An intelligent diagnosis scheme based on generative adversarial learning deep neural networks and its application to planetary gearbox fault pattern recognition

•A new diagnostic model named SDAE-GAN is proposed.•The model combines Generative Adversarial Networks and Stacked Denoising Autoencoders.•The performance of the model is investigated in the planetary gearbox experiment platform.•Results suggest that SDAE-GAN is better than SDAE and other common dia...

Full description

Saved in:
Bibliographic Details
Published in:Neurocomputing (Amsterdam) Vol. 310; pp. 213 - 222
Main Authors: Wang, Zirui, Wang, Jun, Wang, Youren
Format: Journal Article
Language:English
Published: Elsevier B.V 08.10.2018
Subjects:
ISSN:0925-2312, 1872-8286
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract •A new diagnostic model named SDAE-GAN is proposed.•The model combines Generative Adversarial Networks and Stacked Denoising Autoencoders.•The performance of the model is investigated in the planetary gearbox experiment platform.•Results suggest that SDAE-GAN is better than SDAE and other common diagnostic models in classification precision. Planetary gearbox has complex structures and works under various non-stationary operating conditions. The vibration signals of planetary gearbox are complicated and usually polluted by noise and interference. It is difficult to extract effective features of early faults. In addition, there are only a small number of fault samples for planetary gearbox fault diagnosis. All of these increase the difficulty of planetary gearbox fault diagnosis. Aiming at these problems, a novel fault diagnostic method is proposed which combines Generative Adversarial Networks (GAN) and Stacked Denoising Autoencoders (SDAE). The generator of GAN can generate new samples which has similar distribution with original samples from planetary gearbox vibration signals. Then, generated samples are transformed to the discriminator together with original samples which expand the sample size. SDAE is used as the discriminator of GAN which can automatically extract effective fault features from input samples and discriminate their authenticity and fault categories. Through novel adversarial machine learning mechanism, the generator and discriminator are concurrently optimized to enhance the quality of generation samples and the ability of fault mode classification. The experimental results show that the developed SDAE-GAN method for planetary gearbox has good anti-noise ability and achieve better fault diagnosis performance in the case of small samples.
AbstractList •A new diagnostic model named SDAE-GAN is proposed.•The model combines Generative Adversarial Networks and Stacked Denoising Autoencoders.•The performance of the model is investigated in the planetary gearbox experiment platform.•Results suggest that SDAE-GAN is better than SDAE and other common diagnostic models in classification precision. Planetary gearbox has complex structures and works under various non-stationary operating conditions. The vibration signals of planetary gearbox are complicated and usually polluted by noise and interference. It is difficult to extract effective features of early faults. In addition, there are only a small number of fault samples for planetary gearbox fault diagnosis. All of these increase the difficulty of planetary gearbox fault diagnosis. Aiming at these problems, a novel fault diagnostic method is proposed which combines Generative Adversarial Networks (GAN) and Stacked Denoising Autoencoders (SDAE). The generator of GAN can generate new samples which has similar distribution with original samples from planetary gearbox vibration signals. Then, generated samples are transformed to the discriminator together with original samples which expand the sample size. SDAE is used as the discriminator of GAN which can automatically extract effective fault features from input samples and discriminate their authenticity and fault categories. Through novel adversarial machine learning mechanism, the generator and discriminator are concurrently optimized to enhance the quality of generation samples and the ability of fault mode classification. The experimental results show that the developed SDAE-GAN method for planetary gearbox has good anti-noise ability and achieve better fault diagnosis performance in the case of small samples.
