Method and apparatus for determining potential abrasivity in a wellbore

• US 6,386,297 B1
• Filed: 02/23/2000
• Issued: 05/14/2002
• Est. Priority Date: 02/24/1999
• Status: Expired due to Term

• Current Assignee: Baker Hughes Incorporated (Baker Hughes Company)
• Original Assignee: Baker Hughes Incorporated (Baker Hughes Company)
• Inventors: Cooley, Craig Hodges, Curry, David Alexander, Ledgerwood, Leroy William III
• Primary Examiner(s): Bagnell, David
• Assistant Examiner(s): BATES, ZAKIYA W
• Application Number:
  US09/511,617
• Time in Patent Office: 811 Days
• Field of Search: 702/9, 702/6, 702/11, 175/39, 175/40, 175/50, 731/520.2, 731/520.3, 731/524.3, 731/524.4, 731/524.5
• US Class Current: 175/39
• CPC Class Codes: E21B 2200/22 Fuzzy logic, artificial intelligence, neural networks or the like
• CPC Class Codes: E21B 44/00 Automatic control systems specially adapted for drilling operations, i.e. self-operating systems which function to carry out or modify a drilling operation without intervention of a human operator, e.g. computer-controlled drilling systems; Systems specially adapted for monitoring a plurality of drilling variables or conditions
• CPC Class Codes: E21B 49/003 by analysing drilling variables or conditions E21B49/005 takes precedence; systems specially adapted for monitoring a plurality of drilling variables or conditions E21B44/00
• CPC Class Codes: E21C 39/00 Devices for testing in situ the hardness or other properties of minerals, e.g. for giving information as to the selection of suitable mining tools

Abstract

A method is provided for generating an indicator of potential bit abrasion in a particular wellbore. Forensic wellbore data is obtained from at least one previously drilled wellbore which is determined to be comparable to the particular target wellbore. Typically, the comparable wellbore comprises an “offset” wellbore which is proximate the target wellbore, and which has similar geologic features. An inference engine computer program is provided which consists of executable program instructions. It is adapted to utilize a plurality of wellbore parameters, including the forensic wellbore data. The inference engine includes at least one rule matrix which defines a plurality of fuzzy sets. These fuzzy sets establish correspondence between the plurality of wellbore parameters and the indictor of potential bit abrasion. The inference engine computer program is loaded onto a data processing system. At least the forensic wellbore data is supplied as an input to the inference engine computer program. The data processing system is utilized to execute the program instructions of the inference engine computer program. This causes the application of the inputs to the inference engine computer program. The inference engine computer program produces as an output an indication of potential bit abrasion in the particular target wellbore.

Claims

1. A method of providing an indicator of potential for abrasive wear of a drill bit in a particular wellbore, comprising:

(a) obtaining forensic wellbore data from at least one previously drilled wellbore which is determined to be comparable to said particular wellbore;

(b) providing an inference engine computer program consisting of executable program instructions which includes a rule matrix, and which is adapted to utilize a plurality of wellbore parameters, including said forensic wellbore data;

(c) loading said inference engine computer program on to a data processing system;

(d) supplying as an input to said inference engine computer program said forensic wellbore data and at least one other of said plurality of wellbore parameters; and

(e) utilizing said data processing system to execute program instructions of said inference engine computer program to apply said input to said inference engine computer program, with at least a portion of said input being supplied to said rule matrix, and to produce as an output an indicator of potential for abrasive wear of a drill bit in said particular wellbore.

2. The method of providing an indicator of potential for abrasive wear of a drill bit according to claim 1, further comprising:

(f) obtaining empirical wellbore data from said particular wellbore during drilling operations; and

(g) additionally supplying said empirical wellbore data as an input to said inference engine computer program.

3. The method of providing an indicator of potential for abrasive wear of a drill bit according to claim 2, further comprising:

(h) wherein said indicator of potential for abrasive wear of a drill bit is generated repeatedly during drilling operations; and

(i) wherein said indicator provides an indication of potential for abrasive wear of a drill bit before abrasive wear occurs.

4. The method of providing an indicator of potential for abrasive wear of a drill bit according to claim 3 further comprising:

(j) altering at least one drilling condition in response to said indicator in order to diminish the probability of abrasive wear of a drill bit occurring.

