| Citation: |
SONG Feng, ZHANG Jiachen, LYU Bingyi, et al. Influence of multiple process parameters on the friction coefficient of prepregs and machine learning prediction method[J]. Acta Materiae Compositae Sinica, 2024, 41(11): 5935-5945. DOI: 10.13801/j.cnki.fhclxb.20240304.001
|
During the forming process of composites, the friction-sliding behavior between prepreg ply-ply and ply-tool may lead to defects such as wrinkles and pores, which seriously affect the mechanical properties of the components. However, there are many factors affecting the inter-ply friction of the prepreg plies in the forming process of complex components. The existing theoretical models contain insufficient process parameters, resulting in the accuracy of forming process simulation not meeting high-quality forming requirements. In this paper, a friction test method for carbon fiber prepregs was designed for multiple process parameters. The influence of sliding velocity, normal force, viscosity, surface roughness, contact material, and fiber orientation on the friction coefficient were studied. Taking the typical fiber orientations of as examples, the inter-ply friction mechanism in different fiber orientations was revealed. In order to predict the friction coefficient of prepreg corresponding to multiple process parameters rapidly and accurately, a prediction model for the friction coefficient of prepreg was established using the support vector regression (SVR) method. Taking the prepreg ply-ply friction behavior with relative fiber orientation of and as examples, the experiments and predictions were conducted, and the error was less than 9%.
There are various deformation mechanisms during the forming process of composites, among which the friction-sliding behavior between prepreg ply-ply and ply-tool plays an important role in the structure and product quality. The inter-ply friction coefficient is subject to the coupling effect of multiple process parameters, and improper combinations will lead to defects such as wrinkles and pores, which seriously affect the mechanical properties of components. However, the existing theoretical models contain insufficient process parameters, resulting in the accuracy of forming process simulation not meeting high-quality forming requirements. In this paper, a friction test method was designed to characterize the inter-ply friction performance of carbon fiber prepreg corresponding to multiple process parameters. The fast and accurate prediction of the friction coefficient of carbon fiber prepreg has been achieved based on machine learning algorithms.
A friction test method for carbon fiber prepregs was designed for multiple process parameters, and the influence of sliding velocity, normal force, viscosity, surface roughness, contact material, and fiber orientation on the friction coefficient was studied. Taking the typical fiber orientations of as examples, the inter-ply friction mechanism in different fiber orientations was revealed. A prediction model for the friction coefficient of prepreg was established using the support vector regression (SVR) method. The reliability and applicability of the model have been verified by comparing the training and experimental results.
The results show that due to the hydrodynamic lubrication effect of the resin during sliding and its behavior exhibiting Newtonian shear characteristics, the static/kinetic friction coefficient of the prepreg/metal interface increases with the increase of sliding velocity. With the increase of normal force, the prepreg becomes thinner and the fiber surface becomes smoother. In addition, the resin extrusion overflows to the interface, resulting in a decrease in the roughness of the interface. Therefore, the friction coefficient gradually decreases. As the viscosity increases, stronger external forces are needed to overcome the resin bonding force at the interface, resulting in an increase in tangential friction and a corresponding increase in friction coefficient. Due to the material stiffness and surface roughness, the friction coefficient of different contact materials shows significant differences. Under the same resin viscosity, sliding velocity and normal force, the friction coefficient at the contact interface between the contact material and the prepreg is greater when the contact material has lower stiffness or higher surface roughness. The change in fiber direction has a significant impact on the friction coefficient between the prepreg ply-ply interface. The interface has the highest friction coefficient, and the interface has a higher friction coefficient than the interface. The determination coefficients of the SVR model for predicting the static/kinetic friction coefficients are 0.930 and 0.938, respectively, and the variances of the prediction errors are 0.0016 and 0.0011. Compared with BPNN and RF methods, the SVR model has significant advantages, proving its ability to predict the friction coefficient of carbon fiber prepreg. Taking the interface of prepreg ply-ply in the fiber direction of and as examples, the friction coefficient were predicted using the prediction model. The predicted results and experimental values fit well, verifying the generalization ability of the model.Conclusions：When inter-ply slip occurs in prepreg, process parameters such as sliding velocity, normal force, viscosity, surface roughness, contact material, fiber direction, have a significant impact on the friction coefficient, and the interface with different fiber directions has different friction mechanisms. The proposed friction coefficient prediction model based on SVR with multiple process parameters can quickly and effectively predict the static/kinetic friction coefficients of prepreg under different contact interfaces and process parameters, and has good generalization ability, providing a fast method for predicting friction coefficients.
| [1] |
TEN THIJE R H W, AKKERMAN R, UBBINK M, et al. A lubrication approach to friction in thermoplastic composites forming processes[J]. Composites Part A: Applied Science and Manufacturing, 2011, 42(8): 950-960. DOI: 10.1016/j.compositesa.2011.03.023
|
| [2] |
李哲夫. 航空复合材料热模压预成型缺陷形成机理与仿真预测研究[D]. 上海: 东华大学, 2023.
LI Zhefu. Study on the formation mechanism and simulation prediction of defects in the process of hot-press forming for aerospace composites[D]. Shanghai: Donghua University, 2023(in Chinese).
|
| [3] |
矫维成, 王荣国, 刘文博, 等. 缠绕纤维与芯模表面间滑线系数的测量表征[J]. 复合材料学报, 2012(3): 191-196.
