Composite material finite element model construction method based on random adaptive distribution algorithm

Abstract

The invention provides a composite material finite element model construction method based on a random adaptive distribution algorithm, and belongs to the technical field of finite element modeling. Aiming at the random distribution characteristic of the SMC composite material, generating a random geometric model of the sheet material in a specified range through a random distribution algorithm; the distribution characteristics are adjusted by combining an adaptive optimization algorithm so as to meet the target volume fraction; geometric interference between the sheets is eliminated through a collision detection algorithm based on a separation axis theorem, so that an interference-free geometric model is generated; and finally, carrying out finite element mesh generation and material attribute definition, and constructing an accurate finite element model. According to the method, the mesostructure of the SMC composite material can be truly represented, and the modeling efficiency and the precision of mechanical property prediction can be improved, so that technical support is provided for design and optimization of the SMC composite material.

Description

Technical Field

[0001] The invention belongs to the technical field of finite element modeling, and relates to a method for constructing a finite element model of a composite material based on a random adaptive distribution algorithm. Background Art

[0002] With the development of modern industry, composite materials have gained widespread application in various fields due to their excellent performance. Sheet molding compound (SMC), a key composite material, is widely used in the automotive, aerospace, construction, electrical equipment, and other industries due to its lightweight, high specific strength, and excellent corrosion resistance. In the automotive industry, SMC is commonly used to manufacture body panels, engine hoods, and other components; in the aerospace field, SMC is used to manufacture lightweight structural components; and in the construction and electrical fields, SMC is used to create high-strength, aging-resistant housings and structural components.

[0003] The main components of SMC composites are chopped carbon fibers and a resin matrix. Their microstructure is characterized by the random distribution and stacking of fiber-reinforced sheets (hereafter referred to as sheets) within the resin matrix. This complex mesostructure significantly affects the mechanical properties of SMC, such as strength, stiffness, and impact toughness. However, due to the randomness and heterogeneity of SMC composites, the precise prediction and optimal design of their mechanical behavior are extremely challenging.

[0004] Finite element analysis (FEA) is a powerful tool widely used to study the properties of composite materials. However, traditional finite element modeling methods often use homogenization assumptions or simplified models, making it difficult to accurately reproduce the complex mesostructure of SMC. This simplification can lead to errors in mechanical property predictions, limiting the model's application in engineering practice. Therefore, developing a modeling method that can accurately reproduce the mesostructure of SMC composites is of great significance for studying their mechanical properties and guiding product design.

[0005] In recent years, the development of random distribution algorithms and optimization techniques has opened up new possibilities for modeling SMC composite materials. By introducing random distribution algorithms, a fiber sheet distribution consistent with the actual material can be generated in virtual space. Simultaneously, incorporating adaptive optimization algorithms can further improve model realism and computational efficiency. Furthermore, collision detection algorithms based on the Separating Axis Theorem (SAT) can effectively resolve interference issues in geometric models, providing technical support for generating high-precision, interference-free geometric models. Summary of the Invention

[0006] In order to solve the problem of SMC composite material modeling, the present invention proposes a composite material finite element model construction method based on a random adaptive distribution algorithm. This method realizes geometric model generation, meshing and material property assignment, and performs finite element analysis. It can not only realistically reproduce the mesoscopic structure of SMC composite materials, but also improve modeling efficiency and the accuracy of mechanical property prediction, thereby providing technical support for composite material design and optimization.

[0007] The technical solution adopted in the present invention is as follows:

[0008] A method for constructing a composite material finite element model based on a random adaptive distribution algorithm includes: generating a random geometric model of a sheet within a specified range using a random distribution algorithm based on the random distribution characteristics of the SMC composite material; adjusting the sheet distribution characteristics in combination with an adaptive optimization algorithm to meet the target volume fraction; eliminating geometric interference between sheets using a collision detection algorithm based on the separating axis theorem (SAT), thereby generating an interference-free geometric model; and finally performing finite element meshing and material property definition to construct an accurate SMC composite material finite element model. Specifically, the method includes the following steps:

[0009] Step 1: According to the actual size of the composite material mechanical test specimen, set the basic parameters of the composite material finite element model, including the length, width, thickness of the model, as well as the length, width, and thickness of the sheet; through SEM analysis, calculate the target volume fraction and spatial density of the sheet in the composite material mechanical test specimen as the setting conditions, and set a reasonable volume tolerance range for it.

[0010] Step 2: Generate the position and rotation angle of the sheet in the composite material finite element model space through the random distribution algorithm; combine with the adaptive optimization algorithm to adjust the sheet distribution to meet the target volume fraction and distribution uniformity.

