A method and system for creating a composite material RVE oriented to a concave reinforcement

Abstract

The application discloses a kind of composite RVE creation method and system for concave reinforcement, it is related to composite microstructure modeling and numerical simulation technical field, including steps: according to RVE parameter, establish two-dimensional rectangular RVE region;Based on two-dimensional rectangular RVE region, obtain the reinforcement set allowing initial intersection;Orientation adjustment is executed to each reinforcement, and convex decomposition is executed;Separation termination condition step is iteratively executed, when the maximum penetration depth between all reinforcement pairs is less than the preset separation tolerance, iteration is terminated, and two-dimensional periodic RVE satisfying non-overlapping constraint and periodic boundary condition is obtained.The application can always keep the original return boundary characteristics of concave reinforcement intact, without using convex approximation, so as to ensure that the generated RVE has geometric authenticity, statistical consistency and periodic consistency, and provides a high-credibility geometric model for subsequent composite material homogenization analysis and mesoscopic damage evolution prediction.

Description

Technical Field

[0001] This invention relates to the field of composite material microstructure modeling and numerical simulation technology, and in particular to a method and system for creating RVEs of composite materials for concave reinforcements. Background Technology

[0002] Composite materials possess advantages such as high specific strength, high specific stiffness, low coefficient of thermal expansion, and good thermal conductivity, and have been widely used in aerospace, defense equipment, and automotive manufacturing. In predicting the equivalent properties and optimizing the structure of composite materials, it is essential to construct a Reliable Value Evaluator (RVE) that can accurately characterize the microstructure of the composite material.

[0003] Existing methods for creating real-time vesicle (RVE) structures in composite materials mainly include statistical generation methods such as random sequential adsorption, Monte Carlo methods, molecular dynamics methods, finite element compression methods, and position optimization methods. These methods are suitable for parametric studies and batch numerical calculations, and have become an important technical route for microscopic modeling of composite materials. For convex reinforcements such as circular, elliptical, and general convex polygons, existing methods are relatively mature in terms of intersection determination, non-overlap constraints, and periodic arrangement. However, actual composite reinforcements often exhibit non-convex morphologies such as concave polygons, bilobal shapes, pea shapes, and dumbbell shapes, with obvious concave boundaries and re-entry characteristics, which significantly affect the local field distribution and damage evolution behavior in RVEs.

[0004] In the aforementioned existing technologies for processing concave reinforcements, the difficulty of processing concave reinforcements is usually reduced by simplifying them into convex hull contours, thereby improving the construction of RVE. However, the non-convex boundaries of concave reinforcements make it easy for reinforcements to form multi-point contacts, narrow gaps, and local nesting. When simplifying the convex hull contour, existing technologies cause the concave boundaries to lose key geometric features, making it difficult for intersection detection and position correction methods to stably and accurately achieve the non-overlapping periodic arrangement of all reinforcements. This can easily lead to missed detections, misjudgments, oscillations, or local jamming, reducing the reliability of subsequent homogenization and damage analysis results. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of the prior art by providing a method and system for creating RVEs of composite materials oriented towards concave reinforcements, thereby solving the problems in the prior art.

[0006] The present invention specifically provides the following technical solution: A method for creating RVEs (Reverse Velocities) in composite materials oriented towards concave reinforcements, comprising the following steps: Obtain the RVE parameters of the composite material, and establish a two-dimensional rectangular RVE region based on the composite material RVE parameters; Within the two-dimensional rectangular RVE region, the initial center coordinates and initial orientation angles of multiple concave reinforcements are randomly generated to obtain a set of candidate initially intersecting reinforcements; For each concave reinforcement in the reinforcement set, perform convex decomposition to transform it into a set of interconnected convex subdomains; The iteration of the enhancement pairs is terminated by a separation condition step, specifically: based on the minimum mirror principle, the closest relative position relationship between any two enhancements under periodic boundary conditions is determined, and based on the result of the convex decomposition, the intersection of the convex subdomain sets of the two enhancements is determined; if an intersecting pair of convex subdomains is determined, the penetration depth and corresponding separation direction of the convex subdomains are obtained, and the separation vector between the enhancement pairs is determined; the separation vectors between each enhancement and all neighboring enhancements with intersecting relationships are aggregated to obtain the total translation increment of the enhancement, and the center coordinates are updated according to the total translation increment; periodic mapping is performed on enhancements that exceed the boundary of the two-dimensional rectangular RVE region after the center coordinates are updated, and they re-enter the RVE region; when the maximum penetration depth between all enhancement pairs is less than the preset separation tolerance, the iteration is terminated, and a two-dimensional periodic RVE that satisfies the non-overlapping constraint and periodic boundary conditions is obtained.

