Method and system for determining pipeline circumferential and non-circumferential wall loss defects in a water pipeline

• US 6,359,434 B1
• Filed: 09/30/1998
• Issued: 03/19/2002
• Est. Priority Date: 09/30/1998
• Status: Expired due to Term

• Current Assignee: Pipeline Inspection and Condition Analysis Corporation
• Original Assignee: Hydroscope Canada Inc.
• Inventors: Winslow, Jens, Rajan, Varagur S. V., Yu, Yuwu, Ridley, Rodney K., Staples, Lawrence B.
• Primary Examiner(s): SNOW, WALTER E
• Application Number:
  US09/164,437
• Time in Patent Office: 1,266 Days
• Field of Search: 324/220, 324/219, 324/221, 324/236, 324/237, 324/238, 324/239, 324/240, 324/242, 324/243, 702/38, 702/189, 702/190
• US Class Current: 324/220
• CPC Class Codes: G01N 27/9046 by analysing electrical signals

Abstract

In accordance with the teachings of the present invention, a method is provided for analyzing RFT data from a data file. The method includes parsing the data file into pipe lengths, calculating a phase profile for the data points within each pipe length, locating potential defects in the pipe length using the phase profiles, determining for each defect a total equivalent phase shift as a combination of a circumferential equivalent phase shift and a non-circumferential equivalent phase shift, and using the total equivalent phase shift to analyze the defect.

Claims

1. A method for determining pipeline defects using remote field technology data obtained from an inspection of a pipeline, the data including a plurality of data points, a processor and memory being available for manipulating the data, the method comprising the computer-implemented steps of:

(a) using a plurality of data points each having a phase signal and an amplitude signal;

(b) analyzing the data to determine signals indicative of pipeline defects, including determining which portions of the data represent circumferential wall loss defects and which portions of the data represents non-circumferential wall loss defects, and combining the circumferential and non-circumferential wall data to determine a total wall loss for the defect; and

(c) quantifying the defects using the circumferential wall loss and the non-circumferential wall loss values.

2. The method according to claim 1, further comprising preparing the data prior to analyzing it, the preparing including at least one of filtering, smoothing, and decimating the data.

5. The method according to claim 1, wherein the data is initially parsed into smaller data portions, the analyzing of data being conducted on the smaller data portions.

6. The method according to claim 5, wherein the smaller data portions are individual pipe lengths.

29. The method according to claim 1, wherein quantifying pipe defects includes calibrating the inspection data to correspond dimensionally with the pipeline.

33. The method according to claim 1, wherein quantifying pipe defects includes calculating at least one of an average remaining wall thickness amount per pipe (JARW), an average circumferential remaining wall thickness amount per defect (LARW), and an average total remaining wall thickness amount per defect (DARW).

3. A method for determining pipeline defects using remote field technology data obtained from an inspection of a pipeline, the data including a plurality of data points having phase and amplitude signals, a processor and memory being available for manipulating the data, the method comprising the computer-implemented steps of:

(a) analyzing the data to determine signals indicative of pipeline defects, including determining which portions of the data represent circumferential wall loss defects and which portions of the data represents non-circumferential wall loss defects, and combining the circumferential and non-circumferential wall data to determine a total wall loss for the defect;

(b) quantifying the defects using the circumferential wall loss and the non-circumferential wall loss values; and

(c) preparing the data prior to analyzing it, the preparing including at least one of filtering, smoothing, and decimating the data, wherein preparing the data includes filtering the data to determine a median element, the filtering including

(i) defining a fixed data set size for the filtering;

(ii) applying the data set size sequentially to the data by

(1) determining for each data point in the data set a total distance, the total distance being the sum of the distances between that point and all other data points in that data set; and

(2) identifying the data point in the set having the smallest total distance; and

(iii) using the data point having the smallest total distance as the median element for that data set.

4. A method for determining pipeline defects using remote field technology data obtained from an inspection of a pipeline, the data including a plurality of data points having phase and amplitude signals, a processor and memory being available for manipulating the data, the method comprising the computer-implemented steps of:

(a) analyzing the data to determine signals indicative of pipeline defects, including determining which portions of the data represent circumferential wall loss defects and which portions of the data represents non-circumferential wall loss defects, and combining the circumferential and non-circumferential wall data to determine a total wall loss for the defect;

(b) quantifying the defects using the circumferential wall loss and the non-circumferential wall loss values; and

(c) preparing the data prior to analyzing it, the preparing including at least one of filtering, smoothing, and decimating the data, wherein preparing the data includes smoothing the data using a moving block average filter.

