Method for making cement impedance measurements with characterized transducer

• US 5,146,432 A
• Filed: 08/05/1991
• Issued: 09/08/1992
• Est. Priority Date: 08/05/1991
• Status: Expired due to Term

• Current Assignee: Schlumberger Technology Corporation (Schlumberger NV)
• Original Assignee: Schlumberger Technology Corporation (Schlumberger NV)
• Inventors: Ayter, Sevig, Kimball, Christopher V.
• Primary Examiner(s): Pihulic, Daniel T.
• Application Number:
  US07/740,530
• Time in Patent Office: 400 Days
• Field of Search: 367/13, 367/35, 73/1 DV, 181/105
• US Class Current: 367/13
• CPC Class Codes: G01N 29/09 by measuring mechanical or acoustic impedance
• CPC Class Codes: G01N 29/30 Arrangements for calibrating or comparing, e.g. with standard objects
• CPC Class Codes: G01N 33/383 Concrete, cement

Abstract

Methods for characterizing acoustic transducers and using a characterized acoustic transducer for measuring cement impedance of a cased well are disclosed. The method for characterizing the acoustic transducer generally comprises: arranging the acoustic transducer a predetermined distance from a calibration target of known thickness; transmitting a pulse from the acoustic transducer through a known medium and toward the calibration target and measuring a return signal received therefrom; and fitting, via minimization, a return signal calculation generated by a model of the return signal which has a limited number of parameters, to the received return signal to determine a value for at least one of the parameters and thereby characterize the transducer. The preferred parameters for characterization are the transducer radius and stand-off, such that the transducer may be characterized as having an effective stand-off and an effective radius. The method for measuring cement impedance behind the casing in a well utilizes the method for characterizing the transducer, where the calibration target is preferbly of the same radius and thickness as the casing of the well. Then, downhole, a pulse is transmitted toward the casing in the well and the return signal is measured. Using the effective stand-off and effective radius previously determined as fixed values, the model of the return signal, which utilizes reflective coefficients which, inter alia, are a function of the cement impedance, is fit to the actual return signal to determine an indication of the cement impedance.

Claims

1. A method of characterizing an acoustic transducer, comprising:

a) arranging said acoustic transducer a predetermined distance from a calibration target, wherein said calibration target is of known thickness, and said calibration target is in contact with matter of known impedance along its inner surface and outer surface;

b) transmitting a pulse from said acoustic transducer toward said calibration target and measuring a return signal received therefrom; and

c) fitting a return signal calculation generated by a model of said return signal to said return signal, said model having a limited number of parameters, to determine a value for at least one of said parameters and thereby characterizing said transducer, by substantially minimizing the difference between said return signal calculation and said measured return signal.

2. A method according to claim 1, wherein:

said limited number of parameters include at least one of radius and stand-off, and said acoustic transducer is characterized with a value for at least one of the effective radius and effective stand-off of said acoustic transducer as a function of said thickness of said calibration target.

3. A method according to claim 1, wherein:

said calibration target comprises one of a flat plate, a cylindrical tube, and a segment of a cylindrical tube.

4. A method according to claim 1, wherein:

said calculated return signal (V(f)) is defined according to ##EQU11## wherein W is the planewave kernel of a transducer parameterized by said limited number of parameters,

θ

is the incidence angle of planewaves which can range over different angles,

f is frequency which can range over a frequency spectrum,

L are layers related to said calibration target, said matter along the inner surface of said target, and said matter along the outer surface of said target, and

R(f,θ

|L) are reflection coefficients for the layers L, at frequency f, and incidence angle θ

.

5. A method according to claim 4, wherein:

said model of said return signal and said return signal are limited to a narrow frequency band substantially centered on a resonant frequency of said calibration target such that said planewave kernel is approximated to be substantially independent of frequency in said narrow frequency band.

6. A method according to claim 5, wherein:

said difference between said return signal calculation and said measured return signal is determined in said narrow frequency band.

7. A method according to claim 4, wherein:

said planewave kernel W is a planewave kernel for a Gaussian apodized disk transducer parameterized by its radius and standoff and said value includes values for at least one of the effective radius and effective stand-off of said acoustic transducer as a function of said thickness of said calibration target.

