Optimization method for particle characteristic distribution of heterogeneous composite material

Abstract

The invention provides an optimization method for particle feature distribution of a heterogeneous composite material, and the method specifically comprises the steps: obtaining the geometric features of particles in a microscopic structure image according to the maximum Feret diameter, the aspect ratio and the nearest neighbor distance; grouping the geometrical characteristics of the particles according to a certain interval value, and obtaining the characteristic distribution of the particles in the composite material test piece by utilizing a fitting function; according to the particle size of the to-be-optimized model in the microscopic structure image and the particle volume fraction in the to-be-optimized model, by changing the particle feature distribution, obtaining multiple feature distributions of the to-be-optimized particles in the to-be-optimized model; constructing a particle model by determining the center position of the particle, the vertex number of the particle, the rotation center of the particle and the area of the particle in the to-be-optimized model; and checking the to-be-optimized model by applying an external load on the to-be-optimized model, and determining an optimal particle feature distribution function. According to the method, the quasi-realistic property and the calculation precision of the particle characteristics of the multiphase heterogeneous composite material microscopic junction model are improved.

Description

Technical Field

[0001] The present invention relates to the field of computational material science, and in particular to an optimization method for particle characteristic distribution of multiphase heterogeneous composite materials. Background Art

[0002] The modern aerospace field has put forward higher requirements for the excellent properties of materials such as lightweight structure, high specific strength, high specific stiffness, etc. Traditional single materials can no longer meet the stringent use requirements. Particle-reinforced metal matrix composites are composite materials with metals and their alloys as the matrix and non-metallic particles as reinforcement. They have excellent comprehensive properties such as high strength, high thermal conductivity, low expansion and high wear resistance, and show great application potential in aerospace, transportation and other fields.

[0003] At the microscopic scale, particle-reinforced metal matrix composites exhibit complex heterogeneous structures, inhomogeneity, and discontinuity, while their macroscopic properties are closely related to their microstructures. Revealing the relationship between component-structure-performance of composite materials will help provide theoretical guidance and scientific basis for their design, preparation, and processing.

[0004] The particle size, aspect ratio and spatial position of metal matrix composites have important effects on the overall mechanical and processing properties. At present, the microstructure model only considers the fixed value or average value of a certain feature, while ignoring the probability density distribution of the feature value in different intervals. A comprehensive and specific evaluation of the feature distribution is more conducive to achieving the purpose of real material performance and structural optimization design. Summary of the invention

[0005] In view of the problems existing in the prior art, the present invention provides a method for optimizing the characteristic distribution of particles in a multiphase heterogeneous composite material. By extracting the information on the characteristic distribution of particles in the multiphase heterogeneous composite material, the distribution changes of the particle characteristics in each interval are integrated into the microstructure reconstruction model of the composite material, and the unique characteristic distribution of the internal reinforcement of the multiphase heterogeneous composite material is considered more comprehensively and systematically, thereby guiding the optimization of particle characteristic distribution functions such as the maximum Feret diameter, aspect ratio and nearest neighbor distance, and laying a foundation for the optimal design and preparation, processing and manufacturing and promotion and application of the composite material; by reconstructing a highly realistic composite material model, the reconstructed model is subjected to high-throughput simulation calculation and analysis, thereby revealing the damage and failure mechanisms that are difficult to capture by traditional experimental methods, such as the stress-strain distribution changes, crack initiation and propagation of the multiphase heterogeneous composite material during loading, and overcoming the limitation that it is difficult to conduct a systematic study on the structure-performance relationship in experiments.