7. A method for determining pipeline defects using remote field technology data obtained from an inspection of a pipeline, the data including a plurality of data points having phase and amplitude signals, a processor and memory being available for manipulating the data, the method comprising the computer-implemented steps of:

(a) analyzing the data to determine signals indicative of pipeline defects, including determining which portions of the data represent circumferential wall loss defects and which portions of the data represents non-circumferential wall loss defects, and combining the circumferential and non-circumferential wall data to determine a total wall loss for the defect;

(b) quantifying the defects using the circumferential wall loss and the non-circumferential wall loss values; and

(c) identifying signals indicative of non-analyzable features in the data and ignoring such signals when analyzing the data for defects, the non-analyzable features including at least one of a bell-and-spigot pipe connection, a tee, an elbow, and a valve.

8. The method according to claim 7, wherein the data is parsed into smaller data portions delineated by the non-analyzable features, the analyzing of data being conducted on the smaller data portions.

9. The method according to claim 8, wherein the smaller data portions are individual pipe lengths.

10. A method for determining pipeline defects using remote field technology data obtained from an inspection of a pipeline, the data including a plurality of data points having phase and amplitude signals, a processor and memory being available for manipulating the data, the method comprising the computer-implemented steps of:

(a) analyzing the data to determine signals indicative of pipeline defects, including determining which portions of the data represent circumferential wall loss defects and which portions of the data represents non-circumferential wall loss defects, and combining the circumferential and non-circumferential wall data to determine a total wall loss for the defect;

(b) quantifying the defects using the circumferential wall loss and the non-circumferential wall loss values; and

(c) identifying signals indicative of non-analyzable features in the data, the identifying including comparing at least one template representative of a non-analyzable feature with the pipe data.

11. The method according to claim 10, wherein the at least one template is stored in memory.

12. The method according to claim 11, wherein the memory includes multiple templates selectable by an operator for use in comparing with the data.

13. The method according to claim 10, wherein the at least one template is modified by at least one of adjusting, scaling, and shifting the template values to correspond to the pipe data.

14. The method according to claim 10, wherein an operator defines a portion of the data as a template for use in comparing with the data.

15. The method according to claim 10, wherein the identifying of non-analyzable features includes selecting a template, sequencing through the data to locate portions that are similar to the template, and if a similar trace is found, creating a new template to match the data and rechecking the data using the new template;

the similarity of the trace being determined by a predefined logic.

16. The method according to claim 15, wherein the template is selected from memory and the new template replaces the selected template in memory.

17. The method according to claim 15, wherein sequencing through the data includes calculating the sum of the squared difference between the template and the data set at each data point.

18. The method according to claim 17, wherein the predefined logic includes a characterization of how well the data matched the template, the characterization including a worst match, the new template being defined using the data having the worst match.

19. A method for determining pipeline defects using remote field technology data obtained from an inspection of a pipeline, the data including a plurality of data points having phase and amplitude signals, a processor and memory being available for manipulating the data, the method comprising the computer-implemented steps of:

(a) analyzing the data to determine signals indicative of pipeline defects, including determining which portions of the data represent circumferential wall loss defects and which portions of the data represents non-circumferential wall loss defects, and combining the circumferential and non-circumferential wall data to determine a total wall loss for the defect; and

(b) quantifying the defects using the circumferential wall loss and the non-circumferential wall loss values;

wherein analyzing the data includes

(i) defining a nominal signal and defining a Reference Curve, the Reference Curve being a collection of theoretical phase and amplitude values for decreasing uniform circumferential wall thickness for the pipe;

(ii) determining which data points are positioned near the Reference Curve based on the nominal signal, the data point phase and amplitude signals, and a predefined logic;

(iii) locating defects in the pipe as a function of the data point phase signals; and

(iv) determining for each defect which portions of the data represent circumferential wall loss and which portions represent non-circumferential wall loss using the determinations of which data points are positioned near the Reference Curve.

20. The method according to claim 19, wherein

(a) defining a nominal signal includes defining a nominal phase (P_NOM); and

(b) determining which data points are positioned near the Reference Curve includes defining a Phase Profile for the data;

(c) determining circumferential wall loss and non-circumferential wall loss is based on the Phase Profile; and

(d) quantifying defects includes calculating for each defect a total equivalent-phase-shift (EQPS) as a combination of a circumferential equivalent-phase-shift (EQPS_CIRC) and a non-circumferential equivalent-phase-shift (EQPS_NON-CIRC);

EQPS_CIRC and EQPS_NON-CIRC each being a function of P_NOM and Phase Profile.

