Automatic fiber laying interlayer performance optimization method and system based on molecular dynamics

Abstract

The invention belongs to the field of composite material manufacturing, and particularly discloses an automatic fiber laying interlayer performance optimization method and system based on molecular dynamics, and the method comprises the steps: building an interface bonding model based on a molecular model of continuous carbon fibers, resin and carbon nanotubes, and carrying out the pre-balance treatment of the interface bonding model; performing automatic fiber laying simulation based on preset process parameters to obtain a temperature field and a pressure field of the representative unit; performing molecular dynamics analysis based on the interface combination model and the temperature field and the pressure field of the representative unit; calculating the diffusion coefficient of the carbon nano tube according to a molecular dynamics analysis result, and further determining the interface binding energy of the resin and the carbon nano tube; based on the interface bonding energy of the resin and the carbon nanotubes, automatic fiber laying process parameters are selected, so that the interlayer performance of automatic fiber laying is optimized. According to the method, the interlayer performance of automatic fiber laying can be efficiently and accurately optimized.

Description

Technical Field

[0001] This invention belongs to the field of composite material manufacturing, and more specifically, relates to an automated fiber placement interlayer performance optimization method and system based on molecular dynamics. Background Technology

[0002] Continuous carbon fiber composites possess excellent properties such as high specific strength and high specific stiffness, and are widely used in transportation and other fields. However, during the automated fiber placement process, the short laser heating and pressure application time leads to resin-rich zones between the layers of the automated fiber placement component, severely affecting its interlaminar performance. In these resin-rich zones, there is no fiber reinforcement between layers; they are interconnected through the resin matrix, resulting in weak interlaminar performance. When the laminate is subjected to external forces, microcracks first appear in the weakest direction of the laminate, and as the cracks propagate, they eventually form large cracks between the layers, leading to delamination failure and limiting its application in critical load-bearing structural components.

[0003] Current research on optimizing the interlayer performance of automated fiber placement mainly uses methods such as directly changing process parameters or directly adding carbon nanotubes. However, it has not yet adopted molecular dynamics methods at the microscopic level to study the mechanism of enhanced interlayer performance of automated fiber placement, so as to optimize the interlayer performance of automated fiber placement more efficiently and accurately. Summary of the Invention

[0004] In view of the above-mentioned defects or improvement needs of the existing technology, the present invention provides an automated fiber placement interlayer performance optimization method and system based on molecular dynamics, the purpose of which is to efficiently and accurately optimize the interlayer performance of automated fiber placement.

[0005] To achieve the above objectives, according to a first aspect of the present invention, an automated fiber placement interlayer performance optimization method based on molecular dynamics is proposed, comprising the following steps: Based on the molecular model of continuous carbon fibers, resins, and carbon nanotubes, an interface bonding model was constructed, and the interface bonding model was pre-equilibrium treated. Automatic fiber placement simulation was performed based on preset process parameters to obtain the temperature and pressure fields of representative units. Molecular dynamics analysis was performed based on the interface bonding model and the temperature and pressure fields of representative units. The diffusion coefficient of carbon nanotubes was calculated based on the molecular dynamics analysis results, and then the interfacial bonding energy between the resin and carbon nanotubes was determined. Based on the interfacial bonding energy between resin and carbon nanotubes, the parameters of the automated fiber placement process are selected to optimize the interlayer performance of automated fiber placement.

[0006] As a further preferred option, automatic fiber placement simulation was performed based on preset process parameters to obtain the temperature and pressure fields of representative units; Based on the temperature field distribution obtained from the temperature field model, the temperature field distribution is imported into the mechanical model as a boundary condition. Fixed constraints are set for the bottom layer, binding constraints are set between each layer of the laminate, and the required pressure roller force is applied to the pressure roller above the layer to be laid, forming an overall thermo-mechanical coupling model, and the temperature field and stress field of the representative unit are output.

[0007] As a further preferred approach, molecular dynamics analysis is performed based on the interface bonding model and the temperature and pressure fields of representative units, including: Based on the interface-based model, molecular dynamics analysis was performed using the NPT ensemble. The temperature and pressure fields of representative units were used as boundary conditions for the NPT ensemble to ensure constant pressure and temperature operating conditions. The force field was set as the Compass force field, and the Nose method was used to control the statistical temperature of the NPT ensemble.