8. A method according to claim 7, wherein:

said model of said return signal and said return signal are limited to a narrow frequency band substantially centered on a resonant frequency of said calibration target such that

said planewave kernel is approximated to be substantially independent of frequency in said narrow frequency band.

9. A method according to claim 2, wherein:

said model of said return signal defines a calculated return signal V(f) according to ##EQU12## wherein W is the planewave kernel of a transducer parameterized by said limited number of parameters,

θ

is the incidence angle of the planewave which can range over different angles,

f is a frequency which can range over a frequency spectrum,

L are layers related to said calibration target, said matter along the inner surface of said target, and said matter along the outer surface of said target, and

R(f,θ

|L) are complex planar reflection coefficients for the layers L, at frequency f, and incidence angle θ

.

10. A method according to claim 9, wherein:

said model of said return signal and said return signal are limited to a narrow frequency band substantially centered on a resonant frequency of said calibration target such that said planewave kernel is approximated to be substantially independent of frequency in said narrow frequency band.

11. A method according to claim 6, wherein:

said difference between said return signal calculation and said measured return signal is determined utilizing a mean-squared error calculation.

12. A method according to claim 4, wherein:

said planewave kernel W is a planewave kernel for a Gaussian apodized disk transducer parameterized by its radius and standoff.

13. A method according to claim 12, wherein:

said model of said return signal and said return signal are limited to a narrow frequency band substantially centered on a resonant frequency of said calibration target such that

said planewave kernel is approximated to be substantially independent of frequency in said narrow frequency band.

14. A method according to claim 1, wherein:

said difference between said return signal calculation and said measured return signal is determined utilizing a means-squared error calculation.

15. A method according to claim 14, wherein:

in making said mean-squared error calculation, the error is computed over one or both of,

(i) a range of thicknesses in steps about the assumed thickness of said calibration target, and

(ii) a range of time shifts in steps relative to said return signal,

to substantially minimize the error.

16. A method according to claim 11, wherein:

in making said mean-squared error calculation, the error is computed over one or both of,

(i) a range of thicknesses in steps about the assumed thickness of said calibration target, and

(ii) a range of time shifts in steps relative to said return signal,

to substantially minimize the error.

17. A method according to claim 1, further comprising:

in fitting said return calculation to said return signal and substantially minimizing said difference between said return signal calculation and said measured return signal, utilizing said difference as a quality indicator of the characterization of said acoustic transducer.

18. A method according to claim 17, wherein:

said difference between said return signal calculation and said measured return signal is determined utilizing a mean-squared error calculation, and said mean-squared error calculation of a best fit is said quality indicator.

19. A method according to claim 15, further comprising:

in fitting said return signal calculation to said return signal and minimizing said difference between said return signal calculation and said measured return signal in said narrow frequency band, utilizing said difference as a quality indicator of the characterization of said acoustic transducer.

20. A method according to claim 19, wherein:

said difference between said return signal calculation and said measured return signal is determined utilizing a mean-squared error calculation, and said mean-squared error calculation of a best fit is said quality indicator.

45. A method according to claim 4, wherein:

said return signal calculations and additional return signal calculations (V(f)) are defined according to ##EQU16## wherein W is the planewave kernel of a transducer parameterized by said limited number of parameters,

θ

is the incidence angle of the planewave which can range over different angles,

f is a frequency which can range over a frequency spectrum,

L are layers related to said calibration target and said matter along the inner surface of said target, and said matter along the outer surface of said target for step c), and are relate to said target in said well and said mud for step f),

R(f,θ

|L) are reflection coefficients for the layers L, at frequency f, and incidence angle θ

.

46. A method according to claim 45, wherein:

said model of said return signal and said return signal are limited in a narrow frequency band substantially centered on a resonant frequency of said calibration target such that

said planewave kernel is approximated to be substantially independent of frequency in said narrow frequency band.

47. A method according to claim 45, wherein:

said planewave kernel W is a planewave kernel for a Gaussian apodized disk transducer parameterized by its radius and standoff and said value include values for at least one of the effective radius and effective stand-off of said acoustic transducer as a function of said thickness of said calibration target.

