1 Introduction
composite materials have special performance characteristics that traditional materials do not have, including high strength, high stiffness-to-weight ratio and high wear resistance. Composites can also be engineered with customized properties by changing their microstructure according to specific applications. Therefore, composite materials have been widely used in many fields. However, the application of composite materials depends critically on their manufacturing process. Manufacturing processes can affect the surface integrity of composite parts/structures, which in turn determines their quality, reliability and service life. Numerical models have been widely used to explore material removal, tool-matrix-reinforcement interactions, and workpiece damage during the fabrication of composite materials. The numerical models currently used mainly include finite element models (FEM), discrete element models (DEM), molecular dynamics (MD) simulations, and multi-scale models implemented by combining these methods. Each model has its advantages and limitations. The theoretical basis of the finite element model is continuum mechanics, which cannot well represent the discreteness, fracture and damage processes of materials. Discrete element models require relatively high computational costs, and they require a long time to check parameters. Molecular dynamics simulations are computationally expensive and the model size is limited to the nanometer scale. Multi-scale models can well represent deformation at each scale. However, due to the connection problem at different scales, there are currently few studies using multi-scale models for composite material processing.
In July 2022, a team of academician Zhang Liangchi of Southern University of Science and Technology published a review article titled "On the numerical modeling of composite machining" online in the top journal "Composites Part B: Engineering" in the field of composite materials mechanics. . Combining the research results of relevant papers in the past two decades, the article comprehensively summarizes and analyzes the development trends of digital technology in the field of composite materials precision manufacturing and key issues that urgently need breakthroughs, and looks forward to possible breakthroughs in terms of cross-scale analysis bottlenecks in the development of the field. direction.
2 Composite material processing numerical model
2.1 Finite element model
The finite element model has been widely used in composite material processing analysis. In the finite element model, accurately describing the mechanical properties of the workpiece material is a very important step. There are usually two methods to characterize the basic mechanical properties of composite materials, namely the equivalent homogeneous material model and the multiphase model. In the equivalent homogeneous material model, the composite material is considered to be uniformly distributed, and its mechanical properties can be obtained by averaging the reinforcement and matrix. This processing method can significantly improve computational efficiency, but it cannot reveal the microscale deformation mechanism of composite materials because it ignores the interaction between the reinforcement and the matrix. Compared with the equivalent homogeneous material model, since the reinforcement and matrix are treated separately, the multiphase model can well characterize the interaction between different phases in the composite material and is used to understand the microscopic deformation mechanism during composite material processing. explore. However, its computational cost is much higher than that of an equivalent homogeneous material model.
2.1.1 The equivalent homogeneous material (EHM) model
The equivalent homogeneous material model has been used to evaluate the macroscopic stress distribution, plastic deformation, cutting forces and processing-induced subsurface of fiber-reinforced polymer composites (FRPC) during processing. Characteristics such as damage have also been used to study the effects of fiber orientation, cutting depth and tool geometry on cutting forces and subsurface damage during the cutting of unidirectional fiber composites. As shown in Figure 1, although the equivalent homogeneous material model cannot predict the thrust force during composite processing well, by using a micro-macro modeling method that combines adaptive grid technology and density, it can also obtain experimental results. The measurement results are relatively consistent for cutting forces. In addition, as shown in Figure 2, the chip formation mechanism predicted using the equivalent homogeneous material model is also consistent with the experimental observations.
Figure 1 Comparison of measured and predicted values of (a) cutting force (Fc) and (b) thrust force (Ft).
Figure 2 Comparison of chip formation during carbon fiber reinforced polymer processing: (a) experimental results and (b) finite element model results.
2.1.2 Multiphase (MP) Model
In addition to calculating cutting forces, stress and strain distribution, plastic deformation and material failure, the multiphase model can be used to reveal the interaction between the matrix, reinforcement and tool to understand material removal and surface generation mechanisms. Using multiphase models for composite material processing requires specifying appropriate material properties and constitutive models for each component in the composite material, including matrix (constitutive model, chip separation criterion), reinforcement (type, shape, distribution, etc.) and reinforcement-matrix interfaces (adhesion, friction, and cohesion). Different from the equivalent homogeneous material model, using the multiphase model, the cutting force and thrust force during the cutting process can be well predicted. Sometimes in order to save computing resources and obtain accurate results, the MP model and the EHM model are also combined. As shown in Figure 3, through the joint modeling of MP and EHM, the influence of fiber orientation and thermal effects on the chip formation mechanism is revealed in detail. Influence.