[0007] Preferably, each concave reinforcement in the reinforcement set is subjected to convex decomposition, transforming it into a set of interconnected convex subdomains, specifically as follows: Triangulation is performed on the concave polygons in multiple input concave reinforcements, and convex merging is performed on adjacent triangular units obtained by triangulation to obtain a set of several interconnected convex subdomains; wherein the convex merging is performed while keeping the merged region still a convex domain.

[0008] Preferably, the triangulation adopts a hybrid triangulation strategy combining the divide-and-conquer method and the ear-cutting method. When the divide-and-conquer method cannot find a valid diagonal that meets the conditions, the ear-cutting method is switched to continue the triangulation.

[0009] Preferably, before terminating the iteration, the process further includes: A two-stage circumcircle screening mechanism is introduced during the reinforcement intersection detection process, specifically: First-level screening: Establish a circumcircle at the overall level of the concave reinforcement. If the circumcircles of the two reinforcements do not intersect, it is directly determined that they do not interact. Second-level screening: A circumcircle is established at the convex subdomain level, and exact intersection determination and penetration solution are performed only on convex subdomain pairs that pass both the first and second-level screening.

[0010] Preferably, when performing convex decomposition on each concave reinforcement in the reinforcement set, the method further includes: Establish a deviation objective function between the current fourth-order orientation tensor of all augmentations and the target fourth-order orientation tensor; Based on the bias objective function, the orientation vector of each augmentation is iteratively updated to complete the orientation correction of each augmentation while maintaining the unit orientation vector constraint, thus obtaining the corrected augmentation, which serves as the input for convex decomposition.

[0011] Preferably, when obtaining the candidate initial intersection set of enhancers, the method further includes: The vertex coordinates of each augmentation relative to its centroid are scaled up uniformly by a proportional factor to control the minimum spacing between adjacent augmentations.

[0012] Preferably, the step of establishing a two-dimensional rectangular RVE region based on the RVE parameters further includes: The number of reinforcements is obtained by rounding down the product of the target reinforcement volume fraction of the composite material and the dimensions in different coordinate directions of the RVE; where the dimensions in different coordinate directions of the RVE are the dimensions of the two-dimensional rectangular region.

[0013] This invention provides a composite material RVE creation system for concave reinforcements, comprising: The data acquisition module is used to obtain the RVE parameters of the composite material and to establish a two-dimensional rectangular RVE region based on the RVE parameters of the composite material. The augmentation acquisition module is used to randomly generate the initial center coordinates and initial orientation angles of multiple concave augmentations within the two-dimensional rectangular RVE region, thereby obtaining a set of candidate initially intersecting augmentations; The correction module is used to perform convex decomposition on each concave reinforcement in the reinforcement set, transforming it into a set of several interconnected convex subdomains; The iterative module is used to perform a separation termination condition step on the reinforcement pair iteration. Specifically, it involves: determining the closest relative position relationship between any two reinforcements under periodic boundary conditions based on the minimum mirror principle, and performing an intersection determination on the convex subdomain sets of the two reinforcements based on the result of the convex decomposition; if an intersecting convex subdomain pair is determined, obtaining the penetration depth and corresponding separation direction of the convex subdomain, and determining the separation vector between the reinforcement pairs; aggregating the separation vectors between each reinforcement and all neighboring reinforcements with intersecting relationships to obtain the total translation increment of the reinforcement, and updating the center coordinates according to the total translation increment; performing periodic mapping on reinforcements that exceed the boundary of the two-dimensional rectangular RVE region after the center coordinate update, and re-entering the RVE region; terminating the iteration when the maximum penetration depth between all reinforcement pairs is less than the preset separation tolerance, thus obtaining a two-dimensional periodic RVE that satisfies the non-overlapping constraint and periodic boundary conditions.

