Early detection of unbalance voltage in three phase induction motor based on SVM

Author

Sawitri, D.R. ; Asfani, D.A. ; Purnomo, Mauridhi Hery ; Purnama, I.K.E. ; Ashari, Mohamed

Author_Institution

Electr. Eng. Dept., Dian Nuswantoro Univ., Semarang, Indonesia

fYear

2013

fDate

27-30 Aug. 2013

Firstpage

573

Lastpage

578

Abstract

Unbalance voltage supply in induction motor is a crucial problem. This paper proposes an original system to detect an unbalance voltage condition in induction motor using Support Vector Machine (SVM). Induction motor current, as a signal due to the unbalance voltage supply in three-phase induction motor, is recorded in lab bench. The features of recorded signals are extracted by wavelet transform and Principal Component Analysis algorithm. Then, the quality of detection is classified by SVM, and the average result of detection is 86%.

Keywords

condition monitoring; fault diagnosis; feature extraction; induction motors; principal component analysis; signal processing; support vector machines; wavelet transforms; SVM; feature extraction; principal component analysis; support vector machine; three phase induction motor; unbalance voltage supply detection; wavelet transform; Circuit faults; Feature extraction; Induction motors; Kernel; Noise reduction; Principal component analysis; Support vector machines; Induction Motor; Support Vector Machine (SVM); Unbalance Voltage;

fLanguage

English

Publisher

ieee

Conference_Titel

Diagnostics for Electric Machines, Power Electronics and Drives (SDEMPED), 2013 9th IEEE International Symposium on

Conference_Location

Valencia

Type

conf

DOI

10.1109/DEMPED.2013.6645772

Filename

6645772