[0006] The present invention provides a method for optimizing the characteristic distribution of particles of a multiphase heterogeneous composite material, which comprises the following steps:

[0007] S1. After the composite material specimen is roughly ground, polished, cleaned and dried, an optical microscope is used to obtain a microstructure image of the composite material specimen;

[0008] S2, obtaining geometric features of particles in the microstructure image obtained in step S1 according to the maximum Feret diameter, aspect ratio and nearest neighbor distance;

[0009] S3, grouping the geometric features of the particles according to a certain interval value, and using the fitting function to obtain the distribution of particle features in the composite material specimen;

[0010] S4, obtaining multiple characteristic distributions of the particles to be optimized in the model to be optimized according to the particle size of the model to be optimized in the microstructure image and the volume fraction of the particles in the model to be optimized, and by changing the particle characteristic distribution obtained in step S3;

[0011] S5. On the basis of step S4, a particle model is constructed by determining the center position of the particle in the model to be optimized, the number of vertices of the particle, the rotation center of the particle and the area of ​​the particle. The coordinates of any vertex of the particle (x i ,y i ) is:

[0012]

[0013] Among them, r i and θ i are the polar radius and angle of the ith vertex, respectively, and x0 and y0 are the abscissa and ordinate of the center point of the particle, respectively;

[0014] S6. If the distance between the center point of the newly generated particle obtained in step S5 and any of the generated particles is greater than or equal to the sum of the maximum Feret diameters of the two adjacent particles, the newly generated particle is retained; if the distance between the center point of the newly generated particle obtained in step S5 and any of the generated particles is less than the sum of the maximum Feret diameters of the two adjacent particles, the next step of judgment is performed; if the newly generated particle obtained in step S5 intersects or overlaps with a certain edge of any of the generated particles, the newly generated particle needs to be eliminated; if the newly generated particle obtained in step S5 does not intersect or overlap with each edge of any of the generated particles, the next step of judgment is performed; if the newly generated particle obtained in step S5 contains each other with any of the generated particles, the newly generated particle needs to be eliminated; if the newly generated particle obtained in step S5 does not contain each other with any of the generated particles, the newly generated particle is retained;

[0015] S7. Repeat steps S4 and S5 to obtain the volume of particles in the model to be optimized, until the volume of particles in the model to be optimized is equal to the set volume, to obtain the model to be optimized, and calibrate the model to be optimized by applying an external load to the model to be optimized, and determine the optimal particle characteristic distribution function.

[0016] Preferably, in step S1, the microstructure image includes the geometric morphology of the particles, the spatial position distribution of the particles and the volume fraction of the particles.

[0017] Preferably, in step S3, the Gaussian distribution expressions of the maximum Feret diameter, aspect ratio and nearest neighbor distance of the particles are:

[0018]

[0019]

[0020]

[0021] Among them, X Fmaxi ,φ Ai and d NDi are the maximum Feret diameter, aspect ratio and nearest neighbor distance of the ith particle, respectively, and μ and σ are the mathematical expectation and standard deviation of the Gaussian distribution, respectively.

[0022] Preferably, in step S4, the particles in the model to be optimized have multiple characteristic distributions, including characteristic distribution of maximum Feret diameter of the particles, characteristic distribution of aspect ratio of the particles and characteristic distribution of nearest neighbor distance of the particles.

[0023] Preferably, in step S5, the area expression S of the particle is one for:

[0024]

[0025] Where k is the number of edges of the polygonal particle, x i ,y i , x i+1 and i+1 are the x- and y-coordinate values ​​of the ith and i+1th vertices respectively.

[0026] Compared with the prior art, the present invention has the following advantages:

[0027] 1. The present invention is based on the real microstructure, extracts the information of the characteristic distribution of particles in the multiphase heterogeneous composite material, and the particle shape is close to the real microstructure. By controlling the maximum Feret diameter, aspect ratio and nearest neighbor distance of the particles to be consistent with the real distribution function in the multiphase heterogeneous composite material, a highly realistic composite material model is reconstructed, which has a certain effect on improving the model accuracy.

[0028] 2. In traditional modeling theory, when evaluating the influence of particle characteristics, such as maximum Feret diameter, aspect ratio and nearest neighbor distance on material properties, only the case where the particle characteristics are constants is considered, and the influence of the distribution changes of particle characteristics in each interval is ignored. The present invention is based on a Gaussian random distribution model, which integrates the particle characteristic distribution into the microstructure reconstruction model of the composite material, and more comprehensively and systematically considers the unique characteristic distribution of the internal reinforcement of the multiphase heterogeneous composite material, which can guide the optimization of the particle characteristic distribution function such as maximum Feret diameter, aspect ratio and nearest neighbor distance, and lays a foundation for the optimal design and preparation, processing and manufacturing, and promotion and application of composite materials.

