Machine Learning Based Optimization of Polymer Structure Property Relationships in Composite Material Systems

Notice

This is an unedited manuscript accepted for publication and provided as an Article in Press for early access at the author’s request. The article will undergo copyediting, typesetting, and galley proof review before final publication. Please be aware that errors may be identified during production that could affect the content. All legal disclaimers of the journal apply.

Year : 2026 | Volume : 14 | 03 | Page :
    By

    Prashant V. Thokal,

  • P. William,

  • Ganesh P. Dawange,

  • Dharmendra Kumar Roy,

  • Pravin B. Khatkale,

  1. Assistant Professor, Department of Electrical Engineering, Sanjivani College of Engineering, Kopargaon, Maharashtra, India
  2. Professor (Research), School of Computer Science and Technology, Karunya Institute of Technology and Sciences, Coimbatore, Tamil Nadu, India
  3. Assistant Professor, Department of Structural Engineering, Sanjivani College of Engineering, Kopargaon, Maharashtra, India
  4. Associate Professor, Department of Computer Science and Engineering, Hyderabad Institute of Technology and Management, Medchal, Hyderabad, Telangana, India
  5. Controller of Examination, School of Engineering and Technology, Sanjivani University, Kopargaon, Maharashtra, India

Abstract

In modern engineering applications, polymer-based composite materials have garnered a lot of attention because of their lightweight nature, high strength-to-weight ratio, and changing physical features. In order to maximize the relationships between polymer structure and properties in composite materials, this study suggests a strategy based on reinforcement learning (RL). The research utilized the Polymer Composite Properties Dataset, which contains 12,700 records associated with polymer matrices, reinforcement fillers, interfacial bonding characteristics, curing conditions, thermal properties, processing variables, and mechanical performance indicators. To ensure accurate representation of structure–property correlations, material characterization was carried out using X-ray diffraction (XRD) for phase identification, Fourier transform infrared spectroscopy (FTIR) for chemical interaction analysis, and scanning electron microscopy (SEM) for microstructural analysis. Tensile strength, elongation at break, Young’s modulus, impact toughness, and other mechanical parameters were tested in accordance with the American Society for Testing and Materials (ASTM) recommendations. Thermal stability was assessed by thermogravimetric analysis (TGA). A deep Q-network (DQN)-based RL agent was used to repeatedly modify polymer structural properties, such as chain design, crosslink density, and filler dispersion. The agent was guided by reward functions derived from surrogate models trained on the experimental dataset. The model achieved minimum MSE values of 0.0478, 0.3520, 0.0857, and 3.5042, along with MAE values of 0.1850, 0.5017, 0.2694, and 1.5607, respectively, confirming its superior capability in capturing nonlinear structure–property relationships of polymer composites (PC). The results demonstrate the capability of RL to capture complex nonlinear interactions, accelerate the design of high-performance PC, reduce experimental effort, and enable efficient material innovation.

Keywords: Reinforcement Learning, Polymer Composites, Materials Characterization, Structure–Property Relationship, Optimization.

How to cite this article:
Prashant V. Thokal, P. William, Ganesh P. Dawange, Dharmendra Kumar Roy, Pravin B. Khatkale. Machine Learning Based Optimization of Polymer Structure Property Relationships in Composite Material Systems. Journal of Polymer & Composites. 2026; 14(03):-.
How to cite this URL:
Prashant V. Thokal, P. William, Ganesh P. Dawange, Dharmendra Kumar Roy, Pravin B. Khatkale. Machine Learning Based Optimization of Polymer Structure Property Relationships in Composite Material Systems. Journal of Polymer & Composites. 2026; 14(03):-. Available from: https://journals.stmjournals.com/jopc/article=2026/view=246727


