DIAGNOSIS OF BEARING DEFECTS BY ANFIS IN THE INDUCTION MOTOR

In this paper, we developed a robust technique for fault classification in the induction motor, which was used by the Adaptive Network Reference Information System (ANFIS). A robustness and precision test of the proposed method was performed on a small database. To speed up the response and minimize...

Full description

Bibliographic Details
Main Authors: Nabil TALBI, Abderrezak METATLA, Toufik SEBBAGH, Nadir BOUTASSETA, Mohamed HASSANI
Format: Article
Language:English
Published: Technical University of Cluj-Napoca 2019-12-01
Series:Carpathian Journal of Electrical Engineering
Subjects:
Online Access:http://cee.cunbm.utcluj.ro/wp-content/uploads/CJEE20193.pdf
Description
Summary:In this paper, we developed a robust technique for fault classification in the induction motor, which was used by the Adaptive Network Reference Information System (ANFIS). A robustness and precision test of the proposed method was performed on a small database. To speed up the response and minimize the calculation time, the most sensitive indicators have been used to ensure the learning of the classifier. The database used was carried out at the University of Case Western, this database has two modes of healthy operation and with rolling defects located side coupling. The objective of the technique used in this paper is to extract the indicators and to test them to select the most sensitive to the apparition of defects. The obtained results show the effectiveness and sensitivity of the proposed approach to identify the nature of the defect even at the birth stage.
ISSN:1843-7583
1843-7583