Deep Learning-Based Gear Pitting Severity Assessment Using Acoustic Emission, Vibration and Currents Signals

Ruben Medina; Mariela Cerrada; Diego Cabrera; Rene Vinicio Sánchez; Chuan Li; Jose Valente De Oliveira

doi:10.1109/PHM-Paris.2019.00042

Deep Learning-Based Gear Pitting Severity Assessment Using Acoustic Emission, Vibration and Currents Signals

Ruben Medina, Mariela Cerrada, Diego Cabrera, Rene Vinicio Sánchez, Chuan Li, Jose Valente De Oliveira

Research output: Contribution to conference › Paper

Abstract

A method for gearbox pitting faults severity classification using Deep Learning techniques is reported. The signals are preprocessed for obtaining a 2D time-frequency representation corresponding to the Mel Frequency Cepstral Coefficients. This bi-dimensional representation is the feature space used for classification. A Long Short Term Memory network (LSTM) is used for classifying nine levels of pitting in spur gears. Each signals dataset is used for training and validating a LSTM network. Classification accuracies up to 100 % are obtained during cross-validation with the analyzed signals dataset.

Original language	English
Pages	210-216
Number of pages	7
DOIs	https://doi.org/10.1109/PHM-Paris.2019.00042
State	Published - 12 Jul 2019
Externally published	Yes
Event	Proceedings - 2019 Prognostics and System Health Management Conference, PHM-Paris 2019 - Duration: 1 May 2019 → …

Conference

Conference	Proceedings - 2019 Prognostics and System Health Management Conference, PHM-Paris 2019
Period	1/05/19 → …

Keywords

acoustic emission
deep learning
Faults detection
gearbox
long short term memory networks

Access to Document

10.1109/PHM-Paris.2019.00042

Cite this

@conference{620d7a1ea22340e482822e523787496b,

title = "Deep Learning-Based Gear Pitting Severity Assessment Using Acoustic Emission, Vibration and Currents Signals",

abstract = "A method for gearbox pitting faults severity classification using Deep Learning techniques is reported. The signals are preprocessed for obtaining a 2D time-frequency representation corresponding to the Mel Frequency Cepstral Coefficients. This bi-dimensional representation is the feature space used for classification. A Long Short Term Memory network (LSTM) is used for classifying nine levels of pitting in spur gears. Each signals dataset is used for training and validating a LSTM network. Classification accuracies up to 100 % are obtained during cross-validation with the analyzed signals dataset.",

keywords = "acoustic emission, deep learning, Faults detection, gearbox, long short term memory networks",

author = "Ruben Medina and Mariela Cerrada and Diego Cabrera and S{\'a}nchez, {Rene Vinicio} and Chuan Li and {De Oliveira}, {Jose Valente}",

year = "2019",

month = jul,

day = "12",

doi = "10.1109/PHM-Paris.2019.00042",

language = "English",

pages = "210--216",

note = "Proceedings - 2019 Prognostics and System Health Management Conference, PHM-Paris 2019 ; Conference date: 01-05-2019",

}

TY - CONF

T1 - Deep Learning-Based Gear Pitting Severity Assessment Using Acoustic Emission, Vibration and Currents Signals

AU - Medina, Ruben

AU - Cerrada, Mariela

AU - Cabrera, Diego

AU - Sánchez, Rene Vinicio

AU - Li, Chuan

AU - De Oliveira, Jose Valente

PY - 2019/7/12

Y1 - 2019/7/12

N2 - A method for gearbox pitting faults severity classification using Deep Learning techniques is reported. The signals are preprocessed for obtaining a 2D time-frequency representation corresponding to the Mel Frequency Cepstral Coefficients. This bi-dimensional representation is the feature space used for classification. A Long Short Term Memory network (LSTM) is used for classifying nine levels of pitting in spur gears. Each signals dataset is used for training and validating a LSTM network. Classification accuracies up to 100 % are obtained during cross-validation with the analyzed signals dataset.

AB - A method for gearbox pitting faults severity classification using Deep Learning techniques is reported. The signals are preprocessed for obtaining a 2D time-frequency representation corresponding to the Mel Frequency Cepstral Coefficients. This bi-dimensional representation is the feature space used for classification. A Long Short Term Memory network (LSTM) is used for classifying nine levels of pitting in spur gears. Each signals dataset is used for training and validating a LSTM network. Classification accuracies up to 100 % are obtained during cross-validation with the analyzed signals dataset.

KW - acoustic emission

KW - deep learning

KW - Faults detection

KW - gearbox

KW - long short term memory networks

UR - https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85070519509&origin=inward

UR - https://www.scopus.com/inward/citedby.uri?partnerID=HzOxMe3b&scp=85070519509&origin=inward

UR - http://www.mendeley.com/research/deep-learningbased-gear-pitting-severity-assessment-using-acoustic-emission-vibration-currents-signa

U2 - 10.1109/PHM-Paris.2019.00042

DO - 10.1109/PHM-Paris.2019.00042

M3 - Paper

SP - 210

EP - 216

T2 - Proceedings - 2019 Prognostics and System Health Management Conference, PHM-Paris 2019

Y2 - 1 May 2019

ER -

Deep Learning-Based Gear Pitting Severity Assessment Using Acoustic Emission, Vibration and Currents Signals

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this