Open AccessNeural Network Based Fault Diagnostics in Multi Phase Induction Machine
Author(s) -
Balamurugan Annamalai,
Sivakumaran Thangavel Swaminathan
Publication year - 2020
Publication title -
international journal of innovative technology and exploring engineering
Language(s) - English
Resource type - Journals
ISSN - 2278-3075
DOI - 10.35940/ijitee.d2057.029420
Subject(s) - artificial neural network , extractor , computer science , classifier (uml) , induction motor , pattern recognition (psychology) , fault (geology) , artificial intelligence , control theory (sociology) , engineering , electrical engineering , voltage , seismology , process engineering , geology , control (management)
This article proposes a new method for solving the diagnosis of faults in a multiphase induction motor using a least-squares filter (LMS) and a neural network. The proposed hybrid fault diagnosis method includes an efficient LMS-based feature extractor and an artificial neural network fault classifier. First, the LMS method is used to obtain efficient functions. The performance and efficiency of the presented neural network hybrid classifier is evaluated by testing a total of 600 samples, which are modeled on a failure model. The average correct classification is 96.17% for different fault signals, respectively. The result obtained from the simulation analysis shows the effectiveness of the proposed neural network for the diagnosis of faults in the multiphase induction motor.