S. Amutha,
R. K. Jeyauthmigha,
S. Tephillah,
N. Kanagavalli,
V. S. Prakash,
P. R. Therasa,
S. Gayathri,
- Professor, Department of Computer Science and Engineering, School of Computing, Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology, Chennai, Tamil Nadu, India
- Assistant Professor, Department of Computer Science and Design, SNS College of Technology, Coimbatore, Tamil Nadu, India
- Associate Professor, Department of Electronics and Communication Engineering, St. Joseph’s Institute of Technology, Chennai, Tamil Nadu, India
- Associate Professor, Department of Computer Science and Engineering, Vel Tech High Tech Dr. Rangarajan Dr. Sakunthala Engineering College, Chennai, Tamil Nadu, India
- Professor, Department of Computer Applications, Kristu Jayanti (Deemed to be University), Bengaluru, Karnataka, India
- Associate Professor, Department of Computer Science and Engineering, R.M.K. Engineering College, RSM Nagar, Kavaraipettai, Tamil Nadu, India
- Assistant Professor, Department of Computer Science and Engineering, Sathyabama Institute of Science and Technology, Chennai, Tamil Nadu, India
Abstract
Fiber Reinforced Polymer (FRP) composites have broad spread use in aerospace, automotive, marine and structural applications due to its high specific strength, stiffness and corrosion resistance. The various damage mechanisms, such as matrix cracking, fiber breakage, delamination, and interfacial failure, however, make the forecasting of damage particularly complex. In this work, a hybrid machine learning (ML) and finite element (FE) system is proposed for predicting the damaged behavior of FRP composites. Progressive damage is simulated under different loading conditions using a complete FE model with suitable material constitutive and failure criteria. The numerical simulations are used to recover the primary aspects – such as fiber orientation, volume %, mechanical characteristics, loading circumstances, and damage variables – to provide a whole data set. The machine learning algorithms are then trained to detect the nonlinear relationships between the input parameters and the damage responses. The prediction capacity of ML models is evaluated using statistical measures such as coefficient of determination, root mean square error and mean absolute error. The proposed hybrid method aims at using the physical interpretability of FE simulations and the computational efficiency of ML model to quickly anticipate damage initiation and evolution. The proposed methodology is expected to be a useful tool for composite design, structural health assessment and failure analysis with minimal computation costs and reliable predictive accuracy. The proposed approach can also enable optimization and real time prediction of damage for sophisticated composite constructions.
Keywords: Finite element analysis, damage prediction, progressive damage, delamination failure analysis, composite structures.
[This article belongs to Special Issue under section in Journal of Polymer & Composites (jopc)]
References
1. Wang A, Lan R, Chen Q, Kong W, Liu H, Yue Q, et al. Applications of aramid fiber-reinforced polymer composites in civil engineering: A review. Polymers. 2026;18(9):1102.
2. Diniță A, Ripeanu RG, Ilincă CN, Cursaru D, Matei D, Naim RI, et al. Advancements in fiber-reinforced polymer composites: A comprehensive analysis. Polymers. 2024;16(1):2.
3. Temesgen AG, Kaufmann J, Cebulla H. Mechanical performance of natural fiber reinforced polymer composites with critical assessment of recent advances and future perspectives. Discover Materials. 2026 May 19.
4. Adam HM, Bogale TM, Paramasivam V. Natural Fiber Composites Under Dynamic Loading: A Comprehensive Review of Mechanical Performance and Structural Integrity. Eng Rep. 2026 Apr;8(4):e70722.
5. Davidson M, Graunke R, Green A, Haelsig H, Heinemann L, Antony Jose S, et al. Carbon fiber-reinforced polymer matrix composites: Processing, properties, and applications. Fibers. 2026;14(3):29.
6. Yang L, Huang X, Liao Z, Wei Z, Zou J. Experimental and simulation study on failure of thermoplastic carbon fiber composite laminates under low-velocity impact. Polymers. 2024;16(18):2581.
7. Taherzadeh-Fard A, Cornejo A, Jiménez S, Barbu LG. Numerical analysis of damage in composites: From intra-layer to delamination and data-assisted methods. Mathematics. 2025;13(10):1578.
8. Scutaru ML, Nicolae P, Vlase S, Mitu AM, Sireteanu T, Munteanu VM. Mechanical behavior analysis-based finite element method of composites: A review. Mathematics. 2026;14(13):2248.
9. Santana PB, Gomes HM, Ferreira AJ, Tita V. Uncertainty in composite damage modeling and finite element model updating: A survey. Arch Comput Methods Eng. 2026 Jul 27:1–38.
