A Hybrid Probabilistic-Backpropagation Neural Network Solver for Nonlinear Systems in Reservoir Simulation
DOI:
https://doi.org/10.29017/scog.v48i3.1751Keywords:
nonlinear solver, probabilistic, neural networks, backpropagation, reservoir simulationAbstract
Reservoir simulation requires solving large, sparse systems of nonlinear equations, where iterative Krylov subspace solvers such as the conjugate gradient (CG), stabilized conjugate gradient (BiCG-STAB), and generalized minimal residual (GMRES) are widely applied. However, these methods often have limitations in terms of their stability and accuracy in nonlinear systems. This paper introduces a hybrid probabilistic backpropagation neural network (Prob-BPNN) solver that integrates neural-network-based initialization with probabilistic inference to improve robustness. The solver was benchmarked against CG, BiCG-STAB, and GMRES using two synthetic reservoir models with the GMRES solution at a tolerance of 10-10, serving as the reference solution. The results show that Prob-BPNN consistently achieved production profiles closely matching the reference solution, with errors of MAE ≤ 0.066, RMSE ≤ 0.071, MAPE ≤ 2.04%, and R2 ≥ 0.945. In contrast, CG and BiCG-STAB produced unstable and nonphysical results, with errors exceeding 292% and negative R2 values. In terms of computational performance, Prob- BPNN required 9.96 s in Case 1 and 45.90 s in Case 2, compared to 2.85 s and 1.53 s for GMRES, respectively. Although more computationally expensive, Prob-BPNN delivered convergence on the same residual order of magnitude (below 10-3) as GMRES while avoiding the severe instabilities observed in CG and BiCG-STAB. These findings indicate that the Prob-BPNN is preferable in applications where solver robustness and accuracy are critical, even at the expense of a higher execution time. Future research should focus on reducing computational overhead through parallelization and hybridization strategies to enhance the scalability of large-scale reservoir models.
References
Alakeely, A., and Horne, R. (2022): Simulating Oil and Water Production in Reservoirs with Generative Deep Learning, SPE Reservoir Evaluation & Engineering, 25(04), 751–773.
https://doi.org/10.2118/206126-PA Aliaga, J. I., Anzt, H., Grützmacher, T.,
Quintana-Ortí, E. S., and Tomás, A. E. (2023): Compressed basis GMRES on high-performance graphics processing units, The International Journal of High Performance Computing Applications, 37(2), 82–100.https://doi.org/10.1177/1094342022111 5140
Almajid, M. M., and Abu-Al-Saud, M. O. (2022): Prediction of porous media fluid flow using physics informed neural networks, Journal of Petroleum Science and Engineering,208,109205.https://doi.org/10.1016/j.petrol.2021.109 205
Alpak, F. O., Jammoul, M., and Wheeler, M. F. (2023): Consistent Discretization Methods for Reservoir Simulation on Cut-Cell Grids, Presented at the SPE Reservoir Simulation Conference, D011S003R001.
https://doi.org/10.2118/212213-MS Asif, A., Abd, A. S., and Abushaikha, A. (2025):
Advanced Linearization Methods for Efficient and Accurate Compositional Reservoir Simulations, Computation, 13(8), 191.
https://doi.org/10.3390/computation130 80191
Bakhvalov, N. S. (1966): On the convergence of a relaxation method with natural constraints on the elliptic operator, USSR Computational Mathematics and Mathematical Physics, 6(5), 101–135. https://doi.org/10.1016/0041- 5553(66)90118-2
Benzi, M. (2002): Preconditioning Techniques for Large Linear Systems: A Survey, Journal of Computational Physics, 182(2), 418–477.
