Neural networks applied for fault diagnosis of AC motors

This paper presents an Artificial Neural Network (ANN) technique to recognize the incipient faults of an AC motor such as a synchronous motor. The proposed ANN-based fault detector is developed using the Resilient Error Back Propagation (RPROP) training algorithm. The fast and reliable method for mu...

Full description

Bibliographic Details
Main Authors: K.S., Rama Rao, Yahya , M.A.
Format: Conference or Workshop Item
Language:English
Published: 2008
Subjects:
Online Access:http://scholars.utp.edu.my/id/eprint/294/
http://scholars.utp.edu.my/id/eprint/294/1/paper.pdf
Description
Summary:This paper presents an Artificial Neural Network (ANN) technique to recognize the incipient faults of an AC motor such as a synchronous motor. The proposed ANN-based fault detector is developed using the Resilient Error Back Propagation (RPROP) training algorithm. The fast and reliable method for multilayer neural networks converges much faster than the conventional back propagation algorithm. The main causes to diagnose three major faults are investigated and validated by adopting feed-forward back propagation neural networks. © 2008 IEEE.