48. A method according to claim 47, wherein:

said model of said return signal and said return signal are limited in a narrow frequency band substantially centered on a resonant frequency of said calibration target such that

said planewave kernel is approximated to be substantially independent of frequency in said narrow frequency band.

21. A method of measuring cement impedance of a cased well with a tool having an acoustic transducer, the casing of the well being of substantially known diameter, said method comprising:

a) characterizing said acoustic transducer by testing said acoustic transducer relative to a calibration target having a diameter substantially equal to the diameter of said casing of said well, and characterizing said acoustic transducer as having at least one of an effective radius (a_e), and an effective stand-off (S_e);

b) transmitting a pulse from said acoustic transducer toward said casing in said well and measuring a return signal received therefrom; and

c) fitting a return signal calculation generated by a model of said return signal to said return signal to determine an indication of said cement impedance, by substantially minimizing the difference between said return signal calculation and said return signal, wherein said model of said return signal utilizes at least one of said effective radius and said effective stand-off of said acoustic transducer as determined during said characterizing step, and wherein said model of said return signal includes reflection coefficients which are a function of said cement impedance.

22. A method according to claim 21, wherein:

said tool includes said calibration target, and said characterizing step is accomplished downhole in said cased well.

23. A method according to claim 22, wherein:

said step of characterizing comprises

1) arranging said acoustic transducer a predetermined distance from said calibration target, wherein said calibration target is of known thickness,

2) transmitting a pulse from said acoustic transducer through borehole mud toward said calibration target and measuring the calibration return signal therefrom, and

3) fitting a return signal calculation generated by said model of said return signal to said calibration return signal by substantially minimizing the difference between said return signal and said calibration return signal, to determine at least one of said effective radius and said effective stand-off of said acoustic transducer as a function of said thickness of said calibration target.

24. A method according to claim 23, further comprising:

precharacterizing said tool uphole by testing said acoustic transducer uphole relative to an uphole calibration target, and precharacterizing said acoustic transducer as having an estimated effective radius and an estimated effective stand-off, wherein

during said characterizing step, said estimated effective radius and said estimated effective stand-off are used in characterizing said acoustic transducer as having said effective radius and effective stand-off.

25. A method according to claim 21, wherein:

said step of characterizing comprises

1) arranging said acoustic transducer a predetermined distance from a calibration target, wherein said calibration target is located in a location other than in said well, said calibration target is of known thickness, and said calibration target is in contact with matter of known impedances along its inner and outer surfaces,

2) transmitting a pulse from said acoustic transducer through said matter in contact with said inner surface of said uphole calibration target toward said uphole calibration target and measuring a calibration return signal therefrom, and

3) fitting a return signal calculation generated by a model of said return signal to said calibration return signal by substantially minimizing the difference between said return signal calculation and said calibration return signal, to determine at least one of said effective radius and said effective stand-off of said acoustic transducer as a function of said thickness of said calibration target.

26. A method according to claim 21, wherein:

said calculated return signal (V(f)), used in step c) in determining an indication of cement impedance, is defined according to ##EQU13## wherein wherein W is the planewave kernel of a transducer parameterized by a limited number of parameters, which planewave kernel is determined in said characterizing step,

θ

is the incidence angle of the planewave which can range over different angles,

f is a frequency which can range over a frequency spectrum,

L are layers related to said casing, and matter along the inner surface of said casing, and cement and/or other matter along the outer surface of said casing, and

R(f,θ

|L) are planar reflection coefficients for the layers L, at frequency f, and incidence angle θ

, which planar reflection coefficients are functionally related to the impedance of said cement and/or other matter located along the outer surface of said casing.

27. A method according to claim 26, wherein:

said step c) of fitting a return signal calculation to determine an indication of cement impedance comprises, varying values of said planar reflection coefficients of said model and hence the results of said return signal calculations, and comparing said results of said return signal calculations with said return signal to substantially find a minimum difference, wherein said different values of planar reflection coefficients correspond to different cement impedances behind said casing.

