Neural network based well log synthesis with reduced usage of radioisotopic sources

• US 7,587,373 B2
• Filed: 11/09/2005
• Issued: 09/08/2009
• Est. Priority Date: 06/24/2005
• Status: Active Grant

• Current Assignee: Halliburton Energy Services Incorporated (Halliburton Company)
• Original Assignee: Halliburton Energy Services Incorporated (Halliburton Company)
• Inventors: Grable, Jeffery L., Quirein, John A., Chen, Dingding, Smith, Harry D. Jr.
• Primary Examiner(s): Vincent; David R
• Assistant Examiner(s): Kennedy; Adrian L
• Application Number:
  US11/270,284
• Publication Number:
  US 20070011115A1
• Time in Patent Office: 1,399 Days
• Field of Search: 706/16, 706/15, 706/20, 702/6, 702/1
• US Class Current: 706/16
• CPC Class Codes: G06N 3/045 Combinations of networks
• CPC Class Codes: G06N 3/086 using evolutionary algorithms, e.g. genetic algorithms or genetic programming

Abstract

Logging systems and methods are disclosed to reduce usage of radioisotopic sources. Some embodiments comprise collecting at least one output log of a training well bore from measurements with a radioisotopic source; collecting at least one input log of the training well bore from measurements by a non-radioisotopic logging tool; training a neural network to predict the output log from the at least one input log; collecting at least one input log of a development well bore from measurements by the non-radioisotopic logging tool; and processing the at least one input log of the development well bore to synthesize at least one output log of the development well bore. The output logs may include formation density and neutron porosity logs.

Claims

1. A computer-implemented method for synthesizing one or more well logs, the method comprising:

collecting at least one output log of a first well bore from measurements with a radioisotopic source;

collecting at least one input log of the first well bore from measurements by a non-radioisotopic logging tool;

training a neural network via a computer to predict the output log from the at least one input log;

collecting at least one input log of a second well bore from measurements by the non-radioisotopic logging tool; and

processing the at least one input log of the second well bore to synthesize at least one output log of the second well bore.

2. The method of claim 1, further comprising:

selecting at least one well completion location based at least in part on the at least one output log of the second well bore; and

perforating well bore casing at the at least one well completion location.

3. The method of claim 2, wherein the processing operation provides a density log and a neutron porosity log of the second well bore, and wherein the selecting operation comprises identifying cross-over intervals of the density and neutron porosity logs.

4. The method of claim 1, wherein the first well bore is a training well bore and the second well bore is a development well bore.

5. The method of claim 1, wherein the at least one input log collected from the first and second well bores comprises an open hole log and a cased hole log.

6. The method of claim 5, wherein the cased hole log is determined from measurements by a pulsed neutron capture logging tool.

7. The method of claim 5, wherein the cased hole log comprises a log in a set consisting of resistivity logs, sonic logs, gamma ray spectroscopy logs, production logs, and combinations thereof.

8. The method of claim 5, wherein the open hole log comprises a log in a set consisting of resistivity logs, sonic logs, nuclear magnetic resonance logs, gamma ray spectroscopy logs, spontaneous potential logs, and combinations thereof.

9. The method of claim 1, wherein the at least one output log comprises at least one of a log of neutron porosity and a log of formation density.

10. The method of claim 1, wherein the at least one input log comprises at least one of a log indicative of formation resistivity and a sonic log.

11. The method of claim 1, wherein the at least one input log comprises a log in a set consisting of natural gamma ray logs, induced gamma ray logs, sonic logs, nuclear magnetic resonance logs, and combinations thereof.

12. The method of claim 1, wherein the non-radioisotopic logging tool comprises a pulsed neutron source.

13. The method of claim 12, wherein the non-radioisotopic logging tool is a cased hole logging tool.

14. The method of claim 12, wherein the non-radioisotopic logging tool is a logging while drilling tool.

15. An information storage medium that, when placed in operable relation to a computer, provides the computer with well log synthesis software comprising:

a training process that generates a neural network for synthesizing one or more radioisotopic logging tool logs; and

a transform process that applies the neural network to convert an input set of non-radioisotopic logging tool logs into an output set of one or more predicted radioisotopic logging tool logs in the form of an observable table or chart.

16. The medium of claim 15, wherein the neural network is part of a neural network ensemble, wherein the training process generates a set of neural networks having diversity in inputs and in complexity, and wherein the software further comprises:

a selection process that identifies a combination of neural networks from the set of neural networks to form the neural network ensemble.

17. The medium of claim 15, wherein the output set comprises at least one of a neutron porosity log and a density log.

18. The medium of claim 15, wherein the input set comprises a resistivity log and logs by a pulsed neutron logging tool.

19. The medium of claim 18, wherein the pulsed neutron logging tool is a cased hole logging tool.

20. The medium of claim 16, wherein the selection process identifies a combination of neural networks from the set having a desirable fitness measure, said fitness measure being based at least in part on a measure of negative correlation for each neural network in the combination.

21. A system for synthesizing one or more well logs, the system comprising:

a non-radioisotopic logging tool to obtain measurements in a well bore; and

a computer to receive the measurements from the non-radioisotopic logging tool, wherein the computer synthesizes at least one of a density log and a neutron porosity log for display to a user using the measurements from the non-radioisotopic logging tool.

22. The system of claim 21, wherein the computer implements a neural network to convert multiple non-radioisotopic logs into at least one of a density log and a neutron porosity log.

23. The system of claim 22, further comprising a perforation tool to perforate casing in the well bore at positions selected based at least in part on at least one of a density log and a neutron porosity log synthesized by the computer.

24. A computer implemented well logging method that comprises:

obtaining a first log of a well bore by at least one of a resistivity tool and a sonic tool;

obtaining a pulsed neutron tool log of the well bore; and

using a computer to combine the first log with the pulsed neutron tool log to synthesize and display a radioisotopic source tool log of the well bore.

25. The logging method of claim 24, wherein the combining operation employs a neural network ensemble to synthesize the radioisotopic source tool log.

26. The logging method of claim 24, wherein the radioisotopic source tool log is a neutron porosity log.

27. The logging method of claim 24, wherein the radioisotopic source tool log is a density log.

28. The logging method of claim 24, wherein the first log is a resistivity log.