Physics-informed machine learning for solving partial differential equations in porous media

Abstract

Physical phenomenon in nature is generally simulated by partial differential equations. Among different sorts of partial differential equations, the problem of two-phase flow in porous media has been paid intense attention. As a promising direction, physics-informed neural networks shed new light on the solution of partial differential equations. However, current physics-informed neural networks’ ability to learn partial differential equations relies on adding artificial diffusion or using prior knowledge to increase the number of training points along the shock trajectory, or adaptive activation functions. To address these issues, this study proposes a physics-informed neural network with long short-term memory and attention mechanism, an ingenious method to solve the Buckley-Leverett partial differential equations representing two-phase flow in porous media. The designed network structure overcomes the dependency on artificial diffusion terms and enhances the importance of shallow features. The experimental results show that the proposed method is in good agreement with analytical solutions. Accurate approximations are shown even when encountering shock points in saturated fields of porous media. Furthermore, experiments show our innovative method outperforms existing traditional physics-informed machine learning approaches.

Document Type: Original article

Cited as: Shan, L., Liu, C., Liu, Y., Tu, Y., Dong, L., Hei, X. Physics-informed machine learning for solving partial differential equations in porous media. Advances in Geo-Energy Research, 2023, 8(1): 37-44. https://doi.org/10.46690/ager.2023.04.04

References

Almajid, M. M., Abu-Al-Saud, M. O. Prediction of porous media fluid flow using physics informed neural networks. Journal of Petroleum Science and Engineering, 2022, 208: 109205.

Alwated, B., El-Amin, M. F. Enhanced oil recovery by nanoparticles flooding: From numerical modeling improvement to machine learning prediction. Advances in Geo-Energy Research, 2021, 5(3): 297-317.

Cai, S., Mao, Z., Wang, Z., et al. Physics-informed neural networks (PINNs) for fluid mechanics: A review. Acta Mechanica Sinica, 2021, 37(12): 1727-1738.

Cheng, C., Zhang, G. T. Deep learning method based on physics informed neural network with resnet block for solving fluid flow problems. Water, 2021, 13(4): 423.

Coutinho, E. J. R., Dall’Aqua, M., McClenny, L., et al. Physics-informed neural networks with adaptive localized artificial viscosity. ArXiv preprint, 2022: 2203.08802.

Cuomo, S., Di Cola, V. S., Giampaolo, F., et al. Scientific machine learning through physics-informed neural networks: Where we are and what’s next. Journal of Scientific Computing, 2022, 92(3): 88.

Esmaeilzadeh, S., Azizzadenesheli, K., Kashinath, K., et al. Meshfreeflownet: A physics-constrained deep continuous space-time super-resolution framework. Paper 20489143 Presented in SC20: International Conference for High Performance Computing, Networking, Storage and Analysis, Virtual Event, 9-19 November, 2020.

Fraces, C. G., Tchelepi, H. Physics informed deep learning for flow and transport in porous media. Paper SPE 203934 Presented at the SPE Reservoir Simulation Conference, Virtual Event, 26 October-1 November, 2021.

Fukami, K., Fukagata, K., Taira, K. Super-resolution reconstruction of turbulent flows with machine learning. Journal of Fluid Mechanics, 2019, 870: 106-120.

Fuks, O., Tchelepi, H. A. Limitations of Physics Informed Machine Learning for Nonlinear Two-Phase Transport in Porous Media. Journal of Machine Learning for Modeling and Computing, 2020, 1(1): 19-37.

Glorot, X., Bengio, Y. Understanding the difficulty of training deep feedforward neural networks. Paper Presented in Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics, Sardinia, Italy, 13-15 May, 2010.

Haghighat, E., Bekar, A. C., Madenci, E., et al. A nonlocal physics-informed deep learning framework using the peridynamic differential operator. Computer Methods in Applied Mechanics and Engineering, 2021, 385: 114012.

