On the probability of detecting wall thinning defects with dispersive circumferential guided waves
File(s) 1-s2.0-S0021782416301167-main.pdf (522.32 KB)
Published version
Author(s)
Howard, R
Cegla, F
Type
Journal Article
Abstract
Two ultrasonic techniques are well established for pipe inspection and monitoring: highly sensitive ultrasonic spot thickness measurements, which can be presented in C-scan form, or low frequency guided waves that rapidly screen large areas for big defects. Recently there has been a growing interest in pipe inspection and monitoring systems exploring the middle ground between these two techniques by using higher frequency guided waves over shorter distances. In this paper the use of an axial transducer array (more suitable for monitoring applications) or a single axially scanned transducer (more suitable for inspections) that sends and receives dispersive circumferential guided waves around a pipe has been studied. The presence of a defect is detected as a result of a change in the wave travel time around the pipe circumference as a result of the thickness reduction. Both measurement modalities have a pitch between adjacent transducers/measurements. By approximating the pipe to a plate, a finite element model assisted method to evaluate the detection capabilities (probability of detection-POD) of this short range guided wave technique as a function of scan or transducer pitch is presented. The performance of three guided wave modes (A0, S0, and SH1) are compared in a 10mm thick plate. The results help to optimize the pitch and defect sensitivity of the setup. For the parameters investigated in this study, it was found that the S0 mode, at 2MHz.mm, is the most suitable for detecting wide shallow defects. For the same detection capabilities a much wider pitch can be used for S0 mode transducers as compared to SH1 mode transducers. Whereas the SH1 mode, at 3MHz.mm, is better suited to detecting narrow and deep defects using a relatively small pitch. The S0 mode is much less sensitive to these defects. The A0 mode was excluded from the POD analysis because it had a much higher variability in average thickness measurements, at comparable SNRs, compared to the other two modes.
Date Issued
2016-11-24
Date Acceptance
2016-11-23
Citation
NDT & E International, 2016, 86, pp.73-82
ISSN
0963-8695
Publisher
Elsevier
Start Page
73
End Page
82
Journal / Book Title
NDT & E International
Volume
86
Copyright Statement
© 2016 The Authors. Published by Elsevier Ltd.
This is an open access article under the CC BY license (http://creativecommons.org/licenses/BY/4.0/).
This is an open access article under the CC BY license (http://creativecommons.org/licenses/BY/4.0/).
Sponsor
Engineering and Physical Sciences Research Council
Grant Number
EP/I017704/1
Subjects
Science & Technology
Technology
Materials Science, Characterization & Testing
Materials Science
Ultrasound
Guided waves
Pipeline corrosion
Monitoring
Screening
Probability of detection
TOMOGRAPHY
INSPECTION
PLATES
PIPES
REFLECTION
GENERATION
CORROSION
MEDIA
MODES
SKIP
Acoustics
09 Engineering
Publication Status
Published
