15th Triennial World Congress of the International Federation of Automatic Control
  Barcelona, 21–26 July 2002 
FAULT DETECTION AND IDENTIFICATION VIA MULTI-LEVEL HYPOTHESES TESTING
Asok Ray* Shashi Phoha**
The Pennsylvania State University
University Park, PA 16802, USA
* Email:axr2@psu.edu **Email:sxp26@psu.edu
This paper formulates a recursive algorithm of multi-level hypotheses testing for real-time detection and identification of faults from continuous sensor signals. The usage of the recursive algorithm is illustrated on a data set of temperature sensors, collected from the Instrumentation and Control system of an operating power plant.
Keywords: Fault detection; Fault diagnosis; Hypothesis; Statistical analysis; Stochastic systems
E-mail: axr2@psu.edu
Session slot T-Fr-A01: Identification of Stochastic Systems/Area code 3a : Modelling, Identification and Signal Processing