by the X-ray tomography technology, orderly stacking the two-dimensional gray-scale atlas and assigning three-dimensional coordinates to generate three-dimensional space data corresponding to the two-dimensional gray-scale atlas; establishing a reference coordinate system, and calibrating and reconstructing the three-dimensional space data coordinates by taking the reference coordinate system as a reference to obtain three-dimensional space data with unified coordinates; threshold segmentation is adopted for the two-dimensional gray level atlas to obtain an image entity; And S3, performing DVC calculation on the three-dimensional space data with unified coordinates and the grid model to obtain three-dimensional displacement field distribution data and three-dimensional strain field distribution data inside the carbon fiber composite material sample at each stage of in-situ experiment.
2. The method for calculating DVC of a carbon fiber composite material according to claim 1, wherein in S2: The preprocessing comprises noise reduction and image brightness adjustment; the noise reduction is realized through a bilateral filtering algorithm or through a Filter Sandbox component in AVIZO software; The image brightness adjustment is realized by a method of combining image background subtraction and gray scale normalization or a method based on linear brightness adjustment; when the bilateral filtering algorithm is used for noise reduction, parameters are set to be KERNEL SIZE X=3, KERNEL SIZE Y=3 and SIMILARITY =20; when the image brightness adjustment is carried out by combining the image background subtraction and the gray scale normalization, firstly calculating the background brightness approximation value of each X-ray tomography image in the X-ray tomography image set by an image edge area gray scale value average method or an adaptive threshold method, subtracting the background brightness approximation value corresponding to each X-ray tomography image from the pixel gray scale value of each X-ray tomography image, eliminating the interference of uneven background brightness, respectively carrying out gray scale normalization processing on an X-ray tomography image sequence in the same stage of an in-situ experiment, and mapping the gray scale value of the two-dimensional gray scale image sequence to a range of 0-255.
3. The method for calculating the DVC of the carbon fiber composite material according to claim 1, wherein the three-dimensional coordinate assignment in the step S2 adopts an FDK reconstruction algorithm.
4. The method for calculating DVC of a carbon fiber composite material according to claim 1, wherein in S2: The method specifically comprises the steps of selecting an experimental reference stage, loading three-dimensional space data of the reference stage in AVIZO software, calling a VOLUME RENDERING component through AVIZO software, setting a gray mapping MODE and transparency of 0.7-0.9 based on voxel gray values of the three-dimensional space data, adopting a VOLUME RENDERING algorithm TO analyze and render one by one TO generate a VOLUME RENDERING model, clicking 3 non-collinear characteristic POINTs on the VOLUME RENDERING model through a FIT TO POINT command in the SLICE component, confirming an XY plane and an XYZ axis direction based on the characteristic POINT coordinates TO define a reference coordinate system, activating a RESAMPLE TRANSFORMED IMAGE component in AVIZO software, setting MODE as CROPPED, resampling the three-dimensional space data of the reference stage, reconstructing TO obtain standard three-dimensional space data under the reference coordinate system, outputting a first three-dimensional visual model, and confirming an XY plane position and an XYZ axis direction of the reference coordinate system through the first three-dimensional visual model; loading three-dimensional space data preprocessed in all stages of an in-situ experiment in AVIZO software, calculating an affine transformation matrix by adopting a ITERATIVE OPTIMIZATION ALGORITHM algorithm, and aligning the three-dimensional space data of the rest stages of the in-situ experiment with standard three-dimensional space data under the reference coordinate system; The three-dimensional space data coordinate reconstruction specifically comprises the step of resampling the three-dimensional space data of the rest in-situ experiment stages after coordinate alignment by taking standard three-dimensional space data in the reference coordinate system as a reference.
5. The method of claim 4, wherein the reference stage is a pre-load stage.
6. The method for calculating the DVC of the carbon fiber composite material according to claim 1, wherein in the step S2, the two-dimensional gray scale atlas is subjected to threshold segmentation, specifically, the three-dimensional space data with unified coordinates are loaded in AVIZO software, the three-dimensional space data with unified coordinates are subjected to threshold segmentation, a threshold range is set based on gray scale value differences of all components of the carbon fiber composite material sample, and binarization data of the carbon fiber composite material sample are segmented based on the threshold range, so that an image entity is obtained.
7. The DVC calculation method of a carbon fiber composite material according to claim 1, wherein the grid model in S2 is a tetrahedral grid model, and the establishing the grid model based on the image entity specifically comprises: Based on the binarized data of the image entity, using GENERALIZED MARCHING CUBES ALGORITHM algorithm based on non-binary classification, setting CONSTRAINED SMOOTHING =2, generating triangle approximation value of a sample interface, adopting Laplacian smoothing to carry out grid smoothing, adopting a propulsion leading edge method, filling a region defined by surface data with tetrahedron, and generating a tetrahedron grid model for DVC calculation.
8. The method for calculating the DVC of the carbon fiber composite material according to claim 1, wherein in the step S3, the DVC calculation is performed on the three-dimensional space data with unified coordinates and the grid model, specifically, the three-dimensional space data with unified coordinates and the grid model in each stage of an in-situ experiment are loaded in AVIZO software, a INCREMENTAL GLOBAL DVC calculation function is activated, parameters such as the size of a relevant window, a convergence threshold value and the like are set, and the three-dimensional displacement field distribution data and the three-dimensional strain field distribution data inside the carbon fiber composite material sample in each stage of the in-situ experiment are obtained by starting calculation.
