Gradient-based workflows for conditioning of process-based geologic models

• US 8,612,195 B2
• Filed: 12/03/2009
• Issued: 12/17/2013
• Est. Priority Date: 03/11/2009
• Status: Active Grant

• Current Assignee: Exxonmobil Upstream Research Company (Exxon Mobil Corporation)
• Original Assignee: Exxonmobil Upstream Research Company (Exxon Mobil Corporation)
• Inventors: Sun, Tao, Ghayour, Kaveh, Imhof, Matthias
• Primary Examiner(s): Shah, Kamini S
• Assistant Examiner(s): PIERRE LOUIS, ANDRE
• Application Number:
  US13/148,642
• Publication Number:
  US 20110315376A1
• Time in Patent Office: 1,475 Days
• Field of Search: 703/9, 703/10, 703/6, 166/250.01, 166/250.17, 166/266, 166/336, 166/369
• US Class Current: 703/10
• CPC Class Codes: G01V 11/00 Prospecting or detecting by methods combining techniques covered by two or more of main groups G01V1/00 - G01V9/00

Abstract

A method for correlating predicted data describing a subsurface region with obtained data describing the subsurface region is provided. Data is obtained describing an initial state of the subsurface region. Data describing a subsequent state of the subsurface region is predicted. A likelihood measure that determines whether the predicted data is within an acceptable range of the obtained data is dynamically and/or interactively updated. The predicted data is compared with the obtained data using the likelihood measure and determining a sensitivity of the predicted data if the predicted data is not within an acceptable range of the obtained data as measured by the likelihood measure. Data describing the initial state of the subsurface region is adjusted based on the sensitivity before performing a subsequent iteration of predicting data describing the subsequent state of the subsurface region. The predicted data is outputted.

Claims

1. A computer implemented method for correlating predicted data describing a subsurface region with known data describing the subsurface region, the method comprising:

obtaining data describing an initial state of the subsurface region;

predicting data describing a subsequent state of the subsurface region;

updating a likelihood measure that determines whether the predicted data is within an acceptable range of the obtained data, the updating being performed at least one of dynamically and interactively;

using a computer to compare the predicted data with the obtained data using the likelihood measure;

using a computer to determine a sensitivity of the predicted data if the predicted data is not within an acceptable range of the obtained data as measured by the likelihood measure;

adjusting the data describing the initial state of the subsurface region based on the sensitivity before performing a subsequent iteration of predicting data describing the subsequent state of the subsurface region, wherein the adjusting is performed based on the likelihood measure having the largest change in sensitivity, such that at any point in parameter space only the likelihood measure that produces the largest change in sensitivity is chosen and drive the process until other measures catch up and the likelihood measure changes subsequently; and

outputting the predicted data based on the adjusting.

2. The method for correlating predicted data recited in claim 1, comprising terminating the method if the predicted data is within the acceptable range of the obtained data.

3. The method for correlating predicted data recited in claim 1, wherein determining the sensitivity of the predicted data is performed via an adjoint method.

4. The method for correlating predicted data recited in claim 1, wherein predicting data describing a subsequent state of the subsurface region is performed via a process-based model.

5. The method for correlating predicted data recited in claim 1, wherein the data describing the initial state of the subsurface region comprises a plurality of estimates describing the initial state of the subsurface region.

6. The method for correlating predicted data recited in claim 1, comprising comparing the predicted data with the obtained data using at least a second likelihood measure if the sensitivity does not indicate an acceptable change in the comparison of the predicted data with the obtained data.

7. The method for correlating predicted data recited in claim 6, wherein adjusting the data describing the initial state of the subsurface region is performed based on the likelihood measure that produces the largest change in sensitivity.

8. The method for correlating predicted data recited in claim 1, wherein the predicted data is outputted by displaying the predicted data.

9. The method for correlating predicted data recited in claim 1, further comprising extracting hydrocarbons from the subsurface region using the predicted data if the predicted data is within an acceptable range of the obtained data.

10. A method for producing hydrocarbons from an oil and/or gas field, the method comprising:

obtaining data describing an initial state of a subsurface region containing at least a part of the oil and/or gas field;

predicting data describing a subsequent state of the subsurface region;

updating a likelihood measure that determines whether the predicted data is within an acceptable range of the obtained data, the updating being performed at least one of dynamically and interactively;

comparing the predicted data with the obtained data using the likelihood measure;

determining a sensitivity of the predicted data if the predicted data is not within an acceptable range of the obtained data as measured by the likelihood measure;

adjusting the data describing the initial state of the subsurface region based on the sensitivity before performing a subsequent iteration of predicting data describing the subsequent state of the subsurface region, wherein the adjusting is performed based the likelihood measure having the largest change in sensitivity, such that at any point in parameter space only the likelihood measure that produces the largest change in sensitivity is chosen and drive the process until other measures catch up and the likelihood measure changes subsequently; and

extracting hydrocarbons from the oil and/or gas field using the predicted data if the predicted data is within an acceptable range of the obtained data.

11. The method for producing hydrocarbons recited in claim 10, wherein determining the sensitivity of the predicted data is performed via an adjoint model.

12. The method for producing hydrocarbons recited in claim 10, wherein predicting data describing a subsequent state of the subsurface region is performed via a process-based model.

13. The method for producing hydrocarbons recited in claim 10, wherein the data describing the initial state of the subsurface region comprises a plurality of estimates describing the initial state of the subsurface region.

14. The method for producing hydrocarbons recited in claim 10, comprising comparing the predicted data with the obtained data using at least a second likelihood measure if the sensitivity does not indicate an acceptable change in the comparison of the predicted data with the obtained data.

15. The method for producing hydrocarbons recited in claim 14, wherein adjusting the data describing the initial state of the subsurface region is performed based on the likelihood measure that produces the largest change in sensitivity.

16. A computer system that is configured to correlate predicted data describing a subsurface region with obtained data describing the subsurface region, the computer system comprising:

processor; and

a non-transitory machine-readable storage medium that stores tangible, machine-readable instructions for execution by the processor, the tangible, machine-readable instructions comprising;

code that is configured to obtain data describing an initial state of the subsurface region;

code that is configured to predict data describing a subsequent state of the subsurface region;

code that is configured to update a likelihood measure that determines whether the predicted data is within an acceptable range of the obtained data, the updating being performed at least one of dynamically and interactively;

code that is configured to compare the predicted data with the obtained data using the likelihood measure;

code that is configured to determine a sensitivity of the predicted data if the predicted data is not within an acceptable range of the obtained data as measured by the likelihood measure; and

code that is configured to adjust the data describing the initial state of the subsurface region based on the sensitivity before performing a subsequent iteration of predicting data describing the subsequent state of the subsurface region, wherein the adjusting is performed based on the likelihood measure having the largest change in sensitivity, such that at any point in parameter space only the likelihood measure that produces the largest change in sensitivity is chosen and drive the process until other measures catch up and the likelihood measure changes subsequently.