International Journal of Innovative Research in                 Electrical, Electronics, Instrumentation and Control Engineering

A monthly peer-reviewed online and print journal

ISSN Online 2321-2004
ISSN Print 2321-5526

Since  2013

Abstract: This paper presents a Long Short-Term Memory (LSTM) network-based method to identify faults in a 3-phase induction motor.  Various external faults in a 3-phase induction motor are first described.  A brief introduction to LSTM technique is then presented.  The LSTM is trained to classify external faults using 3-phase voltages and currents collected from a 1/3 hp induction motor in real-time.  MATLAB is used for training and testing the LSTM method. Results show that the proposed LSTM based method is effective in classifying different external faults in the 3-phase induction motor. The performance of the LSTM method is observed to be better than some of the previous methods in model formation and testing accuracy.

Keywords: Long Short-Term Memory, Induction Motor, Fault Identification, Machine Learning, Protective Relays.


PDF | DOI: 10.17148/IJIREEICE.2020.8901

Open chat