[0008] As a further preferred approach, an interface bonding model is constructed based on a molecular model of continuous carbon fibers, resin, and carbon nanotubes, specifically as follows: The molecular models of continuous carbon fibers, resins, and carbon nanotubes are used as separate modules of composite components to construct the AC model, and then the interface bonding model is constructed based on the AC model.

[0009] As a further preferred option, the interface combination model undergoes pre-balancing processing, including: By performing relaxation, annealing, and calculations of the NPT and NVT ensembles at room temperature and pressure, unreasonable structures in the interface bonding model are eliminated, providing an initial structure with minimized energy for subsequent molecular dynamics analysis.

[0010] As a further preferred method, the diffusion coefficient of carbon nanotubes is calculated based on molecular dynamics analysis results, using the following formula:

[0011] In the formula, D represents the diffusion coefficient of carbon nanotubes, and Na represents the number of particles. x Indicates displacement; This represents the position vector of the i-th particle at time t. This represents the position vector of the i-th particle at the initial moment. and Determined based on molecular dynamics analysis results.

[0012] As a further preferred method, the interfacial bonding energy between the resin and carbon nanotubes is determined based on the diffusion coefficient of the carbon nanotubes, and the calculation formula is as follows:

[0013] In the formula, Indicates interface bonding energy, This represents the total energy of the interface. Indicates resin energy; This represents the energy of the carbon nanotube, which is determined based on the carbon nanotube diffusion coefficient.

[0014] As a further preferred embodiment, the process parameters include laying temperature, pressure, and speed.

[0015] According to a second aspect of the present invention, an automated fiber placement interlayer performance optimization system based on molecular dynamics is provided, comprising a processor for executing the above-described automated fiber placement interlayer performance optimization method based on molecular dynamics.

[0016] According to a third aspect of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the above-described method for optimizing interlayer performance of automated fiber placement based on molecular dynamics.

[0017] In summary, compared with the prior art, the above-described technical solutions conceived by this invention mainly possess the following technical advantages: This invention employs a multi-scale analysis method, conducting a coordinated analysis from macroscopic to microscopic levels. It uses the temperature and stress fields of the RVE unit obtained from the thermo-mechanical coupling simulation of the automated fiber placement process as boundary conditions for molecular dynamics analysis. Furthermore, it analyzes the diffusion of carbon nanotubes and the corresponding interlayer binding energy between carbon nanotubes and resin during the entire microscopic motion process, thereby reflecting the interlayer performance of the automated fiber placement. The invention observes the specific numerical analysis results under different process conditions, explaining the mechanism of carbon nanotube-enhanced interlayer performance in automated fiber placement at the microscopic level. Based on this, it selects the optimal placement process parameters, thereby efficiently and accurately optimizing the interlayer performance of automated fiber placement and overcoming the constraints imposed by interlayer mechanical properties on key load-bearing components. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the process for optimizing interlayer performance of automated fiber placement based on molecular dynamics, according to an embodiment of the present invention. Figure 2 This is a schematic diagram of the overall process of automatic fiber placement according to an embodiment of the present invention; Figure 3 In the middle (a) to (d), respectively, schematic diagrams of the polyether ether ketone molecular model, carbon nanotube molecular model, continuous carbon fiber molecular model, and interface bonding model of the present invention are shown. Figure 4 This is a schematic diagram of the simulation results of the thermo-coupling model in an embodiment of the present invention; Figure 5This is a schematic diagram of the specific MSD (mean square displacement) results and curve fitting of an embodiment of the present invention; Figure 6 Figures (a) to (c) are schematic diagrams of interface combination models with different degrees of aggregation in the embodiments of the present invention, namely Model 1 to Model 3.

[0019] In all the accompanying drawings, the same reference numerals are used to denote the same elements or structures, wherein: 1-1-mold, 1-2-laminate, 1-3-RVE unit, 1-4-pressure roller, 1-5-micro system. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0021] This invention provides an automated fiber placement interlayer performance optimization method based on molecular dynamics, such as... Figure 1 As shown, it includes the following steps: S1. Construct molecular models and interface binding models, and perform pre-equilibrium treatment on the interface binding models.