[0014] Compared with the prior art, the present invention has the following significant advantages: The method provided by this invention decomposes a concave reinforcement into a set of several convex subdomains and establishes the entire iterative separation process of non-overlapping constraints entirely on the accurate intersection determination and penetration depth calculation of the convex subdomain pairs. The iteration terminates when the maximum penetration depth between all reinforcement pairs is less than the preset separation tolerance. The entire process strictly satisfies the non-overlapping constraints and periodic boundary conditions, thereby ensuring that the true concave boundary of the reinforcement is strictly preserved and accurately processed without any convex hull simplification. This comprehensively overcomes the omissions, misjudgments, oscillations, and local jamming caused by simplified geometry in the background technology. It stably and accurately realizes the strict non-overlapping arrangement of all reinforcements under periodic boundary conditions, providing an RVE model with both high geometric fidelity and high numerical reliability for subsequent homogenization and damage analysis. Attached Figure Description

[0015] Figure 1 This is the overall flow of a composite material RVE creation method for concave reinforcement according to the present invention; Figure 2 This invention allows for the random initialization configuration of the enhancer to initially intersect; Figure 3 This refers to the process of triangulation, convex decomposition, and convex merging of the concave reinforcement in this invention; Figure 4 This is the periodic minimum mirror interaction detection process in this invention; Figure 5 This invention describes the process of using the GJK and EPA algorithms for intersection determination and penetration depth calculation; wherein, Figure 5 (a) shows the process of intersection determination using GJK. Figure 5 (b) shows the process of solving the penetration depth using the GJK algorithm. Figure 5 (c) shows the process of intersection determination using EPA. Figure 5 (d) represents the process of solving the penetration depth using the EPA algorithm; Figure 6 The orientation tensor a generated using the method of this invention is... t =diag[0.5, 0.5], volume fraction v1=0.60, two-dimensional periodic RVE of concave polygonal reinforcement; Figure 7 The orientation tensor a in this invention t Two-dimensional periodic RVEs with bilobal and pea-shaped reinforcements of size diag[1.0, 0.0] and volume fraction v1=0.60; where, Figure 7 (a) is a two-dimensional periodic RVE with a bilobal reinforcement. Figure 7 (b) is a two-dimensional periodic RVE with a pea-shaped reinforcement. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0017] Reinforcement velocimetry (RVE) typically needs to simultaneously reflect information such as reinforcement volume fraction, geometry, size distribution, spatial distribution, orientation state, and periodic mirroring, thus providing a reliable geometric basis for homogenization analysis and mesoscopic damage simulation. Based on the characteristics of RVE, this invention provides an RVE creation method for concave reinforcement composites. While preserving the non-convex geometric features of the reinforcement, it achieves efficient arrangement and strict non-overlap control under high volume fraction conditions, while simultaneously satisfying periodic boundary constraints, reinforcement shape preservation, and spatial statistical distribution requirements. This provides reliable geometric model support for predicting the equivalent properties of composite materials, microfield analysis, and mesoscopic damage evolution simulation.

[0018] This invention provides a method for creating RVE (Reverse Velocity Enhancement) in composite materials oriented towards concave reinforcements, comprising the following steps: Step 1: Obtain the RVE parameters of the composite material, establish a two-dimensional rectangular RVE region based on the composite material RVE parameters, and determine the number of reinforcements.

[0019] The input parameters should include at least: RVE dimensions l1 and l2, and the volume fraction of the target reinforcement in the composite material. v 1. Target Second-Orientation Tensor a t Area of ​​a single reinforced body s And the geometric description of the reinforcement.

[0020] The two-dimensional rectangular region ω is constructed based on the dimensions of the RVE in different coordinate directions. The specific expression is as follows: ω=[−0.5l1, 0.5l1]×[−0.5l2, 0.5l2]; Where l1 and l2 are the dimensions of RVE in two coordinate directions, respectively.

[0021] Furthermore, the number of reinforcements is obtained by rounding down the product of the target reinforcement volume fraction of the composite material and the dimensions in different coordinate directions of the RVE. n The dimensions of RVE in different coordinate directions are the dimensions of a two-dimensional rectangular region; the specific expression is: ; in, This indicates rounding down to the nearest integer.

