公开(公告)号：US20230184977A1

公开(公告)日：2023-06-15

申请号：US17644397

申请日：2021-12-15

Applicant: SAUDI ARABIAN OIL COMPANY

Inventor： Jewoo Yoo ,  Roald Van Borselen

IPC: G01V1/34 ,  G01V1/30 ,  E21B49/00

CPC classification number: G01V1/345 ,  G01V1/303 ,  E21B49/00 ,  G01V2210/1295 ,  G01V2210/1425 ,  G01V2210/6222 ,  G01V2210/74

Abstract: A system and method for determining a set of first breaks of a seismic dataset are disclosed, the method including obtaining the seismic dataset composed of a plurality of seismic traces and a provisional first break for each seismic trace. The method further includes selecting a plurality of proximal picks for each seismic trace, determining a near-offset pick for each seismic trace starting with shortest offset and sequentially selecting traces in order of increasing offset, and determining a far-offset pick for each seismic trace starting with the farthest offset and sequentially selecting traces in order of decreasing offset. The method further includes determining a set of coincident picks based on the near-offset and the far-offset picks for each seismic trace, fitting a curve to the set of coincident picks, and determining the set of first breaks of the seismic dataset from the curve.

公开(公告)号：US20220283329A1

公开(公告)日：2022-09-08

申请号：US17194106

申请日：2021-03-05

Applicant: SAUDI ARABIAN OIL COMPANY

Inventor： Paul M. Zwartjes ,  Rob Hegge ,  Roald Van Borselen

IPC: G01V1/34 ,  G01V1/16 ,  G01V1/02 ,  G01V1/28 ,  G06N3/08

Abstract: A method may include obtaining seismic data regarding a geological region of interest. The seismic data may include various pre-processed gathers. The method may further include obtaining a machine-learning model that is pre-trained to predict migrated seismic data. The method may further include selecting various training gathers based on a portion of the pre-processed gathers, a migration function, and a velocity model. The method may further include generating a trained model using the training gathers, the machine-learning model, and a machine-learning algorithm. The method may further include generating a seismic image of the geological region of interest using the trained model and a remaining portion of the seismic data.

公开(公告)号：US20240319395A1

公开(公告)日：2024-09-26

申请号：US18189906

申请日：2023-03-24

Applicant: SAUDI ARABIAN OIL COMPANY

Inventor： Jewoo Yoo ,  Paul Zwartjes ,  Robertus F. Hegge ,  Roald Van Borselen

IPC: G01V1/36 ,  E21B44/00 ,  E21B49/00 ,  G01V1/28 ,  G01V1/30 ,  G06N20/00

CPC classification number: G01V1/362 ,  E21B44/00 ,  E21B49/00 ,  G01V1/282 ,  G01V1/302 ,  G06N20/00 ,  E21B7/04 ,  E21B2200/22 ,  G01V2210/614

Abstract: Examples of methods and systems are disclosed. The methods may include obtaining, using a seismic processor, a training seismic dataset, comprising an input seismic dataset with a first extent and an output seismic dataset with a second extent, wherein the second extent is greater than the first extent. The methods may also include training, using the seismic processor and the training seismic dataset, a machine-learning (ML) network to predict the output seismic dataset, at least in part, from the input seismic dataset.

公开(公告)号：US12032111B2

公开(公告)日：2024-07-09

申请号：US17194106

申请日：2021-03-05

Applicant: SAUDI ARABIAN OIL COMPANY

Inventor： Paul M. Zwartjes ,  Rob Hegge ,  Roald Van Borselen

IPC: G01V1/34 ,  G01V1/02 ,  G01V1/16 ,  G01V1/28 ,  G06N3/08

CPC classification number: G01V1/345 ,  G01V1/02 ,  G01V1/16 ,  G01V1/282 ,  G06N3/08 ,  G01V2210/121 ,  G01V2210/1295 ,  G01V2210/1425 ,  G01V2210/324 ,  G01V2210/74

Abstract: A method may include obtaining seismic data regarding a geological region of interest. The seismic data may include various pre-processed gathers. The method may further include obtaining a machine-learning model that is pre-trained to predict migrated seismic data. The method may further include selecting various training gathers based on a portion of the pre-processed gathers, a migration function, and a velocity model. The method may further include generating a trained model using the training gathers, the machine-learning model, and a machine-learning algorithm. The method may further include generating a seismic image of the geological region of interest using the trained model and a remaining portion of the seismic data.