Improved methodology for guided wave inspections of pipelines.
The accuracy of defect size determination in oil and gas pipelines using long-range ultrasonic guided waves may be improved by obtaining additional information about the extent of the defect as well as its amplitude. This is achieved by focusing the ultrasonic energy at a desired distance along the pipe from the transducer and at a desired angular position. Not only is information available about the lateral extent of the defect, but the relationship between its size and reflection amplitude is more precise. Focusing also has the benefit of locally increasing the sensitivity for detection of small defects, as the energy is concentrated in the focal zone and the signal to noise ratio is increased as a result.
The focusing technique allows a concentration of energy in the pipe wall at a specific angle for a given distance from the transducer. Numerical modeling was used to investigate the properties of the focus under different conditions. This is indicated in Figure 1 which shows an isometric view of a computer numerical simulation of a focused guided wave propagating in a pipe. The size of the focal spot can also be controlled by varying certain excitation parameters.
[FIGURE 1 OMITTED]
Figure 1 shows how there is a concentration of energy (in the form of particle displacement) at only one circumferential location. This is again demonstrated in Figure 2, where the same focal spot is viewed as a radial graph showing the distribution of energy around the circumference, at the axial focus position. Here, there are displacements at 45[degrees], but very little displacements at other locations about the circumference. Should a defect be present at the focal location at 45[degrees] as in Figure 2 (gray), there will be a high signal response, compared to a position where there is no defect present (such as at 225[degrees] in Figure 2). In addition to the 2 numerical modeling shown here, experimental trials have validated the focusing algorithms. Figure 3 shows a radial plot of a "target" profile (dotted line) against a numerically predicted focus profile (dashed line) and an experimentally measured focus profile (solid line).
The figure clearly shows how the numerical prediction closely resembles the target profile with the exception of two small side lobes. There is strong agreement between numerical prediction and empirical measurements, including the presence of small side lobes. If a defect is present, the focal spot will only interact (and thus cause a reflection) when it is centered at the same point, as shown in Figure 2. The focus can be scanned around the circumference as well as along the length of the pipe--a high reflection will be received where there is a defect at the focal location, but nothing will be received from where there is no defect present.
[FIGURE 2 OMITTED]
[FIGURE 3 OMITTED]
[FIGURE 4 OMITTED]
Figure 4 shows how there is a much higher response when focusing in line with a defect (right) than when no defect is present. This emphasizes the defect's presence and gives the inspector confidence that the defect is not a false positive artifact.
Not only does the focusing technique achieve this circumferential resolution, but it also increases sensitivity to small defects. Figure 5 shows a single unfocused symmetric wave test of the same defect examined in Figure 4. The signal to background noise ratio for the unfocused test shown in Figure 5 was 8dB. The signal to background noise ratio when the sound energy was focused on the defect rose to 26dB, an increase by a factor of approximately eight. This significant improvement in the clarity of the response makes the correct call/ no-call decision much more reliable so that not only is sensitivity increased but the confidence in reporting defects is also enhanced.
If every point on the pipe is focused upon in this manner, a 3D C-Scan type image can be generated, which allows for a very intuitive projection of the pipe features. Figure 6 shows an example of this, where there is a 3-inch branch at top-dead center of a 16-inch pipe, followed by a girth weld, and finally a flange. The three features are unambiguously shown as the angular resolution of the technique allows circumferential positioning and quantification of features. There is clearly opportunity for other similar three dimensional plots (such as an isometric view, for example) that can further aid interpretation.
[FIGURE 5 OMITTED]
To address the aforementioned problems with "traditional" guided wave inspection, it is necessary to combine the amplitude information yielded from symmetric tests with directionality information obtained from focused tests. A classification scheme has been designed to aid this process, whereby responses are given a score according to the amplitude from a symmetrical test, and their circumferential spread from a focused test. First, each response is given a value from 1 to 3, known as its Defect Category, C. This is calculated by taking the amplitude of the response relative to the expected amplitude of a weld signal (i.e. the dB difference). Distance Amplitude Correction (DAC) curves are used to estimate the expected weld signal amplitude at arbitrary positions. As such, C is defined as follows:
Received Amplitude from C = 1 symmetric test is greater than 12dB less than a weld signal Received Amplitude is C = 2 between 6dB and 12dB lower than a weld Received Amplitude is less C = 3 than 6dB lower than a weld
[FIGURE 6 OMITTED]
Received Amplitude from symmetric test is greater than 12dB less than a weld signal C = 1
Received Amplitude is between 6dB and 12dB lower than a weld C=2
Received Amplitude is less than 6dB lower than a weld
Figure 7 shows how these defect categories can be easily defined using distance amplitude calibration (DAC) curves.
[FIGURE 7 OMITTED]
[FIGURE 8 OMITTED]
Secondly, a "directionality distribution" of the focused test, D, is computed. The directionality distribution is best described graphically, as shown in Figure 8, and is defined as follows:
Energy in response D = 3 concentrated over less than 45[degrees] of the circumference Energy in response D = 2 concentrated between 45[degrees] and 90[degrees] of the circumference Energy in response D = 1 concentrated between 90[degrees] and 315[degrees] of the circumference Energy in response D = 0 distributed evenly over 360[degrees]
The product of C and D will then give operators a "Follow-up priority". Table 1 gives the follow-up priority matrix. If the C x D product is 3 or greater, high priority should be given. If C x D = 2, medium priority, and C x D = 1 low priority. If C x D = 0, there is no directionality to the signal, so that it is interpreted as the reflection from a girth weld.
