Locating acoustic emission sources in complex structures using Gaussian processes

J. Hensman, R. Mills, S.G. Pierce, K. Worden, M. Eaton, Messier-Dowty Ltd (Funder), EPSRC (Funder)

Research output: Contribution to journalArticle

76 Citations (Scopus)

Abstract

A standard technique in the field of non-destructive evaluation is to use acoustic emissions to characterise and locate the damage events that generate them. The location problem is typically posed in terms of the times of flight of the waves and results in an optimisation problem, which can at times be ill-posed. A method is proposed here for learning the relationship between time of flight differences and damage location using data generated by artificially stimulated acoustic emission (AE)-a classic problem of regression. A structure designed to represent a complicated aerospace component was interrogated using a laser to thermoelastically generate AE at multiple points across the structure's surface. Piezoelectric transducers were mounted on the surface of the structure, and the resulting waveforms were recorded. A Gaussian process (GP) with RBF kernels was chosen for regression. Since during AE monitoring not all events can be guaranteed to be detected by all sensors, a GP was trained on data for all possible combinations (subsets) of sensors. The inputs to the GPs were the differences in time of flight between sensors in the set, and the targets were the locations of the source of ultrasonic stimulation. Subsequent (test) data points were located by every possible GP, given the active sensors. It is shown that maps learned on a given structure can generalise effectively to nominally identical structures.
LanguageEnglish
Pages211-223
Number of pages13
JournalMechanical Systems and Signal Processing
Volume24
Issue number1
DOIs
Publication statusPublished - Jan 2010

Fingerprint

Acoustic emissions
Sensors
Piezoelectric transducers
Surface structure
Ultrasonics
Lasers
Monitoring

Keywords

  • acoustic emission
  • gaussian processes
  • laser generated ultrasound

Cite this

Hensman, J., Mills, R., Pierce, S. G., Worden, K., Eaton, M., Messier-Dowty Ltd (Funder), & EPSRC (Funder) (2010). Locating acoustic emission sources in complex structures using Gaussian processes. Mechanical Systems and Signal Processing, 24(1), 211-223. https://doi.org/10.1016/j.ymssp.2009.05.018
Hensman, J. ; Mills, R. ; Pierce, S.G. ; Worden, K. ; Eaton, M. ; Messier-Dowty Ltd (Funder) ; EPSRC (Funder). / Locating acoustic emission sources in complex structures using Gaussian processes. In: Mechanical Systems and Signal Processing. 2010 ; Vol. 24, No. 1. pp. 211-223.
@article{d9105bf427ab4162b2496ad1903000ed,
title = "Locating acoustic emission sources in complex structures using Gaussian processes",
abstract = "A standard technique in the field of non-destructive evaluation is to use acoustic emissions to characterise and locate the damage events that generate them. The location problem is typically posed in terms of the times of flight of the waves and results in an optimisation problem, which can at times be ill-posed. A method is proposed here for learning the relationship between time of flight differences and damage location using data generated by artificially stimulated acoustic emission (AE)-a classic problem of regression. A structure designed to represent a complicated aerospace component was interrogated using a laser to thermoelastically generate AE at multiple points across the structure's surface. Piezoelectric transducers were mounted on the surface of the structure, and the resulting waveforms were recorded. A Gaussian process (GP) with RBF kernels was chosen for regression. Since during AE monitoring not all events can be guaranteed to be detected by all sensors, a GP was trained on data for all possible combinations (subsets) of sensors. The inputs to the GPs were the differences in time of flight between sensors in the set, and the targets were the locations of the source of ultrasonic stimulation. Subsequent (test) data points were located by every possible GP, given the active sensors. It is shown that maps learned on a given structure can generalise effectively to nominally identical structures.",
keywords = "acoustic emission, gaussian processes, laser generated ultrasound",
author = "J. Hensman and R. Mills and S.G. Pierce and K. Worden and M. Eaton and {Messier-Dowty Ltd (Funder)} and {EPSRC (Funder)}",
note = "Also published in Proceedings of ISMA 2008: International Conference on Noise and Vibration Engineering, pp. 3217-3230, January 2008",
year = "2010",
month = "1",
doi = "10.1016/j.ymssp.2009.05.018",
language = "English",
volume = "24",
pages = "211--223",
journal = "Mechanical Systems and Signal Processing",
issn = "0888-3270",
number = "1",

