Physics-based simulation for fluid flow in porous media is a computational technology to predict the temporal-spatial evolution of state variables (e.g. pressure) in porous media, and usually requires high computational expense due to its nonlinearity and the scale of the study domain. This letter describes a deep learning (DL) workflow to predict the pressure evolution as fluid flows in large-scale 3D heterogeneous porous media. In particular, we apply feature coarsening technique to extract the most representative information and perform the training and prediction of DL at the coarse scale, and further recover the resolution at the fine scale by 2D piecewise cubic interpolation. We validate the DL approach that is trained from physics-based simulation data to predict pressure field in a field-scale 3D geologic CO_2 storage reservoir. We evaluate the impact of feature coarsening on DL performance, and observe that the feature coarsening can not only decrease training time by >74% and reduce memory consumption by >75%, but also maintains temporal error <1.5%. Besides, the DL workflow provides predictive efficiency with ~1400 times speedup compared to physics-based simulation.
Neural Networks for Predicting Permeability Tensors of 2D Porous Media: Comparison of Convolution- and Transformer-based Architectures
PoreLab, The Njord Centre, Department of Physics, University of Oslo（PoreLab、Njord中心、物理系、奥斯陆大学）；PoreLab, Department of Physics, Norwegian University of Science and Technology（PoreLab、物理系、挪威科学技术大学）；The Njord Centre, Department of Physics, University of Oslo（Njord中心、物理系、奥斯陆大学）
Machine learning methods for prediction of breakthrough curves in reactive porous media
Fraunhofer ITWM（弗劳恩霍夫工业数学研究所）；Technical University Kaiserslautern（凯泽斯劳滕工业大学）；Bulgarian Academy of Sciences（保加利亚科学院）；Skolkovo Institute of Science and Technology（斯科尔科沃科学技术研究院）；Artificial Intelligence Research Institute（人工智能研究院）
Deep Learning-Driven Prediction of Microstructure Evolution via Latent Space Interpolation
School of Materials Science；Technology, Indian Institute of Technology (BHU), Varanasi, India；Department of Materials Science \& Engineering, Indian Institute of Technology, Kanpur, Kanpur 208016, India
Physics-informed convolutional neural networks for fluid flow through porous media
Dioscuri Center in Topological Data Analysis, Institute of Mathematics, Polish Academy of Sciences（拓扑数据分析迪奥斯库里中心，波兰科学院数学研究所）；Institute of Experimental Physics, Faculty of Physics and Astronomy, University of Wrocław（物理学与天文学系实验物理研究所，沃林堡大学）；Institute of Theoretical Physics, Faculty of Physics and Astronomy, University of Wrocław（天文学与物理学系理论物理研究所，沃林堡大学）；Parallel and Distributed Systems Laboratory, Jožef Stefan Institute（乔泽夫·斯蒂芬研究所并行与分布式系统实验室）
King Abdullah University of Science and Technology（阿卜杜拉国王科技大学）；Xiangtan University（湘潭大学）；Academy of Mathematics and System Sciences, Chinese Academy of Sciences（中国科学院数学与系统科学研究院）；University of Chinese Academy of Sciences（中国科学院大学）；Shenzhen International Center for Industrial and Applied Mathematics（深圳国际工业与应用数学中心）；Shenzhen Research Institute of Big Data（深圳大数据研究院）
CNN-powered micro- to macro-scale flow modeling in deformable porous media
Institute of Mechanics and Computational Mechanics (IBNM), Leibniz University Hannover（莱布尼茨汉诺威大学力学与计算力学研究所（IBNM））；Institute of Applied Mechanics, University of Stuttgart（斯图加特大学应用力学研究所）；Institute of General Mechanics, RWTH Aachen University（亚琛工业大学普通力学研究所）
A Generative Deep Learning and Explainable Machine Learning Framework for Heat Transfer Prediction and Analysis in Porous Structures with Oscillatory Flows
Department of Mechanical and Aerospace Engineering, University of Houston（休斯顿大学机械与航空航天工程系）；Department of Mechanical Engineering, Rice University（莱斯大学机械工程系）；Department of Mechanical Engineering, Embry‑Riddle Aeronautical University（埃默里-里德尔航空大学机械工程系）