Figure 3 Temperature distribution and material removal (a) at the beginning and (b) after long-distance cutting. The
multiphase model has also been used to perform processing simulations of particle-reinforced metal matrix composites (PRMMCs), elucidating particle-matrix debonding, particle fracture and shedding, tool-particle interaction, temperature prediction and processing during processing. Surface defect integrity of components. The study found that the machined surface integrity is significantly affected by the processing parameters and material properties of PRMMCs: higher feed rates lead to more severe residual stresses and damage rates, higher cutting speeds lead to more severe subsurface damage, Larger cutting depth results in more edge defects and higher surface roughness.
The study also found that damage to PRMMCs during processing, such as particle fracture and matrix-particle delamination, is greatly affected by the stress/strain distribution. According to the relative position of the tool and the particles, the interaction between the tool and the particles can be divided into three situations. As shown in Figure 4a, when the particles are located below the tool, the matrix between the particles and the tool undergoes high compressive stress, while the contact surface on the lower right side of the particles and the matrix is subject to tensile stress. Such stress distribution may lead to exfoliation of particles. As shown in Figure 4b, when the tool is in direct contact with the particles, high mutually perpendicular tensile and compressive stresses will be generated in the particles. If the stress is high enough, the particles will break. As shown in Figures 4c and 4d, as the tool is advanced further, the particles are peeled off and plowed through the matrix, leaving a cavity that slides out along the flanks of the tool.
Figure 4 Evolution of the stress field when the particle is located on the cutting path: (a) before contact, (b) at the beginning of contact, (c) at completion of contact and (d) after contact.
As shown in Figure 5, when the particles are located above the cutting path, the particles will be partially peeled off after interacting with the rake face of the tool and move upward with the chips. Particles located above the cutting path are subject to high compressive stress (perpendicular to the rake face) and tensile stress (parallel to the rake face), which may lead to particle fracture and matrix-particle interface debonding.
Figure 5 Stress field evolution when the particle is located above the cutting path: (a) before contact and (b) contact stage.
When the particles are located below the cutting path, as shown in Figure 6, the stress they receive is much smaller, and there is no particle breakage or matrix-particle interface peeling.
Figure 6 Stress field evolution when the particle is located below the cutting path: (a) approaching stage, (b) contact stage and (c) leaving stage.
By considering the elasticity and fracture of particles in PRMMCs, as well as simulating the real microstructure of PRMMCs, the deformation and failure of PRMMCs with random shapes and random particle distribution can be more accurately described. As shown in Figure 7, during the processing of PRMMCs, the particles located on the cutting path are cut off or pressed into the matrix, causing the particles to break and peel off from the matrix. The pressed particles may interact with other particles distributed below, causing more particles to break or peel off, and leaving cavities and deep pits on the machined surface. Particles can also be dragged by the tool, plowing across the machined surface to form large cavities.
Figure 7 The machined surface morphology when the cutting speed is 100m/min and the cutting depth is (a) 25µm and (b) 50µm.
2.2 Discrete element model
The discrete element model is mainly used to study the properties of granular materials and discontinuous materials, as well as the basic mechanism of material removal during the processing of certain materials, such as the impact of adhesion of shed particles during wear and surface Evolution of subsurface damage during polishing. Discrete element modeling was also used to reveal the microscopic failure mechanism of unidirectional FRPC under transverse tension and the influence of fiber orientation in orthogonal cutting. As shown in Figure 8a, during the orthogonal processing of unidirectional carbon fiber-reinforced polymers, when the fiber orientation is 0°, chips are formed by both mode I fracture and mode II fracture of the polymer, and are accompanied by fiber Buckling fracture and delamination; when the fiber orientation is 45°, as shown in Figure 8b, the fibers are stretched and sheared by the cutting edge, and the chips are formed by shearing the fiber/matrix to Free surface formation; when the fiber orientation is 90°, as shown in Figure 8c, material removal begins with the bending of the fibers, which results in Mode I fracture along the fiber/matrix interface. Errors such as fiber shedding and multiple cracks on the machined surface are also observed. and other damage; when the fiber orientation is -45°, as shown in Figure 8d, the fiber is obviously bent during the cutting process and breaks by pulling out. These phenomena are in good agreement with the corresponding experimental observations.
Figure 8 Chip formation mechanism at different fiber directions in orthogonal cutting: (a) 0°, (b) 45°, (d) 90° and (d) -45°.
