System and method for performing inversion on LWD resistivity logs with enhanced resolution

• US 5,867,806 A
• Filed: 10/22/1996
• Issued: 02/02/1999
• Est. Priority Date: 03/13/1996
• Status: Expired due to Term

• Current Assignee: Halliburton Energy Services Incorporated (Halliburton Company)
• Original Assignee: Halliburton Energy Services Incorporated (Halliburton Company)
• Inventors: Jackson, Charles E., Illfelder, Herbert Max Joseph, Merchant, Gulamabbas A., Strickland, Robert W.
• Primary Examiner(s): McElheny, Jr., Donald E.
• Application Number:
  US08/734,772
• Time in Patent Office: 833 Days
• Field of Search: 364/421, 364/422, 324/338, 324/339, 324/368, 324/369, 324/342, 324/343
• US Class Current: 702/7
• CPC Class Codes: G01V 3/38 Processing data, e.g. for analysis, for interpretation, for correction

Abstract

A system and method for performing inversion and for correcting for shoulder bed effect. The method enhances the resolution of wave propagation resistivity logs and corrects the logs for adjacent bed effects, whereby the resulting logs have a vertical response as good as wireline high-resolution induction logs. The method first performs logging while drilling on a formation to obtain LWD resistivity log data. The method then selects one or more control depths at one or more locations of each of a plurality of detected beds in the formation. The present invention includes a novel method for selecting bed boundaries from the log data, and the control depths are then determined based on the determined bed boundaries. The method then estimates the resistivity of each bed only at the selected control depths to produce an estimated resistivity of the beds. The method then computes a simulated log value at each control depth using a current estimate of the resistivity of the beds. The computed simulated log, is then compared to the actual log data at each control depth, and the resistivity of each bed is adjusted using the difference between the actual and simulated values at the control depths. The above method iteratively repeats a plurality of times until the simulated log substantially matches the actual log at the control depths.

Claims

1. A method for performing inversion resistivity on log data to determine a model formation, comprising:

performing logging on a formation to obtain a resistivity log, wherein the resistivity log comprises log data;

selecting one or more control depths at one or more locations of each of a plurality of detected beds in the formation;

estimating the resistivity of each bed only at the selected control depths, wherein said estimating uses the log data, wherein said estimating produces an estimated resistivity for each of the beds;

computing a simulated log value at each control depth using a current estimate of the resistivity of the beds, wherein said computing uses said estimated resistivities at said selected control depths as an initial estimate of the true resistivity;

comparing the computed simulated log to the actual log data at each control depth;

adjusting the resistivity of each bed using the difference between the actual and simulated values at the control depths;

repeating said steps of computing a simulated log value at each control depth using the current estimate of the resistivity of the beds, comparing the computed log to the actual log at each control depth, and adjusting the resistivity of each bed using the difference between the actual and simulated values at the control depths, wherein said repeating is performed if the simulated log does not match the actual log at the control depths;

constructing a corrected log from the depths of the bed boundaries and the resistivity of each bed if the simulated log substantially matches the actual log at the selected control depths.

2. The method of claim 1, further comprising:

selecting one or more bed boundaries prior to said selecting one or more control depths;

wherein said selecting one or more control depths at one or more locations of each of said plurality of detected beds in the formation uses said selected bed boundaries.

3. The method of claim 2, wherein said selecting one or more control depths comprises selecting a control depth in a location between each of said selected bed boundaries.

4. The method of claim 2, wherein said selecting one or more bed boundaries comprises:

performing a curve fit of the log data after said performing logging while drilling;

determining inflection points of the log data after said performing a curve fit of the log data; and

selecting bed boundaries from the inflection points of the curve-fit of the log data.

5. The method of claim 4, wherein said performing a curve fit of the log data comprises performing a curve fit of the apparent conductivity log data.

6. The method of claim 4, wherein said performing a curve fit of the log data comprises fitting a polynomial in depth to the conductivity.