[0022] Before automated fiber layup, continuous carbon fibers are pre-mixed with resin to form a prepreg, which is then laid layer by layer. Because resin-rich zones appear between layers, carbon nanotubes are often added to optimize performance. Carbon nanotubes can be mixed with continuous carbon fibers and resin to form the prepreg, or they can be directly introduced during the layup of the continuous carbon fibers and resin prepreg. Figure 2 As shown, prepreg is laid layer by layer on mold 1-1 by pressure rollers 1-4 to obtain the laid-up laminate 1-2. Specifically, the prepreg is fed to the end of the pressure rollers at a certain laying speed. A laser heat source heats the prepreg until the resin melts, and the prepreg strip is rapidly solidified under high pressure by the pressure rollers, achieving interlayer bonding. After laying, it is cut by a cutting device and re-fed by a feeding device, thus completing one laying and forming cycle, and finally obtaining the required laminate.

[0023] For automated fiber placement, molecular models of continuous carbon fibers, resins, and carbon nanotubes were constructed respectively. Based on these, an AC (Amorphous Cell) model was established, and each molecular model was used as a separate module of the composite component to complete the construction of the AC model, thereby realizing the establishment of the interface bonding model.

[0024] Furthermore, the interface bonding model undergoes pre-equilibrium processing, including: optimizing the geometry of the constructed interface bonding model, performing relaxation and annealing, and using the Forcite module to calculate the NPT ensemble and NVT ensemble at room temperature and pressure, eliminating unreasonable initial structures in the overall interface model, obtaining a structure close to equilibrium, and providing an initial structure with minimum energy for subsequent molecular dynamics calculations.

[0025] S2. Perform automatic fiber placement process analysis and establish a thermo-mechanical coupling model to obtain the temperature and pressure fields of RVE units 1-3 (representative units).

[0026] Specifically, an automatic fiber placement simulation analysis is performed based on preset process parameters, including placement temperature, pressure, and speed. First, a temperature field model is constructed, and the obtained temperature field distribution is imported into the mechanical model as boundary conditions. Fixed constraints are set for the bottom layer, and binding constraints are set between each layer of the laminate. The required pressure roller force is applied above the layer to be laid, thereby obtaining the overall thermo-mechanical coupling model and outputting the temperature field and stress field of the RVE element.

[0027] S3. Molecular dynamics analysis is performed based on the temperature and stress fields of the RVE unit.

[0028] Specifically, the interface model simulates real interface contact from a microscopic perspective, and then molecular dynamics calculations are used to analyze the changes in interface properties under different process parameters. During the molecular dynamics analysis, the temperature and pressure of the RVE unit are imported as boundary conditions to serve as the boundary conditions for the NPT ensemble, ensuring constant pressure and temperature operating conditions. Furthermore, the force field is set to a Compass force field in the molecular dynamics software, and the Nose method is used to control the statistical temperature of the ensemble. The main idea of ​​the Nose method is to add a new degree of freedom to the original system, representing the energy exchange between the system and a large heat bath, as shown in the following formula;

[0029]

[0030]

[0031]

[0032] In the formula q i p i and m i Let represent the generalized coordinates, generalized momentum, and mass of the i-th particle, respectively. The momentum corresponding to the introduced thermal bath degree of freedom is represented by , and g represents the total number of degrees of freedom in the ensemble. Let T represent the Boltzmann constant, T represent the expected target temperature, and Q represent the mass constant associated with the degrees of freedom of the heat bath. The velocity distribution function is represented by v, which represents the magnitude of the molecular velocity. The NPT ensemble (isothermal and isobaric ensemble) is selected for molecular dynamics calculation and analysis. The charge type in the model system is matched, and the total motion time is controlled to be 50Ps. The data output of molecular dynamics calculation is performed at intervals of 0.1Ps.

[0033] S4. Analyze the diffusion coefficient of carbon nanotubes based on molecular dynamics analysis results.

[0034] Specifically, the molecular dynamics analysis results are post-processed, mean square displacement analysis is performed, and the diffusion coefficient of carbon nanotubes under different operating conditions is analyzed according to the following formula:

[0035] In the formula, D represents the diffusion coefficient of carbon nanotubes, and Na represents the number of particles. This represents the position vector of the i-th particle at time t. This represents the position vector of the i-th particle at the initial moment. and Determined based on molecular dynamics analysis results.