[0011] In step 2.1, a random distribution algorithm is used to generate the initial spatial position and layup angle data for each sheet by uniformly distributing them within the boundaries of the composite material finite element model. Constraints are set to ensure that each sheet does not exceed the model boundaries set in step 1. A random number generator is also used to ensure the independent distribution of each sheet. The layup angle is the angle between the sheet centerline and the X-axis of the global coordinate system. Based on the target volume fraction calculated in step 1, the number of generated sheets is dynamically adjusted to meet the dual requirements of randomness and volume fraction.

[0012] In step 2.2, to reflect the randomness of the sheets in space, based on the random distribution of angles within a plane and in combination with an adaptive optimization algorithm, the spatial position and stacking angle of each sheet are iteratively optimized, thereby adjusting the spatial distribution density of the sheets to more closely meet the target volume fraction and uniformity requirements. The specific method is to calculate the volume fraction and uniformity index for the current sheet distribution, define an optimization objective function, and use the spatial density and target volume fraction as optimization targets. Using optimization strategies such as gradient descent or genetic algorithms, the spatial position and stacking angle of the sheets are gradually optimized and adjusted. Ultimately, the optimized sheet distribution accurately meets the spatial density and target volume fraction calculated in step 1.

[0013] In step 2.3, based on the optimized sheet distribution, a three-dimensional random stacking geometry model is generated through a layered stacking method. This layered stacking method faithfully reproduces the true mesostructure characteristics of the SMC composite material, fully embodying the characteristics of randomness and multi-layer stacking.

[0014] Step 3: Detect and eliminate geometric interference between sheets using a collision detection algorithm based on the separating axis theorem; iteratively optimize the interfering parts until there is no interference between sheets within the finite element model of the composite material.

[0015] Specifically:

[0016] The SAT algorithm treats each sheet as a convex hull, calculates the projection interval of each sheet on all axes in the global coordinate system, and determines whether the projection intervals of any two sheets intersect; if the projection intervals of any two sheets on all axes do not intersect, it is determined that there is no collision; if a collision is detected, the spatial position and stacking angle of one of the sheets are dynamically adjusted, and the spatial position and stacking angle of all sheets are iteratively optimized until all sheets within the range of the composite material finite element model are completely free of interference, resulting in an interference-free geometric model.

[0017] Step 4: Sheet model definition.

[0018] Based on the non-interference geometric model generated in step 3, the finite element mesh unit type is selected according to user needs, the model is 3D scanned and meshed; the node, unit set and surface set of the model are defined; the local coordinate system of each sheet is established with the long side of each sheet as the X-axis, the short side as the Y-axis, and the Z-axis perpendicular to the sheet plane. The local coordinate system is used to define the anisotropic material properties of each fiber sheet, so that the model can accurately reflect the mechanical behavior characteristics of the SMC material.

[0019] Step 5: After completing the model definition, the sheet model is embedded into the resin matrix to complete the assembly of the complete SMC composite material finite element model.

[0020] The beneficial effects of the present invention are:

[0021] (1) Improve modeling efficiency and accuracy. By combining random distribution algorithms and adaptive optimization technology, the present invention significantly improves model building efficiency while ensuring the randomness and uniformity of the generated model, which can more realistically reflect the mesostructure characteristics of SMC composite materials.

[0022] (2) Eliminate geometric interference and physical irrationality. By introducing a collision detection algorithm based on the separating axis theorem, the overlap problem between sheets can be effectively avoided, ensuring that the generated geometric model conforms to actual physical laws, thereby improving the accuracy of mechanical property prediction.

[0023] (3) Support for multi-field coupling analysis and engineering optimization. The finite element model generated by the present invention has good adaptability and can be directly applied to complex working condition simulations such as thermal-mechanical coupling and multi-scale analysis, providing important support for composite material design and engineering optimization. At the same time, the parameterized nature of the model makes it easy to extend to other composite material systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 Schematic diagram of random distribution of sheet stacking angles, where (a), (b), and (c) are schematic diagrams of different stacking angles;

[0025] Figure 2 Schematic diagram of the three-dimensional random stacking geometric model;

[0026] Figure 3 Schematic diagram of the three-dimensional random stacking geometric model from a through-layer perspective, where (a) is the axonometric view and (b) is the side view;

[0027] Figure 4 Figure 1 shows a finite element model with an accurate local coordinate system definition, where (a) is an axonometric view, (b) is a side view, and (c) is a schematic diagram of the local coordinate system definition. DETAILED DESCRIPTION

[0028] The following are specific embodiments of the present invention and, in conjunction with the accompanying drawings, further describe the technical solutions of the present invention in detail, but the present invention is not limited to these embodiments.

