You can view the full text of this article for free using the link below.

Title: A knowledge-based diagnosis algorithm for broken rotor bar fault classification using FFT, principal component analysis and support vector machines

Authors: Hayri Arabaci; Mohamed Ali Mohamed

Addresses: Department of Electrical and Electronics Engineering, Selcuk University, Konya, 42075, Turkey ' Department of Mechatronics Engineering, Selcuk University, Konya, 42075, Turkey

Abstract: Despite their ruggedness and reliability, induction motors experience faults due to stresses and manufacturing errors. Early detection of these faults is important in preventing further damages and minimising down-time. In this study, a machine learning algorithm is proposed for detection and classification of broken rotor bar (BRB) faults according to their severity. Removal of high frequency components then amplification was performed on the measured single-phase current. Features were then extracted using FFT and principal component analysis (PCA). Support vector machines (SVM) was used for classification. Two classification schemes were analysed; one classifying in one step and another in two steps. Experiments were performed to evaluate the algorithms by analysing their recognition rates. Six different SVM kernels were studied. Recognition rates as high as 97.9% were achieved. False negative rate as low as 0% was also realised. Furthermore, it was found out that using more principle components does not yield significant improvements.

Keywords: squirrel cage induction motor; IM; SVM; support vector machines; PCA; principal component analysis; BRB; broken rotor bar; fault diagnosis; machine learning.

DOI: 10.1504/IJIEI.2020.105431

International Journal of Intelligent Engineering Informatics, 2020 Vol.8 No.1, pp.19 - 37

Received: 05 Jun 2019
Accepted: 02 Sep 2019

Published online: 20 Feb 2020 *

Full-text access for editors Access for subscribers Free access Comment on this article