Generating geologically realistic 3D reservoir facies models using deep learning of sedimentary architecture with generative adversarial networks

Abstract This paper proposes a novel approach for generating 3-dimensional complex geological facies models based on deep generative models. It can reproduce a wide range of conceptual geological models while possessing the flexibility necessary to honor constraints such as well data. Compared with...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Tuan-Feng Zhang, Peter Tilke, Emilien Dupont, Ling-Chen Zhu, Lin Liang, William Bailey
Formato: article
Lenguaje:EN
Publicado: KeAi Communications Co., Ltd. 2019
Materias:
Q
Acceso en línea:https://doaj.org/article/2d915f0f0add477db016aee0965eb3c5
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:2d915f0f0add477db016aee0965eb3c5
record_format dspace
spelling oai:doaj.org-article:2d915f0f0add477db016aee0965eb3c52021-12-02T09:55:27ZGenerating geologically realistic 3D reservoir facies models using deep learning of sedimentary architecture with generative adversarial networks10.1007/s12182-019-0328-41672-51071995-8226https://doaj.org/article/2d915f0f0add477db016aee0965eb3c52019-06-01T00:00:00Zhttp://link.springer.com/article/10.1007/s12182-019-0328-4https://doaj.org/toc/1672-5107https://doaj.org/toc/1995-8226Abstract This paper proposes a novel approach for generating 3-dimensional complex geological facies models based on deep generative models. It can reproduce a wide range of conceptual geological models while possessing the flexibility necessary to honor constraints such as well data. Compared with existing geostatistics-based modeling methods, our approach produces realistic subsurface facies architecture in 3D using a state-of-the-art deep learning method called generative adversarial networks (GANs). GANs couple a generator with a discriminator, and each uses a deep convolutional neural network. The networks are trained in an adversarial manner until the generator can create “fake” images that the discriminator cannot distinguish from “real” images. We extend the original GAN approach to 3D geological modeling at the reservoir scale. The GANs are trained using a library of 3D facies models. Once the GANs have been trained, they can generate a variety of geologically realistic facies models constrained by well data interpretations. This geomodelling approach using GANs has been tested on models of both complex fluvial depositional systems and carbonate reservoirs that exhibit progradational and aggradational trends. The results demonstrate that this deep learning-driven modeling approach can capture more realistic facies architectures and associations than existing geostatistical modeling methods, which often fail to reproduce heterogeneous nonstationary sedimentary facies with apparent depositional trend.Tuan-Feng ZhangPeter TilkeEmilien DupontLing-Chen ZhuLin LiangWilliam BaileyKeAi Communications Co., Ltd.articleGeological faciesGeomodelingData conditioningGenerative adversarial networksScienceQPetrologyQE420-499ENPetroleum Science, Vol 16, Iss 3, Pp 541-549 (2019)
institution DOAJ
collection DOAJ
language EN
topic Geological facies
Geomodeling
Data conditioning
Generative adversarial networks
Science
Q
Petrology
QE420-499
spellingShingle Geological facies
Geomodeling
Data conditioning
Generative adversarial networks
Science
Q
Petrology
QE420-499
Tuan-Feng Zhang
Peter Tilke
Emilien Dupont
Ling-Chen Zhu
Lin Liang
William Bailey
Generating geologically realistic 3D reservoir facies models using deep learning of sedimentary architecture with generative adversarial networks
description Abstract This paper proposes a novel approach for generating 3-dimensional complex geological facies models based on deep generative models. It can reproduce a wide range of conceptual geological models while possessing the flexibility necessary to honor constraints such as well data. Compared with existing geostatistics-based modeling methods, our approach produces realistic subsurface facies architecture in 3D using a state-of-the-art deep learning method called generative adversarial networks (GANs). GANs couple a generator with a discriminator, and each uses a deep convolutional neural network. The networks are trained in an adversarial manner until the generator can create “fake” images that the discriminator cannot distinguish from “real” images. We extend the original GAN approach to 3D geological modeling at the reservoir scale. The GANs are trained using a library of 3D facies models. Once the GANs have been trained, they can generate a variety of geologically realistic facies models constrained by well data interpretations. This geomodelling approach using GANs has been tested on models of both complex fluvial depositional systems and carbonate reservoirs that exhibit progradational and aggradational trends. The results demonstrate that this deep learning-driven modeling approach can capture more realistic facies architectures and associations than existing geostatistical modeling methods, which often fail to reproduce heterogeneous nonstationary sedimentary facies with apparent depositional trend.
format article
author Tuan-Feng Zhang
Peter Tilke
Emilien Dupont
Ling-Chen Zhu
Lin Liang
William Bailey
author_facet Tuan-Feng Zhang
Peter Tilke
Emilien Dupont
Ling-Chen Zhu
Lin Liang
William Bailey
author_sort Tuan-Feng Zhang
title Generating geologically realistic 3D reservoir facies models using deep learning of sedimentary architecture with generative adversarial networks
title_short Generating geologically realistic 3D reservoir facies models using deep learning of sedimentary architecture with generative adversarial networks
title_full Generating geologically realistic 3D reservoir facies models using deep learning of sedimentary architecture with generative adversarial networks
title_fullStr Generating geologically realistic 3D reservoir facies models using deep learning of sedimentary architecture with generative adversarial networks
title_full_unstemmed Generating geologically realistic 3D reservoir facies models using deep learning of sedimentary architecture with generative adversarial networks
title_sort generating geologically realistic 3d reservoir facies models using deep learning of sedimentary architecture with generative adversarial networks
publisher KeAi Communications Co., Ltd.
publishDate 2019
url https://doaj.org/article/2d915f0f0add477db016aee0965eb3c5
work_keys_str_mv AT tuanfengzhang generatinggeologicallyrealistic3dreservoirfaciesmodelsusingdeeplearningofsedimentaryarchitecturewithgenerativeadversarialnetworks
AT petertilke generatinggeologicallyrealistic3dreservoirfaciesmodelsusingdeeplearningofsedimentaryarchitecturewithgenerativeadversarialnetworks
AT emiliendupont generatinggeologicallyrealistic3dreservoirfaciesmodelsusingdeeplearningofsedimentaryarchitecturewithgenerativeadversarialnetworks
AT lingchenzhu generatinggeologicallyrealistic3dreservoirfaciesmodelsusingdeeplearningofsedimentaryarchitecturewithgenerativeadversarialnetworks
AT linliang generatinggeologicallyrealistic3dreservoirfaciesmodelsusingdeeplearningofsedimentaryarchitecturewithgenerativeadversarialnetworks
AT williambailey generatinggeologicallyrealistic3dreservoirfaciesmodelsusingdeeplearningofsedimentaryarchitecturewithgenerativeadversarialnetworks
_version_ 1718397874304188416