no code implementations • 14 Dec 2023 • Haoyu Tang, Louis J. Durlofsky
Results are presented for a large set of test cases, in which five injection wells and five production wells are placed randomly throughout the model, with a random control variable (bottom-hole pressure) assigned to each well.
no code implementations • 5 Oct 2023 • Su Jiang, Louis J. Durlofsky
In this study, we develop and implement (in DSI) a deep-learning-based parameterization to represent spatio-temporal pressure and CO2 saturation fields at a set of time steps.
no code implementations • 11 Aug 2023 • Yifu Han, Francois P. Hamon, Su Jiang, Louis J. Durlofsky
The trained surrogate model is shown to provide accurate predictions for new realizations over the full range of geological scenarios, with median relative error of 1. 3% in pressure and 4. 5% in saturation.
no code implementations • 21 Jul 2022 • Yusuf Nasir, Louis J. Durlofsky
In both cases, four assets with different well counts, well configurations, and geostatistical descriptions are considered.
no code implementations • 23 Apr 2022 • Su Jiang, Louis J. Durlofsky
The multifidelity surrogate is also applied for history matching using an ensemble-based procedure, where accuracy relative to reference results is again demonstrated.
no code implementations • 24 Mar 2022 • Yusuf Nasir, Louis J. Durlofsky
The DRL-based methodology is shown to result in an NPV increase of 15% (for the 2D cases) and 33% (3D cases) relative to robust optimization over prior models, and to an average improvement of 4% in NPV relative to traditional CLRM.
no code implementations • 14 Mar 2022 • Yong Do Kim, Louis J. Durlofsky
The CNN-RNN proxy is trained using simulation results for 300 different sets of BHP schedules and permeability realizations.
no code implementations • 30 Oct 2021 • Haoyu Tang, Louis J. Durlofsky
In this work, we present an optimization framework in which these simulations are performed with low-fidelity (LF) models.
no code implementations • 4 May 2021 • Meng Tang, Xin Ju, Louis J. Durlofsky
The surrogate model is trained to predict the 3D CO2 saturation and pressure fields in the storage aquifer, and 2D displacement maps at the Earth's surface.
no code implementations • 16 Jul 2020 • Yimin Liu, Louis J. Durlofsky
Geological parameterization enables the representation of geomodels in terms of a relatively small set of variables.
no code implementations • 30 Apr 2020 • Su Jiang, Louis J. Durlofsky
Data-space inversion (DSI) and related procedures represent a family of methods applicable for data assimilation in subsurface flow settings.
no code implementations • 16 Aug 2019 • Meng Tang, Yimin Liu, Louis J. Durlofsky
High-fidelity numerical simulation results for the posterior geomodels (generated by the surrogate-based data assimilation procedure) are shown to be in essential agreement with the recurrent R-U-Net predictions.
no code implementations • 7 Jul 2018 • Yimin Liu, Wenyue Sun, Louis J. Durlofsky
The metrics are based on summary statistics of the nonlinear filter responses of geological models to a pre-trained deep CNN.