It is worth noting that compared with the finite element model, the discrete element model has more advantages in simulating the fracture of brittle matrix materials, but its calculation ability of stress and strain distribution is poor. The main problem with discrete element modeling in composite material processing is that the mathematical formula of the method itself requires special refinement.
2.3 Molecular dynamics simulation
molecular dynamics simulation is mainly used to understand and explain existing experimental results from the perspective of atoms . Using molecular dynamics simulations, some fundamental properties of the nanocomposite can be explored. As shown in Figure 9, using molecular dynamics simulation, it can be concluded that due to the van der Waals force of carbon nanotubes , the carbon nanotube-reinforced epoxy resin composite shows stronger fracture performance and resistance than the pure epoxy resin matrix. Crack growth properties. However, molecular dynamics nanoprocessing simulation is mainly for single crystal materials, and there is currently no research on molecular dynamics simulation of composite material processing.
Figure 9 (a) (b) pure epoxy resin matrix and (c) (d) are the states of carbon nanotube/epoxy resin composite materials under strains of 0.2 and 0.3.
2.4 Multi-scale model
The main purpose of multi-scale modeling is to capture material deformation at different scales and achieve high-precision analysis. Currently, two frameworks are mainly used to combine atomic scale simulation and continuum simulation to achieve atomic size accurate analysis, namely the concurrent framework and the hierarchical framework. In the concurrent framework, the workpiece material is divided into different regions, and simulations of different length scales are run simultaneously through information transfer between different regions. When developing multi-scale modeling based on a concurrent framework, special attention needs to be paid to mechanical processing models of composites and nanocomposites, and should be able to describe the effects of random orientation and waveforms of fibers/ nanotubes at nanometer resolution.
In a hierarchical framework, independent simulations are performed in order of length scales, with the information generated in each scale (usually material properties) passed sequentially from the lower scale to the higher scale. For example, when using a layered framework to handle the processing of A359/SiC/20p composites, molecular dynamics simulation was first used to determine the interface characteristics between aluminum and silicon carbide . Based on this, a contact interface model was established and This model is applied in the finite element model of composite material processing. As shown in Figure 10, the cutting force and subsurface damage depth obtained by this method are in good agreement with the experimental results.
Figure 10 Comparison of experimental and simulation results of (a) cutting force and (b) subsurface damage depth during processing of A359/SiC/20p composite materials.
3 Summary
Composite materials are becoming increasingly important in many engineering fields due to their superior and customizable properties over traditional materials. However, the difficulties encountered in composite material processing are still challenging, mainly reflected in the surface integrity and processing efficiency of processing. In order to gain an in-depth understanding of the processing deformation mechanism of composite materials in future research, three-dimensional analysis based on random cross-scale factors and with micro/nanometer resolution will be crucial. With the deepening of cross-scale research, it is expected that some important issues that have not yet been fully considered in composite material characterization will be solved. Among them, a research direction that is particularly noteworthy is the method of combining digital representation of reliable mechanical models with artificial intelligence (AI), which is expected to become the basis for accurately predicting the surface integrity of composite materials and become an effective method for developing non-destructive manufacturing technology. Tools have become one of the important directions to break through the current bottleneck of cross-scale analysis.
Original document
Liangchi Zhang, Zhonghuai Wu, Chuhan Wu and Qi Wu. On the numerical modeling of composite machining [J]. Composites Part B: Engineering, 2022, 241:110023.
Original link
https://www.sciencedirect .com/science/article/abs/pii/S1359836822004000?via%3Dihub
Welcome to contribute
Call for papers on composite materials mechanics in 2022. The maximum fee for a single article is 1880
focuses on the creation and sharing of knowledge in the field of composite materials mechanics. It is one of the most influential technical exchange platforms in the field of composite materials in China. It updates cutting-edge technologies in the direction of composite materials in the first time, publishing cutting-edge information, simulation cases, technical methods, and codes. The plug-in has helped improve the professionalism and professional skills of countless students, and is deeply loved by the majority of students and young science and technology workers. "Focus on the frontier and lead the future", the Composite Materials Mechanics Public Platform looks forward to your attention!
| Original works may not be reproduced without permission|
Official account homepage reply to the following numbers to obtain the corresponding information
【01】: Historical article【02】: Submission guidelines【03】: Video materials
【04】: Material library【05】 ：Technical training 【06】: Project cooperation
【07】: Business cooperation 【08】: Article reprint 【09】: Join us
【10】: Contact us 【11】: Communication group 【12】: Opinions and suggestions
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-