Home > Research > Publications & Outputs > Enhanced feature extraction method for motor fa...

Associated organisational unit

Links

Text available via DOI:

View graph of relations

Enhanced feature extraction method for motor fault diagnosis using low-quality vibration data from wireless sensor networks

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published

Standard

Enhanced feature extraction method for motor fault diagnosis using low-quality vibration data from wireless sensor networks. / Shu, Q.; Lu, S.; Xia, M. et al.

In: Measurement Science and Technology, Vol. 31, No. 4, 045016, 15.01.2020.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

APA

Vancouver

Shu Q, Lu S, Xia M, Ding J, Niu J, Liu Y. Enhanced feature extraction method for motor fault diagnosis using low-quality vibration data from wireless sensor networks. Measurement Science and Technology. 2020 Jan 15;31(4):045016. Epub 2019 Nov 28. doi: 10.1088/1361-6501/ab5cca

Author

Shu, Q. ; Lu, S. ; Xia, M. et al. / Enhanced feature extraction method for motor fault diagnosis using low-quality vibration data from wireless sensor networks. In: Measurement Science and Technology. 2020 ; Vol. 31, No. 4.

Bibtex

@article{a4ce54cb53614231ae3b13e8c84c5177,
title = "Enhanced feature extraction method for motor fault diagnosis using low-quality vibration data from wireless sensor networks",
abstract = "Wireless sensor networks (WSNs), which are usually powered by batteries, have been extensively used in condition monitoring and fault diagnosis of motors. To extend the battery service life, the length of the acquired and transmitted signal should be short and the sampling resolution should be reduced. In this case, the motor signal quality is low, which affects the fault diagnosis accuracy. To address this issue, this study proposes an enhanced feature extraction method for motor fault diagnosis using low-quality vibration signals acquired from a battery-powered WSN node. First, the vibration signal is converted to an image using a wavelet synchrosqueezed transform technique. Second, the constructed image is enhanced using a histogram equalization. Finally, the enhanced image is inputted into a convolutional neural network (CNN) model, and the motor fault type can be recognized from the CNN output. The effectiveness and efficiency of the proposed method are validated by comparing its performance in the brushless direct motor test rig with the performance of several traditional methods. The relationship between the fault diagnosis accuracy and WSN performances is investigated and discussed. The proposed method shows potential applications for remote motor fault diagnosis using the low-quality vibration signal acquired from a WSN node with limited battery capacity.",
keywords = "CNN, histogram equalization, low-quality vibration data, motor fault diagnosis, WSN, WSST, Condition monitoring, Convolutional neural networks, Electric batteries, Equalizers, Extraction, Failure analysis, Feature extraction, Graphic methods, Image enhancement, Sensor nodes, Effectiveness and efficiencies, Feature extraction methods, Histogram equalizations, Motor fault, Sampling resolution, Vibration data, Wireless sensor network (WSNs), Fault detection",
author = "Q. Shu and S. Lu and M. Xia and J. Ding and J. Niu and Y. Liu",
year = "2020",
month = jan,
day = "15",
doi = "10.1088/1361-6501/ab5cca",
language = "English",
volume = "31",
journal = "Measurement Science and Technology",
issn = "0957-0233",
publisher = "IOP Publishing Ltd.",
number = "4",

}

RIS

TY - JOUR

T1 - Enhanced feature extraction method for motor fault diagnosis using low-quality vibration data from wireless sensor networks

AU - Shu, Q.

AU - Lu, S.

AU - Xia, M.

AU - Ding, J.

AU - Niu, J.

AU - Liu, Y.

PY - 2020/1/15

Y1 - 2020/1/15

N2 - Wireless sensor networks (WSNs), which are usually powered by batteries, have been extensively used in condition monitoring and fault diagnosis of motors. To extend the battery service life, the length of the acquired and transmitted signal should be short and the sampling resolution should be reduced. In this case, the motor signal quality is low, which affects the fault diagnosis accuracy. To address this issue, this study proposes an enhanced feature extraction method for motor fault diagnosis using low-quality vibration signals acquired from a battery-powered WSN node. First, the vibration signal is converted to an image using a wavelet synchrosqueezed transform technique. Second, the constructed image is enhanced using a histogram equalization. Finally, the enhanced image is inputted into a convolutional neural network (CNN) model, and the motor fault type can be recognized from the CNN output. The effectiveness and efficiency of the proposed method are validated by comparing its performance in the brushless direct motor test rig with the performance of several traditional methods. The relationship between the fault diagnosis accuracy and WSN performances is investigated and discussed. The proposed method shows potential applications for remote motor fault diagnosis using the low-quality vibration signal acquired from a WSN node with limited battery capacity.

AB - Wireless sensor networks (WSNs), which are usually powered by batteries, have been extensively used in condition monitoring and fault diagnosis of motors. To extend the battery service life, the length of the acquired and transmitted signal should be short and the sampling resolution should be reduced. In this case, the motor signal quality is low, which affects the fault diagnosis accuracy. To address this issue, this study proposes an enhanced feature extraction method for motor fault diagnosis using low-quality vibration signals acquired from a battery-powered WSN node. First, the vibration signal is converted to an image using a wavelet synchrosqueezed transform technique. Second, the constructed image is enhanced using a histogram equalization. Finally, the enhanced image is inputted into a convolutional neural network (CNN) model, and the motor fault type can be recognized from the CNN output. The effectiveness and efficiency of the proposed method are validated by comparing its performance in the brushless direct motor test rig with the performance of several traditional methods. The relationship between the fault diagnosis accuracy and WSN performances is investigated and discussed. The proposed method shows potential applications for remote motor fault diagnosis using the low-quality vibration signal acquired from a WSN node with limited battery capacity.

KW - CNN

KW - histogram equalization

KW - low-quality vibration data

KW - motor fault diagnosis

KW - WSN

KW - WSST

KW - Condition monitoring

KW - Convolutional neural networks

KW - Electric batteries

KW - Equalizers

KW - Extraction

KW - Failure analysis

KW - Feature extraction

KW - Graphic methods

KW - Image enhancement

KW - Sensor nodes

KW - Effectiveness and efficiencies

KW - Feature extraction methods

KW - Histogram equalizations

KW - Motor fault

KW - Sampling resolution

KW - Vibration data

KW - Wireless sensor network (WSNs)

KW - Fault detection

U2 - 10.1088/1361-6501/ab5cca

DO - 10.1088/1361-6501/ab5cca

M3 - Journal article

VL - 31

JO - Measurement Science and Technology

JF - Measurement Science and Technology

SN - 0957-0233

IS - 4

M1 - 045016

ER -