References

  1. Okpala, C.C., 2026. Machine learning–enabled design of composite materials: Scalable structure–processing–property relationships across applications. International Journal of Technology, Health and Sustainability, 2(1), pp.154-161.
  2. Mittapally, R., 2025. Data-Driven Prediction of Mechanical Properties in 3D-Printed Composites Using Hybrid Machine Learning Models. Journal of Data Science and Information Technology, 2, pp.1-16.https://dx.doi.org/10.55124/jdit.v2i2.274
  3. Hu, W., Jing, E., Qiu, H. and Sun, Z.Y., 2025. Discovering polyimides and their composites with targeted mechanical properties through explainable machine learning. Journal of Materials Informatics, 5(1), pp.N-A.https://doi.org/10.20517/jmi.2024.59
  4. Zhao, Y., Chen, Z. and Jian, X., 2023. A high-generalizability machine learning framework for analyzing the homogenized properties of short fiber-reinforced polymer composites. Polymers, 15(19), p.3962.https://doi.org/10.3390/polym15193962
  5. Mohammed, A.J., Mohammed, A.S. and Mohammed, A.S., 2023. Prediction of tribological properties of UHMWPE/SiC polymer composites using machine learning techniques. Polymers, 15(20), p.4057.https://doi.org/10.3390/polym15204057
  6. Okafor, C.E., Iweriolor, S., Ani, O.I., Ahmad, S., Mehfuz, S., Ekwueme, G.O., Chukwumuanya, O.E., Abonyi, S.E., Ekengwu, I.E. and Chikelu, O.P., 2023. Advances in machine learning-aided design of reinforced polymer composite and hybrid material systems. Hybrid Advances, 2, p.100026.https://doi.org/10.1016/j.hybadv.2023.100026
  7. Yang, S., Bieliatynskyi, A., Trachevskyi, V., Shao, M. and Ta, M., 2022. Technology for improving modern polymer composite materials. Materials Science Poland, 40(3), pp.27-41.https://doi.org/10.2478/msp-2022-0027
  8. Patil, D.S. and Bhoomkar, M.M., 2023. Investigation on mechanical behaviour of fiber-reinforced advanced polymer composite materials, EVERGREEN Joint Journal of Novel Carbon Resource Sciences & Green Asia Strategy, 10 (01), pp55-62.https://doi.org/10.5109/6781040
  9. Kalusivalingam, A.K., Sharma, A., Patel, N. and Singh, V., 2022. Leveraging Reinforcement Learning and Genetic Algorithms for Enhanced Optimization of Sustainability Practices in AI Systems. International Journal of AI and ML, 3(9).
  10. Çevik, Z.A. and Sarıışık, G., 2025. A machine learning-driven optimization of machining parameters for nanoclay-glass fiber polymer-reinforced composites for performance prediction and process enhancement. Journal of Tribology, 147(9), p.091117.https://doi.org/10.1115/1.4068354
  11. Berladir, K., Antosz, K., Ivanov, V. and Mitaľová, Z., 2025. Machine learning-driven prediction of composite materials properties based on experimental testing data. Polymers, 17(5), p.694.https://doi.org/10.3390/polym17050694
  12. Pandit, P., Abdusalamov, R., Itskov, M. and Rege, A., 2024. Deep reinforcement learning for microstructural optimisation of silica aerogels. Scientific Reports, 14(1), p.1511.https://doi.org/10.1038/s41598-024-51341-y
  13. Wan, Z.L., Zhao, W.C., Qiu, H.K., Zhou, S.S., Chen, S.Y., Fu, C.L., Feng, X.Y., Pan, L.J., Wang, K., He, T.C. and Wang, Y.G., 2024. Data-driven exploration of polymer processing effects on the mechanical properties in carbon black-reinforced rubber composites. Chinese Journal of Polymer Science, 42(12), pp.2038-2047.https://doi.org/10.1007/s10118-024-3216-3
  14. Biruk-Urban, K., Bere, P. and Józwik, J., 2023. Machine learning models in drilling of different types of glass-fiber-reinforced polymer composites. Polymers, 15(23), p.4609.https://doi.org/10.3390/polym15234609
  15. Kumar, S.S., Shyamala, P., Pati, P.R., Mishra, D.K., Sharma, V., Tank, G. and Kanan, M., 2025. Investigation and machine learning-based prediction of mechanical properties in hybrid natural fiber composites. Scientific Reports, 15(1), p.33700.https://doi.org/10.1038/s41598-025-18944-5
  16. Jayaram, R.S., Saravanamuthukumar, P., Abdullah, A.B., Krishnamoorthy, R., Kunar, S., Yong, X. and Prabhakar, S., 2025. Machine learning driven optimization of compressive strength of 3D printed bio polymer composite material. PLoS One, 20(8), p.e0330625.https://doi.org/10.1371/journal.pone.0330625
  17. Ulkir, O. and Ersoy, S., 2025. Hybrid experimental–machine learning study on the mechanical behavior of polymer composite structures fabricated via FDM. Polymers, 17(15), p.2012.https://doi.org/10.3390/polym17152012
  18. Sharma, A., Mukhopadhyay, T., Rangappa, S.M., Siengchin, S. and Kushvaha, V., 2022. Advances in computational intelligence of polymer composite materials: machine learning assisted modeling, analysis and design. Archives of Computational Methods in Engineering, 29(5), pp.3341-3385.https://doi.org/10.1007/s11831-021-09700-9
  19. Champa-Bujaico, E., Díez-Pascual, A.M., Redondo, A.L. and Garcia-Diaz, P., 2024. Optimization of mechanical properties of multiscale hybrid polymer nanocomposites: A combination of experimental and machine learning techniques. Composites Part B: Engineering, 269, p.111099.https://doi.org/10.1016/j.compositesb.2023.111099
Ahead of Print Subscription Original Research
Volume 14
03
Received 25/05/2026
Accepted 09/06/2026
Published 15/06/2026
Publication Time 21 Days
21

Login


My IP

PlumX Metrics