https://doi.org/10.1006/jcph.2002.7176 Benzi, M., and Tûma, M. (1999): A comparative
study of sparse approximate inverse preconditioners, Applied Numerical Mathematics, 30(2), 305–340. https://doi.org/10.1016/S0168- 9274(98)00118-4
Bhogeswara, R., and Killough, J. E. (1994): Parallel Linear Solvers for Reservoir Simulation: A Generic Approach for Existing and Emerging Computer Architectures, SPE Computer Applications, 6(01), 5–11. https://doi.org/10.2118/25240-PA
Chen, H., Terada, K., Li, A., and Datta-Gupta, A. (2022): Rapid Simulation of Unconventional Reservoirs Using Multi- Domain Multi-Resolution Discretization Based on the Diffusive Time of Flight, Presented at the SPE/AAPG/SEG
Unconventional Resources Technology Conference,OnePetro. https://doi.org/10.15530/urtec-2022- 3723026
Duran, A., and Tuncel, M. (2014): Spectral Effects of Large Matrices from Oil Reservoir Simulators on Performance of Scalable Direct Solvers, Presented at the SPE Large Scale Computing and Big Data Challenges in Reservoir Simulation Conference and Exhibition, SPE- 172984-MS.
https://doi.org/10.2118/172984-MS Gasmi, C. F., and Tchelepi, H. (April 22, 2021):
Physics Informed Deep Learning for Flow and Transport in Porous Media, arXiv. https://doi.org/10.48550/arXiv.2104.026 29
Gasmi, C. F., and Tchelepi, H. (May 19, 2022): Uncertainty Quantification for Transport in Porous media using Parameterized Physics Informed neural Networks, arXiv. https://doi.org/10.48550/arXiv.2205.127 30
Gasparini, L., Rodrigues, J. R. P., Augusto, D. A., Carvalho, L. M., Conopoima, C., Goldfeld, P., Panetta, J., Ramirez, J. P., Souza, M., Figueiredo, M. O., and Leite,
V. M. D. M. (2021): Hybrid parallel iterative sparse linear solver framework for reservoir geomechanical and flow simulation, Journal of Computational Science, 51, 101330.
https://doi.org/10.1016/j.jocs.2021.1013 30
Gharieb, A., El Hamady, A., Gad, R., Lairet, B. B., and Rivas-Rivas, S. (2024): Optimizing Field Development in Data- Starved Reservoirs: Mitigating Economic Risks Through Stochastic Production Profiling and Fully Implicit Static-Dynamic Simulation, Presented at the Mediterranean Offshore Conference, D011S007R002.
https://doi.org/10.2118/223239-MS Goulianas, K., Margaris, A., Refanidis, I., and
Diamantaras, K. (2018): Solving polynomial systems using a fast adaptive back propagation-type neural network algorithm, European Journal of Applied Mathematics, 29(2), 301–337.
https://doi.org/10.1017/S095679251700 0146
Habib, M., and Joslin, K. (2020): Retrospective Validation of the Robustness of Reservoir Simulation Predictions, Presented at the SPE Canada Heavy Oil Conference, D021S003R003. https://doi.org/10.2118/200023-MS
Han, J.-X., Xue, L., Wei, Y.-S., Qi, Y.-D., Wang, J.-L., Liu, Y.-T., and Zhang, Y.-Q. (2023): Physics-informed neural network-based petroleum reservoir simulation with sparse data using domain decomposition, Petroleum Science, 20(6), 3450–3460.
https://doi.org/10.1016/j.petsci.2023.10. 019
He, K., Tan, S. X.-D., Zhao, H., Liu, X.-X.,
Wang, H., and Shi, G. (2016): Parallel GMRES solver for fast analysis of large linear dynamic systems on GPU platforms, Integration, 52, 10–22. https://doi.org/10.1016/j.vlsi.2015.07.00 5
Hestenes, M. R., and Stiefel, E. (1952): Methods of conjugate gradients for solving linear systems, Journal of Research of the National Bureau of Standards, 49(6),
409. https://doi.org/10.6028/jres.049.044 Hørsholt, S., Nick, H., and Jørgensen, J. B. (2019): A Hierarchical Multigrid Method
for Oil Production Optimization, 12th IFAC Symposium on Dynamics and Control of Process Systems, Including Biosystems DYCOPS 2019, 52(1), 492–
497.
https://doi.org/10.1016/j.ifacol.2019.06. 110
Isaiah, J., Schrader, S., Reichhardt, D., and Link,
C. (2013): Performing Reservoir Simulation with Neural Network Enhanced Data, SPE Digital Energy Conference, SPE, The Woodlands, Texas, USA, SPE-163691-MS.
https://doi.org/10.2118/163691-MS Jammoul, M., Alpak, F. O., and Wheeler, M. F.
