<abstract xmlns="http://www.w3.org/1999/xhtml"><p>The classification and detection of defects play an important role in different disciplines. Research is oriented towards the development of approaches for the early detection and classification of defects in electrical drive systems. This paper, proposes a new approach for the classification of induction motor defects based on image processing and pattern recognition. The proposed defect classification approach was carried out in four distinct stages. In the first step, the stator currents were represented in the 3D space and projected onto the 2D space. In the second step, the projections obtained were transformed into images. In the third step, extraction of features whereas the Histogram of Oriented Gradient (HOG) is used to construct a descriptor based on several sizes of cells. In the fourth step, a method of classifying the induction motor defects based on the Support Vector Machine (SVM) was applied. The evaluation results of the developed approach show the efficiency and the precision of classification of the proposed approach.</p></abstract>

The classification and detection of defects play an important role in different disciplines. Research is oriented towards the development of approaches for the early detection and classification of defects in electrical drive systems. This paper, proposes a new approach for the classification of induction motor defects based on image processing and pattern recognition. The proposed defect classification approach was carried out in four distinct stages. In the first step, the stator currents were represented in the 3D space and projected onto the 2D space. In the second step, the projections obtained were transformed into images. In the third step, extraction of features whereas the Histogram of Oriented Gradient (HOG) is used to construct a descriptor based on several sizes of cells. In the fourth step, a method of classifying the induction motor defects based on the Support Vector Machine (SVM) was applied. The evaluation results of the developed approach show the efficiency and the precision of classification of the proposed approach.

Use of Support Vector Machines for Classification of Defects in the Induction Motor

Department of Mechanical Engineering, Faculty of Technology

Department of Electrical Engineering, Faculty of Sciences and Technologies

Acta Universitatis Sapientiae, Electrical and Mechanical Engineering

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

The classification and detection of defects play an important role in different disciplines. Research is oriented towards the development of approaches for the...

Use of Support Vector Machines for Classification of Defects in the Induction Motor

Online veröffentlicht: 31. Dez. 2019

Seitenbereich: 1 - 21

Eingereicht: 22. Sept. 2018

DOI: https://doi.org/10.2478/auseme-2019-0001

Schlüsselwörter
Induction motor, defect, pattern recognition, SVM, HOG, classification, detection

© 2019 Samia Benzahioul et al., published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Use of Support Vector Machines for Classification of Defects in the Induction Motor

Online veröffentlicht: 31. Dez. 2019

Seitenbereich: 1 - 21

Eingereicht: 22. Sept. 2018

DOI: https://doi.org/10.2478/auseme-2019-0001

SchlüsselwörterInduction motor, defect, pattern recognition, SVM, HOG, classification, detection

© 2019 Samia Benzahioul et al., published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Schlüsselwörter
Induction motor, defect, pattern recognition, SVM, HOG, classification, detection