21. The method according to claim 20, wherein P_NOM is determined by sorting the phase data into ascending order, removing a percentage of the smallest phase data values, and setting P_NOM equal to the smallest remaining phase data value.

22. The method according to claim 20, wherein the predefined logic includes setting the Phase Profile for a data point equal to the data point phase signal for those data points near P_NOM and those data points within a predetermined Reference_Curve_Threshold Margin and Extent_Threshold.

23. The method according to claim 20, wherein the predefined logic includes defining an intersection point having a given phase for groups of data points that are not near the Reference Curve, and setting the Phase Profile for those data points equal to the intersection point phase.

24. The method according to claim 23, wherein the intersection point is the crossing location on the Reference Curve of a line representative of the group of data points that are not near the Reference Curve, the line being determined using at least one of a least squares fit and a linear interpolation of the consecutive data points for that group of data points.

25. The method according to claim 20, wherein

(a) quantifying defects includes identifying the maximum phase signals;

(b) EQPS_CIRC is a function of the difference between P_NOM and the Phase Profile value for each maximum defect data point;

(c) for those data points having a Phase Profile value at the maximum defect data point equal to the maximum defect phase signal, EQPS_NON-CIRC is set to zero; and

(d) for the remaining maximum defect data points, EQPS_NON-CIRC is set equal to an RFT Phase Lag Angle φ

.

26. The method according to claim 19, wherein locating defects includes locating local maximum phase signals within a group of data points.

27. The method according to claim 26, wherein those data points having a maximum phase signal less than a predefined minimum amount are disregarded.

28. The method according to claim 26, wherein those consecutive data points having a maximum phase signal within a predefined minimum distance to another maximum actual phase of larger value are disregarded.

34. The method according to claim 19, wherein quantifying pipe defects includes calculating at least one of an average remaining wall thickness amount per pipe (JARW), an average circumferential remaining wall thickness amount per defect (LARW), and an average total remaining wall thickness amount per defect (DARW).

35. The method according to claim 20, wherein quantifying pipe defects includes calculating at least one of an average remaining wall thickness amount per pipe (JARW), an average circumferential remaining wall thickness amount per defect (LARW), and an average total remaining wall thickness amount per defect (DARW).

36. The method according to claim 35, wherein the average remaining wall thickness amount is a function of an average equivalent-phase-shift amount (EQPS_CAL) and EQPS_CAL.

37. The method according to claim 35, wherein the average circumferential remaining wall thickness amount is a function of EQPS_CIRC and EQPS_CAL.

38. The method according to claim 35, wherein the average circumferential remaining wall thickness amount is a function of EQPS_CIRC and EQPS_CAL.

30. A method for determining pipeline defects using remote field technology data obtained from an inspection of a pipeline, the data including a plurality of data points having phase and amplitude signals, a processor and memory being available for manipulating the data, the method comprising the computer-implemented steps of:

(a) analyzing the data to determine signals indicative of pipeline defects, including determining which portions of the data represent circumferential wall loss defects and which portions of the data represents non-circumferential wall loss defects, and combining the circumferential and non-circumferential wall data to determine a total wall loss for the defect; and

(b) quantifying the defects using the circumferential wall loss and the non-circumferential wall loss values;

wherein quantifying pipe defects includes calibrating the inspection data to correspond dimensionally with the pipeline; and

wherein quantifying pipe defects includes calibrating the data by obtaining information from a representative pipe having no defects, the calibrating including

(i) determining a nominal data signal for the representative pipe;

(ii) placing a defect in the representative pipe and determining the signal change resulting from the defect; and

(iii) using the nominal data signal and the signal change to calibrate the pipeline data.

31. The method according to claim 30, wherein calibrating the data includes defining a calibration amount that results in a selected percentage of the data being a 100 percentage through-wall loss.