7. The method of claim 6, wherein said performing a curve fit of the log data comprises fitting a polynomial in depth to the conductivity for low dip angles;

and

wherein said performing a curve fit of the log data comprises fitting a polynomial in depth to the log of the resistivity for high dip angles.

8. The method of claim 6, wherein said performing a curve fit of the log data comprises fitting a cubic polynomial in depth to the conductivity using a least-squares method.

9. The method of claim 8, wherein said performing a curve fit of the log data comprises fitting a cubic polynomial in depth to the conductivity using a weighted linear least-squares method.

10. The method of claim 9, wherein said weighted linear least-squares method uses a weighting function which is Gaussian in depth, wherein said Gaussian weighting function provides smoothness as said window is moved across said log data.

11. The method of claim 6, wherein said performing a curve fit of the log data is performed a plurality of times on said log data using a windowing method, wherein said windowing method comprises moving a window across said log data at periodic increments, and performing said curve fit on said window of log data at said periodic increments as said window is moved across said log data.

12. The method of claim 11, wherein said determining inflection points of the log data comprises evaluating said fitted cubic polynomial at the center of said window as said window is moved across said log data.

13. The method of claim 11, further comprising:

subtracting an offset from depth prior to said performing said curve fit of the log data for each of said windows, wherein said subtracting causes the transformed depth to be zero at the center of said window.

14. The method of claim 2, wherein said selecting one or more bed boundaries comprises:

examining resistivity with respect to depth using said resistivity log; and

discarding one or more boundaries in response to said examining.

15. The method of claim 2, wherein said selecting one or more bed boundaries comprises:

examining a derivative of the logarithm of the resistivity with respect to depth using said resistivity log; and

discarding one or more boundaries in response to said examining.

16. The method of claim 2, wherein said selecting one or more bed boundaries comprises:

examining an absolute value of the first derivative of the logarithm of the resistivity with respect to depth; and

discarding one or more boundaries in response to said examining.

17. The method of claim 16, wherein said discarding comprises discarding bed boundaries with absolute values below a first threshold.

18. The method of claim 2, wherein said selecting one or more bed boundaries comprises:

examining at least one pair of bed boundaries closer in depth than a pre-defined tolerance;

determining if the first derivatives of control depths of each of said at least one pair of bed boundaries have the same sign;

determining which of said first derivatives of said control depths of each of said at least one pair of bed boundaries is smaller in magnitude; and

discarding a bed boundary of said at least one pair of bed boundaries whose first derivative is smaller in magnitude.

19. The method of claim 1, wherein said performing logging on a formation to obtain a resistivity log comprises performing logging while drilling (LWD) on the formation to obtain a LWD resistivity log.

20. The method of claim 1, wherein said performing logging on a formation to obtain a resistivity log comprises performing induction logging on the formation to obtain the resistivity log.

21. The method of claim 1, wherein said performing logging on a formation to obtain a resistivity log comprises performing wireline logging on the formation to obtain the resistivity log.

22. A method for performing inversion resistivity on log data to determine a model formation, comprising:

performing logging on a formation to obtain a resistivity log, wherein the resistivity log comprises log data;

performing a curve fit of the log data after said performing logging while drilling;

determining inflection points of the log data after said performing a curve fit of the log data;

selecting bed boundaries from the inflection points of the curve-fit of the log data;

selecting one or more control depths at one or more locations of each of a plurality of detected beds in the formation;

estimating the resistivity of each bed only at the selected control depths, wherein said estimating uses the log data, wherein said estimating produces an estimated resistivity for each of the beds;

computing a simulated log value at each control depth using a current estimate of the resistivity of the beds, wherein said computing uses said estimated resistivities at said selected control depths as an initial estimate of the true resistivity;

comparing the computed simulated log to the actual log data at each control depth;

adjusting the resistivity of each bed using the difference between the actual and simulated values at the control depths;

repeating said steps of computing a simulated log value at each control depth using the current estimate of the resistivity of the beds, comparing the computed log to the actual log at each control depth, and adjusting the resistivity of each bed using the difference between the actual and simulated values at the control depths, wherein said repeating is performed if the simulated log does not match the actual log at the control depths;

constructing a corrected log from the depths of the bed boundaries and the resistivity of each bed if the simulated log substantially matches the actual log at the selected control depths.