[0029] 3. The present invention conducts high-throughput simulation and calculation analysis on the reconstructed model, effectively revealing the damage and failure mechanisms that are difficult to capture by traditional experimental methods, such as the changes in stress-strain distribution, crack initiation and propagation of multiphase heterogeneous composite materials during loading. This overcomes the limitation that it is difficult to conduct systematic research on the structure-performance relationship in experiments, and is of great significance for optimizing material structure and performance.

[0030] 4. The present invention is not only applicable to particle-reinforced composite materials, but also to fiber-reinforced composite materials, especially short-fiber-reinforced composite materials. In this case, the polygonal particles in the above steps are replaced by circles or ellipses. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 A flow chart of the method for optimizing the particle characteristic distribution of a multiphase heterogeneous composite material according to the present invention;

[0032] Figure 2 It is a binary image of the microstructure of SiCp / Al of the multiphase heterogeneous composite material in the method for optimizing the characteristic distribution of particles of the multiphase heterogeneous composite material of the present invention;

[0033] Figure 3 A maximum Feret diameter distribution frequency diagram of particles in the method for optimizing the characteristic distribution of particles of a multiphase heterogeneous composite material according to the present invention;

[0034] Figure 4 A frequency diagram of aspect ratio distribution of particles in the method for optimizing characteristic distribution of particles of a multiphase heterogeneous composite material according to the present invention;

[0035] Figure 5 A frequency diagram of the nearest neighbor distance distribution of particles in the method for optimizing the characteristic distribution of particles of a multiphase heterogeneous composite material according to the present invention;

[0036] Figure 6 A schematic diagram of generating a 4-8 vertex single particle model in the method for optimizing the characteristic distribution of particles of a multi-phase heterogeneous composite material according to the present invention;

[0037] Figure 7A schematic diagram of generating a 3-vertex single particle model in the method for optimizing the characteristic distribution of particles of a multi-phase heterogeneous composite material according to the present invention;

[0038] Figure 8a-8b A comparison diagram of the microstructure image and microstructure model of 2024Al / 20%SiC-30μm in the optimization method for particle characteristic distribution of multiphase heterogeneous composite materials of the present invention;

[0039] Figure 9a-9c They are respectively the microstructure models for optimizing the particle size characteristic distribution in the method for optimizing the particle characteristic distribution of the multiphase heterogeneous composite material of the present invention;

[0040] Figure 10a-Figure 10c They are respectively the microstructure models for optimizing the characteristic distribution of particle aspect ratios in the method for optimizing the characteristic distribution of particles of a multiphase heterogeneous composite material of the present invention;

[0041] Figure 11a-Figure 11c They are respectively microstructure models for optimizing the characteristic distribution of particle agglomeration distance in the method for optimizing the characteristic distribution of particles of multiphase heterogeneous composite materials of the present invention. DETAILED DESCRIPTION

[0042] In order to fully describe the technical content, structural features, objectives and effects of the present invention, the following will be described in detail with reference to the accompanying drawings.

[0043] Optimization methods for particle characteristic distribution of heterogeneous composite materials, such as Figure 1 As shown, the following steps are included:

[0044] S1. Prepare the composite material specimen to the corresponding size, and after rough grinding, polishing, cleaning and drying, use an optical microscope to obtain the microstructure image of the composite material specimen, and use image processing software to perform binarization on the image so that the characteristic distribution of particles in the image is revealed from the metal matrix. Specifically, the microstructure image includes the geometric morphology of the particles, the spatial position distribution of the particles, and the volume fraction of the particles.

[0045] S2. According to the maximum Feret diameter, aspect ratio and nearest neighbor distance, the geometric features of the particles in the microstructure image obtained in step S1 are obtained. Specifically, the geometric features of the particles include: particle size, aspect ratio and spatial distribution.

