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

Combining Classification Techniques With Kalman Filters for Aircraft Engine Diagnostics

[+] Author and Article Information
P. Dewallef

ASMA Department,  University of Liège, 1 Chemin des Chevreuils, 4000 Liège, Belgiump.dewallef@ulg.ac.be

C. Romessis

Laboratory of Thermal Turbomachines,  National Technical University of Athens, P.O. Box 64069, Athens 15710, Greececristo@mail.ntua.gr

O. Léonard

ASMA Department,  University of Liège, 1 Chemin des Chevreuils, 4000 Liège, Belgiumo.Leonard@ulg.ac.be

K. Mathioudakis

Laboratory of Thermal Turbomachines,  National Technical University of Athens, P.O. Box 64069, Athens 15710, Greecekmathiou@central.ntua.gr

J. Eng. Gas Turbines Power 128(2), 281-287 (Mar 01, 2004) (7 pages) doi:10.1115/1.2056507 History: Received October 01, 2003; Revised March 01, 2004

A diagnostic method consisting of a combination of Kalman filters and Bayesian Belief Network (BBN) is presented. A soft-constrained Kalman filter uses a priori information derived by a BBN at each time step, to derive estimations of the unknown health parameters. The resulting algorithm has improved identification capability in comparison to the stand-alone Kalman filter. The paper focuses on a way of combining the information produced by the BBN with the Kalman filter. An extensive set of fault cases is used to test the method on a typical civil turbofan layout. The effectiveness of the method is thus demonstrated, and its advantages over individual constituent methods are presented.

Copyright © 2006 by American Society of Mechanical Engineers
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Figure 1

Classification procedure using BBN

Grahic Jump Location
Figure 2

Description of the soft-constrained Kalman filter

Grahic Jump Location
Figure 3

Data preprocessing for operating-point estimation

Grahic Jump Location
Figure 4

Procedure followed to combine classification algorithm with Kalman filter

Grahic Jump Location
Figure 5

Conversion of probability density from a piecewise constant function into a Gaussian function

Grahic Jump Location
Figure 6

Turbofan engine layout. Measurement uncertainties represent three times the standard deviation

Grahic Jump Location
Figure 7

Identification results of individual methods for HPC fault case c (dotted lines refer to actual values)

Grahic Jump Location
Figure 8

Identification results of combined method for HPC fault case c (dotted lines refer to actual values)

Grahic Jump Location
Figure 9

Identification results of individual methods for LPT fault case k (dotted lines refer to actual values)

Grahic Jump Location
Figure 10

Identification results of combined method for LPT fault case k (dotted lines refer to actual values)

Grahic Jump Location
Figure 11

Identification results of SCKF (upper figure) and combined method (lower figure) for LPT fault case l (dotted lines refer to actual values of health parameters)

Grahic Jump Location
Figure 12

Identification results of individual methods for LPT fault case j (dotted lines refer to actual values)

Grahic Jump Location
Figure 13

Identification results of combined method for LPT fault case j (dotted lines refer to actual values)

Grahic Jump Location
Figure 14

Influence of the measurement noise level on the convergence speed

Grahic Jump Location
Figure 15

Identification results of BBN alone (upper figure) and combined method (lower figure) for LPT fault case j (dotted lines refer to actual values)

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 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