Author Wang, Zirui
Wang, Youren
Wang, Jun
Author_xml – sequence: 1
  givenname: Zirui
  surname: Wang
  fullname: Wang, Zirui
  email: zwang162@sheffield.ac.uk
– sequence: 2
  givenname: Jun
  surname: Wang
  fullname: Wang, Jun
– sequence: 3
  givenname: Youren
  surname: Wang
  fullname: Wang, Youren
BookMark eNotkE1OwzAQhS1UJNrCDVj4Agm2EzvJBqmq-JMqsYG15cTj4pLake0WuA1HxVVZPWnmzZunb4FmzjtA6JaSkhIq7nalg8Pg9yUjtC0JLwmrL9Cctg0rWtaKGZqTjvGCVZRdoUWMO0JoQ1k3R78rh61LMI52Cy5hbdXW-WgjjsMH7AH3KoLG3uG8hqCSPQJW-gghqmDViEdQwVm3xRpgwrlHyEMH6cuHz4iV09imrNM02iFf56Dk8TSqbFHhJ6eq0PtvbNRhTHhSKUFwOMDgt86e7Nfo0qgxws2_LtH748Pb-rnYvD69rFebApioUtGYqmN1z4hpuSK86irOOsOYYb0Rg-mHuuaip6CBt6CFgZ6LptdCNU1NtdDVEt2fcyE_OVoIMg4W3ADa5jZJam8lJfLEW-7kmbc88ZaEy8y7-gPNC35G
ContentType Journal Article
Copyright 2018 Elsevier Ltd
Copyright_xml – notice: 2018 Elsevier Ltd
DOI 10.1016/j.neucom.2018.05.024
DatabaseTitleList
DeliveryMethod fulltext_linktorsrc
Discipline Computer Science
EISSN 1872-8286
EndPage 222
ExternalDocumentID S0925231218305617
GroupedDBID ---
--K
--M
.DC
.~1
0R~
123
1B1
1~.
1~5
4.4
457
4G.
53G
5VS
7-5
71M
8P~
9JM
9JN
AABNK
AACTN
AADPK
AAEDT
AAEDW
AAIAV
AAIKJ
AAKOC
AALRI
AAOAW
AAQFI
AAXLA
AAXUO
AAYFN
ABBOA
ABCQJ
ABFNM
ABJNI
ABMAC
ABYKQ
ACDAQ
ACGFS
ACRLP
ACZNC
ADBBV
ADEZE
AEBSH
AEKER
AENEX
AFKWA
AFTJW
AFXIZ
AGHFR
AGUBO
AGWIK
AGYEJ
AHHHB
AHZHX
AIALX
AIEXJ
AIKHN
AITUG
AJOXV
ALMA_UNASSIGNED_HOLDINGS
AMFUW
AMRAJ
AOUOD
AXJTR
BKOJK
BLXMC
CS3
DU5
EBS
EFJIC
EFLBG
EJD
EO8
EO9
EP2
EP3
F5P
FDB
FIRID
FNPLU
FYGXN
G-Q
GBLVA
GBOLZ
IHE
J1W
KOM
LG9
M41
MO0
MOBAO
N9A
O-L
O9-
OAUVE
OZT
P-8
P-9
P2P
PC.
