U-FNO—An enhanced Fourier neural operator-based deep-learning model for multiphase flow. (May 2022)

• Record Type: Journal Article
• Authors: Wen, Gege; Li, Zongyi; Azizzadenesheli, Kamyar; Anandkumar, Anima; Benson, Sally M.
• Is Part Of: Advances in water resources. Volume 163(2022)
• Publication Date: 2022-05
• Subjects: Multiphase flow -- Fourier neural operator -- Convolutional neural network -- Carbon capture and storage -- Deep learning; Hydrology -- Periodicals; Hydrodynamics -- Periodicals; Hydraulic engineering -- Periodicals; 551.48
• DOI: 10.1016/j.advwatres.2022.104180
• Languages: English
• ISSNs: 0309-1708

Abstract

Abstract: Numerical simulation of multiphase flow in porous media is essential for many geoscience applications. Machine learning models trained with numerical simulation data can provide a faster alternative to traditional simulators. Here we present U-FNO, a novel neural network architecture for solving multiphase flow problems with superior accuracy, speed, and data efficiency. U-FNO is designed based on the newly proposed Fourier neural operator (FNO), which has shown excellent performance in single-phase flows. We extend the FNO-based architecture to a highly complex CO 2 -water multiphase problem with wide ranges of permeability and porosity heterogeneity, anisotropy, reservoir conditions, injection configurations, flow rates, and multiphase flow properties. The U-FNO architecture is more accurate in gas saturation and pressure buildup predictions than the original FNO and a state-of-the-art convolutional neural network (CNN) benchmark. Meanwhile, it has superior data utilization efficiency, requiring only a third of the training data to achieve the equivalent accuracy as CNN. U-FNO provides superior performance in highly heterogeneous geological formations and critically important applications such as gas saturation and pressure buildup "fronts" determination. The trained model can serve as a general-purpose alternative to routine numerical simulations of 2D-radial CO 2 injection problems with significant speed-ups than traditional simulators. Highlights: U-FNO model for multiphase flow designed based on Fourier neural operator. Data-efficient multiphase flow predictions for gas saturation and pressure buildup. Results are significantly faster and more accurate than state-of-the-art CNNs.