ATTACHED PAPER SHOULD BE DEVELOPED AS IT IS This paper presents a novel simplified cost effective
binary based mechanism for eliminating problems in existing
condition monitoring and diagnostic techniques. The existing
techniques are engraved with problems like high cost of their
implementation, failure to detect multiplicity of faults, complicated
diagnosis system, expert knowledge, inherent ambiguities etc. The
mechanism developed in this work gives flexibility to monitor faults
at incipient stage and avoid catastrophic cascading of faults that leads
to plant shutdown and consequent long repair/maintenance periods.
The developed system detects working status of sensors, location of
faults, no. of faults at that particular instant, acknowledgement of
faults, and health status of electrical machines connected to the
system. These diagnostic parameters have been extracted by
implementing a simple logic circuitry. The main advantage of this
system is that it can be applied to any rating of electrical machines.
The input parameters for the system are the outputs of the sensors
which are capable to measure variation in parameters like current,
voltage, flux, torque, speed, temperature, and vibration etc. due to
occurrence of faults. The test results of the proposed mechanism have
been found to be highly encouraging, accurate and extremely fast
providing the much needed protection of machines.