9. The method for calculating DVC of a carbon fiber composite material according to claim 1, further comprising: using a conversion SCALAR DVC Output to Volume tool to the three-dimensional displacement field distribution data and the three-dimensional strain field distribution data, correlating the three-dimensional displacement field distribution data and the three-dimensional strain field distribution data with the three-dimensional coordinates of the three-dimensional space data with unified coordinates in the S2, calling a grid view component through AVIZO software to take the grid model as a data mapping carrier, confirming that the grid model and the outline of the carbon fiber composite material sample have no suspension and no recess, and confirming that the grid of the grid model has no empty area, no out-of-range and no unit distortion; And then adjusting a color field tool, setting a range as a correct interval of the three-dimensional displacement field distribution data and the three-dimensional strain field distribution data, adjusting a color map to convert the three-dimensional strain field distribution data into color gradient mapping, superposing the color gradient mapping and the color gradient mapping on a volume rendering model, outputting a second three-dimensional visualization model, and confirming that the displacement direction of the carbon fiber composite material sample is consistent with the loading direction of an in-situ experiment by using the second three-dimensional visualization model, wherein the change rule of a strain area of the carbon fiber composite material sample corresponds to the defect of the carbon fiber composite material sample, and the strain distribution inside the carbon fiber composite material sample accords with the experimental rule.
10. A DVC computing device of a carbon fibre composite material, comprising at least one processor, and a memory communicatively coupled to the at least one processor, the memory storing instructions executable by the at least one processor to enable the at least one processor to perform a DVC computing method of a carbon fibre composite material according to any one of claims 1 to 9.

Description

DVC calculation method and device for carbon fiber composite material Technical Field The invention relates to the technical field of mechanical property test and nondestructive detection of composite materials, in particular to a DVC calculation method and device of a carbon fiber composite material. Background The fiber composite material is widely applied to the fields of aerospace, automobiles, new energy sources and the like due to high specific strength, high specific rigidity and designability. The mechanical properties of the material are highly dependent on microstructure characteristics. In order to understand the damage mechanism of the carbon fiber reinforced polymer matrix composite in depth and optimize the material design, the internal three-dimensional displacement field and the strain field of the carbon fiber reinforced polymer matrix composite need to be precisely quantified, and especially the dynamic response under complex load is required. The traditional research method mostly adopts strain foil, DIC or SEM technology, but only can obtain load-displacement curve of the material in the damage process, strength of the sample, surface strain and macro and micro morphology of fracture after damage. However, the conventional unavailable material is a dynamic evolution process of defects in the stretching process, and a damage evolution mechanism of the material cannot be obtained. DVC (Digital Volume Correlation, digital volume image correlation) technology is an image-based non-destructive deformation measurement technique. The method has the advantages of being capable of characterizing the internal deformation of the material in a non-contact, full-field and three-dimensional mode, and is particularly suitable for heterogeneous materials. Although DVC has great potential, uncontrolled rigid displacement is easy to generate in the experimental process, so that images before and after deformation have global or local translation and rotation, and measurement accuracy is insufficient or results are unreliable. If the grid is not generated according to the actual size of the sample, the edge of the cube cannot be accurately matched with the curved surface or the interface of the fiber, and a 'step-like' artifact is generated in the interface area. The prior patent CN114419284B discloses a three-dimensional reconstruction modeling method of a fiber reinforced composite material based on a CT slice image, which comprises the following steps of acquiring a microscopic slice image of the fiber reinforced composite material by adopting a CT technology, converting the microscopic slice image into a gray level image, binarizing the gray level image of the composite material, identifying the fiber bundle contour of the binary gray level image of the material, smoothing the fiber bundle contour of the gray level image of the material, and three-dimensional reconstruction of the microscopic structure of the fiber reinforced composite material. The patent provides a method for modeling the three-dimensional reconstruction of the microscopic scale, which greatly reduces the cost and error of the artificial three-dimensional reconstruction modeling, can accurately and efficiently quantitatively characterize and model the change of the appearance and the size of the microscopic structure of the material, and has good engineering popularization. However, the invention can only reconstruct the structure form of the microscopic fiber bundle of the composite material, can not reflect the dynamic deformation characteristic of the material in the loading process, can not establish the quantitative relation with the macroscopic mechanical property of the material, and can not realize the quantitative analysis of the internal deformation behavior of the material in different loading stages. The prior patent CN118602969B discloses a three-dimensional strain calculation method and a system for a deformed object, wherein the method comprises the following steps of calculating pixel displacement and deformation gradient of the deformed object after the deformation of the input three-dimensional image by using a zero normalized cross correlation function, decomposing the three-dimensional image into two sub-problems by using an alternate minimization algorithm, defining an augmented Lagrange function, solving a first sub-problem, obtaining a solution of the first sub-problem, solving a second sub-problem, obtaining a solution of the second sub-problem, updating dual variables in the Lagrange function according to the solution of the two sub-problems, adopting the alternate minimization iteration calculation variable to update the Lagrange function, and obtaining a three-dimensional strain random field for the deformation of the object as a global displacement field of the last iteration when the preset iteration stop condition is met. The high-efficiency three-dimensional strain calculation method for the deformed ob