[0022] Step 2: Randomly generate the initial center coordinates and initial orientation angles of multiple concave reinforcements within the RVE region to obtain a set of candidate initial intersecting reinforcements.

[0023] The initial center coordinates are randomly sampled within the RVE region, and the initial orientation angle is randomly sampled within [0, 2π). This step allows for initial intersection between different reinforcements.

[0024] To control the minimum spacing between adjacent reinforcements, a temporary scaling can be applied to the reinforcement geometry during intersection elimination, that is, the vertex coordinates of each reinforcement relative to its centroid are uniformly enlarged by a scaling factor λ. Preferably, the scaling factor λ is between 1.01 and 1.10; more preferably, λ is 1.025 or 1.05.

[0025] Step 3: Perform orientation adjustment on each enhancer in the enhancer set to obtain the corrected enhancer, and perform convex decomposition on each concave enhancer in the corrected enhancer set to transform it into a set consisting of several interconnected convex subdomains.

[0026] Orientation adjustment is performed on all augmentations so that the actual orientation statistics of the augmentation set approach the preset target orientation statistics.

[0027] When performing convex decomposition on each concave reinforcement in the reinforcement set, it also includes: Establish the objective function for the deviation between the current fourth-order orientation tensor of all augmentations and the target fourth-order orientation tensor. The specific expression is: ; in, For the target fourth-order orientation tensor, Let p be the actual fourth-order orientation tensor of all augmentations; based on the bias objective function, normalized gradient descent is then used to apply the orientation vector p of each augmentation. i Iterative updates are performed to correct the orientation of each augmentation while maintaining the unit orientation vector constraint, resulting in the corrected augmentation. As the input for convex decomposition, the specific expression is: ; Where k represents the iteration step, η a This indicates the step size for updating the orientation vector; f Regarding the enhancement orientation vector p i The gradient can be written as: ; Preferably, the orientation correction iteration uses a step size parameter η. aThe termination condition for orientation correction can be set as f ≤ ε a Or reach the maximum number of iterations N a,max ,in ε a 10 can be taken −6 N a,max 10,000 is acceptable.

[0028] Step 4: Perform a separation termination condition step for the augmentation pair iteration, specifically: determine the closest relative position relationship between any two augmentations under periodic boundary conditions based on the minimum mirror principle, and perform an intersection determination on the convex subdomain sets of the two augmentations based on the result of the convex decomposition; if an intersecting convex subdomain pair is determined, obtain the penetration depth and corresponding separation direction of the convex subdomain, and determine the separation vector between the augmentation pairs; aggregate the separation vectors between each augmentation and all neighboring augmentations with intersecting relationships to obtain the total translation increment of the augmentation, and update the center coordinates according to the total translation increment; perform periodic mapping on augmentations that exceed the boundary of the two-dimensional rectangular RVE region after the center coordinates are updated, and re-enter the RVE region; when the maximum penetration depth between all augmentation pairs is less than the preset separation tolerance, terminate the iteration, and obtain a two-dimensional periodic RVE that satisfies the non-overlapping constraint and periodic boundary conditions.

[0029] Preferably, convex decomposition includes the following two stages: (1) Perform triangulation on the concave polygons in the multiple input concave augmentations.

[0030] (2) Perform convex merging on adjacent triangular units obtained by triangulation to reduce the number of final convex subdomains and obtain a set of several interconnected convex subdomains.

[0031] More preferably, the triangulation employs a hybrid partitioning strategy combining divide-and-conquer and ear-cutting methods; when the divide-and-conquer method cannot find a valid diagonal that meets the conditions, it switches to ear-cutting to continue partitioning; convex merging is performed under the conditions that the merged region remains convex, accurately covers the original union, and does not introduce new vertices; corresponding illustrations. Figure 3 .

[0032] Preferably, the circumscribed concave polygon construction includes boundary sampling, redundant point deletion, vertex order correction, and iterative outer wrapping update to ensure that the original curve boundary of the constructed circumscribed concave polygon envelope is maintained.

[0033] For any two augmentations i and j, determine their nearest periodic relative displacement vector based on the minimum mirror image principle. k ij The specific expression is: ; in, L=( l 1, l 2), nint(·) represents taking the nearest integer by component, and ⊙ and ⊘ represent multiplication and division by components, respectively. The fractional coordinate difference between the two augmentations. Minimal mirror interaction illustration. Figure 4 .