5. The method of providing an indicator of potential for abrasive wear of a drill bit according to claim 4, wherein said at least one drilling condition includes at least one of:

(1) mud type;

(2) bit type;

(3) bit hydraulic;

(4) rotary speed; and

(5) weight on bit.

6. The method of providing an indicator of potential for abrasive wear of a drill bit according to claim 1, wherein said forensic wellbore data includes at least log data which indicates sand content.

7. The method of providing an indicator of potential for abrasive wear of a drill bit according to claim 1, wherein said forensic wellbore data includes at least one of:

(a) data which relates to sand content; and

(b) data which relates to formation strength.

8. The method of providing an indicator of potential for abrasive wear of a drill bit according to claim 1, wherein said forensic wellbore data includes both of:

(a) data which relates to sand content; and

(b) data which relates to formation strength.

9. The method of providing an indicator of potential for abrasive wear of a drill bit according to claim 1, wherein said forensic wellbore data includes both of:

(a) log data which provides an indication of likely and content at particular depths; and

(b) log data which provides an indication of likely formation strength at particular depths.

10. The method of providing an indicator of potential for abrasive wear of a drill bit according to claim 9, wherein said data which provides an indication of likely formation strength at a particular depth comprises data which provides an indication of likely unconfined compressive strength of formations at particular depths.

11. The method of providing an indicator of potential for abrasive wear of a drill bit according to claim 1, wherein said indicator of potential for abrasive wear of a drill bit comprises a numerical indicator of potential for abrasive wear of a drill bit.

12. The method of providing an indicator of potential for abrasive wear of a drill bit according to claim 11, wherein said numerical indicator comprises a numerical value in the range between an upper boundary value and a lower boundary value.

13. A method of providing an indicator of potential for abrasive wear of a drill bit comprising:

(a) obtaining forensic wellbore data from at least one previously drilled wellbore which is determined to be comparable to said particular wellbore;

(b) providing a rule matrix computer program consisting of executable program instructions, and adapted to utilize a plurality of wellbore parameters, including said forensic wellbore data;

(c) loading said rule matrix computer program on to a data processing system;

(d) supplying as an input to said rule matrix computer program said forensic wellbore data and at least one other of said plurality of wellbore parameters; and

(e) utilizing said data processing system to execute program instructions of said rule matrix computer program to apply said input to said rule matrix computer program and to produce as an output an indicator of potential for abrasive wear of a drill bit in said particular wellbore.

14. The method of providing an indicator of potential for abrasive wear of a drill bit according to claim 13, wherein said rule matrix establishes correspondence between said plurality of wellbore parameters and said indicator.

15. The method of providing an indicator of potential for abrasive wear of a drill bit according to claim 13, wherein said rule matrix defines a plurality of fuzzy sets.

16. The method of providing an indicator of potential for abrasive wear of a drill bit according to claim 15, wherein said each of said fuzzy sets is identified to at least one corresponding membership function.

17. The method of providing an indicator of potential for abrasive wear of a drill bit according to claim 16, wherein collectively said fuzzy sets define a responsive surface which relates said plurality of wellbore parameters to said indicator of potential for abrasive wear of a drill bit.

18. The method of providing an indicator of potential for abrasive wear of a drill bit according to claim 13:

wherein said rule matrix is composed of a plurality of qualitative descriptions for each said plurality of wellbore parameters; and

wherein each of said plurality of qualitative descriptions has a numerical value associated therewith;

wherein said inference engine computer program combines the effects of said plurality of wellbore parameters through said indicator of potential for abrasive wear of a drill bit in said particular wellbore.

19. The method of providing an indicator of potential for abrasive wear of a drill bit according to claim 18, wherein said numerical value which is associated with each of said plurality of qualitative descriptions is adjustable in order to allow modification of said inference engine.

20. The method of providing an indicator of potential for abrasive wear of a drill bit according to claim 19, wherein said numerical value which is associated with each of said plurality of qualitative descriptions is adjustable in order to allow modification of said inference engine, in order to allow calibration of a response surface which defines a universe of possible values for said indicator of potential for abrasive wear of a drill bit.