[0029] A method for constructing a composite material finite element model based on a random adaptive distribution algorithm comprises the following steps:

[0030] Step 1: According to the actual size of the composite material mechanical test specimen, set the basic parameters of the composite material finite element model, including the length, width, thickness of the model, as well as the length, width, and thickness of the sheet; through SEM analysis, calculate the target volume fraction and spatial density of the composite material specimen as setting conditions, and set a reasonable volume tolerance range for it.

[0031] Step 2: Generate the position and rotation angle of the sheet in the composite material finite element model space in the MATLAB program through the random distribution algorithm; combine with the adaptive optimization algorithm to adjust the sheet distribution to meet the target volume fraction and distribution uniformity.

[0032] First, a random distribution algorithm is used to generate the initial spatial position and layup angle data for each sheet by uniformly distributing them within the boundaries of the composite material finite element model. Constraints are set to ensure that each sheet does not exceed the model boundaries set in step one. A random number generator is also used to ensure the independent distribution of each sheet. The layup angle is the angle between the sheet centerline and the X-axis of the global coordinate system. Based on the target volume fraction calculated in step one, the number of generated sheets is dynamically adjusted to meet the dual requirements of randomness and volume fraction.

[0033] Secondly, in order to reflect the randomness of the sheet in space, according to the random distribution characteristics of the angle in the plane (see Figure 1 ) and, combined with an adaptive optimization algorithm, iteratively optimizes the spatial position and stacking angle of each sheet, thereby adjusting the spatial distribution density of the sheets to more closely match the target volume fraction and uniformity requirements. The specific method involves calculating the volume fraction and uniformity index for the current sheet distribution, defining an optimization objective function, and using spatial density and target volume fraction as optimization targets. Using strategies such as gradient descent or genetic algorithms, the spatial position and stacking angle of the sheets are gradually optimized and adjusted. Ultimately, the optimized sheet distribution meets both the spatial density and target volume fraction calculated in step one.

[0034] Finally, based on the optimized sheet distribution, a three-dimensional random stacking geometric model is generated by layered stacking, such as Figure 2 and Figure 3 This is a three-dimensional model structure from a layer-by-layer perspective. The layered stacking method truly reproduces the real mesoscopic structural characteristics of SMC composite materials, fully demonstrating the characteristics of randomness and multi-layer stacking.

[0035] Step 3: Detect and eliminate geometric interference between sheets using a collision detection algorithm based on the separating axis theorem; iteratively optimize the interfering parts until there is no interference between sheets within the finite element model of the composite material.

[0036] Specifically:

[0037] In order to solve the geometric interference problem between sheets, a three-dimensional collision detection algorithm based on the separating axis theorem (SAT) is used, such as Figure 4The SAT algorithm treats each sheet as a convex hull, calculates the projection interval of each sheet on all axes in the global coordinate system, and determines whether the projection intervals of any two sheets intersect. If the projection intervals of any two sheets on all axes do not intersect, it is determined to be non-collision-free. If a collision is detected, the spatial position and layup angle of one of the sheets are dynamically adjusted, and the spatial position and layup angles of all sheets are iteratively optimized until all sheets within the composite material finite element model are completely free of interference, resulting in an interference-free geometric model. This method effectively avoids the problem of geometric overlap between sheets and ensures the physical rationality of the generated model.

[0038] Step 4: Sheet model definition.

[0039] Import the non-interference geometric model generated in step 3 into Abaqus, select the finite element mesh unit type (such as C3D8R or C3D4 unit) according to user needs, perform 3D scanning on the model and complete meshing; define the node, unit set and surface set of the model in Abaqus; establish a local coordinate system for each sheet with the long side of each sheet as the X-axis, the short side as the Y-axis, and the Z-axis perpendicular to the sheet plane. Use the local coordinate system to define the anisotropic material properties of each fiber sheet, so that the model can accurately reflect the mechanical behavior characteristics of the SMC material.

[0040] Step 5: After completing the model definition, the sheet model is embedded into the resin matrix through the embedded constraint function in Abaqus, thereby realizing the assembly of the complete SMC composite material finite element model.

[0041] The finite element model of the SMC composite material was validated by simulating the compression molding process and observing the stress and deformation distribution of the SMC overall and individual models during loading. The simulation results demonstrated that the model generated in this example can faithfully reproduce the mechanical behavior of the SMC composite material, verifying its reliability and accuracy in engineering applications.

[0042] Those skilled in the art will readily appreciate other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, and the true scope and spirit of the invention are indicated by the claims of this application.

Claims

1. A method for constructing a composite material finite element model based on a random adaptive distribution algorithm, characterized in that: include: A random distribution algorithm is used to generate a random geometric model of the sheet within a specified range; an adaptive optimization algorithm is used to adjust the sheet distribution characteristics to meet the target volume fraction. The geometric interference between sheets is eliminated through a collision detection algorithm based on the separating axis theorem to generate an interference-free geometric model; finally, finite element meshing and material property definition are performed to construct a finite element model of the SMC composite material.