32. A method for determining pipeline defects using remote field technology data obtained from an inspection of a pipeline, the data including a plurality of data points having phase and amplitude signals, a processor and memory being available for manipulating the data, the method comprising the computer-implemented steps of:

(a) analyzing the data to determine signals indicative of pipeline defects, including determining which portions of the data represent circumferential wall loss defects and which portions of the data represents non-circumferential wall loss defects, and combining the circumferential and non-circumferential wall data to determine a total wall loss for the defect; and

(b) quantifying the defects using the circumferential wall loss and the non-circumferential wall loss values;

wherein quantifying pipe defects includes calibrating the inspection data to correspond dimensionally with the pipeline; and

wherein quantifying pipe defects includes calibrating the data including

(i) exposing a portion of the pipeline having no defects;

(ii) determining nominal phase and amplitude values directly from the exposed portion of the pipe having no defects; and

(iii) determining phase and amplitude values for a location outside of the exposed pipeline in free air;

wherein the nominal and free air phase and amplitude values are used to calibrate the data.

39. A method for determining pipeline defects using remote field technology data obtained from an inspection of a pipeline, the data including a plurality of data points having phase and amplitude signals, a processor and memory being available for manipulating the data, the method comprising the computer-implemented steps of:

(a) parsing the data into smaller portions representative of individual pipe lengths;

(b) analyzing the data to determine signals indicative of pipeline defects, including determining which portions of the data represent circumferential wall loss defects and which portions of the data represents non-circumferential wall loss defects, and combining the circumferential and non-circumferential wall data to determine a total wall loss for the defect;

the analyzing including

(i) defining a nominal signal and defining a Reference Curve, the Reference Curve being a collection of theoretical phase and amplitude values for decreasing uniform circumferential wall thickness for the pipe;

(ii) determining which data points are positioned near the Reference Curve based on the nominal signal, the data point phase and amplitude signals, and a predefined logic;

(iii) locating defects in the pipe as a function of the data point phase signals; and

(iv) determining for each defect which portions of the data represent circumferential wall loss and which portions represent non-circumferential wall loss using the determinations of which data points are positioned near the Reference Curve; and

(c) quantifying the defects using the circumferential wall loss and the non-circumferential wall loss values, the quantifying including calibrating the data to correspond dimensionally with the pipeline.

40. The method according to claim 39, wherein

(a) defining a nominal signal includes defining a nominal phase (P_NOM); and

(b) determining which data points are positioned near the Reference Curve includes defining a Phase Profile for the data;

(c) determining circumferential wall loss and non-circumferential wall loss is based on the Phase Profile; and

(d) quantifying defects includes calculating for each defect a total equivalent-phase-shift (EQPS) as a combination of a circumferential equivalent-phase-shift (EQPS_CIRC) and a non-circumferential equivalent-phase-shift (EQPS_NON-CIRC);

EQPS_CIRC and EQPS_NON-CIRC each being a function of P_NOM and Phase Profile.

41. The method according to claim 40, wherein quantifying pipe defects includes calculating at least one of an average remaining wall thickness amount per pipe (JARW), an average circumferential remaining wall thickness amount per defect (LARW), and an average total remaining wall thickness amount per defect (DARW).

42. A method of determining changes in wall thickness of a pipeline using measured remote field data including a plurality of data points having phase and amplitude signals, the method comprising:

(a) defining a nominal signal and defining a Reference Curve, the Reference Curve being a collection of theoretical phase and amplitude values for decreasing uniform circumferential wall thickness for the pipeline; and

(b) determining which measured data points are positioned near the Reference Curve based on the nominal signal and a predefined logic;

wherein for those data points located near the Reference Curve, circumferential changes in wall thickness are determined according to the difference between the data point phase value and the nominal phase value; and

(c) determining which measured data points are not positioned near the Reference Curve including identifying an intersection phase signal for those data points;

wherein for those data points not located near the Reference Curve, non-circumferential changes in wall thickness are determined according to the difference between the data point phase value and the intersection phase signal.

43. A method of determining non-circumferential changes in wall thickness of a pipeline using measured remote field data including a plurality of data points having phase and amplitude signals, the method comprising:

(a) defining a Reference Curve for the pipeline, the Reference Curve being a collection of theoretical phase and amplitude values for decreasing uniform circumferential wall thickness for the pipeline; and

(b) determining which measured data points are representative of non-circumferential changes in wall thickness by determining which measured data points are not positioned near the Reference Curve based on a predefined logic;

wherein for those data points not located near the Reference Curve,

(i) assigning an intersection phase value representative of a nominal circumferential phase amount for those data points; and

(ii) defining non-circumferential changes in wall thickness according to the difference between the measured data point phase values and the intersection phase signal.