Acoustic emission localization on ship hull structures using a deep learning approach

George  Georgoulas; Vassilios Kappatos; George  Nikolakopoulos

Acoustic emission localization on ship hull structures using a deep learning approach

George Georgoulas
, Vassilios Kappatos
, George Nikolakopoulos

Luleå University of Technology

Publikation: Kapitel i bog/rapport/konference-proceeding › Konferencebidrag i proceedings › Forskning › peer review

Abstract

In this paper, deep belief networks were used for localization of acoustic emission events on ship hull structures. In order to avoid complex and time consuming implementations, the proposed approach uses a simple feature extraction module, which significantly reduces the extremely high dimensionality of the raw signals/data. In simulation experiments, where a stiffened plate model was partially sunk into the water, the localization rate of acoustic emission events in a noise-free environment is greater than 94 %, using only a single sensor.

Originalsprog	Engelsk
Titel	Vibroengineering PROCEDIA
Forlag	JVE International Ltd.
Publikationsdato	2016
Sider	56‑61
Status	Udgivet - 2016
Begivenhed	23rd International Conference of Vibroengineering on Modeling, Identification and Fault Detection in Oil and Gas Equipment and Infrastructures - Istanbul, Tyrkiet Varighed: 7. okt. 2016 → 9. okt. 2016 Konferencens nummer: 23 http://www.jveconferences.com/about-istanbul-conference

Konference

Konference	23rd International Conference of Vibroengineering on Modeling, Identification and Fault Detection in Oil and Gas Equipment and Infrastructures
Nummer	23
Land/Område	Tyrkiet
By	Istanbul
Periode	07/10/2016 → 09/10/2016
Internetadresse	http://www.jveconferences.com/about-istanbul-conference

Navn	Vibroengineering Procedia
Vol/bind	9
ISSN	2345-0533

Dokumenter og links

http://www.jvejournals.com/Vibro/article/VP-17769.html

SDU link

Tjek adgangen til materialet i Syddansk Universitetsbiblioteks søgemaskine

Citationsformater

@inproceedings{96a34c094aa24e38bbbf21cbdd01dae9,

title = "Acoustic emission localization on ship hull structures using a deep learning approach",

abstract = "In this paper, deep belief networks were used for localization of acoustic emission events on ship hull structures. In order to avoid complex and time consuming implementations, the proposed approach uses a simple feature extraction module, which significantly reduces the extremely high dimensionality of the raw signals/data. In simulation experiments, where a stiffened plate model was partially sunk into the water, the localization rate of acoustic emission events in a noise-free environment is greater than 94 \%, using only a single sensor.",

keywords = "Acoustic emission, Deep belief networks, Deep learning, Ship hull, Source localizations",

author = "George Georgoulas and Vassilios Kappatos and George Nikolakopoulos",

year = "2016",

language = "English",

series = "Vibroengineering Procedia",

publisher = "JVE International Ltd. ",

pages = "56‑61",

booktitle = "Vibroengineering PROCEDIA",

address = "Lithuania",

note = "23rd International Conference of Vibroengineering on Modeling, Identification and Fault Detection in Oil and Gas Equipment and Infrastructures, VIBROENGINEERING Conference ; Conference date: 07-10-2016 Through 09-10-2016",

url = "http://www.jveconferences.com/about-istanbul-conference",

}

Georgoulas, G, Kappatos, V & Nikolakopoulos, G 2016, Acoustic emission localization on ship hull structures using a deep learning approach. i Vibroengineering PROCEDIA. JVE International Ltd. , Vibroengineering Procedia, bind 9, s. 56‑61, 23rd International Conference of Vibroengineering on Modeling, Identification and Fault Detection in Oil and Gas Equipment and Infrastructures, Istanbul, Tyrkiet, 07/10/2016. <http://www.jvejournals.com/Vibro/article/VP-17769.html>

Acoustic emission localization on ship hull structures using a deep learning approach. / Georgoulas, George ; Kappatos, Vassilios; Nikolakopoulos, George .
Vibroengineering PROCEDIA. JVE International Ltd. , 2016. s. 56‑61 (Vibroengineering Procedia, Bind 9).

Publikation: Kapitel i bog/rapport/konference-proceeding › Konferencebidrag i proceedings › Forskning › peer review

TY - GEN

T1 - Acoustic emission localization on ship hull structures using a deep learning approach

AU - Georgoulas, George

AU - Kappatos, Vassilios

AU - Nikolakopoulos, George

N1 - Conference code: 23

PY - 2016

Y1 - 2016

N2 - In this paper, deep belief networks were used for localization of acoustic emission events on ship hull structures. In order to avoid complex and time consuming implementations, the proposed approach uses a simple feature extraction module, which significantly reduces the extremely high dimensionality of the raw signals/data. In simulation experiments, where a stiffened plate model was partially sunk into the water, the localization rate of acoustic emission events in a noise-free environment is greater than 94 %, using only a single sensor.

AB - In this paper, deep belief networks were used for localization of acoustic emission events on ship hull structures. In order to avoid complex and time consuming implementations, the proposed approach uses a simple feature extraction module, which significantly reduces the extremely high dimensionality of the raw signals/data. In simulation experiments, where a stiffened plate model was partially sunk into the water, the localization rate of acoustic emission events in a noise-free environment is greater than 94 %, using only a single sensor.

KW - Acoustic emission

KW - Deep belief networks

KW - Deep learning

KW - Ship hull

KW - Source localizations

M3 - Article in proceedings

T3 - Vibroengineering Procedia

SP - 56‑61

BT - Vibroengineering PROCEDIA

PB - JVE International Ltd.

T2 - 23rd International Conference of Vibroengineering on Modeling, Identification and Fault Detection in Oil and Gas Equipment and Infrastructures

Y2 - 7 October 2016 through 9 October 2016

ER -

Acoustic emission localization on ship hull structures using a deep learning approach

Abstract

Konference

Dokumenter og links

SDU link

Fingeraftryk

Citationsformater