John Summerscales' monograph (free download) [1] reviews AESL up to circa 1993 and considers system calibration, source location philosophies, location in isotropic media and location in anisotropic composite media.  This page is intended as a follow-on to that monograph for developments since the early 'nineties.

Holroyd [2] suggested that information on the physical location of an acoustic emission source can be extracted in a number of unsophisticated ways :

• note the arrival sequence at two or more sensors,
• move the transducer until the signal has the highest magnitude, and
• use the slower event rise times resulting from longer propagation distances.

Shehadeh et al [3] considered the role of arrival time estimation in AESL on steel pipelines and found a low-amplitude low frequency high-speed wave (LALFHS, suspected to be an extensional Lamb wave mode) and a relatively high-amplitude high-frequency low-speed wave (HAHFLS: flexural wave).  They sought to undertake automatic wave arrival time estimation for source location using several standard techniques: (i) cross-correlation of decomposed signals, (ii) maximum cross-correlation function, (iii) thresholding and (iv) wavelet transforms, and two new methods: (v) sliding window energy technnique, and (vi) cross-correlation combined with the Gabor wavelet transform (GWT).  Techniques which are over-reliant on LALFHS tend to over-estimate the distance and vice versa. The windowed energy technique gave the lowest error and all techniques except GWT gave good location accuracy.  There is an option to use a single sensor as the sole, or supporting, means for linear source location where the two separate arrival times and speeds can be identified.

Baxter et al [4] proposed a novel solution for AE source location in complicated geometric structures.  “Delta T” (ΔT) source location uses an artificial source at a number of locations and records differences in times-of-arrival (TOA) information.  The method is claimed to not require knowledge of the sensor location or wavespeed.  A five-step process is employed:

• determine the area of interest,
• construct a grid on the component.  The grid orientation does not necessarily align with the sensor orientation,
• stimulate each node on the grid with a Hsu-Nielsen source and establish the TOA at each sensor,
• calculate the ΔT maps, and
• generate real data and identify a convergence point indicating the source location.

Results from their initial trials showed considerable improvement over the conventional TOA source location method. The location error reduced from 4.8% using the conventional TOA technique to 1.8% using ΔT location.  Eaton et al [5] used the ΔT technique for AESL on cross-plied carbon fibre composites and again claimed "improvement in location accuracy over the traditional TOA approach in nearly all cases".

Jomdecha et al [6] developed a low-cost AESL system to study the four main types of corrosion (uniform-, pitting-, crevice- and stress-corrosion cracking) found in the petrochemical industry.  AE parameters (hits, counts, tiime and amplitude) were correlated to identify each of the four types of corrosion.

Ciampa and Meo [7] proposed an algorithm based on the differences between stress waves from six surface sensors.  They used the Continuous Wavelet Transform (CWT) squared modulus to identify the times of arrival of the A_o flexural mode Lamb wave and obtain the coordinates of the source and the flexural wave velocity.  A set of non-linear equations were solved by combining global Line Search, and back-tracking techniques associated with a local Newton's iterative method.  The proposed method differs from the triangulation algorithms in not requiring a priori knowledge of (a) thickness, (b) laminate stacking sequence or (c) the anisotropy group velocity.  The system was shown to be applicable to composite structures (quasi-isotropic angle-plied CFRP laminates and quasi-isotropic angle-plied CFRP skin/Nomex honeycomb core sandwich panels.  Location accuracy was ~3mm for the laminate and 2mm for the sandwich panel.

References

1. John Summerscales, Acoustic Emission Source Location in Fibre-Reinforced Composite Materials, ACMC, Plymouth, 2013.  ISBN 978-1-870918-04-6..
2. T Holroyd, Coxmoor's Machine and Systems Condition Monitoring Series: acoustic emission and ultrasonics handbook, Coxmoor Publishing Company, Moreton-in-Marsh, 2000.  ISBN 978-1-90189-207-9.  PU CSH Library.
3. M Shehadeh, JA Steel and RL Reuben, Acoustic emission source location for steel pipe and pipeline applications: the role of arrival time estimation, Proc.IMechE.Part E: Journal of Process Mechanical Engineering, May 2006, E220(2), 121-133.
4. MG Baxter, R Pullin, KM Holford and SL Evans, Delta T source location for acoustic emission, Mechanical Systems and Signal Processing, April 2007, 21(3), 1512–1520.
5. MJ Eaton, R Pullin and KM Holford, Acoustic emission source location in composite materials using Delta T mapping, Composites Part A: Applied Science and Manufacturing, June 2012, 43(6), 856-863.
6. C Jomdecha, A Prateepasen and P Kaewtrakulpong, Study on source location using an acoustic emission system for various corrosion types, NDT&E International, December 2007, 40(8), 584-593.
7. F Ciampa and M Meo, A new algorithm for acoustic emission localization and flexural group velocity determination in anisotropic structures, Composites Part A: Applied Science and Manufacturing, December 2010, 41(12), 1777-1786.