This classification scheme has been tested on "blind" trials of 16 inch coal tar-coated cased pipe containing artificial defects, manufactured so as to mimic those typically experienced in field conditions with good success. During the R&D project phase, for the purpose of having performance targets, the technical team defined the threshold for "moderate", or "severe" defects as shown in Table 2. (For characteristics with dimensions less than that defined for "moderate," those defects are characterized as "small".) The above methodology was employed and the follow-up priorities were calculated for each defect. Table 3 shows the relationship of actual defect size with the assigned follow-up priority classification given to the defects.
Note how all welds were correctly identified as welds, and all severe defects were allocated a high follow-up priority. Moderate defects were also detected with conservative estimations of follow-up priority usually assigned to them. There were no "small" defects in the sample specimens.
A phased transducer array has been used successfully to steer and to focus ultrasonic guided waves.
Laboratory and field results show that sensitivity to defects is increased by focusing.
The spatial information provided from the focused tests allows defect extent to be estimated, permitting the classification of defect severity. A classification scheme has been developed.
Preliminary application of this classification scheme to the results from field tests on a variety of known defects gave an accurate prediction of defect severity, thereby providing pipeline operators with the information to make confident decisions about the continuing fitness for service of their pipelines.
The focusing concept has been developed in collaboration with Penn State University and FBS Inc. Some of the results presented were generated in a development project funded by U.S. DoT/PHMSA, NYSEARCH/Northeast Gas Association and gas industry operators. Development work has also been carried out under the EPSRC Engineering Doctorate Program. The authors gratefully acknowledge the help and support of the above organizations.
Alleyne, DN and MJS Lowe, The reflection of guided waves from circumferential notches in pipes. ASME J Appl Mech, 1998.65: p. 635-641.
Ditri, JJ and JL Rose, Excitation of guided elastic wave modes in hollow cylinders by applied surface tractions. J. Appl. Phys, 1992. 72(7): p. 2589-2597.
Kwun, H. and H. Dynes, Long range guided wave inspection of pipe using the magnetostrictive sensor technology-feasibility of defect characterization., Int Soc for Optical Eng (SPIE) on Nondestructive evaluation of utilities and pipelines, 1998. II: p. 28-34.
Li, J and JL Rose, Implementing Guided Wave Mode Control by Use of a Phased Transducer Array. IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control, Volume 48, Issue 3, May 2001 p. 761-767.
Li, J and JL Rose, Angular-Profile Tuning of Guided Waves in Hollow Cylinders Using a Circumferential Phased Array. IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control Volume 49, Issue 12, December 2002; p. 1720-1729.
Mudge, PJ, Lank, AM and Alleyne, DN, A long range method of detection of corrosion under insulation in process pipework. 5th European Union Hydrocarbons Symposium, Edinburgh, Nov. 26-28, 1996.
Rose, JL, Rajana, KM and Carr, FT, Ultrasonic Guided Wave Inspection Concepts for Steam Generator Tubing. Materials Evaluation, February 1994: p. 307-311.
Rose, J.L, A baseline and vision of ultrasonic guided wave inspection potential Journal of Pressure Vessel Technology, 2002: p. 273-282.
Rose JL, Guided Wave Ultrasonic Pipe Inspection-The Next Generation. 8th European NDT Conference, Barcelona June 17-21, 2002.
Authors: Peter Mudge is a Technology Fellow at TWI, working on guided waves. He has more than 30 years experience in nondestructive testing and is a past president of the British Institute of NDT.
Phil Catton is studying for an Engineering Doctorate in development of guided wave techniques at Brunell Surrey Universities and at TWI Ltd.
Daphne D'Zurko is Executive Director of NYSEARCH, a voluntary R & D organization for the gas LDCs in North America. She has 19 years of experience in gas industry R & D.
Joseph L. Rose is the Paul Morrow Professor in Engineering Science and Mechanics at The Pennsylvania State University. He is also chief scientist of FBS, Inc.
Table 1: Classification scheme of follow up priorities. Follow up priority Defect Category, C Directionality 0 0 0 0 Distribution, D 1 1 2 3 2 2 4 6 3 3 6 9 Table 2: Operators' proposed classification scheme for testing purposes for threshold sizing of "moderate" and "severe" defects. Confidence level Depth Length Width Severity Near 100% > 80% thru wall > 3/4" > 3/4" Severe 80% <= x <= 100% 50% < x < 80% thru wall > 3/4" > 3/4" Severe 70% <= x <= 80% 20% < x 50% thru wall > 3/4" > 3/4" Moderate 80% <= x <= 100% > 80% thru wall < 3/4" < 3/4" Severe 70% <= x <= 80% 50% < x < 80% thru wall < 3/4" < 3/4" Moderate Table 3: Results of field trials for the classification scheme. Gray cells indicate the "correct" classification. Actual Defect Classification Shaded cells show the ideal classification Weld Small Moderate Severe Number in 5 0 5 23 Sample Interpretation Weld 5 of defects Low Priority 1 according to Medium Priority 1 classification Hiqh Priority 3 23 scheme Note that all 23 severe defects were allocated "high priority", and all welds were correctly identified.
|Printer friendly Cite/link Email Feedback|
|Comment:||Improved methodology for guided wave inspections of pipelines.|
|Author:||Catton, Phil; Mudge, Peter; D'Zurko, Daphne; Rose, Joseph|
|Publication:||Pipeline & Gas Journal|
|Date:||Jun 1, 2008|
|Previous Article:||Nation's pipeline construction boom brings rewards and challenges.|
|Next Article:||Improving interpretation of SCC anomalies found by ultrasonic ILI.|