[0046] Further, in the method of the present invention, the maximum Feret diameter X Fmax To describe the size of the particles, the Feret diameter is defined as the line segment distance between the two parallel lines of the particle projection contour measured along a certain direction, while the maximum Feret diameter is defined as the maximum length of these line segments and is used as the equivalent diameter.

[0047] Aspect Ratio Describes the shape of the particle, which is defined as the ratio of the minimum Feret diameter to the maximum Feret diameter. The specific expression is:

[0048]

[0049] The nearest neighbor distance characterizes the spatial position distribution of particles. It evaluates the degree of particle agglomeration in space by calculating the minimum distance between the center point of a particle and the center point of its adjacent particles.

[0050] S3, set the scale in the image processing software to sort the maximum Feret diameter, aspect ratio and coordinates of the center point of the particles in the binary image, and export the sorted data for grouping statistics. Group the geometric features of the particles according to a certain interval value, extract the particle characteristic distribution, and use the fitting function to obtain the particle characteristic distribution in the composite material specimen. In this step, the composite microstructure image is processed using image processing software, and characteristic distribution data such as particle size, aspect ratio and nearest neighbor distance are extracted, and the fitting function is used to describe the particle characteristic distribution inside the composite material.

[0051] In a preferred embodiment of the present invention, the Gaussian distribution expressions of the maximum Feret diameter, aspect ratio and nearest neighbor distance of the particles are:

[0052]

[0053]

[0054]

[0055] Among them, X Fmaxi ,φ Ai and d NDi are the maximum Feret diameter, aspect ratio and nearest neighbor distance of the ith particle, respectively, and μ and σ are the mathematical expectation and standard deviation of the Gaussian distribution, respectively.

[0056] S4. According to the particle size of the model to be optimized in the microstructure image and the volume fraction of the particles in the model to be optimized, and by changing the particle characteristic distribution obtained in step S3, a plurality of characteristic distributions of the particles to be optimized in the model to be optimized are obtained.

[0057] The various characteristic distributions of particles in the model to be optimized include the characteristic distribution of the maximum Feret diameter of the particles, the characteristic distribution of the aspect ratio of the particles and the characteristic distribution of the nearest neighbor distance of the particles. The characteristic distribution of the maximum Feret diameter of the particles includes single size, multi-size equal distribution and multi-size Gaussian distribution. The characteristic distribution of the aspect ratio of the particles includes single aspect ratio, multi-aspect ratio equal distribution and multi-aspect ratio Gaussian distribution. The characteristic distribution of the nearest neighbor distance of the particles includes nearest neighbor distance of 15μm, nearest neighbor distance of 25μm, nearest neighbor distance of 35μm and Gaussian distribution.

[0058] S5. On the basis of step S4, a particle model is constructed by determining the center position of the particle in the model to be optimized, the number of vertices of the particle, the rotation center of the particle and the area of ​​the particle.

[0059] The coordinates of any vertex of the particle (x i ,y i ) is:

[0060]

[0061] Among them, r i and θ i are the polar radius and angle of the ith vertex, respectively, and x0 and y0 are the abscissa and ordinate of the center point of the particle, respectively.

[0062] The area expression of the particle S one for:

[0063]

[0064] Where k is the number of edges of the polygonal particle, x i ,y i , x i+1 and i+1 are the x- and y-coordinate values ​​of the ith and i+1th vertices respectively.

[0065] S6. Determine whether the newly generated particle interferes with the previously generated particle: use the sum of the maximum Feret's diameters of the two particles as the threshold value, if the distance between the center point of the newly generated particle obtained in step S5 and any of the generated particles is greater than or equal to the sum of the maximum Feret's diameters of the two adjacent particles, then retain the newly generated particle; if the distance between the center point of the newly generated particle obtained in step S5 and any of the generated particles is less than the sum of the maximum Feret's diameters of the two adjacent particles, then it is necessary to determine the possible edge intersection and particle mutual inclusion to prevent edge intersection and mutual inclusion between particles; if the newly generated particle obtained in step S5 intersects or overlaps with a certain edge of any of the generated particles, then the newly generated particle needs to be eliminated; if the newly generated particle obtained in step S5 does not intersect or overlap with each edge of any of the generated particles, then determine whether the newly generated particle and the previously generated particle are mutually included to prevent particle inclusion; if the newly generated particle obtained in step S5 and any of the generated particles are mutually included, then the newly generated particle needs to be eliminated; if the newly generated particle obtained in step S5 and any of the generated particles are not mutually included, then retain the newly generated particle.