JIAO Weicheng, WANG Rongguo, LIU Wenbo, et al. Measurement of slippage coefficient between fiber and mandrel surface for non-geodesic filament winding[J]. Acta Materiae Compositae Sinica, 2012(3): 191-196(in Chinese).
|
| [4] |
MARTIN C J, SEFERIS J C, WILHELM M A. Frictional resistance of thermoset prepregs and its influence on honeycomb composite processing[J]. Composites Part A: Applied Science and Manufacturing, 1996, 27(10): 943-951. DOI: 10.1016/1359-835X(96)00037-1
|
| [5] |
ÅKERMO M, LARBERG Y R, SJÖLANDER J, et al. Influence of interply friction on the forming of stacked UD prepreg[C]. The International Conference on Composite Materials. Palais Des Congrès, Montréal: Concordia University, 2013: 919-928.
|
| [6] |
BENDEMRA H, COMPSTON P, CROTHERS P J. Optimisation study of tapered scarf and stepped-lap joints in composite repair patches[J]. Composite Structures, 2015, 130: 1-8. DOI: 10.1016/j.compstruct.2015.04.016
|
| [7] |
LIU S, SINKE J, DRANSFELD C. An inter-ply friction model for thermoset based fibre metal laminate in a hot-pressing process[J]. Composites Part B: Engineering, 2021, 227: 109400. DOI: 10.1016/j.compositesb.2021.109400
|
| [8] |
DUTTA A, HAGNELL M K, ÅKERMO M. Interply friction between unidirectional carbon/epoxy prepreg plies: Influence of fibre orientation[J]. Composites Part A: Applied Science and Manufacturing, 2023, 166: 107375. DOI: 10.1016/j.compositesa.2022.107375
|
| [9] |
LARBERG Y R, ÅKERMO M. On the interply friction of different generations of carbon/epoxy prepreg systems[J]. Composites Part A: Applied Science and Manufacturing, 2011, 42(9): 1067-1074. DOI: 10.1016/j.compositesa.2011.04.010
|
| [10] |
HALLANDER P, ÅKERMO M, MATTEI C, et al. An experimental study of mechanisms behind wrinkle development during forming of composite laminates[J]. Composites Part A: Applied Science and Manufacturing, 2013, 50: 54-64. DOI: 10.1016/j.compositesa.2013.03.013
|
| [11] |
SOSE A T, JOSHI S Y, KUNCHE L K, et al. A review of recent advances and applications of machine learning in tribology[J]. Physical Chemistry Chemical Physics, 2023, 25(6): 4408-4443. DOI: 10.1039/D2CP03692D
|
| [12] |
PATURI U M R, PALAKURTHY S T, REDDY N S. The role of machine learning in tribology: A systematic review[J]. Archives of Computational Methods in Engineering, 2023, 30: 1345-1397. DOI: 10.1007/s11831-022-09841-5
|
| [13] |
BAŞ H, KARABACAK Y E. Triboinformatic modeling of the friction force and friction coefficient in a cam-follower contact using machine learning algorithms[J]. Tribology International, 2023, 181: 108336. DOI: 10.1016/j.triboint.2023.108336
|
| [14] |
WU B, QIN D, HU J, et al. Experimental data mining research on factors influencing friction coefficient of wet clutch[J]. Journal of Tribology, 2021, 143(12): 121802. DOI: 10.1115/1.4050140
|
| [15] |
WANG Q, WANG X, ZHANG X, et al. Tribological properties study and prediction of PTFE composites based on experiments and machine learning[J]. Tribology International, 2023, 188: 108815. DOI: 10.1016/j.triboint.2023.108815
|
| [16] |
EGALA R, JAGADEESH G V, SETTI S G. Experimental investigation and prediction of tribological behavior of unidirectional short castor oil fiber reinforced epoxy composites[J]. Friction, 2021, 9: 250-272. DOI: 10.1007/s40544-019-0332-0
|
| [17] |
NIRMAL U. Prediction of friction coefficient of treated betelnut fibre reinforced polyester (T-BFRP) composite using artificial neural networks[J]. Tribology International, 2010, 43(8): 1417-1429. DOI: 10.1016/j.triboint.2010.01.013
|
| [18] |
American Society for Testing and Materials. Standard test method for static and kinetic coefficients of friction of plastic film and sheeting: ASTM D1894[S]. West Conshohoken: American Society for Testing and Materials Standard International, 1998.
|
| [19] |
ZHANG W, ZHOU H, HUANG B, et al. Characterization of tool-ply friction behavior for treated jute/PLA biocomposite prepregs in thermoforming[J]. Composites Part A: Applied Science and Manufacturing, 2024, 177: 107875. DOI: 10.1016/j.compositesa.2023.107875
|
| [20] |
RASHIDI A, MONTAZERIAN H, YESILCIMEN K, et al. Experimental characterization of the inter-ply shear behavior of dry and prepreg woven fabrics: Significance of mixed lubrication mode during thermoset composites processing[J]. Composites Part A: Applied Science and Manufacturing, 2020, 129: 105725. DOI: 10.1016/j.compositesa.2019.105725
|
| [21] |
薛建凯. 一种新型的群智能优化技术的研究与应用: 麻雀搜索算法[D]. 上海: 东华大学, 2021.
XUE Jiankai. Research and application of a novel swarm intelligence optimization technique: Sparrow search algorithm[D]. Shanghai: Donghua University, 2021(in Chinese).
|
Supported by: Beijing Renhe Information Technology Co., Ltd. 
DownLoad: