A LSTM Neural Network Approach using Vibration Signals for Classifying Faults in a Gearbox

Ruben Medina; Jean Carlo Macancela; Pablo Lucero; Diego Cabrera; Chuan Li; Mariela Cerrada; Rene Vinicio Sanchez; Rafael E. Vasquez

doi:10.1109/SDPC.2019.00045

A LSTM Neural Network Approach using Vibration Signals for Classifying Faults in a Gearbox

Ruben Medina, Jean Carlo Macancela, Pablo Lucero, Diego Cabrera, Chuan Li, Mariela Cerrada, Rene Vinicio Sanchez, Rafael E. Vasquez

Research and Development Group in Industrial Technologies (GIDTEC)

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

4 Scopus citations

Abstract

A deep learning based method for classifying multi-class faults in a gearbox is presented. A set of 900 vibration signals representing the normal condition and nine faults comprises the dataset used in this research. The recorded vibration signals are pre-processed for extracting the first and second derivatives as well as the first five Intrinsic Mode Functions (IMFs) by applying the Empirical Mode Decomposition (EMD) method. A 2D representation of these signals is the feature space used for classifying ten conditions of a gearbox using a Long Short Term Memory (LSTM) neural network. The 2D feature space is subdivided along the temporal axis in segments of the same size as the LSTM network. These segments are classified and a voting systems is proposed for attaining the signal classification. A 10-fold cross-validation is used for evaluating the proposed deep learning model. An average accuracy up to 99.4 % for classifying the faults is attained during the cross-validation.

Original language	English
Title of host publication	Proceedings - 2019 International Conference on Sensing, Diagnostics, Prognostics, and Control, SDPC 2019
Editors	Chuan Li, Shaohui Zhang, Jianyu Long, Diego Cabrera, Ping Ding
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	208-214
Number of pages	7
ISBN (Electronic)	9781728101996
DOIs	https://doi.org/10.1109/SDPC.2019.00045
State	Published - Aug 2019
Event	2019 International Conference on Sensing, Diagnostics, Prognostics, and Control, SDPC 2019 - Beijing, China Duration: 15 Aug 2019 → 17 Aug 2019

Publication series

Name	Proceedings - 2019 International Conference on Sensing, Diagnostics, Prognostics, and Control, SDPC 2019

Conference

Conference	2019 International Conference on Sensing, Diagnostics, Prognostics, and Control, SDPC 2019
Country/Territory	China
City	Beijing
Period	15/08/19 → 17/08/19

Bibliographical note

Funding Information:
ACKNOWLEDGMENT This work was supported in part by the MOST Science and Technology Partnership Program (KY201802006), and Universidad Politécnica Salesiana through the research group GIDTEC.

Publisher Copyright:
© 2019 IEEE.

Copyright:
Copyright 2020 Elsevier B.V., All rights reserved.

Keywords

Deep learning
Faults classification
Gearboxes
Long short term memory networks
Vibration signals

Access to Document

10.1109/SDPC.2019.00045

Cite this

Medina, R., Macancela, J. C., Lucero, P., Cabrera, D., Li, C., Cerrada, M., Sanchez, R. V., & Vasquez, R. E. (2019). A LSTM Neural Network Approach using Vibration Signals for Classifying Faults in a Gearbox. In C. Li, S. Zhang, J. Long, D. Cabrera, & P. Ding (Eds.), Proceedings - 2019 International Conference on Sensing, Diagnostics, Prognostics, and Control, SDPC 2019 (pp. 208-214). Article 9169022 (Proceedings - 2019 International Conference on Sensing, Diagnostics, Prognostics, and Control, SDPC 2019). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/SDPC.2019.00045

Medina, Ruben ; Macancela, Jean Carlo ; Lucero, Pablo et al. / A LSTM Neural Network Approach using Vibration Signals for Classifying Faults in a Gearbox. Proceedings - 2019 International Conference on Sensing, Diagnostics, Prognostics, and Control, SDPC 2019. editor / Chuan Li ; Shaohui Zhang ; Jianyu Long ; Diego Cabrera ; Ping Ding. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 208-214 (Proceedings - 2019 International Conference on Sensing, Diagnostics, Prognostics, and Control, SDPC 2019).