21. An apparatus for providing an indicator of potential for abrasive wear of a drill bit in a particular wellbore, comprising:

(a) a data processing system adapted for execution of program instructions;

(b) an inference engine computer program composed of executable program instructions, and including;

(1) an inference engine program module which includes a rule matrix which establishes correspondence between a plurality of wellbore parameters and an indicator of potential for abrasive wear of a drill bit and which combines the effects of at least two distinct wellbore parameters utilizing said rule matrix in order to produce as an output an indicator of potential for abrasive wear of a drill bit; and

(2) an input program module for receiving data corresponding to at least two of said plurality of wellbore parameters and supplying said data to said inference engine program module.

22. The apparatus for providing an indicator of potential for abrasive wear of a drill bit according to claim 21, wherein said inference engine further includes:

(3) an output program which provides in a human-readable format said indicator of potential for abrasive wear of a drill bit.

23. The apparatus for providing an indicator of potential for abrasive wear of a drill bit according to claim 22, wherein said indicator for potential abrasive wear of a drill bit is presented in human-readable format of a single numeric value from a range of available numeric values between an upper numeric limit and a lower numeric limit.

24. The apparatus for providing an indicator of potential for abrasive wear of a drill bit according to claim 21, wherein said inference engine computer program further includes:

(3) program instructions for recursively computing said indicator of potential for abrasive wear of a drill bit during drilling operations.

25. The apparatus for providing an indicator of potential for abrasive wear of a drill bit according to claim 21, wherein said input program module of said inference engine computer program comprises:

(2) an input program module for receiving data corresponding to at least two following distinct wellbore parameters;

(a) forensic wellbore data from at least one previously drilled wellbore which is determined to be comparable to a subject wellbore; and

(b) empirical wellbore data related to said subject wellbore; and

for supplying said data to said inference engine program module.

26. The apparatus for providing an indicator of potential for abrasive wear of a drill bit according to claim 21, wherein said inference engine program module includes a rule matrix which establishes correspondence between the two following distinct wellbore parameters:

(a) forensic wellbore data from at least one previously drilled wellbore which is determined to be comparable to a subject wellbore; and

(b) other wellbore data related to said subject wellbore; and

and an indicator of potential for abrasive wear of a drill bit, by combining the effects of the forensic wellbore data and the other wellbore data utilizing said rule matrix in order to produce as an output and indicator of potential for abrasive wear of a drill bit of said subject wellbore.

27. A method of drilling a wellbore, comprising:

(a) obtaining forensic wellbore data from at least one previously drilled wellbore which is determined to be comparable to a target wellbore;

(b) providing an inference engine computer program consisting of executable program instructions which includes a rule matrix, and which is adapted to utilize a plurality of wellbore parameters, including said forensic wellbore data;

(c) loading said inference engine computer program on to a data processing system;

(d) supplying as an input to said inference engine computer program said forensic wellbore data and at least one other of said plurality of wellbore parameters;

(e) utilizing said data processing system to execute program instructions of said inference engine computer program to apply said input to said inference engine computer program, with at least a portion of said input being supplied to said rule matrix, and to produce as an output an indicator of potential for abrasive wear of a drill bit in said target wellbore;

(f) providing a plurality of available rock bits for use in drilling particular portions of said target wellbore, with particular ones of said plurality of available rock bits having different resistance to abrasive wear which corresponds generally to said indicator of potential for abrasive wear of a drill bit;

(g) selecting a particular one of said plurality of available rock bits based at least in part upon said potential for abrasive wear of a drill bit as predicted by said indication of potential for abrasive wear of a drill bit;

(h) connecting said particular one of said plurality of available rock bits to a drilling string; and

(i) performing drilling operations.

28. The method drilling a wellbore according to claim 27, further comprising:

(j) obtaining empirical wellbore data from said target wellbore during drilling operations; and

(k) additionally supplying said empirical wellbore data as an input to said inference engine computer program.

29. The method of drilling a wellbore according to claim 27, wherein said forensic wellbore data includes at least log data relating to formation types.

30. The method of drilling a wellbore according to claim 27, wherein said forensic wellbore data includes at least one of:

(a) data which relates to sand content; and

(b) data which relates to formation strength.

