A Hybrid Multi-Physics Ensemble Deep Learning Framework for Simultaneous Prediction of Thermal and Electrical Conductivity in Functional Polymer-Based Nanocomposites

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

    V. Gokula Krishnan,

  • Arvind Kumar Tiwari,

  • R. Vadivel,

  • R. Srinivasan,

  • N. Sivakumar,

  • S. Kaviarasan,

  1. Post-Doctoral Research Fellow, Department of Computer Science and Engineering, Lincoln University College, Kota Bharu, Malaysia
  2. Adjunct Professor, Department of Computer Science and Engineering, Lincoln University College, Kota Bharu, Malaysia
  3. Associate Professor, Department of Artificial Intelligence and Data Science, Nitte Meenakshi Institute of Technology (NMIT), NITTE (Deemed to be University), Bengaluru, Karnataka, India
  4. Professor, Department of Computer Science and Engineering, Veltech Rangarajan Dr. Sagunthala R & D Institute of Science and Technology, Avadi, Tamil Nadu, India
  5. Associate Professor, Department of Artificial Intelligence and Data Science, Panimalar Engineering College, Chennai, Tamil Nadu, India
  6. Assistant Professor, Department of Computer Science and Engineering, Vel Tech Rangarajan Dr Sagunthala R& D Institute of Science and Technology, Avadi, Tamil Nadu, India

Abstract

Polymer-based nanocomposites have become promising materials for applications in energy storage, flexible electronics, biomedical devices, aerospace components, and advanced engineering because of their tunable thermal and electrical properties. Reliable prediction of these properties is essential for accelerating material design; however, existing analytical models and conventional machine learning techniques often fail to represent the complex interactions among filler characteristics, polymer matrices, processing conditions, and interfacial transport phenomena. This work presents HMEP-Net, a Hybrid Multi-Physics Ensemble Prediction Network developed to improve the simultaneous prediction of thermal and electrical conductivity in functional polymer nanocomposites. The proposed framework integrates physics-guided conductivity descriptors with deep feature extraction through an auto encoder, followed by CNN-BiLSTM learning, an attention-based feature fusion mechanism, and an adaptive ensemble meta-learning strategy. By combining physical knowledge with data-driven learning, the framework captures both transport mechanisms and nonlinear material relationships more effectively than existing approaches. The predictive capability of HMEP-Net was evaluated against Linear Regression, Support Vector Regression, Random Forest, XGBoost, Deep Neural Networks, CNN-LSTM, and Attention-CNN-BiLSTM models. The proposed method achieved the highest prediction accuracy, yielding an MAE of 0.041, RMSE of 0.061, and R² of 0.985 for thermal conductivity, together with an MAE of 0.047, RMSE of 0.069, and R² of 0.983 for electrical conductivity. Robustness analysis further demonstrated stable performance under 20% noises, with R² values remaining above 0.96 for both prediction tasks. These results indicate that HMEP-Net provides a reliable and computationally efficient framework for conductivity prediction and supports the accelerated design and optimization of next-generation polymer-based nanocomposite materials.

Keywords: Polymer Nanocomposites, Thermal Conductivity Prediction, Electrical Conductivity Prediction, Deep Learning, Attention Mechanism, Ensemble Learning and Materials Informatics.

How to cite this article:
V. Gokula Krishnan, Arvind Kumar Tiwari, R. Vadivel, R. Srinivasan, N. Sivakumar, S. Kaviarasan. A Hybrid Multi-Physics Ensemble Deep Learning Framework for Simultaneous Prediction of Thermal and Electrical Conductivity in Functional Polymer-Based Nanocomposites. Journal of Polymer & Composites. 2026; 14(03):-.
How to cite this URL:
V. Gokula Krishnan, Arvind Kumar Tiwari, R. Vadivel, R. Srinivasan, N. Sivakumar, S. Kaviarasan. A Hybrid Multi-Physics Ensemble Deep Learning Framework for Simultaneous Prediction of Thermal and Electrical Conductivity in Functional Polymer-Based Nanocomposites. Journal of Polymer & Composites. 2026; 14(03):-. Available from: https://journals.stmjournals.com/jopc/article=2026/view=249324