(2023): An Enriched Galerkin Discretization Scheme for Two Phase Flow on Non-Orthogonal Grids, Presented at the SPE Reservoir Simulation Conference, D011S003R002. https://doi.org/10.2118/212238-MS
Jiang, J., and Pan, H. (2022): Dissipation-Based Nonlinear Solver for Fully Implicit Compositional Simulation, SPE Journal,
27(04),1989–2014.https://doi.org/10.2118/209233-PA
Kang, Z., Deng, Z., Han, W., and Zhang, D.(2018): Parallel Reservoir Simulation with OpenACC and Domain Decomposition, Algorithms, 11(12), 213. https://doi.org/10.3390/a11120213
Karniadakis, G. E., Kevrekidis, I. G., Lu, L., Perdikaris, P., Wang, S., and Yang, L. (2021): Physics-informed machine learning, Nature Reviews Physics, 3(6), 422–440.
https://doi.org/10.1038/s42254-021- 00314-5
Khait, M., and Voskov, D. V. (2017): Operator- based linearization for general purpose reservoir simulation, Journal of Petroleum Science and Engineering, 157, 990–998.
https://doi.org/10.1016/j.petrol.2017.08. 009
Kim, Y., Jang, H., Kim, J., and Lee, J. (2017): Prediction of storage efficiency on CO2 sequestration in deep saline aquifers using artificial neural network, Applied Energy, 185, 916–928.
Kristanto, D., Hariyadi, H., Pramudyohadi, E. W., Kurniawan, A., Nursidik, U. S., Asmorowati, D., Widiyaningsih, I., and Cahyaningtyas, N. (2025): Optimization of Co2 Injection Through Cyclic Huff and Puff to Improve Oil Recovery, Scientific Contributions Oil and Gas, 48(2), 53–67.https://doi.org/10.29017/scog.v48i2.1659
Li, L., and Abushaikha, A. (2022): A fully- implicit parallel framework for complex reservoir simulation with mimetic finite difference discretization and operator- based linearization, Computational Geosciences,26(4),915–931. https://doi.org/10.1007/s10596-021- 10096-5
Lie, K.-A. (2019): An Introduction to Reservoir Simulation Using MATLAB/GNU Octave: User Guide for the MATLAB Reservoir Simulation Toolbox (MRST), Cambridge University Press, Cambridge. https://doi.org/10.1017/9781108591416
Mamo, N. B., and Dennis, Y. A. (2020): Artificial neural network based production forecasting for a hydrocarbon reservoir under water injection, Petroleum Exploration and Development, 47(2), 383–392. https://doi.org/10.1016/S1876-
3804(20)60055-6
Manea, A. M., Sewall, J. ., and Tchelepi, H. A. (2016): Parallel Multiscale Linear Solver for Highly Detailed Reservoir Models, SPE Journal, 21(06), 2062–2078.
https://doi.org/10.2118/173259-PA Mithani, A. H., Rosland, E. A., Jamaludin, M. A.,
W Ismail, W. R., Lajawi, M. T., and A Salam, I. H. (2022): Reservoir Souring Simulation and Modelling Study for Field with Long History of Water Injection: History Matching, Prediction, Presented at the SPE Asia Pacific Oil & Gas Conference and Exhibition, D021S010R003.
https://doi.org/10.2118/210778-MS
Raissi, M., Perdikaris, P., and Karniadakis, G. E.(2019): Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations, Journal of Computational Physics,378,686–707. https://doi.org/10.1016/j.jcp.2018.10.04 5
Saad, Y., and Schultz, M. H. (1986): GMRES: A Generalized Minimal Residual Algorithm for Solving Nonsymmetric Linear Systems, SIAM Journal on Scientific and Statistical Computing, 7(3), 856–869.