[0066] S7. Repeat steps S4 and S5 to obtain the volume of the particles in the model to be optimized until the volume of the particles in the model to be optimized is equal to the set volume, and the model to be optimized is obtained. The model to be optimized is calibrated by applying an external load to the model to be optimized. Through high-throughput simulation, the damage and failure process of the multiphase heterogeneous composite material and the particle strengthening mechanism, the internal stress and strain distribution, and the crack initiation and propagation process of the multiphase composite material during processing or when subjected to external loads are analyzed, the damage and failure mechanism is studied, and the characteristic distribution of particles to achieve optimal performance is determined.

[0067] The model building method of the present invention can be applied to a variety of particle-reinforced composite materials, filling the gap in the particle characteristic distribution theory, laying a foundation for optimizing the particle characteristic distribution, and having guiding significance for the actual design, preparation, processing and application of multiphase composite materials; it overcomes the limitations of experimental conditions, helps to reveal the structure-performance relationship of composite materials, improves the mechanical properties of composite materials, and promotes their wide application in various fields.

[0068] The following is a further description of a method for optimizing the characteristic distribution of particles of a multiphase heterogeneous composite material according to the present invention in conjunction with an embodiment:

[0069] The composite material test piece in the method of the present invention can be prepared by purchasing raw materials and processing them experimentally, or directly purchasing finished products.

[0070] S1. Prepare the composite material specimen to the corresponding size, and after rough grinding, polishing, cleaning and drying, obtain the metallographic micrograph of the workpiece material, and after clearing it, perform binarization processing based on the threshold segmentation method to obtain Figure 2 The binary image shown, where the black phase is the reinforcing particles and the white phase is the metal matrix.

[0071] S2, use the maximum Feret diameter X Fmax Describe the size of the particles:

[0072] X Fmax =max(X F-1 ,X F-2 ,…X F-z )

[0073] Here, the parameter z is the number of angular intervals taken on the particle contour.

[0074] Use Aspect Ratio Describes the shape of the particles. The value varies from 0 to 1. The larger the aspect ratio, the closer the particle shape is to being short and thick. The smaller the aspect ratio, the closer the particle shape is to being slender.

[0075] The nearest neighbor distance distribution is used to describe the agglomeration of particles in space. The center points of the particles are extracted from the binary image, and the distance between the center points is used instead of the distance between the particles.

[0076] S3. Set a scale in the image processing software to sort out the characteristic data such as the maximum Feret diameter, aspect ratio and nearest neighbor distance of the particles in the binary image, and export them for grouping statistics. In this embodiment, the particle size is a value ranging from 0 to 90 μm, and the average value is 24.18 μm; the aspect ratio is distributed from 0.1 to 0.9, mainly concentrated in 0.3 to 0.7, with an average value of 0.53; the nearest neighbor distance is distributed from 0 to 50 μm, mainly concentrated in 10 to 25 μm. The Feret diameter is divided into seven groups, the aspect ratio is divided into eight groups, and the nearest neighbor distance is divided into twelve groups.

[0077] S4. Based on the statistical data of the particle characteristic distribution in step S3, a fitting function is used to fit the distribution. In this embodiment, the distribution of the maximum Feret diameter, aspect ratio and nearest neighbor distance of the particles in the composite material specimen is described by a Gaussian distribution function:

[0078]

[0079]

[0080]

[0081] Among them, X Fmaxi ,φ Ai and d NDi are the maximum Feret diameter, aspect ratio and nearest neighbor distance of the ith particle, μ and σ are the mathematical expectation and standard deviation of the Gaussian distribution, respectively. The statistical results and fitting functions of the final particle characteristic distribution (maximum Feret diameter, aspect ratio and nearest neighbor distance) are as follows: Figure 3 , Figure 4 and Figure 5 shown.