@inproceedings{daaa99ec5a864894981626b543013c88,

title = "A LSTM Neural Network Approach using Vibration Signals for Classifying Faults in a Gearbox",

abstract = "A deep learning based method for classifying multi-class faults in a gearbox is presented. A set of 900 vibration signals representing the normal condition and nine faults comprises the dataset used in this research. The recorded vibration signals are pre-processed for extracting the first and second derivatives as well as the first five Intrinsic Mode Functions (IMFs) by applying the Empirical Mode Decomposition (EMD) method. A 2D representation of these signals is the feature space used for classifying ten conditions of a gearbox using a Long Short Term Memory (LSTM) neural network. The 2D feature space is subdivided along the temporal axis in segments of the same size as the LSTM network. These segments are classified and a voting systems is proposed for attaining the signal classification. A 10-fold cross-validation is used for evaluating the proposed deep learning model. An average accuracy up to 99.4 % for classifying the faults is attained during the cross-validation. ",

keywords = "Deep learning, Faults classification, Gearboxes, Long short term memory networks, Vibration signals",

author = "Ruben Medina and Macancela, {Jean Carlo} and Pablo Lucero and Diego Cabrera and Chuan Li and Mariela Cerrada and Sanchez, {Rene Vinicio} and Vasquez, {Rafael E.}",

note = "Funding Information: ACKNOWLEDGMENT This work was supported in part by the MOST Science and Technology Partnership Program (KY201802006), and Universidad Polit{\'e}cnica Salesiana through the research group GIDTEC. Publisher Copyright: {\textcopyright} 2019 IEEE. Copyright: Copyright 2020 Elsevier B.V., All rights reserved.; 2019 International Conference on Sensing, Diagnostics, Prognostics, and Control, SDPC 2019 ; Conference date: 15-08-2019 Through 17-08-2019",

year = "2019",

month = aug,

doi = "10.1109/SDPC.2019.00045",

language = "English",

series = "Proceedings - 2019 International Conference on Sensing, Diagnostics, Prognostics, and Control, SDPC 2019",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "208--214",

editor = "Chuan Li and Shaohui Zhang and Jianyu Long and Diego Cabrera and Ping Ding",

booktitle = "Proceedings - 2019 International Conference on Sensing, Diagnostics, Prognostics, and Control, SDPC 2019",

address = "United States",

}

Medina, R, Macancela, JC, Lucero, P, Cabrera, D, Li, C, Cerrada, M , Sanchez, RV & Vasquez, RE 2019, A LSTM Neural Network Approach using Vibration Signals for Classifying Faults in a Gearbox. in C Li, S Zhang, J Long, D Cabrera & P Ding (eds), Proceedings - 2019 International Conference on Sensing, Diagnostics, Prognostics, and Control, SDPC 2019., 9169022, Proceedings - 2019 International Conference on Sensing, Diagnostics, Prognostics, and Control, SDPC 2019, Institute of Electrical and Electronics Engineers Inc., pp. 208-214, 2019 International Conference on Sensing, Diagnostics, Prognostics, and Control, SDPC 2019, Beijing, China, 15/08/19. https://doi.org/10.1109/SDPC.2019.00045

A LSTM Neural Network Approach using Vibration Signals for Classifying Faults in a Gearbox. / Medina, Ruben; Macancela, Jean Carlo; Lucero, Pablo et al.
Proceedings - 2019 International Conference on Sensing, Diagnostics, Prognostics, and Control, SDPC 2019. ed. / Chuan Li; Shaohui Zhang; Jianyu Long; Diego Cabrera; Ping Ding. Institute of Electrical and Electronics Engineers Inc., 2019. p. 208-214 9169022 (Proceedings - 2019 International Conference on Sensing, Diagnostics, Prognostics, and Control, SDPC 2019).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - A LSTM Neural Network Approach using Vibration Signals for Classifying Faults in a Gearbox

AU - Medina, Ruben

AU - Macancela, Jean Carlo

AU - Lucero, Pablo

AU - Cabrera, Diego

AU - Li, Chuan

AU - Cerrada, Mariela

AU - Sanchez, Rene Vinicio

AU - Vasquez, Rafael E.