[0036] S5. Analyze the interfacial binding energy based on the diffusion coefficient of carbon nanotubes.

[0037] Specifically, the overall interfacial energy is analyzed, then the molecular energy of each component is calculated and analyzed, the carbon nanotube energy is determined based on the carbon nanotube diffusion coefficient, and finally the interfacial bonding energy between the resin and the carbon nanotube is obtained. The calculation formula is as follows:

[0038] In the formula, Indicates interface bonding energy, This represents the total energy of the interface. Indicates the energy of carbon nanotubes. This indicates the resin's energy.

[0039] The increased diffusion coefficient of carbon nanotubes leads to a corresponding increase in interfacial binding energy, resulting in improved interlayer performance in automated fiber placement. Based on this method, the interlayer performance corresponding to different placement process parameters can be rapidly obtained, allowing for efficient optimization of automated fiber placement process parameters and thus improving interlayer performance.

[0040] The following are specific examples: S1. Construction of Molecular and Interface Models: Using the data model library built into MS (Materials Studio) software, the unit cell models of graphene and carbon nanotubes were imported. Based on these unit cell models, (6,6) armchair-shaped carbon nanotubes with a length of 30 Å were constructed as the carbon nanotube molecular model. The graphene unit cell model was expanded to represent 12 layers of continuous carbon fibers, serving as the continuous carbon fiber molecular model. For the resin, PEEK (polyetheretherketone) was used. Based on the chemical structure of PEEK, a molecular model of PEEK was constructed, and a long-chain PEEK molecule with a degree of polymerization of 20 was established. This molecule was then filled into a 50 Å × 50 Å × 50 Å box to establish the resin molecular model, with a density of 1.3 g / cm³. 3 The AC model was constructed by treating each molecular model as a separate module of the composite component, thereby establishing an interface-based model. The overall model result is as follows: Figure 3 As shown.

[0041] The constructed interface model is geometrically optimized, and the NPT and NVT ensembles are calculated at room temperature and pressure using the Forcite module. This eliminates unreasonable initial structures in the overall interface model, obtains a structure close to equilibrium, and provides an initial structure with minimum energy for subsequent molecular dynamics calculations.

[0042] S2. Perform automatic fiber placement process analysis and construct a thermo-mechanical coupling model, outputting the temperature field and stress field of the RVE unit under different placement process parameters. Examples of simulation numerical results are shown below. Figure 4 As shown in Table 1, the relevant numerical parameters in the simulation process are as follows.

[0043] Table 1. Specific values ​​of main parameters

[0044] S3. During the molecular dynamics calculations, the temperature and pressure of the RVE unit are imported as boundary conditions to serve as the boundary conditions for the NPT ensemble, ensuring constant pressure and temperature operating conditions. The molecular dynamics force field is set as the Compass force field, and the Nose method is used to control the statistical temperature of the ensemble.

[0045] S4. Post-process the molecular dynamics calculation results to calculate the diffusion coefficient of carbon nanotubes. The specific results and corresponding curve fitting are as follows: Figure 5 As shown, under the same process conditions, as the temperature increases, the slope of the mean square displacement also increases, that is, the diffusion coefficient of carbon nanotubes also increases.

[0046] S5. Analyze the overall interface energy, calculate and analyze the molecular energy of each component, and obtain interface bonding models with different degrees of aggregation, such as... Figure 6 As shown in the figure, the corresponding interface bonding energy is shown in Table 2.

[0047] Table 2. Calculation of binding energy of interface models with different degrees of aggregation.

[0048] As the diffusion degree of carbon nanotubes increases, the interfacial binding energy also increases. This indicates that the problem of resin-rich regions between layers in the original automated fiber placement process can be solved by adding carbon nanotubes. Furthermore, when the laminate undergoes delamination, the higher the corresponding interfacial binding energy, the less likely the layers are to separate, and the additional energy required to pull the carbon nanotubes out of the resin can greatly improve the interfacial performance of the laminate. According to Table 2, Model 3 has the highest interfacial binding energy, meaning that when using its corresponding placement process parameters for automated fiber placement, the interfacial performance is optimal.