[0082] S5. On the basis of step S4, a particle model is constructed by determining the center position of the particle in the model to be optimized, the number of vertices of the particle, the rotation center of the particle and the area of ​​the particle.

[0083] In this embodiment, the control variable method is adopted to change only the characteristic distribution function of the particles in the model to be optimized, and the model size, particle volume fraction, and other particle characteristic distributions of the model to be optimized are controlled to be constant values.

[0084] In this embodiment, under the premise that the average maximum Feret diameter of the particles is 30 μm, three characteristic distribution functions are considered for the particle size distribution: single size, multi-size equal distribution (the number of particles in different particle size ranges is equal) and multi-size Gaussian distribution (consistent with the actual microstructure).

[0085] On the premise that the average aspect ratio of the particles is 0.45, three characteristic distribution functions are considered for the aspect ratio distribution of the particles: single aspect ratio, equal distribution of multiple aspect ratios (the number of particles in different particle aspect ratio intervals is equal), and Gaussian distribution of multiple aspect ratios (consistent with the real microstructure).

[0086] The spatial position distribution of the particles is characterized by the degree of particle agglomeration, and the aggregation degree of the particles is quantified by controlling the nearest neighbor distance of the particle center points. In this embodiment, it is assumed that each agglomerate consists of five particles, and each agglomerate consists of a central particle and four particles surrounding it. The nearest neighbor distances between the particles in the agglomerate are set to 15, 25, 35 μm and Gaussian distribution respectively.

[0087] For particles with the number of vertices ranging from 4 to 8, the particle center point (x0, y0) is selected as the origin to establish a Cartesian coordinate system. Coordinate points A (y A >y0) and B (y B <y0) are selected on the y-axis, and coordinate points C (x C <x0) and D (x D >x0) are selected on the x-axis, and the criterion is satisfied: (l OA =l OB )>=(l OC =l OD )). Where l AB and l CD are the maximum and minimum Feret diameters of the particle respectively. l CD / l AB is the aspect ratio of the particle.

[0088] From Figure 6 0 or 1 point is selected from each of the shown sub-intervals M1, M2, M3 and M4, and then all the selected points are connected to form a polygonal particle. The coordinate value (x i , y i ) of the i-th vertex is:[[]]

[0089]

[0090] where r i and θ i are the polar radius and angle of the i-th vertex respectively, and x0 and y0 are the abscissa and ordinate of the center point of the particle respectively.

[0091] Finally, with the particle center as the rotation center, the particle is rotated by an arbitrary angle to obtain particles with arbitrary orientations and shapes.

[0092] For particles with the number of vertices of 3, as Figure 7 shown, points C and D are selected on the x-axis, l CDis equal to the minimum Feret diameter, and point C coincides with the origin O. Select points A and B on the y-axis, and l OA = l OB , l AB Equal to the maximum Feret diameter.

[0093] S6. Determine whether the newly generated particle interferes with the previously generated particle: use the sum of the maximum Feret's diameters of the two particles as the threshold value, if the distance between the center point of the newly generated particle obtained in step S5 and any of the generated particles is greater than or equal to the sum of the maximum Feret's diameters of the two adjacent particles, then retain the newly generated particle; if the distance between the center point of the newly generated particle obtained in step S5 and any of the generated particles is less than the sum of the maximum Feret's diameters of the two adjacent particles, then it is necessary to determine the possible edge intersection and particle mutual inclusion to prevent edge intersection and mutual inclusion between particles; if the newly generated particle obtained in step S5 intersects or overlaps with a certain edge of any of the generated particles, then the newly generated particle needs to be eliminated; if the newly generated particle obtained in step S5 does not intersect or overlap with each edge of any of the generated particles, then determine whether the newly generated particle and the previously generated particle are mutually included to prevent particle inclusion; if the newly generated particle obtained in step S5 and any of the generated particles are mutually included, then the newly generated particle needs to be eliminated; if the newly generated particle obtained in step S5 and any of the generated particles are not mutually included, then retain the newly generated particle.