10. Ghattas B, Manzon D. Machine learning alternatives to response surface models. Mathematics. 2023;11(15):3406.
11. Srivastava A, Jha S. Data-driven machine learning: A new approach to process and utilize biomedical data. In: Predictive Modeling in Biomedical Data Mining and Analysis. Academic Press; 2022 Jan 1. p. 225–252.
12. Sadeghian M, Palevicius A, Sablinskas J, Griskevicius P. From pixels to predictions: Integrating machine learning and digital image correlation for damage identification in engineering materials. Materials. 2026;19(1):77.
13. Smolnicki M, Lesiuk G, Duda S, De Jesus AM. A review on finite-element simulation of fibre metal laminates. Arch Comput Methods Eng. 2023 Mar;30(2):749–763.
14. Sorour SS, Saleh CA, Shazly M. Integrating machine learning and symbolic regression for predicting damage initiation in hybrid FRP bolted connections. Sci Rep. 2025;15(1):18564.
15. Zhang Y, Van Paepegem W, De Corte W. An enhanced progressive damage model for laminated fiber-reinforced composites using the 3D Hashin failure criterion: A multi-level analysis and validation. Materials. 2024;17(21):5176.
16. Liu Y, Xie J, Chen X, Zhang J, Wan M, Sun Z, et al. Characterization of progressive damage behaviour and failure mechanism of carbon fiber reinforced composite laminates. Sci Rep. 2025;15(1):13791.
17. Ning H, Pillay S, Xu C, Vaidya U. Compression-after-impact behaviors of fiber reinforced polymer matrix composites: A comprehensive review. J Compos Mater. 2026;60(9):817–868.
18. Owais M, Idriss LK. Analyzing factors affecting failure mode characteristics of CFRP strengthened reinforced concrete beams using deep residual neural networks and density-based sensitivity analysis. Innov Infrastruct Solut. 2026 Feb;11(2):82.
19. Deng J, Zhou J, Wu T, Liu Z, Wu Z. Review and assessment of fatigue delamination damage of laminated composite structures. Materials. 2023;16(24):7677.
20. Paśnik J, de Moura M, Samborski S, Różyło P, Dębski H. Comparative study of different FEM-based delamination modeling techniques for mechanically coupled laminated composite beams. Fatigue Fract Eng Mater Struct. 2026;49(4):1557–1567.
21. Wang M, Hao H, Liu Q, Miao X, Lai Z, Yuan T, et al. Characterization on mode-I/II interlaminar strength and fracture toughness of co-cured fiber-metal laminates. Polymers. 2025;17(21):2937.
22. Santos JD, Crespo Beltrán F, Berrezueta M, Torres A, Álvarez AG, Valarezo A. Characterization of the interlaminar fracture toughness of an additive manufacturing continuous glass fiber-reinforced thermoplastic composite. Polymers. 2026;18(12):1438.
23. Jueyendah S, Yaman Z, Dere T, Çavuş TF. Comparative study of linear and non-linear ML algorithms for cement mortar strength estimation. Buildings. 2025;15(16):2932.
24. Zhou W, Yan Z, Zhang L. A comparative study of 11 non-linear regression models highlighting autoencoder, DBN, and SVR, enhanced by SHAP importance analysis in soybean branching prediction. Sci Rep. 2024;14(1):5905.
25. Tanase M, Bolboaca R, Veres C. Machine learning in sustainable fiber-reinforced polymers: A bibliometric and critical assessment. Polym Bull. 2026 Aug;83(8):446.
26. Malashin I, Martysyuk D, Tynchenko V, Gantimurov A, Nelyub V, Borodulin A. Data-driven optimization of discontinuous and continuous fiber composite processes using machine learning: A review. Polymers. 2025;17(18):2557.
27. Aydın F, Duru A. The Use of Machine Learning in Aluminum and Magnesium Matrix Composites: A Review. Arch Comput Methods Eng. 2026 Feb 19:1–51.
28. Li Y, Isleem HF, Shao A, Khishe M. A Hybrid Finite-Element–Machine-Learning Framework for Predicting Load Capacity and Eccentricity Effects in Circular Concrete-Filled Steel Tubes. Int J Concr Struct Mater. 2026 May 27;20(1):50.
29. Huang Y, Tang J, Lin D, Sun M, Shu J, Liu W, et al. A hybrid FE-ML approach for critical buckling moment prediction in dented pipelines under complex loadings. Materials. 2025;18(20):4721.
Journal of Polymer & Composites
| Volume | 14 | |
| Special Issue | 04 | |
| Received | 09/09/2026 | |
| Accepted | 19/09/2026 | |
| Published | 23/09/2026 | |
| Publication Time | 14 Days |