A size-controlled AFGAN model for ship acoustic fault expansion

Linke Zhang, Na Wei, Xuhao Du, Shuping Wang

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Identifying changes in the properties of acoustical sources based on a small number of sample data from measurements has been a challenge for decades. Typical problems are the increasing sound power from a vibrating source, decreasing transmission loss of a structure, and decreasing insertion loss of vibration mounts. Limited access to structural and acoustical data from complex acoustical systems makes it challenging to extract complete information of the system and, in practice, often only a small amount of test data is available for detecting changes. Although sample expansion via interpolation can be implemented based on the priori knowledge of the system, the size of the expanded samples also affects identification performance. In this paper, a generative adversarial network (GAN) is employed to expand the acoustic fault vibration signals, and an Acoustic Fault Generative Adversarial Network (AFGAN) model is proposed. Moreover, a size-controlled AFGAN is designed, which includes two sub-models: the generator sub-model generates expanded samples and also determines the optimal sample size based on the information entropy equivalence principle, while the discriminator sub-model outputs the probabilities of the input samples belonging to the real samples and provides the generator with information to guide sample size considerations. Some real data experiments have been conducted to verify the effectiveness of this method.

Original languageEnglish
Article number2292
JournalApplied Sciences (Switzerland)
Volume9
Issue number11
DOIs
Publication statusPublished - 1 Jun 2019

Fingerprint

ships
Ships
Acoustics
expansion
acoustics
Discriminators
Insertion losses
Vibrations (mechanical)
Large scale systems
Interpolation
Entropy
generators
Acoustic waves
vibration
discriminators
transmission loss
insertion loss
interpolation
equivalence
Experiments

Cite this

Zhang, Linke ; Wei, Na ; Du, Xuhao ; Wang, Shuping. / A size-controlled AFGAN model for ship acoustic fault expansion. In: Applied Sciences (Switzerland). 2019 ; Vol. 9, No. 11.
@article{441b8b04d4de453da9838454c0169da0,
title = "A size-controlled AFGAN model for ship acoustic fault expansion",
abstract = "Identifying changes in the properties of acoustical sources based on a small number of sample data from measurements has been a challenge for decades. Typical problems are the increasing sound power from a vibrating source, decreasing transmission loss of a structure, and decreasing insertion loss of vibration mounts. Limited access to structural and acoustical data from complex acoustical systems makes it challenging to extract complete information of the system and, in practice, often only a small amount of test data is available for detecting changes. Although sample expansion via interpolation can be implemented based on the priori knowledge of the system, the size of the expanded samples also affects identification performance. In this paper, a generative adversarial network (GAN) is employed to expand the acoustic fault vibration signals, and an Acoustic Fault Generative Adversarial Network (AFGAN) model is proposed. Moreover, a size-controlled AFGAN is designed, which includes two sub-models: the generator sub-model generates expanded samples and also determines the optimal sample size based on the information entropy equivalence principle, while the discriminator sub-model outputs the probabilities of the input samples belonging to the real samples and provides the generator with information to guide sample size considerations. Some real data experiments have been conducted to verify the effectiveness of this method.",
keywords = "Acoustic fault identification, Generative adversarial network, Sample expansion, Size control",
author = "Linke Zhang and Na Wei and Xuhao Du and Shuping Wang",
year = "2019",
month = "6",
day = "1",
doi = "10.3390/app9112292",
language = "English",
volume = "9",
journal = "Applied Sciences",
issn = "2076-3417",
publisher = "MDPI AG",
number = "11",

}

A size-controlled AFGAN model for ship acoustic fault expansion. / Zhang, Linke; Wei, Na; Du, Xuhao; Wang, Shuping.

In: Applied Sciences (Switzerland), Vol. 9, No. 11, 2292, 01.06.2019.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A size-controlled AFGAN model for ship acoustic fault expansion

AU - Zhang, Linke

AU - Wei, Na

AU - Du, Xuhao

AU - Wang, Shuping

PY - 2019/6/1

Y1 - 2019/6/1

N2 - Identifying changes in the properties of acoustical sources based on a small number of sample data from measurements has been a challenge for decades. Typical problems are the increasing sound power from a vibrating source, decreasing transmission loss of a structure, and decreasing insertion loss of vibration mounts. Limited access to structural and acoustical data from complex acoustical systems makes it challenging to extract complete information of the system and, in practice, often only a small amount of test data is available for detecting changes. Although sample expansion via interpolation can be implemented based on the priori knowledge of the system, the size of the expanded samples also affects identification performance. In this paper, a generative adversarial network (GAN) is employed to expand the acoustic fault vibration signals, and an Acoustic Fault Generative Adversarial Network (AFGAN) model is proposed. Moreover, a size-controlled AFGAN is designed, which includes two sub-models: the generator sub-model generates expanded samples and also determines the optimal sample size based on the information entropy equivalence principle, while the discriminator sub-model outputs the probabilities of the input samples belonging to the real samples and provides the generator with information to guide sample size considerations. Some real data experiments have been conducted to verify the effectiveness of this method.

AB - Identifying changes in the properties of acoustical sources based on a small number of sample data from measurements has been a challenge for decades. Typical problems are the increasing sound power from a vibrating source, decreasing transmission loss of a structure, and decreasing insertion loss of vibration mounts. Limited access to structural and acoustical data from complex acoustical systems makes it challenging to extract complete information of the system and, in practice, often only a small amount of test data is available for detecting changes. Although sample expansion via interpolation can be implemented based on the priori knowledge of the system, the size of the expanded samples also affects identification performance. In this paper, a generative adversarial network (GAN) is employed to expand the acoustic fault vibration signals, and an Acoustic Fault Generative Adversarial Network (AFGAN) model is proposed. Moreover, a size-controlled AFGAN is designed, which includes two sub-models: the generator sub-model generates expanded samples and also determines the optimal sample size based on the information entropy equivalence principle, while the discriminator sub-model outputs the probabilities of the input samples belonging to the real samples and provides the generator with information to guide sample size considerations. Some real data experiments have been conducted to verify the effectiveness of this method.

KW - Acoustic fault identification

KW - Generative adversarial network

KW - Sample expansion

KW - Size control

UR - http://www.scopus.com/inward/record.url?scp=85067234919&partnerID=8YFLogxK

U2 - 10.3390/app9112292

DO - 10.3390/app9112292

M3 - Article

VL - 9

JO - Applied Sciences

JF - Applied Sciences

SN - 2076-3417

IS - 11

M1 - 2292

ER -