Title page for ETD etd-04282008-100002

Type of Document

Thesis

Author

Quiroga, Jabid Eduardo

Author's Email Address

jabib@uis.edu.co

URN

etd-04282008-100002

Title

Stator Winding Fault Detection for a Pmsm Using Fuzzy Logic Classifier and Neural Network Model Identification

Degree

Master of Science

Department

Mechanical Engineering, Department of

Advisory Committee

Advisor Name	Title
David Cartes	Committee Chair
Chris Edrington	Committee Co-Chair
Jonathan Clark	Committee Member
Juan Ordonez	Committee Member

Keywords

Fuzzy Logic
Neural Networks
Fault Detection
PMSM

Date of Defense

2008-03-27

Availability

unrestricted

Abstract

A negative sequence analysis coupled with fuzzy logic and neural network based approaches are applied to stator winding short circuit fault detection in a permanent magnet synchronous motors (PMSM).
A fuzzy logic based approach is implemented to generate a robust detection using the filtered negative sequence current and negative sequence impedance. The filtered negative sequence current is obtained by separating the high frequency components caused by the load fluctuation from the total negative sequence current. The filtered negative sequence current provides a quantitative evaluation on severity of the stator fault.
A MLP neural network is implemented as current predictors. The detection stage is carrying out using the negative sequence analysis of the residuals, obtained according to the difference between the actual values of currents and the current predictors. The negative sequence component of the residuals provides not only the information for detecting the fault condition but also a measurement of the level of severity of the winding short.
Validation of the methods are performed online using a PMSM experimental setup with dSPACE and Matlab/Simulink environment. The use of fuzzy logic classifier and neural network model identification improves the sensitivity of fault detection while reducing false alarm rate under load fluctuations.

Files

Filename		Size	Approximate Download Time (Hours:Minutes:Seconds)
Filename		Size	28.8 Modem	56K Modem	ISDN (64 Kb)	ISDN (128 Kb)	Higher-speed Access
	Thesis_ref_dif_edrington_cartes.pdf	1.41 Mb	00:06:33	00:03:22	00:02:56	00:01:28	00:00:07