Q38
RIG
ROL
RPZ
SDF
SDG
SDP
SES
SPC
SPCBC
SSN
SSV
SSZ
T5K
ZMT
~G-
ID FETCH-LOGICAL-e263t-7f3924b20f85a05393529f22f2bf6cfbc4456b1ede58ed6feb567bd6a7741d6d3
ISICitedReferencesCount 321
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000437299800019&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 0925-2312
IngestDate Fri Feb 23 02:30:23 EST 2024
IsPeerReviewed true
IsScholarly true
Keywords Fault diagnosis
Generative adversarial networks
Deep stacked denoising autoencoders
Adversarial machine learning
Planetary gearbox
Small samples
Language English
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-e263t-7f3924b20f85a05393529f22f2bf6cfbc4456b1ede58ed6feb567bd6a7741d6d3
PageCount 10
ParticipantIDs elsevier_sciencedirect_doi_10_1016_j_neucom_2018_05_024
PublicationCentury 2000
PublicationDate 2018-10-08
PublicationDateYYYYMMDD 2018-10-08
PublicationDate_xml – month: 10
  year: 2018
  text: 2018-10-08
  day: 08
PublicationDecade 2010
PublicationTitle Neurocomputing (Amsterdam)
PublicationYear 2018
Publisher Elsevier B.V
Publisher_xml – name: Elsevier B.V
References Ratliff, Burden, Sastry (bib0033) 2013
Sun, Wang (bib0003) 2017; 38
Zhang, Su, Pu (bib0014) 2014; 50
Sawalhi, Randall (bib0002) 2014; 42
Tamilselvan, Wang (bib0019) 2013; 115
Vincent, Larochelle, Lajoie (bib0030) 2010; 11
Khazaee, Ahmadi, Omid (bib0037) 2014; 228
Bengio (bib0028) 2009; 2
Li, Yang, Li (bib0022) 2017; 91
Erhan, Bengio, Courville (bib0029) 2010; 11
Zhang, Khawaja, Patrick (bib0007) 2008; 13
Golafshan, Sanliturk (bib0013) 2016; 70–71
Qing, Guiming, Qingfei (bib0012) 2011
2015.
Zhao, Jia (bib0006) 2017; 94
Weidong, Shuseng (bib0011) 2015; 36
Laha (bib0008) 2017; 100
Swami, Sharma, Jain (bib0034) 2015; 70
Jia, Lei, Guo (bib0017) 2018; 272
Gan, Wang, Zhu (bib0020) 2016; 72–73
Feng, Liang, Chu (bib0004) 2013; 38
Sun, Shao, Yan (bib0021) 2016; 52
Yang, Pan, Li (bib0010) 2015; 34
K. Diederik and B. Jimmy. ADAM: a method for stochastic optimization. arXiv preprint arXiv
Lei, Lin, Zuo, He (bib0001) 2014; 48
Chen, Zi, He (bib0005) 2013; 38
Liu, Qu, Zuo (bib0038) 2013; 67
Coates, Y. Ng, Lee (bib0035) 2011; 15
Xu, Zhao, Ma, Hou (bib0009) 2016; 31
Lei, Jia, Zhou (bib0016) 2015; 51
Wang, Gou, Duan (bib0032) 2017; 43
Guo, Chen, Shen (bib0015) 2016; 93
2014.
Shao, Jiang, Zhao (bib0024) 2017; 95
Goodfellow, Pougetabadie, Mirza (bib0025) 2014; 3
A. Radford, L. Metz, and S. Chintala. Unsupervised representation learning with deep convolutional generative adversarial networks. arXiv preprint arXiv
Hinton, Salakhutdinov (bib0023) 2006; 313
2016.
Hao, Yi, Qu (bib0018) 2018; 275
A. Odena, C. Olah, and J. Shlens. Conditional image synthesis with auxiliary classifier GANs. arXiv preprint arXiv
Lei, Jia, Lin (bib0036) 2016; 63
References_xml – volume: 94
  start-page: 129
  year: 2017
  end-page: 147
  ident: bib0006
  article-title: A novel strategy for signal denoising using reweighted SVD and its applications to weak fault feature enhancement of rotating machinery
  publication-title: Mech. Syst. Signal Process.
– volume: 51
  start-page: 49
  year: 2015
  end-page: 56
  ident: bib0016
  article-title: A deep learning-based method for machinery health monitoring with big data
  publication-title: J. Mech. Eng.
– volume: 313
  start-page: 504
  year: 2006
  end-page: 507
  ident: bib0023
  article-title: Reducing the dimensionality of data with neural networks
  publication-title: Science
– start-page: 1
  year: 2011
  end-page: 4
  ident: bib0012
  article-title: The research of pattern recognition of gear pump based on EMD and KPCA-SVM
  publication-title: Proceedings of the International Conference on System Science, Engineering Design and Manufacturing Informatization(ICSEM)
– volume: 115
  start-page: 124
  year: 2013
  end-page: 135
  ident: bib0019
  article-title: Failure diagnosis using deep belief learning based health state classification
  publication-title: Reliab. Eng. Syst. Saf.