31. The method of drilling a wellbore according to claim 27, wherein said forensic wellbore data includes both of:

(a) data which relates to sand content; and

(b) data which relates to formation strength.

32. The method of drilling a wellbore according to claim 27, wherein said forensic wellbore data includes both of:

(a) log data which provides an indication of likely sand content at particular depths; and

(b) log data which provides an indication of likely formation strength at particular depths.

33. The method of drilling a wellbore according to claim 32, wherein said data which provides an indication of likely formation strength at a particular depth comprises data which provides an indication of likely unconfined compressive strength of formations at particular depths.

34. The method of drilling a wellbore according to claim 27, wherein said indicator of potential for abrasive wear of a drill bit comprises a numerical indicator of potential for abrasive wear of a drill bit.

35. The method of drilling a wellbore according to claim 34, wherein said numerical indicator comprises a numerical value in the range between an upper boundary value and a lower boundary value.

36. The method of drilling a wellbore according to claim 27, further comprising:

(l) wherein said indicator of potential for abrasive wear of a drill bit is generated repeatedly during drilling operations; and

(m) wherein said indicator provides an indication of potential for abrasive wear of a drill bit before drilling occurs.

37. The method of drilling a wellbore according to claim 36, further comprising:

(n) altering at least one drilling condition in response to said indicator in order to diminish the probability of abrasive wear of a drill bit occurring.

38. The method of drilling a wellbore according to claim 37, wherein said at least one drilling condition includes at least one of:

(1) mud type;

(2) bit type;

(3) bit hydraulics;

(4) rotary speed; and

(5) weight on bit.

39. A method of drilling a wellbore, comprising:

(a) obtaining forensic wellbore data from at least one previously drilled wellbore which is determined to be comparable to a target wellbore;

(b) providing an inference engine computer program consisting of executable program instructions, and adapted to utilize a plurality of wellbore parameters, including said forensic wellbore data said inference engine comprising a rule matrix;

(c) loading said inference engine computer program on to a data processing system;

(d) supplying as an input to said inference engine computer program said forensic wellbore data and at least one other of said plurality of wellbore parameters;

(e) utilizing said data processing system to execute program instructions of said inference engine computer program to apply said input to said inference engine computer program and to produce as an output an indicator of potential for abrasive wear of a drill bit in said target wellbore;

(f) providing a plurality of available rock bits for use in drilling particular portions of said target wellbore, with particular ones of said plurality of available rock bits having different resistance to abrasive wear which corresponds generally to said indicator of potential for abrasive wear of a drill bit;

(g) selecting a particular one of said plurality of available rock bits based at least in part upon said potential for abrasive wear of a drill bit as predicted by said indication of potential for abrasive wear of a drill bit;

(h) connecting said particular one of said plurality of available rock bits to a drilling string; and

(i) performing drilling operations.

40. The method of drilling a wellbore according to claim 39, wherein said rule matrix establishes correspondence between said plurality of wellbore parameters and said indicator.

41. The method of drilling a wellbore according to claim 39, wherein said rule matrix defines a plurality of fuzzy sets.

42. The method of drilling a wellbore according to claim 41, wherein said each of said fuzzy sets is identified to at least one corresponding membership function.

43. The method of drilling a wellbore according to claim 42, wherein collectively said fuzzy sets define a responsive surface which relates said plurality of wellbore parameters to said indicator of potential for abrasive wear of a drill bit.

44. The method of drilling a wellbore according to claim 39:

wherein said rule matrix is composed of a plurality of qualitative descriptions for each said plurality of wellbore parameters; and

wherein each of said plurality of qualitative descriptions has a numerical value associated therewith; and

wherein said inference engine computer program combines the effects of said plurality of wellbore parameters through said indicator of potential for abrasive wear of a drill bit in said target wellbore.

45. The method of drilling a wellbore according to claim 44, wherein said numerical value which is associated with each of said plurality of qualitative descriptions is adjustable in order to allow modification of said inference engine.

46. The method of drilling a wellbore according to claim 45, wherein said numerical value which is associated with each of said plurality of qualitative descriptions is adjustable in order to allow modification of said inference engine in order to allow calibration of a response surface which defines a universe of possible values for said indicator of potential for abrasive wear of a drill bit.