Uncertainty quantification and inverse modeling for subsurface flow in 3D heterogeneous formations using a theory-guided convolutional encoder-decoder network

We build surrogate models for dynamic 3D subsurface single-phase flow problems with multiple vertical producing wells. The surrogate model provides efficient pressure estimation of the entire formation at any timestep given a stochastic permeability field, arbitrary well locations and penetration lengths, and a timestep matrix as inputs. The well production rate or bottom hole pressure can then be determined based on Peaceman's formula. The original surrogate modeling task is transformed into an image-to-image regression problem using a convolutional encoder-decoder neural network architecture. The residual of the governing flow equation in its discretized form is incorporated into the loss function to impose theoretical guidance on the model training process. As a result, the accuracy and generalization ability of the trained surrogate models are significantly improved compared to fully data-driven models. They are also shown to have flexible extrapolation ability to permeability fields with different statistics. The surrogate models are used to conduct uncertainty quantification considering a stochastic permeability field, as well as to infer unknown permeability information based on limited well production data and observation data of formation properties. Results are shown to be in good agreement with traditional numerical simulation tools, but computational efficiency is dramatically improved.

[1]  R. Ghanem,et al.  Stochastic Finite Elements: A Spectral Approach , 1990 .

[2]  Rui Xu,et al.  Weak Form Theory-guided Neural Network (TgNN-wf) for Deep Learning of Subsurface Single and Two-phase Flow , 2020, J. Comput. Phys..

[3]  Lei Yang,et al.  An improved Sobel edge detection , 2010, 2010 3rd International Conference on Computer Science and Information Technology.

[4]  Z. Qu,et al.  Three-dimensional pore-scale study of methane gas mass diffusion in shale with spatially heterogeneous and anisotropic features , 2020, Fuel.

[5]  Jun Yao,et al.  Apparent gas permeability in an organic-rich shale reservoir , 2016 .

[6]  Nagiza F. Samatova,et al.  Theory-Guided Data Science: A New Paradigm for Scientific Discovery from Data , 2016, IEEE Transactions on Knowledge and Data Engineering.

[7]  Paris Perdikaris,et al.  Physics-Constrained Deep Learning for High-dimensional Surrogate Modeling and Uncertainty Quantification without Labeled Data , 2019, J. Comput. Phys..

[8]  Dongxiao Zhang,et al.  Jointly updating the mean size and spatial distribution of facies in reservoir history matching , 2015, Computational Geosciences.

[9]  L. Durlofsky,et al.  Deep-learning-based surrogate model for reservoir simulation with time-varying well controls , 2020, Journal of Petroleum Science and Engineering.

[10]  Haibin Chang,et al.  Theory-guided Auto-Encoder for Surrogate Construction and Inverse Modeling , 2020, ArXiv.

[11]  Heng Li,et al.  Deep Learning of Subsurface Flow via Theory-guided Neural Network , 2019, ArXiv.

[12]  S. Kollmannsberger,et al.  Physics-Informed Neural Networks , 2021, Deep Learning in Computational Mechanics.

[13]  Ilias Bilionis,et al.  Deep UQ: Learning deep neural network surrogate models for high dimensional uncertainty quantification , 2018, J. Comput. Phys..

[14]  Dongxiao Zhang,et al.  An efficient, high-order perturbation approach for flow in random porous media via Karhunen-Loève and polynomial expansions , 2004 .

[15]  Dongxiao Zhang,et al.  A multi-continuum multiple flow mechanism simulator for unconventional oil and gas recovery , 2015 .

[16]  Jichun Wu,et al.  Deep Autoregressive Neural Networks for High‐Dimensional Inverse Problems in Groundwater Contaminant Source Identification , 2018, Water Resources Research.

[17]  Zbigniew Michalewicz,et al.  Particle Swarm Optimization for Single Objective Continuous Space Problems: A Review , 2017, Evolutionary Computation.

[18]  George E. Karniadakis,et al.  Hidden physics models: Machine learning of nonlinear partial differential equations , 2017, J. Comput. Phys..

[19]  Jichun Wu,et al.  Deep Convolutional Encoder‐Decoder Networks for Uncertainty Quantification of Dynamic Multiphase Flow in Heterogeneous Media , 2018, Water Resources Research.

[20]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[21]  Nanzhe Wang,et al.  Efficient well placement optimization based on theory-guided convolutional neural network , 2022, Journal of Petroleum Science and Engineering.

[22]  Yue Shi,et al.  A modified particle swarm optimizer , 1998, 1998 IEEE International Conference on Evolutionary Computation Proceedings. IEEE World Congress on Computational Intelligence (Cat. No.98TH8360).

[23]  Hamdi A. Tchelepi,et al.  Stochastic Formulation for Uncertainty Analysis of Two-Phase Flow in Heterogeneous Reservoirs , 2000 .

[24]  Pore‐Scale Study of Water Adsorption and Subsequent Methane Transport in Clay in the Presence of Wettability Heterogeneity , 2020, Water Resources Research.

[25]  Haibin Chang,et al.  Efficient Uncertainty Quantification for Dynamic Subsurface Flow with Surrogate by Theory-guided Neural Network , 2020, Computer Methods in Applied Mechanics and Engineering.

[26]  Riccardo Poli,et al.  Particle swarm optimization , 1995, Swarm Intelligence.

[27]  K. Aziz,et al.  Petroleum Reservoir Simulation , 1979 .

[28]  Dongxiao Zhang,et al.  Solution of diffusivity equations with local sources/sinks and surrogate modeling using weak form Theory-guided Neural Network , 2021, Advances in Water Resources.

[29]  Haibin Chang,et al.  History matching of facies distribution with the EnKF and level set parameterization , 2010, J. Comput. Phys..

[30]  Paris Perdikaris,et al.  Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations , 2019, J. Comput. Phys..

[31]  Nicholas Zabaras,et al.  Bayesian Deep Convolutional Encoder-Decoder Networks for Surrogate Modeling and Uncertainty Quantification , 2018, J. Comput. Phys..