References

  • Asih GIN, Rafryanto AF, Hartati S, Jiang X, Anggraini A, Yudhowijoyo A, et al. Recent advances of polymer nanocomposites in emerging applications. Compos Funct Mater. 2025;1(1):20250105.
  • Chikwendu OC, Emeka UC, Onyekachi E. The optimization of polymer-based nanocomposites for advanced engineering applications. World J Adv Res Rev. 2025;25(1):755-763.
  • Hiremath P, Bhat SK, K JP, Rao PK, Ambiger KD, BRN M, et al. Data-driven prediction of polymer nanocomposite tensile strength through Gaussian process regression and Monte Carlo simulation with enhanced model reliability. J Compos Sci. 2025;9(7):364.
  • Thomas LM, MS S. Multiwalled carbon nanotubes reinforced polyvinyl alcohol nanocomposites: synergistic enhancement on multifunctional performance and deep learning-driven predictions for next-generation flexible electronics. Polym Adv Technol. 2025;36(8):e70294.
  • Yang Z, Meng Z. A review of computational modeling of polymer composites and nanocomposites. Polymers. 2026;18(4):443.
  • Raja RV, Kumar KA, Krishnan VG. Condition based ensemble deep learning and machine learning classification technique for integrated potential fishing zone future forecasting. Int J Recent Innov Trends Comput Commun. 2023;11(2). doi:10.17762/ijritcc.v11i2.6131.
  • Rajendra Kumar P, Madhav Rao B, Maddireddy CKR, Thakor S, Vaja CR, Prakash K, et al. Dielectric characterization and machine learning-based predictions in polymer composites with mixed nanoparticles. J Macromol Sci B. 2025;64(8):917-931.
  • Antunes M, Arencón D. Recent developments in the mechanical behavior of polymer-based composites. Polymers. 2026;18(5):598.
  • Liu Y, Zheng W, Ai H, Cheng L, Guo R, Song X. Machine learning for performance prediction and optimization of polymer composites: unveiling the dominant role of thermally conductive pathways. ACS Appl Mater Interfaces. 2025;17(51):70143-70155.
  • Kumar A, Kulthe MG. A review on polymer-based nanocomposites as radar absorbing materials. Polym Plast Technol Mater. 2026;65(9):977-1003.
  • Jl L, Venkateswaran N. Artificial intelligence assisted enhanced thermal and electrical performance of epoxy-GO-MgO nanocomposites via hybrid filler integration. In: Proceedings of the 2nd International Conference on Artificial Intelligence and Knowledge Discovery in Concurrent Engineering (ICECONF); 2025 Oct. IEEE; 2025. p. 1-9.
  • Bakhtiyarov SI, Gventsadze LD, Kutelia ER, Gventsadze DI, Bakhtiyarova AS, White SM. Modeling of thermomechanical properties of PTFE-based nanocomposite for space application: thermal expansion. J Space Saf Eng. 2025.
  • Sun Y, Hao C, Han Y, Zhou L. A novel prediction method for thermal conductivity of AlN/polymer composites through quantifying the configuration of thermally conductive chains. Mater Today Phys. 2025;101918.
  • Oladele IO, Adelani SO, Taiwo AS, Akinbamiyorin IM, Olanrewaju OF, Orisawayi AO. Polymer-based nanocomposites for supercapacitor applications: a review on principles, production and products. RSC Adv. 2025;15(10):7509-7534.
  • Liu Y, Fang Z, Liu Y, Zhao N. Toward high-thermal-conductivity polymer-based materials: breakthroughs and barriers. Polym Compos. 2026;47(4):2995-3038.
  • Raza Y, Raza H, Ahmed A, Quazi MM, Jamshaid M, Anwar MT, et al. Integration of response surface methodology (RSM), machine learning (ML), and artificial intelligence (AI) for enhancing properties of polymeric nanocomposites: a review. Polym Compos. 2025;46(15):13591-13627.