[0034] Preferably, the intersection determination of convex subdomains adopts the GJK algorithm, which constructs the Minkowski difference C between two convex subdomains, specifically expressed as: ; Among them, A u and B v Let a and b be two convex subdomains to be detected, and a and b be position vectors.

[0035] And determine whether the origin lies within the Minkowski difference to achieve this.

[0036] When determining that a pair of convex subdomains intersects, the EPA algorithm is used to calculate the penetration depth of that pair of convex subdomains. δ uv and its corresponding separation direction unit normal n uv Thus, the separation vector of the pair of convex subdomains is obtained. s uv The specific expression is: ; When multiple pairs of intersecting convex subdomains exist within the same pair of concave reinforcements, the pair of convex subdomains with the largest penetration depth is selected as the separation criterion between the pair of concave reinforcements. (See diagram for intersection determination and penetration depth calculation.) Figure 5 .

[0037] Preferably, the total translational increment of reinforcement i Represented as: ; Where, N i Let i be the set of adjacent reinforcements that currently intersect or come into contact with reinforcement i. Let i be the separation vector of the two enhancers i and j; its position at step k+1. and the position at step k The update relationship is represented as: ; Where, η c This indicates the update step size of the augmentation center.

[0038] For augmentations that extend beyond the RVE region boundary after the update, a periodic mapping is performed, and the back-mapping relationship is expressed as follows: ; in, This represents the value after mapping.

[0039] Repeat the steps, and set the maximum number of iterations N. sep,max ; Set reference length l ref : ; And set separation tolerance : ; When the maximum penetration depth of all reinforcement pairs is less than ε sep (i.e. δ) max / l ref ≤α sep The iteration stops when N(t) is reached, thus obtaining a two-dimensional periodic RVE that satisfies the strict non-overlapping constraint and periodic boundary conditions; the maximum number of iterations N(t) is [not specified]. sep,max 1000 can be taken. When N is reached. sep,max Still not satisfied with ε sep At that time, adjust the temporary scaling factor λ and the augmentation center update step size η. c Then, repeat steps 2 to 4, or re-initialize the augmentation body positions randomly and continue iterating.

[0040] A two-stage circumcircle screening mechanism is introduced during the reinforcement intersection detection process.

[0041] The first stage establishes a circumcircle at the overall level of the concave reinforcement. If the circumcircles of two reinforcements do not intersect, it is directly determined that the two reinforcements do not interact. The second stage establishes a circumcircle at the level of the convex subdomain. The precise intersection determination and penetration solution are performed only on the convex subdomain pairs that have passed the second stage screening.

[0042] In one embodiment: Periodic RVE creation of concave polygon augmentations: Construct a square RVE region with side lengths satisfying l1=l2. Assume the target augmentation volume fraction is v1=60.0%, and the target augmentation second-order orientation tensor... a t for: ; In this embodiment, the reproducible parameter settings are as follows: (a) Number of reinforcing bodies according to (a) Rounding down to the nearest integer; (b) Temporary scaling factor λ = 1.025; (c) Orientation correction step size η. a =1.0, orientation correction termination tolerance is ε a =10 −6 The maximum number of iterations is N. a,max=10000; (d) The center update step size is η. c =1.0, the separation termination tolerance is ε. sep =10 −6 l ref The maximum number of iterations is N. sep,max =1000.

[0043] First, the center positions of a corresponding number of concave polygonal reinforcements are randomly generated within the RVE region, and each reinforcement is assigned a random initial orientation angle, allowing reinforcements to overlap in the initial configuration.

[0044] Then, based on the preset target orientation tensor, the orientation of all augmentations is corrected using the normalized gradient descent method, so that the actual orientation tensors of all augmentations tend to the target orientation tensor.

[0045] Next, convex decomposition is performed on each concave polygon augmentation. First, a divide-and-conquer method is used for triangulation; when the divide-and-conquer method cannot find a valid diagonal that meets the conditions, the ear-cutting method is switched to continue triangulation; after obtaining the triangulation results, convex merging is performed on adjacent triangular units sharing edges to obtain a smaller number of convex subdomains. This process is described in [link to documentation]. Figure 3 .