[0094] S7, repeating steps S4 and S5, when the total area (volume) of the generated particles reaches the set area (volume), the construction of the particle model is completed, and the matrix model is obtained by using Boolean subtraction. Figure 8b It is a microstructure model based on the real microstructure with a particle volume fraction of 20% and a particle size of 30μm. Figure 8a It is a microscopic image of the composite material specimen. In order to obtain the optimal particle characteristic distribution function, the control variable method is used to establish the maximum Feret diameter, aspect ratio and nearest neighbor distance characteristic distribution optimization model. Only the distribution function of the optimized feature changes, and other geometric features are fixed values, as shown in Figures 9 to 11. Figure 9a The maximum Feret diameter distribution optimization microstructure model is established when the particle aspect ratio is 0.45, the volume fraction is 20%, the nearest neighbor distance distribution satisfies the Gaussian distribution function, and the maximum Feret diameter of the particle is a single value. Figure 9b The maximum Feret diameter distribution optimization microstructure model is established when the particle aspect ratio is 0.45, the volume fraction is 20%, and the nearest neighbor distance distribution satisfies the Gaussian distribution function, and the maximum Feret diameter of the particle is a multi-size Gaussian distribution. Fig.9cThe maximum Feret diameter distribution optimized microstructure model is established when the particle aspect ratio is 0.45, the volume fraction is 20%, the nearest neighbor distance distribution satisfies the Gaussian distribution function, and the maximum Feret diameter of the particles is a multi-size equal distribution. Fig.10a The aspect ratio distribution optimization microstructure model is established when the maximum Feret diameter of the particle is 30 μm, the volume fraction is 20%, the nearest neighbor distance satisfies the Gaussian distribution function, and the particle aspect ratio is a single value. Fig.10b The aspect ratio distribution optimization microstructure model is established when the maximum Feret diameter of the particle is 30 μm, the volume fraction is 20%, the nearest neighbor distance satisfies the Gaussian distribution function, and the particle aspect ratio is a multi-aspect ratio Gaussian distribution. Fig.10c An aspect ratio distribution optimized microstructure model is established when the maximum Feret diameter of the particles is 30 μm, the volume fraction is 20%, the nearest neighbor distance satisfies the Gaussian distribution function, and the aspect ratio of the particles is evenly distributed with multiple aspect ratios. Fig.11a The nearest neighbor distance distribution optimized microstructure model was established for particles with an aspect ratio of 0.45, a maximum Feret diameter of 30 μm, a volume fraction of 20%, and a particle agglomeration distance of 15 μm. Fig.11b The nearest neighbor distance distribution optimized microstructure model was established for particles with an aspect ratio of 0.45, a maximum Feret diameter of 30 μm, a volume fraction of 20%, and a particle agglomeration distance of 25 μm. Fig.11c The nearest neighbor distance distribution optimized microstructure model was established for particles with an aspect ratio of 0.45, a maximum Feret diameter of 30 μm, a volume fraction of 20%, and a particle agglomeration distance of 35 μm.

[0095] Using simulation analysis methods, external loads are applied to the constructed multiphase heterogeneous composite microstructure model to study the stress-strain distribution changes, crack initiation and propagation processes of the composite during processing or loading, and analyze its damage and failure mechanism. Through high-throughput numerical calculations, the simulation results of different particle characteristic distribution models are compared to determine the optimal particle characteristic distribution.

[0096] In summary, the modeling method considering the distribution of particle characteristics proposed in this embodiment has a good effect on optimizing the structure of multiphase heterogeneous composite materials.

[0097] (1) This embodiment analyzes the real microstructure of the multiphase heterogeneous composite material, extracts the information of the characteristic distribution of particles in the multiphase heterogeneous composite material, and constructs a highly realistic microstructure model of the particle-reinforced composite material;

[0098] (2) The concept of particle characteristic distribution proposed in this embodiment takes into account the heterogeneity of the internal reinforcement of the multiphase heterogeneous composite material in a more comprehensive and systematic manner, and can guide the optimization of particle characteristic distribution functions such as maximum Feret diameter, aspect ratio and nearest neighbor distance, providing a new idea for optimizing material structure and performance.

