DIGITAL LIBRARY: CAMX 2024 | SAN DIEGO, CA | SEPTEMBER 9-12

Get This Paper

Predicting Mechanical Properties of High-Temperature Fiber-Reinforced Composites Using Machine Learning Approaches

Description

Title: Predicting Mechanical Properties of High-Temperature Fiber-Reinforced Composites Using Machine Learning Approaches

Authors: Md Shafinur Murad, Abdulhammed K. Hamzat,Eylem Asmatulu, Ersin Bahçeci, Mete Bakir, Ramazan Asmatulu

DOI: 10.33599/nasampe/c.24.0245

Abstract: Fiber-reinforced composites (FRCs) have emerged as a critical material in various industrial applications such as aerospace, defense, medical, energy, and automotive due to their superior strength-to-weight ratio, corrosion resistance, and versatility. The utilization of composite materials in various engineering disciplines relies heavily on the accurate determination of their mechanical properties. However, there are discrepancies in the actual mechanical properties available for designers and manufacturers because of some fabrication process, such as fiber and matrix types, fiber orientations, and stacking sequences, and curing condition. This study explores the application of various machine learning algorithms as a novel modeling tool for the prediction of mechanical properties of fiber-reinforced high temperature composites. Machine learning models such as Linear Regression (LR), Support Vector Regression (SVR), and Multilayer Perceptron (MLP), have been used to develop the prediction model for the tensile and flexural properties of fiber-reinforced composite. Accordingly, homogeneous sets of data were collected from a series of experiments conducted in our laboratory. Reinforcement and matrix type, ply orientation, fiber and matrix content, and stacking sequence are the attributes employed for mechanical property estimation. The performance of the models was estimated using several statistical parameters such as Mean Absolute Error (MAE), Coefficient of Determination (R2), and Root Mean Square Error (RMSE). The test results indicate that support vector regression exhibits a good prediction accuracy with an R-value of 0.996 for flexural strength prediction, while Linear regression showed superior performance with an R2 value of 0.998 for tensile strength prediction. Hyperparameter optimization was carried out to improve the performance of the implemented models. This model can serve as a robust prediction tool specifically tensile and flexural strength of fiber-reinforced composites, thereby modernizing how these properties are evaluated before choosing for intended engineering applications.

References: [1] P. Guo, W. Meng, M. Xu, V. C. Li, and Y. Bao, “Predicting mechanical properties of high-performance fiber-reinforced cementitious composites by integrating micromechanics and machine learning,” Materials, vol. 14, no. 12, p. 3143, 2021. [2] Z. Yu, S. Ye, Y. Sun, H. Zhao, and X.-Q. Feng, “Deep learning method for predicting the mechanical properties of aluminum alloys with small data sets,” Mater Today Commun, vol. 28, p. 102570, 2021. [3] F. Kibrete, T. Trzepieciński, H. S. Gebremedhen, and D. E. Woldemichael, “Artificial intelligence in predicting mechanical properties of composite materials,” Journal of Composites Science, vol. 7, no. 9, p. 364, 2023. [4] W. D. Cheng, C. Z. Cai, Y. Luo, Y. H. Li, and C. J. Zhao, “Mechanical properties prediction for carbon nanotubes/epoxy composites by using support vector regression,” Modern Physics Letters B, vol. 29, no. 05, p. 1550016, 2015. [5] S. Bhattacharya, K. Kalita, R. Čep, and S. Chakraborty, “A comparative analysis on prediction performance of regression models during machining of composite materials,” Materials, vol. 14, no. 21, p. 6689, 2021. [6] A. Mahajan et al., “Comparison of ML algorithms for prediction of tensile strength of polymer matrix composites,” Mater Today Proc, 2022. [7] R. Ramprasad, R. Batra, G. Pilania, A. Mannodi-Kanakkithodi, and C. Kim, “Machine learning in materials informatics: recent applications and prospects,” NPJ Comput Mater, vol. 3, no. 1, p. 54, 2017. [8] K. A. Krishnan, R. Anjana, and K. E. George, “Effect of alkali‐resistant glass fiber on polypropylene/polystyrene blends: Modeling and characterization,” Polym Compos, vol. 37, no. 2, pp. 398–406, 2016. [9] F. Khademi, M. Akbari, S. M. Jamal, and M. Nikoo, “Multiple linear regression, artificial neural network, and fuzzy logic prediction of 28 days compressive strength of concrete,” Frontiers of Structural and Civil Engineering, vol. 11, pp. 90–99, 2017. [10] C. Yang, Y. Kim, S. Ryu, and G. X. Gu, “Prediction of composite microstructure stress-strain curves using convolutional neural networks,” Mater Des, vol. 189, p. 108509, 2020. [11] M. Maurizi, C. Gao, and F. Berto, “Predicting stress, strain and deformation fields in materials and structures with graph neural networks,” Sci Rep, vol. 12, no. 1, p. 21834, 2022. [12] A. Sharan and M. Mitra, “Prediction of static strength properties of carbon fiber-reinforced composite using artificial neural network,” Model Simul Mat Sci Eng, vol. 30, no. 7, p. 075001, 2022. [13] A. Barbosa, P. Upadhyaya, and E. Iype, “Neural network for mechanical property estimation of multilayered laminate composite,” Mater Today Proc, vol. 28, pp. 982–985, 2020. [14] M. V Pathan et al., “Predictions of the mechanical properties of unidirectional fibre composites by supervised machine learning,” Sci Rep, vol. 9, no. 1, p. 13964, 2019. [15] J. Liu, Y. Zhang, Y. Zhang, S. Kitipornchai, and J. Yang, “Machine learning assisted prediction of mechanical properties of graphene/aluminium nanocomposite based on molecular dynamics simulation,” Mater Des, vol. 213, p. 110334, 2022. [16] K. P. S. Attwal and A. S. Dhiman, “Exploring data mining tool-Weka and using Weka to build and evaluate predictive models,” Adv Appl Math Sci, vol. 19, no. 6, pp. 451–469, 2020. [17] D. Stiawan, M. A. S. Arifin, M. Y. Idris, and R. Budiarto, “IoT botnet malware classification Using Weka Tool and scikit-learn machine learning,” in 2020 7th International Conference on Electrical Engineering, Computer Sciences and Informatics (EECSI), 2020, pp. 15–20. [18] C. N. Villavicencio, J. J. E. Macrohon, X. A. Inbaraj, J.-H. Jeng, and J.-G. Hsieh, “Covid-19 prediction applying supervised machine learning algorithms with comparative analysis using weka,” Algorithms, vol. 14, no. 7, p. 201, 2021. . [19] T.K.S. Murali, M.S. Murad, M. Bakir, and R. Asmatulu “PAN-Based Fiber Reinforced Carbon-Carbon Composites for Improved Fire Retardancy and Thermal and Electrical Conductivities for Harsh Environments,” Journal of Composite Materials, 2024 (in press). [20] M.S. Murad, E. Asmatulu, E., O. Er, M. Gursoy, B. Safaker, E. Bahceci, M. Bakir, and R. Asmatulu “Improving Flame Retardancy of Fiber Reinforced Composites via Modified Fire-Resistant Resins and Metallic Thin Film Coatings,” CAMX Conference, Anaheim, CA, October 17-20, 2022, 8 pages.

Conference: CAMX 2024 | San Diego CA

Publication Date: 2024/9/9

SKU: TP24-0000000245

Pages: 10

Price: $20.00

Get This Paper