Thermal Conductivity Analysis of Functional Gradient Materials Based on Physically Informed Neural Networks

Authors

  • Yong Ping Feng Dalian Jiaotong University, China
  • Jinxiu Hu Dalian Jiaotong University, China

DOI:

https://doi.org/10.62411/jimat.v3i1.16162

Keywords:

Physics-Informed Neural Networks , Augmented Lagrangian Method, Functionally Graded Materials, Heat Conduction Analysis

Abstract

This paper investigates transient heat conduction problems in functionally graded materials (FGMs) using Physics-Informed Neural Networks (PINNs) and Augmented Lagrangian Physics-Informed Neural Networks (AL-PINNs). By constructing residual loss functions for the governing equations and boundary initial conditions, this method trains the neural network without relying on sample data, thereby enhancing the model’s generalisation capability and reducing the reliance on the pre-processing tasks, such as the derivation of differential equations, complex modelling, and mesh generation, required by traditional numerical methods. This paper investigates the applicability of PINN and AL-PINN in solving transient heat conduction problems in FGMs and analyses the impact of network architecture on prediction accuracy. The results indicate that AL-PINN exhibits lower error than conventional PINN when addressing transient FGM heat conduction problems with complex boundaries. Although the training cost of PINN-based methods is still higher than that of the conventional Finite Element Method for small-scale benchmark problems, the proposed framework offers greater flexibility for problems with irregular geometries and variable material properties, without requiring mesh generation.

References

K. J. Gowtham, C. G. Pavithra, and B. J. Gireesha, “PINNs and Vieta-Lucas polynomial collocation simulation for heat transfer and thermal efficiency analysis in fully wetted semispherical fins with temperature-dependent thermal properties: Applications in air conditioning systems,” Int. J. Refrig. , vol. 181, pp. 359–369, 2026, doi: 10. 1016/j. ijrefrig. 2025. 10. 024.

S. Morchio and M. Fossa, “Modelling and validation of a new hybrid scheme for predicting the performance of U-pipe borehole heat exchangers during distributed thermal response test experiments,” Appl. Therm. Eng. , vol. 186, p. 116514, 2021, doi: 10. 1016/j. applthermaleng. 2020. 116514.

X. Wang, Z. -Y. Yin, W. Wu, and H. -H. Zhu, “Neural network-augmented differentiable finite element method for boundary value problems,” Int. J. Mech. Sci. , vol. 285, p. 109783, 2025, doi: 10. 1016/j. ijmecsci. 2024. 109783.

K. Li, W. Jiang, and H. Li, “Isogeometric boundary element method for solving 2D multi-media heat conduction problems,” Results Appl. Math. , vol. 28, p. 100639, 2025, doi: 10. 1016/j. rinam. 2024. 100639.

Y. Shang, H. Ban, and D. Liu, “Simultaneous identification of boundary heat flux and thermal conductivity in inverse heat conduction problems using physics-informed neural networks,” Therm. Sci. Eng. Prog. , vol. 65, p. 103905, 2025, doi: 10. 1016/j. tsep. 2024. 103905.

A. Di Meglio, N. Massarotti, and P. Nithiarasu, “A physics-driven and machine learning-based digital twinning approach to transient thermal systems,” Int. J. Numer. Methods Heat Fluid Flow, vol. 34, no. 6, pp. 2229–2256, 2024, doi: 10. 1108/HFF-09-2023-0545.

Y. Cui, X. Zhou, C. Fan, H. Lin, and S. Deng, “Application of physics-informed neural network-based hybrid prediction models in campus thermal environment prediction,” Build. Environ. , vol. 287, part B, p. 113880, 2026, doi: 10. 1016/j. buildenv. 2025. 113880.

R. Zhang and L. Feng, “Application of deep learning in the prediction of heat conduction characteristics and structural optimisation of absorbing materials for solar collectors,” Microchem. J., vol. 221, p. 117028, 2026, doi: 10. 1016/j. microc. 2025. 117028.

M. Raissi, P. Perdikaris, and G. E. Karniadakis, “Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations,” J. Comput. Phys. , vol. 378, pp. 686–707, 2019, doi: 10. 1016/j. jcp. 2018. 10. 045.

G. E. Karniadakis, I. G. Kevrekidis, L. Lu, P. Perdikaris, S. Wang, and L. Yang, “Physics-informed machine learning,” Nat. Rev. Phys. , vol. 3, no. 6, pp. 422–440, 2021, doi: 10. 1038/s42254-021-00314-5.

M. Raissi, A. Yazdani, and G. E. Karniadakis, “Hidden fluid mechanics: Learning velocity and pressure fields from flow visualisations,” Science, vol. 367, no. 6481, pp. 1026–1030, 2020, doi: 10. 1126/science. aaw1159.

X. Jin, S. Cai, H. Li, and G. E. Karniadakis, “NSFnets (Navier-Stokes flow nets): Physics-informed neural networks for the incompressible Navier-Stokes equations,” J. Comput. Phys. , vol. 426, p. 109951, 2021, doi: 10. 1016/j. jcp. 2020. 109951.

L. Sun, H. Gao, S. Pan, and J. X. Wang, “Surrogate modelling for fluid flows based on physics-constrained deep learning without simulation data,” Comput. Methods Appl. Mech. Eng. , vol. 361, p. 112732, 2020, doi: 10. 1016/j. cma. 2019. 112732.

L. Yang, X. Meng, and G. E. Karniadakis, “B-PINNs: Bayesian physics-informed neural networks for forward and inverse PDE problems with noisy data,” J. Comput. Phys. , vol. 425, p. 109913, 2021, doi: 10. 1016/j. jcp. 2020. 109913.

