High Frequency Guided Wave Virtual Array SAFT

Roberts, R. A.; Pardini, Allan F.; Diaz, Aaron A.

doi:10.1063/1.1570138

Cited by 1 publication

(1 citation statement)

References 2 publications

(3 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the other hand, array processing techniques that use individual signals are generally used for imaging and focusing applications [21][22][23][24][25][26][27]. However, these algorithms are computationally expensive, and might require the individual excitation of transducers [26], or a phased-array focusing of the transducers [23], rather than synthetic focusing, where all of the transducers in the array are excited at the same time [21].…”

Section: Introductionmentioning

confidence: 99%

Detection of Defects Using Spatial Variances of Guided-Wave Modes in Testing of Pipes

2018

View full text Add to dashboard Cite

In the past decade, guided-wave testing has attracted the attention of the non-destructive testing industry for pipeline inspections. This technology enables the long-range assessment of pipelines’ integrity, which significantly reduces the expenditure of testing in terms of cost and time. Guided-wave testing collars consist of several linearly placed arrays of transducers around the circumference of the pipe, which are called rings, and can generate unidirectional axisymmetric elastic waves. The current propagation routine of the device generates a single time-domain signal by doing a phase-delayed summation of each array element. The segments where the energy of the signal is above the local noise region are reported as anomalies by the inspectors. Nonetheless, the main goal of guided-wave inspection is the detection of axisymmetric waves generated by the features within the pipes. In this paper, instead of processing a single signal obtained from the general propagation routine, we propose to process signals that are directly obtained from all of the array elements. We designed an axisymmetric wave detection algorithm, which is validated by laboratory trials on real-pipe data with two defects on different locations with varying cross-sectional area (CSA) sizes of 2% and 3% for the first defect, and 4% and 5% for the second defect. The results enabled the detection of defects with low signal-to-noise ratios (SNR), which were almost buried in the noise level. These results are reported with regard to the three different developed methods with varying excitation frequencies of 30 kHz, 34 kHz, and 37 kHz. The tests demonstrated the advantage of using the information received from all of the elements rather than a single signal.

show abstract