Application of a serial denoising autoencoder for geological plausibility of a channelized reservoir in history matching

S. Kim; B. Min; J. Choe

doi:10.3997/2214-4609.201902194

Application of a serial denoising autoencoder for geological plausibility of a channelized reservoir in history matching

S. Kim, B. Min, J. Choe

Department of Climate and Energy Systems Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Scopus citations

Abstract

Denoising autoencoder (DAE) is utilized to preserve and improve geological reality and plausibility in a channelized reservoir model during history matching by ensemble smoother with multiple data assimilation (ES-MDA). As one of history matching methods, ES-MDA calibrates reservoir properties such as rock facies corresponding to production history. While ES-MDA modifies reservoir parameters, it recognizes them only as figures not honoring to geological features. Thus, conservation of geological characteristics during calibration of reservoir parameters is challenging in ES-MDA. DAE is trained to restore lost connectivity and pattern of an original geological concept and it is applied to posterior reservoir models after an assimilation by ES-MDA. ES-MDA combined with DAE shows not only geologically enhanced channel models but also well-matched production prediction.

Original language	English
Title of host publication	4th EAGE Conference on Petroleum Geostatistics
Publisher	European Association of Geoscientists and Engineers, EAGE
ISBN (Electronic)	9789462822962
DOIs	https://doi.org/10.3997/2214-4609.201902194
State	Published - 2019
Event	4th EAGE Conference on Petroleum Geostatistics - Florence, Italy Duration: 2 Sep 2019 → 6 Sep 2019

Publication series

Name	4th EAGE Conference on Petroleum Geostatistics

Conference

Conference	4th EAGE Conference on Petroleum Geostatistics
Country/Territory	Italy
City	Florence
Period	2/09/19 → 6/09/19

Bibliographical note

Funding Information:
The authors acknowledge the project of Korea Institute of Geoscience and Mineral Resources (Project No. GP2017-024) and the project of Korea Institute of Energy Technology Evaluation and Planning granted financial resources from the Ministry of Trade, Industry, and Energy, Republic of Korea (No. 20172510102090). Dr. Baehyun Min was partially supported by the National Research Foundation of Korea (No. 2018R1A6A1A08025520 and No. 2019R1C1C1002574).

Publisher Copyright:
© EAGE 2019.

Access to Document

10.3997/2214-4609.201902194

Cite this

Kim, S., Min, B., & Choe, J. (2019). Application of a serial denoising autoencoder for geological plausibility of a channelized reservoir in history matching. In 4th EAGE Conference on Petroleum Geostatistics Article ThP17 (4th EAGE Conference on Petroleum Geostatistics). European Association of Geoscientists and Engineers, EAGE. https://doi.org/10.3997/2214-4609.201902194

@inproceedings{878409ec26234e4786babc5eb2925fae,

title = "Application of a serial denoising autoencoder for geological plausibility of a channelized reservoir in history matching",

abstract = "Denoising autoencoder (DAE) is utilized to preserve and improve geological reality and plausibility in a channelized reservoir model during history matching by ensemble smoother with multiple data assimilation (ES-MDA). As one of history matching methods, ES-MDA calibrates reservoir properties such as rock facies corresponding to production history. While ES-MDA modifies reservoir parameters, it recognizes them only as figures not honoring to geological features. Thus, conservation of geological characteristics during calibration of reservoir parameters is challenging in ES-MDA. DAE is trained to restore lost connectivity and pattern of an original geological concept and it is applied to posterior reservoir models after an assimilation by ES-MDA. ES-MDA combined with DAE shows not only geologically enhanced channel models but also well-matched production prediction.",

author = "S. Kim and B. Min and J. Choe",

note = "Funding Information: The authors acknowledge the project of Korea Institute of Geoscience and Mineral Resources (Project No. GP2017-024) and the project of Korea Institute of Energy Technology Evaluation and Planning granted financial resources from the Ministry of Trade, Industry, and Energy, Republic of Korea (No. 20172510102090). Dr. Baehyun Min was partially supported by the National Research Foundation of Korea (No. 2018R1A6A1A08025520 and No. 2019R1C1C1002574). Publisher Copyright: {\textcopyright} EAGE 2019.; 4th EAGE Conference on Petroleum Geostatistics ; Conference date: 02-09-2019 Through 06-09-2019",

