Show simple item record

dc.contributor.authorYan, Bicheng
dc.contributor.authorHarp, Dylan Robert
dc.contributor.authorChen, Bailian
dc.contributor.authorPawar, Rajesh
dc.date.accessioned2022-01-17T08:45:58Z
dc.date.available2022-01-17T08:45:58Z
dc.date.issued2021-12
dc.identifier.citationYan, B., Harp, D. R., Chen, B., & Pawar, R. (2021). A physics-constrained deep learning model for simulating multiphase flow in 3D heterogeneous porous media. Fuel, 122693. doi:10.1016/j.fuel.2021.122693
dc.identifier.issn0016-2361
dc.identifier.doi10.1016/j.fuel.2021.122693
dc.identifier.urihttp://hdl.handle.net/10754/674979
dc.description.abstractPhysics-based simulators for multiphase flow in porous media emulate nonlinear processes with coupled physics, and usually require extensive computational resources for software development, maintenance and simulation execution. As a result, a huge demand exists for fast modeling of coupled processes in a wide range of subsurface applications including geological CO2 sequestration, hydrocarbon recovery and geothermal energy extraction. In this work, an efficient physics-constrained deep learning model is developed for solving multiphase flow in 3-Dimensional (3D) heterogeneous porous media. The model fully leverages the spatial topology predictive capability of convolutional neural networks, specifically U-Net with successive contracting and expansive steps, and is coupled with an efficient continuity-based smoother to predict flow responses that need spatial continuity. Furthermore, the transient regions are penalized to steer the training process such that the model can accurately capture flow in these regions. The model takes inputs including properties of porous media, fluid properties and well controls, and predicts the temporal-spatial evolution of the state variables (pressure and saturation). While maintaining the continuity of fluid flow, the 3D spatial domain is decomposed into 2D images for reducing training cost, and the decomposition results in an increased number of training data samples and better training efficiency. Additionally, a surrogate model is separately constructed as a postprocessor to calculate well flow rate based on the predictions of state variables from the deep learning model. We use the example of CO2 injection into saline aquifers, and apply the physics-constrained deep learning model that is trained from physics-based simulation data and emulates the physics process. The model performs prediction with a speedup of ∼ 1400 times compared to physics-based simulations, and the average temporal errors of predicted pressure and saturation plumes are 0.27% and 0.099% respectively. Furthermore, water production rate is efficiently predicted by a surrogate model for well flow rate, with a mean error less than 5%. Therefore, with its unique scheme to cope with the fidelity in fluid flow in porous media, the physics-constrained deep learning model can become an efficient predictive model for computationally demanding inverse problems or other coupled processes.
dc.description.sponsorshipThe authors acknowledge the financial support by US DOE's Fossil Energy Program Office through the project, Science-informed Machine Learning to Accelerate Real Time (SMART) Decisions in Subsurface Applications. Funding for SMART is managed by the National Energy Technology Laboratory (NETL). The authors also thank Dr. Seyyed A. Hosseini from University of Texas at Austin for providing the reservoir simulation data for CO geological sequestration, and thank Dr. Diana Bacon from Pacific Northwest National Laboratory for providing parsing tools to process simulation data.
dc.publisherElsevier BV
dc.relation.urlhttps://linkinghub.elsevier.com/retrieve/pii/S001623612102559X
dc.rightsNOTICE: this is the author’s version of a work that was accepted for publication in Fuel. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Fuel, [, , (2021-12)] DOI: 10.1016/j.fuel.2021.122693 . © 2021. This manuscript version is made available under the CC-BY-NC-ND 4.0 license http://creativecommons.org/licenses/by-nc-nd/4.0/
dc.titleA physics-constrained deep learning model for simulating multiphase flow in 3D heterogeneous porous media
dc.typeArticle
dc.contributor.departmentPhysical Science and Engineering Division, King Abdullah University of Science and Technology (KAUST), Thuwal 23955, Saudi Arabia
dc.identifier.journalFuel
dc.rights.embargodate2023-12-01
dc.eprint.versionPost-print
dc.contributor.institutionEarth and Environmental Sciences, Los Alamos National Laboratory, Los Alamos, NM 87545, USA
dc.identifier.pages122693
dc.identifier.arxivid2105.09467
kaust.personYan, Bicheng
dc.identifier.eid2-s2.0-85120916915


Files in this item

Thumbnail
Name:
JFUE-D-21-06636_R1-Accepted.pdf
Size:
2.503Mb
Format:
PDF
Description:
Accepted Manuscript
Embargo End Date:
2023-12-01

This item appears in the following Collection(s)

Show simple item record