23. A method for determining bed boundaries in a formation from resistivity log data, comprising:

performing logging on a formation to obtain a resistivity log, wherein the resistivity log comprises log data;

performing a curve fit of the log data after said performing logging while drilling;

determining inflection points of the log data after said performing a curve fit of the log data; and

determining bed boundaries from the inflection points of the curve-fit of the log data.

24. The method of claim 23, wherein said performing a curve fit of the log data comprises performing a curve fit of the apparent conductivity log data.

25. The method of claim 24, wherein said performing a curve fit of the log data comprises fitting a polynomial in depth to the conductivity.

26. The method of claim 25, wherein said performing a curve fit of the log data comprises fitting a polynomial in depth to the conductivity for low dip angles;

and

wherein said performing a curve fit of the log data comprises fitting a polynomial in depth to the log of the resistivity for high dip angles.

27. The method of claim 25, wherein said performing a curve fit of the log data comprises fitting a cubic polynomial in depth to the conductivity using a least-squares method.

28. The method of claim 27, wherein said performing a curve fit of the log data comprises fitting a cubic polynomial in depth to the conductivity using a weighted linear least-squares method.

29. The method of claim 28, where in said weighted linear least-squares method uses a weighting function which is Gaussian in depth, wherein said Gaussian weighting function provides smoothness as said window is moved across said log data.

30. The method of claim 25, wherein said performing a curve fit of the log data is performed a plurality of times on said log data using a windowing method, wherein said windowing method comprises moving a window across said log data at periodic increments, and performing said curve fit on said window of log data at said periodic increments as said window is moved across said log data.

31. The method of claim 30, wherein said determining inflection points of the log data comprises evaluating said fitted cubic polynomial at the center of said window as said window is moved across said log data.

32. The method of claim 31, further comprising:

subtracting an offset from depth prior to said performing said curve fit of the log data for each of said windows, wherein said subtracting causes the transformed depth to be zero at the center of said window.

33. The method of claim 23, wherein said performing logging on a formation to obtain a resistivity log comprises performing logging while drilling (LWD) on the formation to obtain a LWD resistivity log.

34. The method of claim 23, wherein said performing logging on a formation to obtain a resistivity log comprises performing induction logging on the formation to obtain the resistivity log.

35. The method of claim 23, wherein said performing logging on a formation to obtain a resistivity log comprises performing wireline logging on the formation to obtain the resistivity log.

36. The method of claim 23, wherein said determining bed boundaries further comprises:

examining resistivity with respect to depth using said resistivity log; and

discarding one or more boundaries in response to said examining.

37. The method of claim 23, wherein said determining bed boundaries further comprises:

examining a derivative of the logarithm of the resistivity with respect to depth using said resistivity log; and

discarding one or more boundaries in response to said examining.

38. The method of claim 23, wherein said determining bed boundaries further comprises:

examining an absolute value of the first derivative of the logarithm of the resistivity with respect to depth; and

discarding one or more boundaries in response to said examining.

39. The method of claim 38, wherein said discarding comprises discarding bed boundaries with absolute values below a first threshold.

40. The method of claim 23, wherein said selecting one or more bed boundaries comprises:

examining at least one pair of bed boundaries closer in depth than a pre-defined tolerance;

determining if the first derivatives of control depths of each of said at least one pair of bed boundaries have the same sign;

determining which of said first derivatives of said control depths of each of said at least one pair of bed boundaries is smaller in magnitude; and

discarding a bed boundary of said at least one pair of bed boundaries whose first derivative is smaller in magnitude.