Wyniki wyszukiwania:

Major catastrophic failures in large scale
engineering systems (e.g., aircraft, power plants and
turbo-machinery) can possibly be averted if the
malignant anomalies are detected at an early stage.
This dissertation experimentally validates a novel
method called Symbolic Time Series Analysis(STSA) for
anomaly detection in electromechanical systems,
derived from time series data of pertinent measured
variable(s).
In this dissertation, the performance of
this anomaly detection method is compared with that
of other existing pattern recognition techniques from
the perspectives of early detection...

The main objective of the work reported in this
monograph is the fabrication of an experimental
apparatus to detect slowly evolving anomalies (i.e.,
decrease in stiffness due to fatigue failure)
in complex mechanical systems at an early stage by
observing time series data of the available
ultrasonic and performance measuring sensors. The
information on evolving anomaly will serve as an
input to the life extending control policy that will,
in turn, generate corrective actions to mitigate
fatigue crack damage in the specimen structures and
thereby extend the remaining life of the system
without...