[0099] (3) This embodiment overcomes the limitations of experiments that make it difficult to systematically study the structure-performance relationship and to capture the damage failure process, and predicts the failure of composite materials with different particle characteristic distributions, thereby optimizing the material design, processing and preparation process.

[0100] The embodiments described above are only descriptions of the preferred implementation modes of the present invention, and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A method for optimizing the characteristic distribution of particles of a heterogeneous composite material, characterized in that: It includes the following steps: S1. After the composite material specimen is roughly ground, polished, cleaned and dried, an optical microscope is used to obtain a microstructure image of the composite material specimen; S2, obtaining geometric features of particles in the microstructure image obtained in step S1 according to the maximum Feret diameter, aspect ratio and nearest neighbor distance; S3, grouping the geometric features of the particles according to a certain interval value, and using the fitting function to obtain the distribution of particle features in the composite material specimen; S4, obtaining multiple characteristic distributions of the particles to be optimized in the model to be optimized according to the particle size of the model to be optimized in the microstructure image and the volume fraction of the particles in the model to be optimized, and by changing the particle characteristic distribution obtained in step S3; S5. On the basis of step S4, a particle model is constructed by determining the center position of the particle in the model to be optimized, the number of vertices of the particle, the rotation center of the particle, and the area of ​​the particle. The coordinates of any vertex of the particle (x i ,y i ) is: Among them, r i and θ i are the polar radius and angle of the ith vertex, respectively, and x0 and y0 are the abscissa and ordinate of the center point of the particle, respectively; S6. If the distance between the center point of the newly generated particle obtained in step S5 and any of the generated particles is greater than or equal to the sum of the maximum Feret diameters of the two adjacent particles, the newly generated particle is retained; if the distance between the center point of the newly generated particle obtained in step S5 and any of the generated particles is less than the sum of the maximum Feret diameters of the two adjacent particles, the next step of judgment is performed; if the newly generated particle obtained in step S5 intersects or overlaps with a certain edge of any of the generated particles, the newly generated particle needs to be eliminated; if the newly generated particle obtained in step S5 does not intersect or overlap with each edge of any of the generated particles, the next step of judgment is performed; if the newly generated particle obtained in step S5 contains each other with any of the generated particles, the newly generated particle needs to be eliminated; if the newly generated particle obtained in step S5 does not contain each other with any of the generated particles, the newly generated particle is retained; S7. Repeat steps S4 and S5 to obtain the volume of particles in the model to be optimized, until the volume of particles in the model to be optimized is equal to the set volume, to obtain the model to be optimized, and calibrate the model to be optimized by applying an external load to the model to be optimized, and determine the optimal particle characteristic distribution function.

2. The method for optimizing the characteristic distribution of particles of a multiphase heterogeneous composite material according to claim 1, characterized in that: In step S1, the microstructure image includes the geometric morphology of the particles, the spatial position distribution of the particles and the volume fraction of the particles.

3. The method for optimizing the characteristic distribution of particles of a multiphase heterogeneous composite material according to claim 1, characterized in that: In step S3, the Gaussian distribution expressions of the maximum Feret diameter, aspect ratio and nearest neighbor distance of the particles are: Among them, X Fmaxi ,φ Ai and d NDi are the maximum Feret diameter, aspect ratio and nearest neighbor distance of the ith particle, respectively, and μ and σ are the mathematical expectation and standard deviation of the Gaussian distribution, respectively.

4. The method for optimizing the characteristic distribution of particles of a multiphase heterogeneous composite material according to claim 1, characterized in that: In step S4, multiple characteristic distributions of the particles in the model to be optimized include characteristic distribution of the maximum Feret diameter of the particles, characteristic distribution of the aspect ratio of the particles, and characteristic distribution of the nearest neighbor distance of the particles.

5. The method for optimizing the characteristic distribution of particles of a multiphase heterogeneous composite material according to claim 1, characterized in that: In step S5, the area expression S of the particle is one for: Where k is the number of edges of the polygonal particle, x i ,y i , x i+1 and i+1 are the x- and y-coordinate values ​​of the ith and i+1th vertices, respectively.

Images