– volume: 11
  start-page: 3371
  year: 2010
  end-page: 3408
  ident: bib0030
  article-title: Stacked denoising autoencoders: learning useful representations in a deep network with a local denoising criterion
  publication-title: J. Mach. Learn. Res.
– volume: 15
  start-page: 215
  year: 2011
  end-page: 223
  ident: bib0035
  article-title: An analysis of single-layer networks in unsupervised feature learning
  publication-title: J. Mach. Learn. Res.
– volume: 31
  start-page: 219
  year: 2016
  end-page: 226
  ident: bib0009
  article-title: Denoising method based on dual-tree complex wavelet transform and MCA and its application in gear fault diagnosis
  publication-title: J. Aerosp. Power
– volume: 38
  start-page: 165
  year: 2013
  end-page: 205
  ident: bib0004
  article-title: Recent advances in time frequency analysis methods for machinery fault diagnosis: a review with application examples
  publication-title: Mech. Syst. Signal Process.
– reference: , 2014.
– volume: 95
  start-page: 187
  year: 2017
  end-page: 204
  ident: bib0024
  article-title: A novel deep autoencoder feature learning method for rotating machinery fault diagnosis
  publication-title: Mech. Syst. Signal Process.
– volume: 43
  start-page: 321
  year: 2017
  end-page: 332
  ident: bib0032
  article-title: Generative adversarial networks: the state of the art and beyond
  publication-title: Acta Autom. Sin.
– reference: K. Diederik and B. Jimmy. ADAM: a method for stochastic optimization. arXiv preprint arXiv:
– volume: 13
  start-page: 558
  year: 2008
  end-page: 565
  ident: bib0007
  article-title: Blind deconvolution denoising for helicopter vibration signals
  publication-title: IEEE/ASME Trans. Mechatron.
– volume: 272
  start-page: 619
  year: 2018
  end-page: 628
  ident: bib0017
  article-title: A neural network constructed by deep learning technique and its application to intelligent fault diagnosis of machines
  publication-title: Neurocomputing
– volume: 228
  start-page: 21
  year: 2014
  end-page: 32
  ident: bib0037
  article-title: Classifier fusion of vibration and acoustic signals for fault diagnosis and classification of planetary gears based on Dempster–Shafer evidence theory
  publication-title: J. Process Mech. Eng.
– volume: 38
  start-page: 020892.1
  year: 2017
  end-page: 14
  ident: bib0003
  article-title: Advance in the study on fault diagnosis of helicopter planetary gears
  publication-title: Acta Aeronautica Et Astronautica Sinica
– volume: 2
  start-page: 1
  year: 2009
  end-page: 127
  ident: bib0028
  article-title: Learning deep architectures for AI
  publication-title: Found. Trends® Mach. Learn.
– volume: 3
  start-page: 2672
  year: 2014
  end-page: 2680
  ident: bib0025
  article-title: Generative adversarial networks
  publication-title: Adv. Neural Inf. Process. Syst.
– reference: , 2015.
– volume: 38
  start-page: 549
  year: 2013
  end-page: 568
  ident: bib0005
  article-title: Adaptive redundant multiwavelet denoising with improved neighboring coefficients for gearbox fault detection
  publication-title: Mech. Syst. Signal Process.
– start-page: 917
  year: 2013
  end-page: 924
  ident: bib0033
  article-title: Characterization and computation of local Nash equilibria in continuous games
  publication-title: Proceedings of the Fifty First Annual Allerton Conference on Communication, Control, and Computing
– reference: A. Odena, C. Olah, and J. Shlens. Conditional image synthesis with auxiliary classifier GANs. arXiv preprint arXiv:
– reference: A. Radford, L. Metz, and S. Chintala. Unsupervised representation learning with deep convolutional generative adversarial networks. arXiv preprint arXiv:
– volume: 63
  start-page: 1
  year: 2016
  ident: bib0036
  article-title: An intelligent fault diagnosis method using unsupervised feature learning towards mechanical big data
  publication-title: IEEE Trans. Ind. Electron.
