Gas turbine fault diagnostics using fuzzy concepts
by
Priya Alexander
Department of Power, Propulsion and Aerospace Engineering, School of Engineering, Whittle building, Cranfield University, MK43 0AL, United Kingdom
Coauthors: Riti Singh, S.D. Probert

In this paper, a new concept of Fuzzy logic is proposed to solve Gas Turbine Engine Diagnostic problems. Handling uncertainty with reduced rules (if-then) and ability to quantify multiple component faults are discussed as relative strength of this approach. Either single or double or multiple sensor measurements are used here to carry out the analysis. In sequence, an evaluation and learning process take place in two phases in order. In evaluation phase, an inference mechanism is enabled to match the incipient fault information with established fault pattern. In learning phase, a range limit is defined explicitly for each sensor, beyond which, the system classifies either sensor or component faults. Finally, the paper concludes with the appraisal of Fuzzy logic based gas path analysis.

Date received: January 23, 2004

Copyright © 2004 by the author(s). The author(s) of this document and the organizers of the conference have granted their consent to include this abstract in Atlas Conferences Inc. Document # cakt-34.