N1 - Funding Information: ACKNOWLEDGMENT This work was supported in part by the MOST Science and Technology Partnership Program (KY201802006), and Universidad Politécnica Salesiana through the research group GIDTEC. Publisher Copyright: © 2019 IEEE. Copyright: Copyright 2020 Elsevier B.V., All rights reserved.

PY - 2019/8

Y1 - 2019/8

N2 - A deep learning based method for classifying multi-class faults in a gearbox is presented. A set of 900 vibration signals representing the normal condition and nine faults comprises the dataset used in this research. The recorded vibration signals are pre-processed for extracting the first and second derivatives as well as the first five Intrinsic Mode Functions (IMFs) by applying the Empirical Mode Decomposition (EMD) method. A 2D representation of these signals is the feature space used for classifying ten conditions of a gearbox using a Long Short Term Memory (LSTM) neural network. The 2D feature space is subdivided along the temporal axis in segments of the same size as the LSTM network. These segments are classified and a voting systems is proposed for attaining the signal classification. A 10-fold cross-validation is used for evaluating the proposed deep learning model. An average accuracy up to 99.4 % for classifying the faults is attained during the cross-validation.

AB - A deep learning based method for classifying multi-class faults in a gearbox is presented. A set of 900 vibration signals representing the normal condition and nine faults comprises the dataset used in this research. The recorded vibration signals are pre-processed for extracting the first and second derivatives as well as the first five Intrinsic Mode Functions (IMFs) by applying the Empirical Mode Decomposition (EMD) method. A 2D representation of these signals is the feature space used for classifying ten conditions of a gearbox using a Long Short Term Memory (LSTM) neural network. The 2D feature space is subdivided along the temporal axis in segments of the same size as the LSTM network. These segments are classified and a voting systems is proposed for attaining the signal classification. A 10-fold cross-validation is used for evaluating the proposed deep learning model. An average accuracy up to 99.4 % for classifying the faults is attained during the cross-validation.

KW - Deep learning

KW - Faults classification

KW - Gearboxes

KW - Long short term memory networks

KW - Vibration signals

UR - http://www.scopus.com/inward/record.url?scp=85091531084&partnerID=8YFLogxK

U2 - 10.1109/SDPC.2019.00045

DO - 10.1109/SDPC.2019.00045

M3 - Conference contribution

AN - SCOPUS:85091531084

T3 - Proceedings - 2019 International Conference on Sensing, Diagnostics, Prognostics, and Control, SDPC 2019

SP - 208

EP - 214

BT - Proceedings - 2019 International Conference on Sensing, Diagnostics, Prognostics, and Control, SDPC 2019

A2 - Li, Chuan

A2 - Zhang, Shaohui

A2 - Long, Jianyu

A2 - Cabrera, Diego

A2 - Ding, Ping

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2019 International Conference on Sensing, Diagnostics, Prognostics, and Control, SDPC 2019

Y2 - 15 August 2019 through 17 August 2019

ER -

Medina R, Macancela JC, Lucero P, Cabrera D, Li C, Cerrada M et al. A LSTM Neural Network Approach using Vibration Signals for Classifying Faults in a Gearbox. In Li C, Zhang S, Long J, Cabrera D, Ding P, editors, Proceedings - 2019 International Conference on Sensing, Diagnostics, Prognostics, and Control, SDPC 2019. Institute of Electrical and Electronics Engineers Inc. 2019. p. 208-214. 9169022. (Proceedings - 2019 International Conference on Sensing, Diagnostics, Prognostics, and Control, SDPC 2019). doi: 10.1109/SDPC.2019.00045

A LSTM Neural Network Approach using Vibration Signals for Classifying Faults in a Gearbox

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this