Stochastic representation and conditioning of process-based geological model by deep generative and recognition networks

Journal Article · 26 February 2024 · Petroleum Geoscience

DOI: https://doi.org/10.1144/petgeo2022-032 · OSTI ID:1988001

Cheung, S. W.  ^[1]; Kushwaha, A. ^[2]; Sun, H. ^[3]; Wu, X. -H. ^[3]

---

1. Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States). Center for Applied Scientific Computing
2. SambaNova Systems, Palo Alto, CA (United States)
3. ExxonMobil Upstream Research Company, Spring, TX (United States)

Accurate and realistic geological modeling is the core of oil and gas development and production. In recent years, process-based methods are developed to produce highly realistic geological models by simulating the physical processes that reproduce the sedimentary events and develop the geometry. However, the complex dynamic processes are extremely expensive to simulate, making process-based models difficult to be conditioned to field data. In this work, we propose a comprehensive generative adversarial network framework as a machine-learning-assisted approach for mimicking the outputs of process-based geological models with fast generation. The main objective of our work is to obtain a continuous parametrization of the highly realistic process-based geological models which enables us to calibrate the models and condition the models to data. Numerical results are presented to illustrate the capability of our proposed methodology.

• Research Organization: Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States)
• Sponsoring Organization: USDOE National Nuclear Security Administration (NNSA)
• Grant/Contract Number: AC52-07NA27344
• OSTI ID: 1988001
• Report Number(s): LLNL-JRNL-826205; ISSN 2041-496X; 1040744
• Journal Information: Petroleum Geoscience, Vol. 30, Issue 1; ISSN 1354-0793
• Publisher: The Geological Society of London - EAGECopyright Statement
• Country of Publication: United States
• Language: English