  • Lakshmaiya N, Kilari N, Chukka NDKR, Kaliappan S, Rajendran A, Maranan R. Bayesian-optimized deep neural network surrogate for orientation-driven anisotropic thermal conductivity prediction in hybrid polymer nanocomposites. Results Eng. 2026;110049.
  • Sharma H, Arora G, Kumar R, Debnath S, Siengchin S. Data-driven insights into the high-temperature behavior of polymer/carbon nanotubes nanocomposites. Iran Polym J. 2026;35(5):955-965.
  • Madani M, Abbasi M. A comprehensive data-driven approach for robust investigation of viscosity of nano-polymer composites utilized as enhanced oil recovery agents. Mater Des. 2026;115741.
  • Zhang F, Li Z, Wang X, Yu ZZ, Yang D. Plants-inspired cellulose nanocomposites-based soft intelligent actuators with exceptional thermal conductivity and electromagnetic interference shielding. Adv Mater. 2026;38(2):e13893.
  • Du H, Xie W, Wang C, Gao C, Fang C, Zhang J, et al. A unified Monte-Carlo to effective-medium study on the electrical conductivity of CNT polymer nanocomposites with electron tunneling. Compos Commun. 2026;102764.
  • Hegazy MA, Nada N, Elhaes H, Ibrahim MA, Ezzat HA. Advanced polymer-based humidity sensor for aerospace applications implementing ZnO-G: theoretical and experimental study. Sci Rep. 2026;16(1):6339.
  • Santosh DT, Anuradha N, Kolukuluri M, Gupta G, Pathak MK, Krishnan VG, et al. Development of IoT based intelligent irrigation system using particle swarm optimization and XGBoost techniques. Bull Electr Eng Inform. 2024;13(3):1927-1934.
  • Nanocomposites data [Internet]. Available from: https://www.kaggle.com/datasets/wejdengazehi/nanocomposites-data
  • Material science dataset [Internet]. Available from: https://www.kaggle.com/datasets/allanwandia/material-science
  • Satyanarayana P, Ravi J, Mahalakshmi T, Satyanarayana KVV, Krishnan GV. Performance analysis of DSR and cache customized DSR steering protocols in wireless mobile ADHOC networks. In: Proceedings of the 5th International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC); 2021. IEEE; 2021. p. 456-461. doi:10.1109/I-SMAC52330.2021.9641042.
  • Manickaraj K, Aruchamy K, Palanisamy S, Jayamani M, et al. Improving mechanical performance of hybrid polymer composites: incorporating banana stem leaf and jute fibers with tamarind shell powder. BioResources. 2025;20(1):1998-2025. doi:10.15376/biores.20.1.1998-2025.
  • Khalid MY, et al. Physics-informed machine learning for multifunctional polymer nanocomposites. J Polym Res. 2024. doi:10.1007/s10965-024-04162-6.
  • Palanisamy S, Januarti J, Alavudeen A, et al. Effect of alkaline treatment on the thermal and mechanical properties of natural fiber reinforced polymer composites. Sci Rep. 2025;15:14085. doi:10.1038/s41598-025-99227-x.
  • Krishnan GV, Ram NS. Analyze traffic forecast for decentralized multi agent system using I-ACO routing algorithm. J Ambient Intell Humaniz Comput. 2018. doi:10.1007/s12652-018-0981-2.
  • Palanisamy S, Kalimuthu M, Dharmalingam S, Alavudeen A, Nagarajan R, Ismail SO, et al. Effects of fiber loadings and lengths on mechanical properties of Sansevieria cylindrica fiber reinforced natural rubber biocomposites. Mater Res Express. 2023;10(8):085503. doi:10.1088/2053-1591/acefb0.
Ahead of Print Subscription Original Research
Volume 14
03
Received 20/06/2026
Accepted 06/07/2026
Published 08/07/2026
Publication Time 18 Days
6

Login


My IP

PlumX Metrics