Condition monitoring of equipments is necessary for various reasons such as to increase their availability, safety, reliability and preventing the capital loss due to their unavailability. Currently, condition based maintenance is considered as the most efficient maintenance strategy in industries. Various signals such as vibration, current and acoustic are measured to monitor the current status of the equipments. To obtain accurate condition information from these signals, it is necessary to analyze them using different mathematical techniques. Two categories of mathematical techniques are used, namely, signal processing and machine learning methods. This talk will elaborate on how different signal processing and machine learning techniques are useful for condition monitoring applications. A case study related to a bearing fault monitoring in an induction motor will be presented. Also, how these techniques could be used to monitor different electrical equipments such as power transformers and circuit breakers will be discussed.
Bubathi Muruganatham received his B.E degree in Electrical and Electronics Speaker Engineering from Anna University Chennai in 2007 and PhD degree from Homi Bhabha National Institute, Mumbai in 2014. His PhD thesis was in the area of 'bearing condition monitoring of a induction motor'. He has carried out his PhD work as Research Fellow at Indira Gandhi Centre for Atomic Research, Tamil Nadu. He has worked as a Consultant with Prognostic Systems lab of GE Global Research Centre, Bangalore from December 2013 to November 2014. Also, he has worked as Electrical Engineer in ESSAR Steel Ltd, Surat from June 2007 to August 2008 . He has four international journal