28. A method according to claim 27, wherein:

said minimum difference between said return signal calculation and said measured return signal is determined utilizing a mean-squared error calculation.

29. A method according to claim 27, further comprising:

in fitting said return signal calculation to said return signal and substantially minimizing said difference between said return signal calculation and said measured return signal, utilizing said difference as a quality indicator of said cement impedance.

30. A method according to claim 23, wherein:

said calculated return signal (V(f)), used in step c) in determining an indication of cement impedance, and used in step a)

     3) in determining at least one of an effective standoff and effective radius is defined according to ##EQU14## wherein W is the planewave kernel of a transducer parameterized by a limited number of parameters, the values for the parameters being varied to provide different planewave kernel values of fitting and minimizing in step a)

     3), with the best fitting value being used in step c) in determining an indication of cement impedance,

θ

is the incidence angle of the planewave which can range over different angles,

f is a frequency which can range over a frequency spectrum,

L are layers related to said casing, and matter along the inner surface of said casing, and cement and/or other matter along the outer surface of said casing when used in step c), and relating to said calibration target, and said borehole mud when used in step a), and

R(f,θ

|L) are planar reflection coefficients for the layers L, at frequency f, and incidence angle θ

, which planar reflection coefficients are predetermined when used in step a)

     3), and which planar reflection coefficients are functionally related to the impedance of said cement and/or other matter located along the outer surface of said casing when used in step c).

31. A method according to claim 30, wherein:

said step c) of fitting a return signal calculation to determine an indication of cement impedance comprises, varying values of said planar reflection coefficients of said model and hence the results of said return signal calculations, and comparing said results of said return signal calculations with said return signal to find a substantially minimum difference, wherein said different values of planar reflection coefficients correspond to different cement impedances behind said casing.

32. A method according to claim 31, wherein:

said substantially minimum difference between said return signal calculation and said measured return signal is determined utilizing a mean-squared error calculation.

33. A method according to claim 31, further comprising:

in fitting said return signal calculation to said return signal and substantially minimizing said difference between said return signal calculation and said measured return signal, utilizing said difference as a quality indicator of said cement impedance.

34. A method according to claim 30, wherein:

said model of said return signal and said return signal are limited to a narrow frequency band substantially centered on a resonant frequency of said calibration target such that

said planewave kernel is approximated to be substantially independent of frequency in said narrow frequency band.

35. A method according to claim 33, wherein:

said transducer is a Gaussian apodized disk transducer, and said planewave kernel W is a planewave kernel for a Gaussian apodized disk transducer parameterized by its radius and standoff.

36. A method according to claim 35, wherein:

said model of said return signal and said return signal are limited to a narrow frequency band substantially centered on a resonant frequency of said calibration target such that

said planewave kernel is approximated to be substantially independent of frequency in said narrow frequency band.

37. A method according to claim 25, wherein:

said calculated return signal (V(f)), used in step c) in determining an indication of cement impedance, and used in step a)

     3) in determining at least one of an effective standoff and effective radius is defined according to ##EQU15## wherein W is the planewave kernel of a transducer parameterized by a limited number of parameters, the values for the parameters being varied to provide different planewave kernel values of fitting and substantially minimizing in step a)

     3), with the best fitting value being used in step c) in determining an indication of cement impedance,

θ

is the incidence angle of the planewave which can range over different angles,

f is a frequency which can range over a frequency spectrum,

L are layers related to said casing, and matter along the inner surface of said casing, and cement and/or other matter along the outer surface of said casing when used in step c), and relating to said calibration target, said matter in contact with the inner surface of said calibration target, and said matter in contact with the outer surface of said calibration target when used in step a), and

R(f,θ

|L) are planar reflection coefficients for the layers L, at frequency f, and incidence angle θ

, which planar reflection coefficients are predetermined when used in step a)

     3), and which planar reflection coefficients are functionally related to the impedance of said cement and/or other matter located along the outer surface of said casing when used in step c).

38. A method according to claim 37, wherein:

said step c) of fitting a return signal calculation to determine an indication of cement impedance comprises, varying values of said planar reflection coefficients of said model and hence the results of said return signal calculations, and comparing said results of said return signal calculations with said return signal to find a substantially minimum difference, wherein said different values of planar reflection coefficients correspond to different cement impedances behind said casing.