}

Hensman, J, Mills, R, Pierce, SG, Worden, K, Eaton, M, Messier-Dowty Ltd (Funder) & EPSRC (Funder) 2010, 'Locating acoustic emission sources in complex structures using Gaussian processes' Mechanical Systems and Signal Processing, vol. 24, no. 1, pp. 211-223. https://doi.org/10.1016/j.ymssp.2009.05.018

Locating acoustic emission sources in complex structures using Gaussian processes. / Hensman, J.; Mills, R.; Pierce, S.G.; Worden, K.; Eaton, M.; Messier-Dowty Ltd (Funder); EPSRC (Funder).

In: Mechanical Systems and Signal Processing, Vol. 24, No. 1, 01.2010, p. 211-223.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Locating acoustic emission sources in complex structures using Gaussian processes

AU - Hensman, J.

AU - Mills, R.

AU - Pierce, S.G.

AU - Worden, K.

AU - Eaton, M.

AU - Messier-Dowty Ltd (Funder)

AU - EPSRC (Funder)

N1 - Also published in Proceedings of ISMA 2008: International Conference on Noise and Vibration Engineering, pp. 3217-3230, January 2008

PY - 2010/1

Y1 - 2010/1

N2 - A standard technique in the field of non-destructive evaluation is to use acoustic emissions to characterise and locate the damage events that generate them. The location problem is typically posed in terms of the times of flight of the waves and results in an optimisation problem, which can at times be ill-posed. A method is proposed here for learning the relationship between time of flight differences and damage location using data generated by artificially stimulated acoustic emission (AE)-a classic problem of regression. A structure designed to represent a complicated aerospace component was interrogated using a laser to thermoelastically generate AE at multiple points across the structure's surface. Piezoelectric transducers were mounted on the surface of the structure, and the resulting waveforms were recorded. A Gaussian process (GP) with RBF kernels was chosen for regression. Since during AE monitoring not all events can be guaranteed to be detected by all sensors, a GP was trained on data for all possible combinations (subsets) of sensors. The inputs to the GPs were the differences in time of flight between sensors in the set, and the targets were the locations of the source of ultrasonic stimulation. Subsequent (test) data points were located by every possible GP, given the active sensors. It is shown that maps learned on a given structure can generalise effectively to nominally identical structures.

AB - A standard technique in the field of non-destructive evaluation is to use acoustic emissions to characterise and locate the damage events that generate them. The location problem is typically posed in terms of the times of flight of the waves and results in an optimisation problem, which can at times be ill-posed. A method is proposed here for learning the relationship between time of flight differences and damage location using data generated by artificially stimulated acoustic emission (AE)-a classic problem of regression. A structure designed to represent a complicated aerospace component was interrogated using a laser to thermoelastically generate AE at multiple points across the structure's surface. Piezoelectric transducers were mounted on the surface of the structure, and the resulting waveforms were recorded. A Gaussian process (GP) with RBF kernels was chosen for regression. Since during AE monitoring not all events can be guaranteed to be detected by all sensors, a GP was trained on data for all possible combinations (subsets) of sensors. The inputs to the GPs were the differences in time of flight between sensors in the set, and the targets were the locations of the source of ultrasonic stimulation. Subsequent (test) data points were located by every possible GP, given the active sensors. It is shown that maps learned on a given structure can generalise effectively to nominally identical structures.

KW - acoustic emission

KW - gaussian processes

KW - laser generated ultrasound

U2 - 10.1016/j.ymssp.2009.05.018

DO - 10.1016/j.ymssp.2009.05.018

M3 - Article

VL - 24

SP - 211

EP - 223

JO - Mechanical Systems and Signal Processing

T2 - Mechanical Systems and Signal Processing

JF - Mechanical Systems and Signal Processing

SN - 0888-3270

IS - 1

ER -