Determination of Reservoir Lithology from Seismic Data by a 2D Hidden Markov Random Field Model

Runhai Feng; Stefan Luthi; Dries Gisolf; Allard Martinius

doi:10.3997/2214-4609.201801076

Determination of Reservoir Lithology from Seismic Data by a 2D Hidden Markov Random Field Model

Runhai Feng, Stefan Luthi, Dries Gisolf, Allard Martinius

Applied Geology

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

9 Downloads (Pure)

Abstract

In this study, geological prior information is incorporated in the classification of reservoir lithologies using the Markov Random Field (MRF) technique. The prediction of hidden lithologies in seismic data is based on measured
observations such as seismic inversion results, which are associated with the latent categorical variables derived from the distribution of Gaussian assumptions. The Hidden Markov Random Field (HMRF) approach can connect
similar lithologies laterally (horizontally) while ensure a geologically reasonable stratigraphic (vertical) ordering. It is, therefore, able to exclude randomly appearing lithologies caused by errors in the inversion. In HMRF, the prior
information consists of a Gibbs distribution function and transition probability matrices. The Gibbs distribution connects similar lithologies and does not need a geological definition derived from non-case-related information.
The transition matrices provide preferential transitions between different lithologies and an estimation of these matrices implicitly depends on the depositional environments and juxtaposition rules between different lithologies.

Original language	English
Title of host publication	80th EAGE Conference and Exhibition 2018, 11-14 June, Copenhagen, Denmark
Number of pages	5
DOIs	https://doi.org/10.3997/2214-4609.201801076
Publication status	Published - 2018
Event	80th EAGE Conference and Exhibition 2018: Opportunities presented by the energy transition - Copenhagen, Denmark Duration: 11 Jun 2018 → 14 Jun 2018 Conference number: 80 https://events.eage.org/2018/EAGE%20Annual%202018

Conference

Conference	80th EAGE Conference and Exhibition 2018
Abbreviated title	EAGE 2018
Country/Territory	Denmark
City	Copenhagen
Period	11/06/18 → 14/06/18
Internet address	https://events.eage.org/2018/EAGE%20Annual%202018

Bibliographical note

Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

Access to Document

10.3997/2214-4609.201801076

We_C_11Final published version, 611 KB

Cite this

@inproceedings{e82de98fd2eb4ed683535139b6eb097c,

title = "Determination of Reservoir Lithology from Seismic Data by a 2D Hidden Markov Random Field Model",

abstract = "In this study, geological prior information is incorporated in the classification of reservoir lithologies using the Markov Random Field (MRF) technique. The prediction of hidden lithologies in seismic data is based on measuredobservations such as seismic inversion results, which are associated with the latent categorical variables derived from the distribution of Gaussian assumptions. The Hidden Markov Random Field (HMRF) approach can connectsimilar lithologies laterally (horizontally) while ensure a geologically reasonable stratigraphic (vertical) ordering. It is, therefore, able to exclude randomly appearing lithologies caused by errors in the inversion. In HMRF, the priorinformation consists of a Gibbs distribution function and transition probability matrices. The Gibbs distribution connects similar lithologies and does not need a geological definition derived from non-case-related information.The transition matrices provide preferential transitions between different lithologies and an estimation of these matrices implicitly depends on the depositional environments and juxtaposition rules between different lithologies.",

author = "Runhai Feng and Stefan Luthi and Dries Gisolf and Allard Martinius",

note = "Green Open Access added to TU Delft Institutional Repository {\textquoteleft}You share, we take care!{\textquoteright} – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.; 80th EAGE Conference and Exhibition 2018 : Opportunities presented by the energy transition, EAGE 2018 ; Conference date: 11-06-2018 Through 14-06-2018",

year = "2018",

doi = "10.3997/2214-4609.201801076",

language = "English",

booktitle = "80th EAGE Conference and Exhibition 2018, 11-14 June, Copenhagen, Denmark",

url = "https://events.eage.org/2018/EAGE%20Annual%202018",

}

TY - GEN

T1 - Determination of Reservoir Lithology from Seismic Data by a 2D Hidden Markov Random Field Model

AU - Feng, Runhai

AU - Luthi, Stefan

AU - Gisolf, Dries

AU - Martinius, Allard

N1 - Conference code: 80

PY - 2018

Y1 - 2018

N2 - In this study, geological prior information is incorporated in the classification of reservoir lithologies using the Markov Random Field (MRF) technique. The prediction of hidden lithologies in seismic data is based on measuredobservations such as seismic inversion results, which are associated with the latent categorical variables derived from the distribution of Gaussian assumptions. The Hidden Markov Random Field (HMRF) approach can connectsimilar lithologies laterally (horizontally) while ensure a geologically reasonable stratigraphic (vertical) ordering. It is, therefore, able to exclude randomly appearing lithologies caused by errors in the inversion. In HMRF, the priorinformation consists of a Gibbs distribution function and transition probability matrices. The Gibbs distribution connects similar lithologies and does not need a geological definition derived from non-case-related information.The transition matrices provide preferential transitions between different lithologies and an estimation of these matrices implicitly depends on the depositional environments and juxtaposition rules between different lithologies.

AB - In this study, geological prior information is incorporated in the classification of reservoir lithologies using the Markov Random Field (MRF) technique. The prediction of hidden lithologies in seismic data is based on measuredobservations such as seismic inversion results, which are associated with the latent categorical variables derived from the distribution of Gaussian assumptions. The Hidden Markov Random Field (HMRF) approach can connectsimilar lithologies laterally (horizontally) while ensure a geologically reasonable stratigraphic (vertical) ordering. It is, therefore, able to exclude randomly appearing lithologies caused by errors in the inversion. In HMRF, the priorinformation consists of a Gibbs distribution function and transition probability matrices. The Gibbs distribution connects similar lithologies and does not need a geological definition derived from non-case-related information.The transition matrices provide preferential transitions between different lithologies and an estimation of these matrices implicitly depends on the depositional environments and juxtaposition rules between different lithologies.

U2 - 10.3997/2214-4609.201801076

DO - 10.3997/2214-4609.201801076

M3 - Conference contribution

BT - 80th EAGE Conference and Exhibition 2018, 11-14 June, Copenhagen, Denmark

T2 - 80th EAGE Conference and Exhibition 2018

Y2 - 11 June 2018 through 14 June 2018

ER -

Determination of Reservoir Lithology from Seismic Data by a 2D Hidden Markov Random Field Model

Abstract

Conference

Bibliographical note

Access to Document

Fingerprint

Cite this