39. A method according to claim 38, wherein:

said substantially minimum difference between said return signal calculation and said measured return signal is determined utilizing a mean-squared error calculation.

40. A method according to claim 38, further comprising:

in fitting said return signal calculation to said return signal and minimizing said difference between said return signal calculation and said measured return signal, utilizing said difference as a quality indicator of said cement impedance.

41. A method according to claim 37, wherein:

said model of said return signal and said return signal are limited to a narrow frequency band substantially centered on a resonant frequency of said calibration target such that

said planewave kernel is approximated to be substantially independent of frequency in said narrow frequency band.

42. A method according to claim 37, wherein:

said transducer is a Gaussian apodized disk transducer, and said planewave kernel W is a planewave kernel for a Gaussian apodized disk transducer parameterized by its radius and standoff.

43. A method according to claim 42, wherein:

said model of said return signal and said return signal are limited to a narrow frequency band substantially centered on a resonant frequency of said calibration target such that

said planewave kernel is approximated to be substantially independent of frequency in said narrow frequency band.

50. A method according to claim 25, further comprising:

generating a calibration database for said acoustic transducer by repeating said characterizing step for at least one of a plurality of different target parameters.

51. A method according to claim 50, wherein:

said different target parameters include calibration target thickness and calibration target radius.

52. A method according to claim 25, further comprising:

generating a calibration database for said acoustic transducer by repeating said characterizing step for at least one of a plurality of matter parameters.

53. A method according to claim 52, wherein:

said different matter parameters include matter pressure, matter temperature, and matter impedance.

54. A method according to claim 51, wherein:

said step of generating a calibration database further comprises repeating said characterizing step for at least one of a plurality of matter parameters including matter pressure, matter temperature, and matter impedance.

55. A method according to claim 32, wherein:

in making said mean-squared error calculation, the error is computed over one or both of,

(i) a range of thicknesses in steps about the assumed thickness of the casing of said well, and

(ii) a range of time shifts in steps relative to said return signal,

to substantially minimize the error.

56. A method according to claim 39, wherein:

in making said mean-squared error calculation, the error is computed over one or both of,

(i) a range of thicknesses in steps about the assumed thickness of the casing of said well, and

(ii) a range of time shifts in steps about the assumed time origin,

to substantially minimize the error.

57. A method according to claim 28, wherein:

in making said mean-squared error calculation, the error is computed over one or both of,

(i) a range of thicknesses in steps about the assumed thickness of the casing of said well, and

(ii) a range of time shifts in steps relative to said return signal,

to substantially minimize the error.

44. A method of measuring the mud impedance of mud in a cased well with a tool having an acoustic transducer, the casing of the well being of substantially known diameter, said method comprising:

a) arranging said acoustic transducer a predetermined distance from a calibration target, wherein said calibration target is of known thickness, and said calibration target is in contact with matter of known impedance along its inner surface and outer surface;

b) transmitting a pulse from said acoustic transducer toward said calibration target and measuring a return signal received therefrom;

c) fitting return signal calculations generated by a return signal model to said first signal, said model having a limited number of parameters, to determine a value for at least one of said parameters, and thereby characterizing said transducer, by substantially minimizing the difference between said return signal calculation and said measured first return signal;

d) placing said tool in said mud in said well;

e) transmitting a pulse from said acoustic transducer through said mud toward a target in said well, said target in said well being of substantially the same thickness as said calibration target, and having said mud along at least one surface of said target, and measuring a second return signal received therefrom; and

f) determining an indication of said mud impedance by minimizing the difference between said second return signal and additional return signal calculations generated by said model, wherein said additional return signals are a function of different mud impedances.

49. A method according to claim 44, comprising:

iteratively repeating steps c) and f), by using said mud impedance determination of step f) to change said return signal calculations of step c), and thereby adjusting said value for said parameter which characterizes said transducer, and using said adjusted value of said parameter to change said additional return signal calculations of step f), thereby adjusting mud impedance determination.