year = "2019",

doi = "10.3997/2214-4609.201902194",

language = "English",

series = "4th EAGE Conference on Petroleum Geostatistics",

publisher = "European Association of Geoscientists and Engineers, EAGE",

booktitle = "4th EAGE Conference on Petroleum Geostatistics",

}

Kim, S, Min, B & Choe, J 2019, Application of a serial denoising autoencoder for geological plausibility of a channelized reservoir in history matching. in 4th EAGE Conference on Petroleum Geostatistics., ThP17, 4th EAGE Conference on Petroleum Geostatistics, European Association of Geoscientists and Engineers, EAGE, 4th EAGE Conference on Petroleum Geostatistics, Florence, Italy, 2/09/19. https://doi.org/10.3997/2214-4609.201902194

Application of a serial denoising autoencoder for geological plausibility of a channelized reservoir in history matching. / Kim, S.; Min, B.; Choe, J.
4th EAGE Conference on Petroleum Geostatistics. European Association of Geoscientists and Engineers, EAGE, 2019. ThP17 (4th EAGE Conference on Petroleum Geostatistics).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Application of a serial denoising autoencoder for geological plausibility of a channelized reservoir in history matching

AU - Kim, S.

AU - Min, B.

AU - Choe, J.

N1 - Funding Information: The authors acknowledge the project of Korea Institute of Geoscience and Mineral Resources (Project No. GP2017-024) and the project of Korea Institute of Energy Technology Evaluation and Planning granted financial resources from the Ministry of Trade, Industry, and Energy, Republic of Korea (No. 20172510102090). Dr. Baehyun Min was partially supported by the National Research Foundation of Korea (No. 2018R1A6A1A08025520 and No. 2019R1C1C1002574). Publisher Copyright: © EAGE 2019.

PY - 2019

Y1 - 2019

N2 - Denoising autoencoder (DAE) is utilized to preserve and improve geological reality and plausibility in a channelized reservoir model during history matching by ensemble smoother with multiple data assimilation (ES-MDA). As one of history matching methods, ES-MDA calibrates reservoir properties such as rock facies corresponding to production history. While ES-MDA modifies reservoir parameters, it recognizes them only as figures not honoring to geological features. Thus, conservation of geological characteristics during calibration of reservoir parameters is challenging in ES-MDA. DAE is trained to restore lost connectivity and pattern of an original geological concept and it is applied to posterior reservoir models after an assimilation by ES-MDA. ES-MDA combined with DAE shows not only geologically enhanced channel models but also well-matched production prediction.

AB - Denoising autoencoder (DAE) is utilized to preserve and improve geological reality and plausibility in a channelized reservoir model during history matching by ensemble smoother with multiple data assimilation (ES-MDA). As one of history matching methods, ES-MDA calibrates reservoir properties such as rock facies corresponding to production history. While ES-MDA modifies reservoir parameters, it recognizes them only as figures not honoring to geological features. Thus, conservation of geological characteristics during calibration of reservoir parameters is challenging in ES-MDA. DAE is trained to restore lost connectivity and pattern of an original geological concept and it is applied to posterior reservoir models after an assimilation by ES-MDA. ES-MDA combined with DAE shows not only geologically enhanced channel models but also well-matched production prediction.

UR - http://www.scopus.com/inward/record.url?scp=85073107437&partnerID=8YFLogxK

U2 - 10.3997/2214-4609.201902194

DO - 10.3997/2214-4609.201902194

M3 - Conference contribution

AN - SCOPUS:85073107437

T3 - 4th EAGE Conference on Petroleum Geostatistics

BT - 4th EAGE Conference on Petroleum Geostatistics

PB - European Association of Geoscientists and Engineers, EAGE

T2 - 4th EAGE Conference on Petroleum Geostatistics

Y2 - 2 September 2019 through 6 September 2019

ER -

Application of a serial denoising autoencoder for geological plausibility of a channelized reservoir in history matching

Abstract

Publication series

Conference

Bibliographical note

Access to Document

Fingerprint

Cite this