https://doi.org/10.1137/0907058
Stüben, K., Clees, T., Klie, H., Lu, B., and Wheeler, M. F. (2007): Algebraic Multigrid Methods (AMG) for the Efficient Solution of Fully Implicit Formulations in Reservoir Simulation, Presented at the SPE Reservoir Simulation Symposium, SPE-105832- MS. https://doi.org/10.2118/105832-MS
Sugihardjo, S. (2022): CCUS-Aksi Mitigasi Gas Rumah Kaca dan Peningkatan Pengurasan Minyak CO2-EOR, Lembaran publikasi minyak dan gas bumi,56(1),21–35.https://doi.org/10.29017/LPMGB.56.1.916
Swadesi, B., Sanmurjana, M., Muhammad Rizky Rahmadsyah Lubis, Dedi Kristanto, Ndaru Cahyaningtyas, and Indah Widiyaningsih (2025): A Simulation Study on Polymer Mobility Design Strategies and Their Impact on Oil Recovery Efficiency and Displacement Mechanisms, Scientific Contributions Oil and Gas, 48(2), 111–129. https://doi.org/10.29017/scog.v48i2.1661
Tokuda, T., and Hashimoto, R. (2023): Development of an Efficient Parallelization Scheme for Fully Implicit Discontinuous Deformation Analysis (DDA), Presented at the 15th ISRM Congress, retrieved September 24, 2025from internet: , ISRM- 15CONGRESS-2023-306.
van der Vorst, H. A. (1992): Bi-CGSTAB: A Fast and Smoothly Converging Variant of Bi- CG for the Solution of Nonsymmetric Linear Systems, SIAM Journal on Scientific and Statistical Computing, 13(2), 631–644.
https://doi.org/10.1137/0913035 Wenger, J., and Hennig, P. (October 2020):
Probabilistic Linear Solvers for Machine Learning, arXiv.
https://doi.org/10.48550/arXiv.2010.096 91
Yan, L., Zhou, Z., Li, B., Fang, Q., Kai, Y., and Shi, Z. (2025): A Reservoir Dynamic Prediction Model Based on the REROSIM Method: Digital Twin and Simulation Studies Driven by Digital Transformation, Presented at the Middle East Oil, Gas and Geosciences Show (MEOS GEO), D031S096R002.
https://doi.org/10.2118/227319-MS Yang, L., Fan, M., Wang, X., Liu, Z., Li, N., Yu,
W., and Wang, Y. (2025): Efficiency Comparison of Direct and Krylov Subspace Iterative Parallel Solvers for Hydraulic Fracture Propagation Simulations in Shale Reservoirs, Presented at the 59th U.S. Rock Mechanics/Geomechanics Symposium, D031S025R002.
https://doi.org/10.56952/ARMA-2025- 0403
Zhang, J., Braga-Neto, U., and Gildin, E. (2024): Physics-Informed Neural Networks for Multiphase Flow in Porous Media Considering Dual Shocks and Interphase Solubility, Energy & Fuels, 38(18), 17781–17795.
https://doi.org/10.1021/acs.energyfuels. 4c02888
Zhao, Y., Fukaya, T., Zhang, L., and Iwashita, T. (2022): Numerical Investigation into the Mixed Precision GMRES(m) Method Using FP64 and FP32, Journal of Information Processing, 30(0), 525–537. https://doi.org/10.2197/ipsjjip.30.525
Zubarev, D. I. (2009): Pros and Cons of Applying Proxy-Models as a Substitute for Full Reservoir Simulations, Presented at the SPE Annual Technical Conference and Exhibition,SPE-124815-MS. https://doi.org/10.2118/124815-MS
Downloads
Published
Issue
Section
License
Copyright (c) 2025 © Copyright by Authors. Published by LEMIGAS
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors are free to Share — copy and redistribute the material in any medium or format for any purpose, even commercially Adapt — remix, transform, and build upon the material for any purpose, even commercially.
The licensor cannot revoke these freedoms as long as you follow the license terms, under the following terms Attribution — You must give appropriate credit , provide a link to the license, and indicate if changes were made . You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.
No additional restrictions — You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.
