Analyzing seismic data

• US 10,422,900 B2
• Filed: 08/30/2013
• Issued: 09/24/2019
• Est. Priority Date: 11/02/2012
• Status: Active Grant

• Current Assignee: Exxonmobil Upstream Research Company (Exxon Mobil Corporation)
• Original Assignee: Exxonmobil Upstream Research Company (Exxon Mobil Corporation)
• Inventors: Imhof, Matthias, Dimitrov, Pavel, Gibson, Laurie D.
• Primary Examiner(s): Ward, Eric A.
• Assistant Examiner(s): Ashbahian, Eric K
• Application Number:
  US14/434,726
• Publication Number:
  US 20150293247A1
• Time in Patent Office: 2,216 Days
• CPC Class Codes: G01V 1/301 for determining seismic cross-sections or geostructures
• CPC Class Codes: G01V 2210/64 Geostructures, e.g. in 3D data cubes
• CPC Class Codes: G01V 2210/642 Faults
• CPC Class Codes: G01V 2210/74 Visualisation of seismic data

Abstract

A system and methods for analyzing seismic data are provided herein. The method includes identifying, via a computing device, a representation of a seismic data set (1802) and determining a number of feature descriptors corresponding to each of a number of aggregates within the representation (1804). The method also includes identifying a query relating to the representation and one or more vocabulary definitions relating to the query (1806), analyzing the representation to compute a likelihood that each of the aggregates satisfies the query (1808), and returning a result of the query (1810).

Claims

1. A method for analyzing seismic data, comprising:

identifying, via a computing device, a representation of a seismic data set, said representation comprising a plurality of aggregates;

partitioning the plurality of aggregates within the representation according to a plurality of levels, wherein each level of the plurality of levels comprises one or more aggregates of the same specified resolution for that level;

for the one or more aggregates of each level, determining a feature descriptor corresponding to each such aggregate within each such level;

formulating a query relating to the representation and one or more vocabulary definitions relating to the query to the feature descriptors;

computing a likelihood that each of the plurality of aggregates satisfies the query with a probabilistic relationship between the vocabulary and feature descriptors;

returning a result of the query;

based at least in part on the returned result, identifying one or more prospects in a subsurface area represented by the seismic data; and

causing a well to be drilled in the subsurface area based on the one or more identified prospects.

2. The method of claim 1, wherein each aggregate comprises one or more cells, and each cell comprises one or more voxels representing specific seismic data.

3. The method of claim 1, wherein determining a feature descriptor corresponding to each aggregate within each level comprises:

analyzing a lowest level of the representation to determine feature descriptors corresponding to each of the one or more aggregates within the lowest level; and

upscaling the feature descriptors for the lowest level to identify feature descriptors corresponding to each of the one or more aggregates within a next lowest level of the representation.

4. The method of claim 3, wherein each of the one or more aggregates within the lowest level of the representation comprises one cell.

5. The method of claim 1, wherein computing the likelihood that each of the plurality of aggregates satisfies the query comprises:

comparing feature descriptors for a first aggregate to feature descriptors for a plurality of surrounding aggregates; and

determining a likelihood that the first aggregate satisfies the query based on the comparison.

6. The method of claim 1, comprising:

validating the result of the query;

if the result cannot be validated, determining modifications to the feature descriptors, definitions of likelihood that each of the plurality of aggregates satisfies the query, the query, or the one or more vocabulary definitions, or any combinations thereof; and

implementing the determined modifications.

7. The method of claim 1, comprising identifying the query and the one or more vocabulary definitions in response to input from a user of the computing device.

8. The method of claim 1, wherein the prospects comprise geological features or geophysical features, or both.

9. The method of claim 1, wherein determining a feature descriptor corresponding to each said aggregate comprises determining at least one of a mean value, a variance, a texture, or a gradient, or any combinations thereof, for the seismic data represented by the aggregate.

10. The method of claim 1, wherein the one or more vocabulary definitions comprises definitions of words within the query, definitions of rules relating to the query, or definitions of maps between the plurality of feature descriptors and corresponding probabilities, or any combinations thereof.

11. The method of claim 1, wherein the feature descriptors comprise numerical values, and wherein the method includes determining geological features or geophysical features, or both, of a subsurface represented by the seismic data based on the numerical values.

12. The method of claim 1, wherein the feature descriptor relates to one of said prospects of the subsurface represented by the seismic data;

and wherein the method further comprises

computing the likelihood that each of the plurality of aggregates corresponds to a region of the subsurface comprising the desired prospect.

13. The method of claim 1, wherein the one or more prospects comprise channels, salt bodies, faults, or geophysical abnormalities, or any combinations thereof.

14. The method of claim 1, wherein the one or more prospects comprise regions of the subsurface comprising hydrocarbons.

15. The method of claim 1, wherein the representation comprises a recursive geometric representation, and wherein the plurality of aggregates within the recursive geometric representation is determined according to a recursive geometric decomposition of the seismic data set.

16. The method of claim 1, wherein each of the plurality of feature descriptors comprises a mean, a residue, a variance, a gradient magnitude, a gradient direction, a histogram, a local texture, or a second moment, or any combinations thereof.

17. A method for identifying or characterizing prospects within a subsurface represented by a seismic data set, comprising:

identifying, via a computing device, a recursive geometric representation of a seismic data set, said recursive geometric representation comprising a plurality of aggregates;

partitioning the plurality of aggregates within the recursive geometric representation according to a plurality of levels, wherein each level of the plurality of levels comprises one or more aggregates, each aggregate of the same specified resolution for that level;

for the one or more aggregates of each level, determining a feature descriptor corresponding to each such aggregate within each such level;

formulating a query relating to an identification or a characterization, or both, of the prospects within the subsurface;

identifying one or more vocabulary definitions relating the query to the feature descriptors;

identifying the prospects and/or characterizing the prospects based at least in part upon analysis of the recursive geometric representation and the query; and

causing a well to be drilled based on the one or more identified and/or characterized prospects.

18. The method of claim 17, comprising:

determining modifications to the feature descriptors, the query, or the one or more vocabulary definitions, or any combinations thereof, that result in an improvement in the identification of the prospects or the characterization of the prospects, or both; and

implementing the determined modifications.

19. The method of claim 17, wherein each of the plurality of aggregates comprises one cell.

20. The method of claim 19, wherein the recursive geometric representation comprises a plurality of two-dimensional layers of cells stacked in a hierarchical manner.

21. The method of claim 20, wherein the recursive geometric representation is recursive in at least one dimension.

22. The method of claim 17, wherein analyzing the recursive geometric representation to identify or characterize the prospects comprises determining a likelihood that each of the plurality of aggregates within the recursive geometric representation corresponds to a region of the subsurface comprising one or more prospects.

23. The method of claim 17, wherein the prospects comprise regions of the subsurface comprising hydrocarbons.

24. The method of claim 17, wherein the recursive geometric representation comprises a recursive hexagonal representation.

25. The method of claim 17, wherein determining a feature descriptor corresponding to each aggregate within each level comprises:

analyzing a lowest level of the recursive geometric representation to determine feature descriptors corresponding to each of the one or more aggregates within the lowest level; and

upscaling the feature descriptors for the lowest level to identify feature descriptors corresponding to each of the one or more aggregates within a next lowest level of the representation.