– volume: 70
  start-page: 1
  year: 2015
  end-page: 12
  ident: bib0034
  article-title: Speech enhancement by noise driven adaptation of perceptual scales and thresholds of continuous wavelet transform coefficients
  publication-title: Speech Commun.
– volume: 11
  start-page: 625
  year: 2010
  end-page: 660
  ident: bib0029
  article-title: Why does unsupervised pre-training help deep learning?
  publication-title: J. Mach. Learn. Res.
– volume: 70–71
  start-page: 36
  year: 2016
  end-page: 50
  ident: bib0013
  article-title: SVD and Hankel matrix based de-noising approach for ball bearing fault detection and its assessment using artificial faults
  publication-title: Mech. Syst. Signal Process.
– volume: 42
  start-page: 368
  year: 2014
  end-page: 376
  ident: bib0002
  article-title: Gear parameter identification in a wind turbine gearbox using vibration signals
  publication-title: Mech. Syst. Signal Process.
– volume: 72–73
  start-page: 92
  year: 2016
  end-page: 104
  ident: bib0020
  article-title: Construction of hierarchical diagnosis network based on deep learning and its application in the fault pattern recognition of rolling element bearings
  publication-title: Mech. Syst. Signal Process.
– volume: 91
  start-page: 295
  year: 2017
  end-page: 312
  ident: bib0022
  article-title: A fault diagnosis scheme for planetary gearboxes using modified multi-scale symbolic dynamic entropy and mRMR feature selection
  publication-title: Mech. Syst. Signal Process.
– volume: 50
  start-page: 70
  year: 2014
  end-page: 77
  ident: bib0014
  article-title: Gear fault diagnosis method using ensemble empirical mode decomposition energy distribution and gray similar incidence
  publication-title: J. Mech. Eng.
– volume: 52
  start-page: 65
  year: 2016
  end-page: 71
  ident: bib0021
  article-title: Induction motor fault diagnosis based on deep neural network of sparse auto-encoder
  publication-title: J. Mech. Eng.
– volume: 100
  start-page: 157
  year: 2017
  end-page: 163
  ident: bib0008
  article-title: Enhancement of fault diagnosis of rolling element bearing using maximum kurtosis fast nonlocal means denoising
  publication-title: Measurement
– volume: 36
  start-page: 1861
  year: 2015
  end-page: 1870
  ident: bib0011
  article-title: Overall-improved fault diagnosis approach based on support vector machine
  publication-title: Chin. J. Sci. Instrum.
– volume: 275
  start-page: 2111
  year: 2018
  end-page: 2125
  ident: bib0018
  article-title: A novel adaptive fault detection methodology for complex system using deep belief networks and multiple models: a case study on cryogenic propellant loading system
  publication-title: Neurocomputing
– volume: 67
  start-page: 1217
  year: 2013
  end-page: 1230
  ident: bib0038
  article-title: Fault level diagnosis for planetary gearboxes using hybrid kernel feature selection and kernel fisher discriminant analysis
  publication-title: Int. J. Adv. Manuf. Technol.
– volume: 48
  start-page: 292
  year: 2014
  end-page: 305
  ident: bib0001
  article-title: Condition monitoring and fault diagnosis of planetary gearboxes: a review
  publication-title: Measurement
– reference: , 2016.
– volume: 93
  start-page: 490
  year: 2016
  end-page: 502
  ident: bib0015
  article-title: Hierarchical adaptive deep convolution neural network and its application to bearing fault diagnosis
  publication-title: Measurement
– volume: 34
  start-page: 49
  year: 2015
  end-page: 54
  ident: bib0010
  article-title: A novel incremental semi-supervised VPMCD gear fault on-line diagnosis method
  publication-title: J. Vib. Shock
SSID ssj0017129
Score 2.6464396
Snippet •A new diagnostic model named SDAE-GAN is proposed.•The model combines Generative Adversarial Networks and Stacked Denoising Autoencoders.•The performance of...
