Organization :
King Mongkuts University of Technology North Bangkok. Faculty of Technical Education. Department of Teacher Training in Mechanical Engineering
Abstract:
This paper proposes a new method of fault
detection and quantification for induction motors.
The method is based on Extended Discrete-
Fourier Transform (EDFT). Stator phase currents
of 3 phase induction motors are used as inputs.
The method is verified on experimental test
with 3 different motor conditions: healthy,
stator fault, and rotor fault motor at full load
condition. The method plots the relation between
normalized frequency and absolute amplitude
(power spectrum). The plots of power spectrum
density and relative frequency resolutions of stator
currents are also applied to differentiate the faults.
The experimental tests show that the proposed
method can differentiate each condition clearly by
observing the change in specific harmonics.
The method can also show the level of fault
severities by observing the percent of growth in the
harmonics, but the measurement noises may slightly
affect the fault differentiation. However, based
on observation the proposed method can show
ability of fault detection and quantification for this
kind of the motor with good accuracy