[0046] Subsequently, for any two enhancers, the nearest periodic mirror position is first determined according to the minimum mirror principle. Then, a two-level circumcircle screening process is used to eliminate enhancer pairs and convex subdomain pairs that clearly do not interact. The GJK algorithm is used to determine whether the remaining candidate convex subdomain pairs intersect. If they intersect, the EPA algorithm is further used to calculate the penetration depth and separation direction. In the case where multiple intersecting convex subdomain pairs exist between the same pair of concave enhancers, the convex subdomain pair with the largest penetration depth is selected as the separation criterion between the pair of enhancers. This process is described in [link to documentation]. Figure 4 and Figure 5 .

[0047] Subsequently, for each augmenter, the separation vector between it and all neighboring augmenters is accumulated to obtain its total translation increment, which is then synchronously applied to all convex subdomains of that augmenter. When an augmenter exceeds the RVE boundary after updating, it is mapped back into the RVE region according to periodic boundary conditions. This process is repeated, with a maximum number of iterations Nsep,max. The maximum penetration depth of all augmenter pairs is less than a preset tolerance ε. sep The iteration stops at a certain point, resulting in a strictly non-overlapping, compact two-dimensional periodic RVE (see...). Figure 6 If the tolerance is still not met after reaching the iteration limit, the augmentation body position is reinitialized and the iteration process continues.

[0048] In another embodiment: For bilobal and pea-shaped augmenters with curved boundaries, the two-dimensional periodic RVE creation first performs adaptive boundary sampling based on the original boundary curvature and arc length information to construct the circumscribed concave polygon that encloses the original curved boundary. Then, hybrid triangulation and convex merging are performed on the circumscribed concave polygon to obtain the convex subdomain representation for intersection detection and migration update.

[0049] Except for the target orientation tensor and the augmentation geometry type, all other calculation parameters and termination criteria can be kept consistent with the first embodiment.

[0050] Given the target orientation tensor a t Under the conditions of =diag[1.0, 0.0] and target volume fraction v1=0.60, the random initialization, orientation correction, periodic intersection detection, penetration depth calculation, symmetric aggregation migration update, and periodic mapping are completed according to the steps. Finally, a two-dimensional periodic RVE of a bilobal or pea-shaped reinforcement that satisfies strict non-overlap constraints and periodic boundary conditions is obtained (see Figure 7 ).

[0051] This invention first addresses the challenge of intersection detection caused by the non-convex boundaries of concave reinforcements. It transforms each concave reinforcement into a set of interconnected convex subdomains through convex decomposition, thereby reducing the complex non-convex intersection determination problem to a mature and reliable convex polygon intersection detection problem, laying a geometric foundation for subsequent high-precision separation calculations. Based on this, and incorporating the minimum mirror principle under periodic boundary conditions, it accurately determines the closest relative positional relationship between any two reinforcements (including cross-boundary mirrors). Furthermore, it obtains the penetration depth and corresponding separation direction based on the intersection determination of the convex subdomain set, thus constructing an accurate separation vector. This makes the previously complex non-convex intersection determination problem much simpler. Contact states with point contact and narrow gaps, which are prone to being missed or misjudged, are reliably quantified. Furthermore, addressing the oscillation or local jamming problem caused by the simultaneous coupling constraints of multiple neighboring reinforcements on a single reinforcement with a high volume fraction, this invention aggregates the separation vectors between each reinforcement and all intersecting neighboring reinforcements to obtain the total translation increment. Based on this, the center coordinates are updated, and periodic mapping is performed on reinforcements exceeding the boundary, thereby achieving coordinated decoupling and stable iteration of multi-directional constraints. Finally, the iteration terminates when the maximum penetration depth between all reinforcement pairs is less than the preset separation tolerance, strictly satisfying the non-overlapping constraint and periodic boundary conditions. Throughout the process, the original return boundary of the concave reinforcement is always completely preserved, without the need for convex hull approximation, thus ensuring the uniformity of the generated RVE in terms of geometric realism, statistical consistency, and periodic consistency, providing a highly reliable geometric model for subsequent homogenization analysis and mesoscopic damage evolution prediction.