Hanna, J. M., Aguado, J. V., Comas-Cardona, S., et al. Residual-based adaptivity for two-phase flow simulation in porous media using physics-informed neural networks. Computer Methods in Applied Mechanics and Engineering, 2022, 396: 115100.

Jagtap, A. D., Karniadakis, G. E. Extended physics-informed neural networks (XPINNs): A generalized space-time domain decomposition based deep learning framework for nonlinear partial differential equations. Paper Presented at AAAI Spring Symposium: MLPS, Stanford, CA, 22-24 March, 2021.

Jin, X., Cai, S., Li, H., et al. NSFnets (Navier-Stokes flow nets): Physics-informed neural networks for the incompressible Navier-Stokes equations. Journal of Computational Physics, 2021, 426: 109951.

Karniadakis, G. E., Kevrekidis, I. G., Lu, L., et al. Physics-informed machine learning. Nature Reviews Physics, 2021, 3(6): 422-440.

Kemeth, F. P., Bertalan, T., Thiem, T., et al. Learning emergent partial differential equations in a learned emergent space. Nature Communications, 2022, 13(1): 3318.

Khoo, Y., Lu, J., Ying, L. Solving parametric PDE problems with artificial neural networks. European Journal of Applied Mathematics, 2021, 32(3): 421-435.

Lu, T., Li, Z., Lai, F., et al. The effect of flow resistance on water saturation profile for transient two-phase flow in fractal porous media. Advances in Geo-Energy Research, 2018, 2(1): 63-71.

Lu, L., Meng, X., Mao, Z., et al. DeepXDE: A deep learning library for solving differential equations. SIAM Review, 2021, 63(1): 208-228.

Maulik, R., San, O., Rasheed, A., et al. Data-driven deconvolution for large eddy simulation of Kraichnan turbulence. Physics of Fluids, 2018, 30(12): 125109.

Meng, X., Li, Z., Zhang, D., et al. PPINN: Parareal physics-informed neural network for time-dependent PDEs. Computer Methods in Applied Mechanics and Engineering, 2020, 370: 113250.

Raissi, M., Perdikaris, P., Karniadakis, G. E. Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations. Journal of Computational Physics, 2019, 378: 686-707.

Ranade, R., Hill, C., Pathak, J. DiscretizationNet: A machine-learning based solver for Navier-Stokes equations using finite volume discretization. Computer Methods in Applied Mechanics and Engineering, 2021, 378: 113722.

Rodriguez-Torrado, R., Ruiz, P., Cueto-Felgueroso, L., et al. Physics-informed attention-based neural network for hyperbolic partial differential equations: Application to the Buckley-Leverett problem. Scientific Reports, 2022, 12(1): 7557.

Vaswani, A., Shazeer, N., Parmar, N., et al. Attention is all you need. Paper Presented at the 31st Conference on Neural Information Processing Systems, Long Beach, CA, 4-9 December, 2017.

Vinuesa, R., Brunton, S. L. Enhancing computational fluid dynamics with machine learning. Nature Computational Science, 2022, 2(6): 358-366.

Wang, S., Teng, Y., Perdikaris, P. Understanding and mitigating gradient flow pathologies in physics-informed neural networks. SIAM Journal on Scientific Computing, 2021, 43(5): A3055-A3081.

Wu, J., Xiao, H., Paterson, E. Physics-informed machine learning approach for augmenting turbulence models: A comprehensive framework. Physical Review Fluids, 2018, 3(7): 074602.

Xu, R., Zhang, D., Rong, M., et al. Weak form theory-guided neural network (TgNN-wf) for deep learning of subsurface single-and two-phase flow. Journal of Computational Physics, 2021, 436: 110318.

Yang, Y., Perdikaris, P. Adversarial uncertainty quantification in physics-informed neural networks. Journal of Computational Physics, 2019, 394: 136-152.

Yang, Y., Zhou, Y., Blunt, M. J., et al. Advances in multiscale numerical and experimental approaches for multiphysics problems in porous media. Advances in Geo-Energy Research, 2021, 5(3): 233-238.