0
TECHNICAL PAPERS: Gas Turbines: Controls, Diagnostics, and Instrumentation

Setting Up of a Probabilistic Neural Network for Sensor Fault Detection Including Operation With Component Faults

[+] Author and Article Information
C. Romesis, K. Mathioudakis

Laboratory of Thermal Turbomachines, National Technical University of Athens, Athens 15710, Greece

J. Eng. Gas Turbines Power 125(3), 634-641 (Aug 15, 2003) (8 pages) doi:10.1115/1.1582493 History: Received December 01, 2001; Revised March 01, 2002; Online August 15, 2003
Copyright © 2003 by ASME
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Layout and station numbering of a turbofan engine
Grahic Jump Location
Probabilistic neural network architecture for turbofan sensor fault diagnosis
Grahic Jump Location
Patterns of deltas for different situations: (a) healthy engine and sensors, (b) faulty engine, healthy sensor, (c) healthy engine, faulty sensor (XNLP), (d) faulty engine and sensor (XNLP)
Grahic Jump Location
The procedure for generating the sets of patterns
Grahic Jump Location
The procedure for evaluating sensor fault cases
Grahic Jump Location
PNN performance matrix for sensor fault detection
Grahic Jump Location
PNN output for several levels of T3 sensor fault
Grahic Jump Location
Minimum detectable levels of sensor faults
Grahic Jump Location
Minimum detectable levels of sensor faults in terms of measurement standard deviations
Grahic Jump Location
Procedure followed for multiple sensor fault detection
Grahic Jump Location
PNN success rate for cases of multiple sensor faults
Grahic Jump Location
Performance of the PNN on detection of sensor’s XNHP and WFE fault on a turbofan with single-component fault
Grahic Jump Location
Minimum detectable levels of XNHP sensor fault, for several levels of HP turbine fault (SW41 deviation)
Grahic Jump Location
Performance of the PNN on sensor fault detection on a turbofan with multiple component faults
Grahic Jump Location
Performance of the PNN on sensor fault detection on a drifting deteriorated turbofan
Grahic Jump Location
Detection of a drift in sensor P13: (a) time variation of P13, (b) probabilities produced by the network

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In