Neural network fault diagnosis for the NASP fuel distribution system

Author

Sacks, R. ; Clark, J. ; Rap, R. ; Wagner, E.A.

Author_Institution

Accurate Autom. Corp., Chattanooga, TN, USA

fYear

1995

fDate

8-10 Aug. 1995

Firstpage

368

Lastpage

369

Abstract

A fault diagnosis system for the cryogenic fuel system of the National Aerospace Plane has been developed. This system is designed to detect and isolate vent, valve, and sensor failures in the fuel distribution system, throughout the duration of a typical single-stage-to-orbit flight. The system was simulated and subjected to a full suite of tests in which noise and modeling errors were introduced into the fuel distribution model.

Keywords

aerospace expert systems; aerospace propulsion; aerospace simulation; aircraft; aircraft testing; automatic test software; avionics; diagnostic expert systems; fault diagnosis; fuel; neural nets; NASP fuel distribution system; National Aerospace Plane; SOFTITE; cryogenic fuel system; feedback control unit; modeling errors; neural network fault diagnosis; noise errors; sensor failure; simulator; single-stage-to-orbit flight; valve failure; vent failure; Fault diagnosis; Feedback control; Fuels; Helium; Hydrogen; Neural networks; Sensor systems; Temperature sensors; Valves; Vents;

fLanguage

English

Publisher

ieee

Conference_Titel

AUTOTESTCON '95. Systems Readiness: Test Technology for the 21st Century. Conference Record

Conference_Location

Atlanta, GA, USA

Print_ISBN

0-7803-2621-0

Type

conf

DOI

10.1109/AUTEST.1995.522697

Filename

522697