[0052] Based on the above method, this invention proposes a composite material RVE creation system for concave reinforcements, comprising: The data acquisition module is used to acquire the RVE parameters of the composite material and establish a two-dimensional rectangular RVE region based on the composite material RVE parameters. The reinforcement acquisition module is used to randomly generate the initial center coordinates and initial orientation angles of multiple concave reinforcements within the two-dimensional rectangular RVE region to obtain a set of candidate initial intersecting reinforcements. The correction module is used to perform convex decomposition on each concave reinforcement in the reinforcement set, transforming it into a set of several interconnected convex subdomains. The iteration module is used to perform a separation termination condition step on the reinforcement pairs iteratively, specifically: determining the closest relative position relationship between any two reinforcements under periodic boundary conditions based on the minimum mirror principle, and based on the result of the convex decomposition. The process involves determining the intersection of the convex subdomains of two augmenters; if an intersecting pair of convex subdomains is found, the penetration depth and corresponding separation direction of the convex subdomains are obtained, and the separation vector between the augmenter pairs is determined; the separation vectors between each augmenter and all neighboring augmenters with intersecting relationships are aggregated to obtain the total translation increment of the augmenter, and the center coordinates are updated based on the total translation increment; periodic mapping is performed on augmenters that exceed the boundary of the two-dimensional rectangular RVE region after the center coordinates are updated, and they re-enter the RVE region; when the maximum penetration depth between all augmenter pairs is less than the preset separation tolerance, the iteration is terminated, and a two-dimensional periodic RVE that satisfies the non-overlapping constraint and periodic boundary condition is obtained.

[0053] The present invention provides a computer device, including a memory and a processor. The memory stores a program, and when the program is executed by the processor, the processor performs the steps of the above-described method for creating a composite material RVE oriented towards a concave reinforcement.

[0054] According to the disclosed embodiments, the computer device can communicate with one or more external devices (e.g., keyboard, pointing device, Bluetooth communication, etc.) or with any device that enables the computing device to communicate with one or more other computing devices (e.g., router, demodulator, etc.).

[0055] The present invention provides a storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the above-described method for creating a composite material RVE oriented towards a concave reinforcement.

[0056] According to the disclosed embodiments, the storage medium can be a non-volatile computer-readable storage medium, such as, but not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this invention, the storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0057] The above description, in conjunction with specific preferred embodiments, provides a more detailed explanation of the present invention. For those skilled in the art, various simple deductions or substitutions can be made without departing from the concept of the present invention, and all such deductions or substitutions should be considered to fall within the scope of protection of the present invention.

Claims

1. A method for creating RVE (Reverse Velocity Enhancement) in composite materials oriented towards concave reinforcements, characterized in that, Includes the following steps: Obtain the RVE parameters of the composite material, and establish a two-dimensional rectangular RVE region based on the composite material RVE parameters; Within the two-dimensional rectangular RVE region, the initial center coordinates and initial orientation angles of multiple concave reinforcements are randomly generated to obtain a set of candidate initially intersecting reinforcements; For each concave reinforcement in the reinforcement set, perform convex decomposition to transform it into a set of interconnected convex subdomains; The step of performing the separation termination condition for the enhancement pairs in the iteration is as follows: based on the minimum mirror principle, determine the closest relative position relationship between any two enhancement pairs under the periodic boundary condition, and based on the result of the convex decomposition, determine the intersection of the convex subdomain sets of the two enhancement pairs. If intersecting convex subdomain pairs are determined, the penetration depth and corresponding separation direction of the convex subdomains are obtained, and the separation vector between the augmentation pairs is determined. The separation vectors between each augmentation and all neighboring augmentations with intersecting relationships are aggregated to obtain the total translation increment of the augmentation, and the center coordinates are updated according to the total translation increment. For augmentations whose center coordinates are updated and exceed the boundary of the two-dimensional rectangular RVE region, periodic mapping is performed to re-enter the RVE region. When the maximum penetration depth between all augmentation pairs is less than the preset separation tolerance, the iteration is terminated, and a two-dimensional periodic RVE that satisfies the non-overlapping constraint and periodic boundary condition is obtained.

2. The method for creating RVE in composite materials oriented towards concave reinforcements as described in claim 1, characterized in that, For each concave reinforcement in the reinforcement set, a convex decomposition is performed, transforming it into a set of interconnected convex subdomains, specifically: Triangulation is performed on the concave polygons in multiple input concave reinforcements, and convex merging is performed on adjacent triangular units obtained by triangulation to obtain a set of several interconnected convex subdomains; wherein the convex merging is performed while keeping the merged region still a convex domain.