C. Rao, H. Sun, and Y. Liu, “Physics-informed deep learning for computational elastodynamics without labelled data,” J. Eng. Mech. , vol. 147, no. 8, p. 04021043, 2020, doi: 10. 1061/(ASCE)EM. 1943-7889. 0001947.

Y. Shin, J. Darbon, and G. E. Karniadakis, “On the convergence of physics-informed neural networks for linear second-order elliptic and parabolic type PDEs,” Commun. Comput. Phys. , vol. 28, no. 5, pp. 2042–2074, 2020, doi: 10. 4208/cicp. OA-2020-0193.

L. Lu, X. Meng, Z. Mao, and G. E. Karniadakis, “DeepXDE: A deep learning library for solving differential equations,” SIAM Rev. , vol. 63, no. 1, pp. 208–228, 2021, doi: 10. 1137/19M1274067.

G. Pang, L. Lu, and G. E. Karniadakis, “fPINNs: Fractional physics-informed neural networks,” SIAM J. Sci. Comput. , vol. 41, no. 4, pp. A2603–A2626, 2019, doi: 10. 1137/18M1204014.

S. Cai, Z. Wang, S. Wang, P. Perdikaris, and G. E. Karniadakis, “Physics-informed neural networks for heat transfer problems,” J. Heat Transfer, vol. 143, no. 6, p. 060801, 2021, doi: 10. 1115/1. 4050542.

N. Zobeiry and K. D. Humfeld, “A physics-informed machine learning approach for solving heat transfer equations in advanced manufacturing and engineering applications,” Eng. Appl. Artif. Intell., vol. 101, p. 104232, 2021, doi: 10. 1016/j. engappai. 2021. 104232.

S. A. Niaki, E. Haghighat, T. Campbell, A. Poursartip, and R. Vaziri, “Physics-informed neural networks for modelling the thermochemical curing process of composite-tool systems during manufacture,” Comput. Methods Appl. Mech. Eng. , vol. 384, p. 113959, 2021, doi: 10. 1016/j. cma. 2021. 113959.

H. Guo, X. Zhuang, P. Chen, T. Rabczuk, “Physics-informed deep learning for three-dimensional transient heat transfer analysis of functionally graded materials,” Comput. Mech. , vol. 72, no. 3, pp. 537–555, 2023, doi: 10. 1007/s00466-023-02315-5.

P. Das, M. A. Islam, F. R. Rijvi, M. R. Rashid, “Physics-informed operator learning integrated with analytical modeling for transient heat conduction in 2D orthotropic functionally graded materials with arbitrary gradation,” Mater. Des. , vol. 259, p. 114186, 2026, doi: 10. 1016/j. matdes. 2025. 114186.

A. D. Jagtap and G. E. Karniadakis, “Extended physics-informed neural networks (XPINNs): A generalised space-time domain decomposition-based deep learning framework for nonlinear partial differential equations,” Commun. Comput. Phys. , vol. 28, no. 5, pp. 2002–2041, 2020, doi: 10. 4208/cicp. OA-2020-0164.

A. D. Jagtap, E. Kharazmi, and G. E. Karniadakis, “Conservative physics-informed neural networks on discrete domains for conservation laws: Applications to forward and inverse problems,” Comput. Methods Appl. Mech. Eng. , vol. 365, p. 113028, 2020, doi: 10. 1016/j. cma. 2020. 113028.

S. Wang, Y. Teng, and P. Perdikaris, “Understanding and mitigating gradient flow pathologies in physics-informed neural networks,” J. Comput. Phys. , vol. 443, p. 110568, 2021, doi: 10. 1016/j. jcp. 2021. 110568.

S. Wang, X. Yu, and P. Perdikaris, “When and why PINNs fail to train: A neural tangent kernel perspective,” J. Comput. Phys. , vol. 449, p. 110768, 2022, doi: 10. 1016/j. jcp. 2021. 110768.

L. McClenny and U. Braga-Neto, “Self-adaptive physics-informed neural networks using a soft attention mechanism,” J. Comput. Phys. , vol. 474, p. 111741, 2023, doi: 10. 1016/j. jcp. 2022. 111741.

L. Lu, R. Pestourie, W. Yao, Z. Wang, F. Verdugo, and S. G. Johnson, “Physics-informed neural networks with hard constraints for inverse design,” SIAM J. Sci. Comput. , vol. 43, no. 6, pp. B1105–B1132, 2021, doi: 10. 1137/20M1352327.

E. Haghighat, M. Raissi, A. Moure, H. Gomez, and R. Juanes, “A physics-informed deep learning framework for inversion and surrogate modelling in solid mechanics,” Comput. Methods Appl. Mech. Eng. , vol. 379, p. 113741, 2021, doi: 10. 1016/j. cma. 2021. 113741.

Son, H. , Jang, J. W. , Han, W. J. , & Hwang, H. J. (2023). Enhanced physics-informed neural networks with augmented Lagrangian relaxation method (AL-PINNs). Neurocomputing, 548, 126424.

Downloads

Published

2026-07-03

Issue

Vol. 3 No. 1 (2026): April

Section

Articles

License

Authors who publish their articles in this journal agree to the following conditions:

  • Copyright remains with the author and gives the JIMAT journal the right as first priority to publish the article under a Creative Commons Attribution License which allows articles to be shared with acknowledgment of the author of the article and this journal as the place of publication.
  • Authors can distribute their published articles non-exclusively (for example: in university repositories or in books) with notification or acknowledgment of publication in JIMAT.
  • Authors are permitted to list their work online (for example: on a personal website or in a university repository) before and after the submission process (see The Effect of Open Access).