Application of Cepstrum and Neural Network to Induction Motor Fault Detection
Three phase induction motors are the most widely used electrical machines in industries. For this reason detection
of induction motor failures is very important. Bearing problems is one of the major causes for drive failures. The method
used parks transformation as a tool for the classification of the faults. Present contribution reports experimental results for
monitoring of bearing as well as interturn faults in induction motor. Stator current signals are transform into Id and Iq using
parks transformation. These signals are further analyzed using cepstrum. Statistical parameters computed from cepstrum
coefficients are fed as input to artificial neural network for motor fault classification. The results show the effectiveness of
cepstrum and ANN in detecting the bearing and interturn conditions.
Index terms - Induction motor, Fault diagnosis, Cepstrum, Park’s transformation, artificial neural network.