The method presented in this thesis for induction motor fault diagnosis is based on the analysis of the envelope of the three phase stator current. This diagnostic method can classify two types of induction motor faults: broken rotor bars and inter-turn short circuits in the stator windings. Experimental results show that the three phase current envelope is a powerful feature for motor fault classification. The envelope signal is extracted from the experimentally acquired stator current signals and is used in conjunction with machine learning techniques based on Gaussian Mixture Models [34] (GMMs) and Reconstructed Phase Spaces (RPSs) [34-36] to identify motor faults.
In addition, this diagnostic method not only classifies an induction motor as healthy or faulty, but also identifies the severity of the fault through the identification of the number of broken rotor bars or the number (or percentage) of short-circuited turns in stator windings. This constitutes a powerful means of monitoring motor fault severities, which could possibly predict the time of onset of complete failure of a motor, and thus help prevent unexpected shutdowns of industrial processes. The second advantage of this method is that the classification process needs only the three-phase stator current sensors, usually available in ac drives. Thus, extra electrical and mechanical installations, sensors,
and mathematical models of an induction motor are not required.
OTHER SUPERIOR PRODUCTS
Yasakawa Motor, Driver SG- Mitsubishi Motor HC-,HA-
Westinghouse Modules 1C-,5X- Emerson VE-,KJ-
Honeywell TC-,TK- Fanuc motor A0-
Rosemount transmitter 3051- Yokogawa transmitter EJA-
Contact person: Anna
E-mail: wisdomlongkeji@163.com
Cellphone: +0086-13534205279