[0049] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for optimizing interlayer performance of automated fiber placement based on molecular dynamics, characterized in that, Includes the following steps: Based on the molecular model of continuous carbon fibers, resins, and carbon nanotubes, an interface bonding model was constructed, and the interface bonding model was pre-equilibrium treated. Automatic fiber placement simulation was performed based on preset process parameters to obtain the temperature and pressure fields of representative units. Molecular dynamics analysis was performed based on the interface bonding model and the temperature and pressure fields of representative units. The diffusion coefficient of carbon nanotubes was calculated based on the molecular dynamics analysis results, and then the interfacial bonding energy between the resin and carbon nanotubes was determined. Based on the interfacial bonding energy between resin and carbon nanotubes, the parameters of the automated fiber placement process are selected to optimize the interlayer performance of automated fiber placement.

2. The method for optimizing interlayer performance of automated fiber placement based on molecular dynamics as described in claim 1, characterized in that, Automatic fiber placement simulation was performed based on preset process parameters to obtain the temperature and pressure fields of representative units. Based on the temperature field distribution obtained from the temperature field model, the temperature field distribution is imported into the mechanical model as a boundary condition. Fixed constraints are set for the bottom layer, binding constraints are set between each layer of the laminate, and the required pressure roller force is applied to the pressure roller above the layer to be laid, forming an overall thermo-mechanical coupling model, and the temperature field and stress field of the representative unit are output.

3. The method for optimizing interlayer performance of automated fiber placement based on molecular dynamics as described in claim 1, characterized in that, Molecular dynamics analysis was performed based on the interface bonding model and the temperature and pressure fields of representative units, including: Based on the interface-based model, molecular dynamics analysis was performed using the NPT ensemble. The temperature and pressure fields of representative units were used as boundary conditions for the NPT ensemble to ensure constant pressure and temperature operating conditions. The force field was set as the Compass force field, and the Nose method was used to control the statistical temperature of the NPT ensemble.

4. The method for optimizing interlayer performance of automated fiber placement based on molecular dynamics as described in claim 1, characterized in that, Based on the molecular model of continuous carbon fibers, resins, and carbon nanotubes, an interface bonding model is constructed, specifically as follows: The molecular models of continuous carbon fibers, resins, and carbon nanotubes are used as separate modules of composite components to construct the AC model, and then the interface bonding model is constructed based on the AC model.

5. The method for optimizing interlayer performance of automated fiber placement based on molecular dynamics as described in claim 1, characterized in that, Pre-balancing of the interface-based model includes: By performing relaxation, annealing, and calculations of the NPT and NVT ensembles at room temperature and pressure, unreasonable structures in the interface bonding model are eliminated, providing an initial structure with minimized energy for subsequent molecular dynamics analysis.

6. The method for optimizing interlayer performance of automated fiber placement based on molecular dynamics as described in claim 1, characterized in that, The diffusion coefficient of carbon nanotubes was calculated based on molecular dynamics analysis results, using the following formula: In the formula, D represents the diffusion coefficient of carbon nanotubes, and Na represents the number of particles. x Indicates displacement; This represents the position vector of the i-th particle at time t. This represents the position vector of the i-th particle at the initial moment. and Determined based on molecular dynamics analysis results.

7. The method for optimizing interlayer performance of automated fiber placement based on molecular dynamics as described in claim 1, characterized in that, The interfacial bonding energy between the resin and carbon nanotubes is determined based on the diffusion coefficient of carbon nanotubes. The calculation formula is as follows: In the formula, Indicates interface bonding energy, This represents the total energy of the interface. Indicates resin energy; This represents the energy of the carbon nanotube, which is determined based on the carbon nanotube diffusion coefficient.

8. The method for optimizing interlayer performance of automated fiber placement based on molecular dynamics as described in any one of claims 1-7, characterized in that, The process parameters include laying temperature, pressure, and speed.

9. An automated fiber placement interlayer performance optimization system based on molecular dynamics, characterized in that, Includes a processor for executing the molecular dynamics-based automated fiber placement interlayer performance optimization method as described in any one of claims 1-8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the molecular dynamics-based automated fiber placement interlayer performance optimization method as described in any one of claims 1-8.

Images