3. The method for creating RVE in composite materials oriented towards concave reinforcements as described in claim 2, characterized in that, The triangulation adopts a hybrid triangulation strategy that combines the divide-and-conquer method with the ear-cutting method. When the divide-and-conquer method cannot find a valid diagonal that meets the conditions, it switches to the ear-cutting method to continue the triangulation.

4. The method for creating RVE in composite materials oriented towards concave reinforcements as described in claim 1, characterized in that, Before the termination of the iteration, the following is also included: A two-stage circumcircle screening mechanism is introduced during the reinforcement intersection detection process, specifically: First-level screening: Establish a circumcircle at the overall level of the concave reinforcement. If the circumcircles of the two reinforcements do not intersect, it is directly determined that they do not interact. Second-level screening: A circumcircle is established at the convex subdomain level, and exact intersection determination and penetration solution are performed only on convex subdomain pairs that pass both the first and second-level screening.

5. The method for creating RVE in a composite material oriented towards a concave reinforcement as described in claim 1, characterized in that, When performing convex decomposition on each concave reinforcement in the reinforcement set, the method further includes: Establish a deviation objective function between the current fourth-order orientation tensor of all augmentations and the target fourth-order orientation tensor; Based on the bias objective function, the orientation vector of each augmentation is iteratively updated to complete the orientation correction of each augmentation while maintaining the unit orientation vector constraint, thus obtaining the corrected augmentation, which serves as the input for convex decomposition.

6. The method for creating RVE in composite materials oriented towards concave reinforcements as described in claim 1, characterized in that, When obtaining the set of candidate initial intersection enhancers, the method further includes: The vertex coordinates of each augmentation relative to its centroid are scaled up uniformly by a proportional factor to control the minimum spacing between adjacent augmentations.

7. The method for creating RVE in a composite material oriented towards a concave reinforcement as described in claim 1, characterized in that, The step of establishing a two-dimensional rectangular RVE region based on RVE parameters also includes: The number of reinforcements is obtained by rounding down the product of the target reinforcement volume fraction of the composite material and the dimensions in different coordinate directions of the RVE; where the dimensions in different coordinate directions of the RVE are the dimensions of the two-dimensional rectangular region.

8. A composite material RVE creation system for concave reinforcements, characterized in that, include: The data acquisition module is used to obtain the RVE parameters of the composite material and to establish a two-dimensional rectangular RVE region based on the RVE parameters of the composite material. The augmentation acquisition module is used to randomly generate the initial center coordinates and initial orientation angles of multiple concave augmentations within the two-dimensional rectangular RVE region, thereby obtaining a set of candidate initially intersecting augmentations; The correction module is used to perform convex decomposition on each concave reinforcement in the reinforcement set, transforming it into a set of several interconnected convex subdomains; The iteration module is used to perform a separation termination condition step on the reinforcement pair iteration. Specifically, it determines the closest relative position relationship between any two reinforcements under periodic boundary conditions based on the minimum mirror principle, and performs an intersection determination on the convex subdomain sets of the two reinforcements based on the result of the convex decomposition. If intersecting convex subdomain pairs are determined, the penetration depth and corresponding separation direction of the convex subdomains are obtained, and the separation vector between the augmentation pairs is determined. The separation vectors between each augmentation and all neighboring augmentations with intersecting relationships are aggregated to obtain the total translation increment of the augmentation, and the center coordinates are updated according to the total translation increment. For augmentations whose center coordinates are updated and exceed the boundary of the two-dimensional rectangular RVE region, periodic mapping is performed to re-enter the RVE region. When the maximum penetration depth between all augmentation pairs is less than the preset separation tolerance, the iteration is terminated, and a two-dimensional periodic RVE that satisfies the non-overlapping constraint and periodic boundary condition is obtained.

9. A computer device, characterized in that, The device includes a memory and a processor, wherein the memory stores a program that, when executed by the processor, causes the processor to perform the steps of a composite material RVE creation method for a concave reinforcement as described in any one of claims 1 to 7.

10. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the composite material RVE creation method for concave reinforcement as described in any one of claims 1 to 7.

Images