DEEP LEARNING-BASED PREDICTION AND EXPERIMENTAL EVALUATION OF MECHANICAL PROPERTIES IN BASALT-FIBER-SiC-REINFORCED HYBRID EPOXY COMPOSITES

SM Ravikumar; JM Prabhudass; A Adinaryanan; N Yuvaraj; S Dinesh; A Elakkiya

doi:10.17222/mit.2025.1545

SM Ravikumar Department of Mechanical Engineering, Vandayar Engineering College, Thanjavur, TamilNadu, India, 613501
JM Prabhudass Department of Mechanical Engineering, Sri Sairam Institute of Technology, Tambaram, Chennai, Tamil Nadu, India 600044
A Adinaryanan Faculty of Mechanical Engineering, AMET University, Chennai, Tamil Nadu, India 603112
N Yuvaraj Department of Computer Science Engineering, Adhi College of Engineering and Technology, Sankarapuram, Chennai, Tamil Nadu, India 631605
S Dinesh Department of Mechanical Engineering, Dhanalakshmi College of Engineering, Chennai – 601301
A Elakkiya Department of Computer Applications, Karpagam Academy of Higher Education, Coimbatore, TamilNadu, India 641021

DOI: https://doi.org/10.17222/mit.2025.1545

Keywords: Basalt fiber, Silicon carbide filler, Mechanical properties, Deep Neural Network , Hybrid composite for structural applications

Abstract

This study investigates the predictive capabilities of deep neural network (DNN) models in estimating the mechanical properties of basalt-fiber-reinforced hybrid composites (BFRHCs). A total of 27 composite samples were fabricated using the compression-molding technique, incorporating varying numbers of basalt-fiber layers, fiber orientation angles (0°, 45°, 90°), and silicon carbide (SiC) filler contents (0, 3 and 6 w/%). The fabricated specimens were subjected to mechanical characterization by ASTM standards to evaluate their tensile strength, flexural strength, impact strength, hardness, shear strength, and interdelamination resistance. A DNN model was developed and trained on the experimental dataset to capture the complex, nonlinear relationships between input fabrication parameters and output mechanical properties. Model performance was assessed using the coefficient of determination (R²), mean absolute error (MAE), and root-mean-square error (RMSE). The DNN achieved high predictive accuracy with R² values exceeding 0.95 for most properties, demonstrating its effectiveness in forecasting composite performance. The results confirm that deep-learning frameworks such as DNNs offer a powerful and reliable approach to predicting the behavior of hybrid composites, reducing the need for extensive experimental trials and supporting efficient material design and optimization in structural applications.

References

1Y.S. Amruth, K.N. Mohandas, and L. Sunith Babu, “Mechanical characteristics study on basalt/epoxy laminates,” Materials Today: Proceedings, 46 (10), 4407–4411 (2021). https://doi.org/10.1016/j.matpr.2020.09.670H.
2W. Yang, H. Liu, and H. Wang, “Experimental study on mechanical properties of basalt fiber reinforced nano-SiO₂ concrete after high temperature,” Frontiers in Materials, 11(4), 1-15 (2024). https://doi.org/10.3389/fmats.2024.1415144
3S. Rai and P. Rawat, “Three-point bending, interlaminar shear, and impact strength in bioinspired helicoidal basalt fiber/epoxy composites: A comparative experimental analysis,” Polymer Composites, 45 (10), 9407–9420 (2024). https://doi.org/10.1002/pc.28417
4B.R. Kavya, H.S. Sureshchandra, S.J. Prashantha, and A.S. Shrikanth, “Prediction of mechanical properties of glass and basalt fiber reinforced concrete using ANN,” Asian Journal of Civil Engineering, 23, 877–886 (2022). https://doi.org/10.1007/s42107-022-00455-z
5A. Hasanzadeh, N.I. Vatin, M. Hematibahar, M. Kharun, and I. Shooshpasha, “Prediction of mechanical properties of basalt fiber–reinforced high–performance concrete using machine learning,” Materials, 15, 7165 (2022).
6L. Zhang, L., Yang, C., Lai, “Prediction of tensile strength of basalt continuous fiber from chemical composition using machine learning models,” Polymer Composites, 44, 6634–6645 (2023)
7C. Cakiroglu, Y. Aydın, G. Bekdaş, and Z.W. Geem, “Interpretable predictive modelling of basalt fiber reinforced concrete splitting tensile strength using ensemble machine learning methods and SHAP approach,” Materials, 16 (13), 4578 (2023).
8A. Almansour, S. Alhajri, and F. Alshammari, “Application of deep neural network for composite strength prediction,” Proceedings in Machine Learning Applications, 29, 115–306 (2022).
9G. R. Ganesh and P. Anand, “Experimental investigation on mechanical properties of Basalt/Jute/SiC reinforced hybrid polymer composites,” Materials Today: Proceedings, 59 (3), 1636–1642 (2022). https://doi.org/10.1016/j.matpr.2022.03.327
10G.S. Gangappa and S.S. Kulkarni, “Experimentation and validation of basalt & jute fiber reinforced in polymer matrix hybrid composites,” Materials Today: Proceedings, 38 (Part 5), 2372–2379 (2021). https://doi.org/10.1016/j.matpr.2020.07.081
11X. Su, J. Wang, and C. Li, “SVM and ANN prediction of fiber-reinforced polymer composites,” Materials & Design, 219, 110780 (2022).
12K. Patel, H. Shah, and R. Gupta, “Evaluation of SiC particulate epoxy nanocomposites: mechanical and microstructural analysis,” ACS Omega, 8, 11956–11964 (2023).
13M. Chopra and R. Joshi, “AI-driven design of multi-layer composite laminates: a machine learning optimization study,” Composite Structures, 324(2), 125723 (2024).
14A. Dey, S. Kundu, and G. Subha, “Influence of nano ceramic fillers on epoxy toughness: an experimental and ML study,” Polymer Composites, 42(1), 4612–4624 (2021).
15S. Nasir, A. Khan, and R. Ahmed, “Surrogate modeling for aerospace-grade composites using DNN,” Materials Science & Engineering A, 834, 142661 (2023).
16K. Poyyathappan, D. S., A. Ganeshkumar, and M. Arul, “Harnessing machine learning for wear prediction in sustainable fibre-reinforced hybrid composites,” Transactions of FAMENA, 49 (2), (2025). https://doi.org/10.21278/TOF.492069324
17U. Fatima, S. Khan, and S. Rashid, “Ceramics and fillers affecting composite hardness: experimental and predictive modeling,” Materials Today Communications, 33, 104144 (2023).
18A. Singh, N. Kumari, R. Kumar, “Effect of fiber orientation on interfacial and delamination strength in fiber composites,” Composite Interfaces, 29, 550–568 (2022).
19R. Kumar and P. Sinha, “Multi-objective optimization of composite design using ANN,” Composite Materials, 7(2), 364 -370 (2022).
20N. Mani, D. Subbiah, A. Moorthy, and A. Arunagiri, “Optimization and prediction of machining parameters in nanoparticle-reinforced FMLs using AI techniques,” Matéria (Rio de Janeiro), 30, (2025). https://doi.org/10.1590/1517-7076-RMAT-2024-0645
21X. Wang, Q. Kan, M. Petru, and G. Kang, “Composition–property relationships of basalt fibers using symbolic regression and physics-informed neural network,” Composites Part A: Applied Science and Manufacturing, 185, 108324 (2024).
22X. Liu, F. Tao, and W. Yu, “A review of ANN constitutive modeling of composite materials,” Composites Part B: Engineering, 224, 109152 (2021).