TY - GEN
T1 - Gearbox Fault Diagnostics
T2 - 2nd International Conference on Innovative Technology, Engineering and Sciences, iCITES 2020
AU - Hasan, Md Jahid
AU - Rashid, Mamunur
AU - Nasir, Ahmad Fakhri Ab
AU - Abdullah, Muhammad Amirul
AU - Razman, Mohd Azraai Mohd
AU - Musa, Rabiu Muazu
AU - P. P. Abdul Majeed, Anwar
N1 - Publisher Copyright:
© 2021, The Author(s), under exclusive license to Springer Nature Switzerland AG.
PY - 2021
Y1 - 2021
N2 - Gearbox defects have been considered as one of the major causes of failure in rotating machinery. It is important to identify and diagnose the actual reasons behind the failure of gearbox for a reliable operation of equipment that relies such a system. In this paper, a fault diagnosis method based on entropy-based feature and support vector machine (SVM) has been proposed for detecting the faults in bearings and gear set in the gearbox. Initially, different features in terms of time domain as well as time-frequency domain have been extracted and classified via SVM. The proposed method has been validated by the publicly available online dataset which is consists of nine classes (eight types of faults and healthy) with load and unloaded conditions. The optimum validation accuracy (98.84% for 20-0 load condition and 99.87% for 30-2 load conditions) has been obtained by the entropy-based feature extracted from transformed continuous wavelet transform (CWT) signal. The outcome of this study is very encouraging since it emphasizes to avoid the computational complexity in feature extraction as well as classification.
AB - Gearbox defects have been considered as one of the major causes of failure in rotating machinery. It is important to identify and diagnose the actual reasons behind the failure of gearbox for a reliable operation of equipment that relies such a system. In this paper, a fault diagnosis method based on entropy-based feature and support vector machine (SVM) has been proposed for detecting the faults in bearings and gear set in the gearbox. Initially, different features in terms of time domain as well as time-frequency domain have been extracted and classified via SVM. The proposed method has been validated by the publicly available online dataset which is consists of nine classes (eight types of faults and healthy) with load and unloaded conditions. The optimum validation accuracy (98.84% for 20-0 load condition and 99.87% for 30-2 load conditions) has been obtained by the entropy-based feature extracted from transformed continuous wavelet transform (CWT) signal. The outcome of this study is very encouraging since it emphasizes to avoid the computational complexity in feature extraction as well as classification.
KW - Continuous Wavelet Transform (CWT)
KW - Gearbox fault detection
KW - Machine learning
KW - Support vector machine (SVM)
UR - http://www.scopus.com/inward/record.url?scp=85104404820&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-70917-4_39
DO - 10.1007/978-3-030-70917-4_39
M3 - Conference Proceeding
AN - SCOPUS:85104404820
SN - 9783030709167
T3 - Advances in Intelligent Systems and Computing
SP - 399
EP - 406
BT - Advances in Robotics, Automation and Data Analytics - Selected Papers from iCITES 2020
A2 - Mat Jizat, Jessnor Arif
A2 - Khairuddin, Ismail Mohd
A2 - Mohd Razman, Mohd Azraai
A2 - Ab. Nasir, Ahmad Fakhri
A2 - Abdul Karim, Mohamad Shaiful
A2 - Jaafar, Abdul Aziz
A2 - Hong, Lim Wei
A2 - Abdul Majeed, Anwar P.
A2 - Liu, Pengcheng
A2 - Myung, Hyun
A2 - Choi, Han-Lim
A2 - Susto, Gian-Antonio
PB - Springer Science and Business Media Deutschland GmbH
Y2 - 22 December 2020 through 22 December 2020
ER -