SourceID elsevier
SourceType Publisher
StartPage 213
SubjectTerms Adversarial machine learning
Deep stacked denoising autoencoders
Fault diagnosis
Generative adversarial networks
Planetary gearbox
Small samples
Title An intelligent diagnosis scheme based on generative adversarial learning deep neural networks and its application to planetary gearbox fault pattern recognition
URI https://dx.doi.org/10.1016/j.neucom.2018.05.024
Volume 310
WOSCitedRecordID wos000437299800019&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVESC
  databaseName: Elsevier SD Freedom Collection Journals 2021
  customDbUrl:
  eissn: 1872-8286
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0017129
  issn: 0925-2312
  databaseCode: AIEXJ
  dateStart: 19950101
  isFulltext: true
  titleUrlDefault: https://www.sciencedirect.com
  providerName: Elsevier
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtZ1Li9swEMdFuttDL32XdvtgDr0Fl0R-H03Z0u5hKXRLQy_Gei0JWSck9pKP08-xn64zluSY5NIWeglB4ER4foxGo5m_GHufYSiUyYkKkkRUQWQyGYgqNgHG8iqPZJhE0l42kV5eZrNZ_nU0uvO9MLfLtK6z3S5f_1dT4xgam1pn_8Lc_Y_iAH5Ho-Mnmh0__8jwRd1pQFidzYaSq1RLN9-OcR-rb_SY1i1FZwTXneJ0VzpU0bXM26q7wWPpkyVK6_WY9C5xsLbV4tv-rGFw8E3h65pqZhuqwLvGx8VqNzZVu2xItpUyjuO-TslRsPCiUS0uoN3FEi5lUdyQcoMiTPsUxQ-X0_4537Tzw8GLtj4cQg_mO9xcPmPqxGX3SbajRhubreRxgKGoddza-uos5V0X_NCZh65I1rvjcLCyc9sBfbRo2PzF4gO-Uaogokl1aq62uftAjvsbTYVmgr6Qdl_pPXbK0zhHj3pafDmfXfRnWOmUW6VHN3XfuNlVFx7_1yAaGkQ4V4_ZQ7c1gcIi9YSNdP2UPfLXfoBbBZ6xX0UNA8KgJwwsYdARBqsa9oTBgDDwhAERBpYw8IQBEgZIGAwIg2YFPWHgCIOOMHCEwYCw5-z7p_Orj58Dd9FHoHkSNkFqMEqPBJ-YLK5wVchxV5Abzg0XJpFGyAjDfDHVSseZVonRIk5SoZIK9y5TlajwBTupV7V-yUBlSpPGnIwMSfDnmdA8jKUUMsm5NNUrlvq3XLoY08aOJdJQ-pLHRWntU5J9yklcon3O_vnJ1-zBHvQ37KTZtPotuy9vm_l2885h8xsjOq4I
linkProvider Elsevier
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=An+intelligent+diagnosis+scheme+based+on+generative+adversarial+learning+deep+neural+networks+and+its+application+to+planetary+gearbox+fault+pattern+recognition&rft.jtitle=Neurocomputing+%28Amsterdam%29&rft.au=Wang%2C+Zirui&rft.au=Wang%2C+Jun&rft.au=Wang%2C+Youren&rft.date=2018-10-08&rft.pub=Elsevier+B.V&rft.issn=0925-2312&rft.eissn=1872-8286&rft.volume=310&rft.spage=213&rft.epage=222&rft_id=info:doi/10.1016%2Fj.neucom.2018.05.024&rft.externalDocID=S0925231218305617
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0925-2312&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0925-2312&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0925-2312&client=summon