2. The method for constructing a composite material finite element model based on a random adaptive distribution algorithm according to claim 1, characterized in that: The specific steps include: Step 1: According to the actual size of the composite material mechanical test specimen, set the basic parameters of the composite material finite element model, calculate the target volume fraction and spatial density of the composite material mechanical test specimen as setting conditions, and set a reasonable volume tolerance range for it; Step 2: Generate the position and rotation angle of the sheet in the composite material finite element model space through a random distribution algorithm; combine with an adaptive optimization algorithm to adjust the sheet distribution to meet the target volume fraction and distribution uniformity; Step 3: Using a collision detection algorithm based on the separating axis theorem, the geometric interference between the sheets is detected and eliminated. The interfering parts are iteratively optimized until there is no interference between the sheets within the finite element model of the composite material, thus obtaining an interference-free geometric model. Step 4: Sheet model definition: Based on the non-interference geometric model generated in step 3, select the finite element mesh unit type according to user needs, perform 3D scanning on the model and complete meshing; define the model's nodes, unit sets, and surface sets; use the local coordinate system of each sheet to define the anisotropic material properties of each fiber sheet, so that the model can accurately reflect the mechanical behavior characteristics of the SMC material; Step 5: After completing the model definition, the sheet model is embedded into the resin matrix to complete the assembly of the complete SMC composite material finite element model.

3. The method for constructing a composite material finite element model based on a random adaptive distribution algorithm according to claim 2, characterized in that: In the step 1, the basic parameters of the composite material finite element model include the length, width, and thickness of the model, as well as the length, width, and thickness of the sheet.

4. The method for constructing a composite material finite element model based on a random adaptive distribution algorithm according to claim 2, characterized in that: In the step 1, the target volume fraction and spatial density of the composite material mechanical test specimen are calculated through SEM scanning electron microscopy analysis.

5. The method for constructing a composite material finite element model based on a random adaptive distribution algorithm according to claim 2, characterized in that: The step 2 is specifically as follows: In step 2.1, a random distribution algorithm is used to generate the initial spatial position and layup angle data for each sheet by uniformly distributing them within the boundaries of the composite material finite element model. Constraints are set to ensure that each sheet does not exceed the model boundaries set in step 1. A random number generator is used to ensure the independent distribution of each sheet. Based on the target volume fraction calculated in step 1, the number of generated sheets is dynamically adjusted to meet the dual requirements of randomness and volume fraction. Step 2.2: Using an adaptive optimization algorithm, the spatial position and stacking of each sheet are optimized iteratively, thereby adjusting the spatial distribution density of the sheets to make them closer to the target volume fraction and uniformity distribution requirements; In step 2.3, based on the optimized sheet distribution, a three-dimensional random stacking geometric model is generated through layered stacking. The layered stacking method truly reproduces the real mesoscopic structural characteristics of the SMC composite material and fully reflects the characteristics of randomness and multi-layer stacking.

6. The method for constructing a composite material finite element model based on a random adaptive distribution algorithm according to claim 5, characterized in that: The stacking angle refers to the angle between the center line of the sheet and the X-axis of the global coordinate system.

7. The method for constructing a composite material finite element model based on a random adaptive distribution algorithm according to claim 5, characterized in that: The specific method of step 2.2 is as follows: calculate the volume fraction and uniformity index under the current sheet distribution, define the optimization objective function, take the spatial density and target volume fraction as the optimization targets, and gradually optimize and adjust the spatial position and stacking angle of the sheet; finally, the optimized sheet distribution meets the spatial density and target volume fraction calculated in step 1 at the same time.

8. The method for constructing a composite material finite element model based on a random adaptive distribution algorithm according to claim 7, characterized in that: The optimization strategy uses gradient descent method or genetic algorithm.

9. The method for constructing a composite material finite element model based on a random adaptive distribution algorithm according to claim 2, characterized in that: The step three is specifically as follows: Treat each sheet as a convex hull, calculate the projection interval of each sheet on all axes in the global coordinate system, and determine whether the projection intervals of any two sheets intersect; If the projection intervals of any two sheets on all axes have no intersection, it is determined that there is no collision; if a collision is detected, the spatial position and stacking angle of one of the sheets are dynamically adjusted, and the spatial position and stacking angle of all sheets are iteratively optimized until all sheets are completely free of interference.

10. The method for constructing a composite material finite element model based on a random adaptive distribution algorithm according to claim 2, characterized in that: In the above-mentioned step 4, the local coordinate system has the long side of the sheet as the X-axis, the short side as the Y-axis, and the Z-axis perpendicular to the sheet plane.

Images