This paper introduces a feature-extraction method to characterize gas turbine engine dynamics. The extracted features are used to develop a fault diagnosis and prognosis method for the fuel supply system in gas turbine engines. The engine start-up profiles of the core speed (N2) and the exhaust gas temperature collected with high-speed sampling rate are obtained and processed into a more compact data set by identifying critical-to-characterization instances. The fuzzy-clustering method is applied to the smaller number of parameters, and the fault is detected by differentiating the clusters matching the failures. In this work, the actual flight data collected in the field was used to develop and validate the system, and the results are shown for the test on nine engines that experienced fuel supply system failure. The developed fault diagnosis system detected the failure successfully in all nine cases. For the earliest detection cases, the alarms start to trigger 26 days before the system completely fails and 7 days in advance for the last detection.
ChaseD., ‘Predictive trend monitoring for commercial engines’ European Operators Conference, 2004.
2.
AshbyM.ByerR., ‘An approach for conducting a cost benefit analysis of aircraft engine prognostics and health management functions’ IEEE Aerospace Conference, Big Sky, Montana, March 2002.
3.
HessA.CalvelloG.DabneyT.FrithT., ‘PHM the key enabler for the joint strike force (JSF) autonomic logistics support concept’ (Virginia Beach, Virginia: MFPT Technologies to Meet Corporate Goals2004).
4.
LittJ.KurtkayaM.DuyarA., ‘Sensor fault detection and diagnosis for a T700 turboshaft engine’ AIAA J. Guid. Control Dyn.183 (1995): 640–642.
5.
KobayashiT.SimonD. L., ‘Evaluation of an enhanced bank of Kalman filters for in-flight aircraft engine sensor fault diagnostics’ (Vienna, Austria: ASME Turbo Expo2004).
6.
BiswasG.BloorG., ‘A model based approach to constructing performance degradation monitoring systems’ IEEE Aerospace Conference, Big Sky, Montana, March 2006.
7.
MylaraswamyD.VenkatasubramanianV., ‘A hybrid framework for large scale process fault diagnosis’ 30th European Symposiumon Computer-Aided Process Engineering, Trondheim, Norway, 1997.
8.
de KleerJ.KurienJ., ‘Fundamentals of model-based diagnosis’ 15th International Workshop on Principles of Diagnosis, Carcassonne, France, June 2004.
9.
KinnaertM., ‘Fault diagnosis based on analytical models for linear and nonlinear systems — a tutorial’ 15th International Workshop on Principles of Diagnosis, Carcassonne, France, June 2004.
10.
KimK.BallC.NwadiogbuE., ‘Fault diagnosis in turbine engines using unsupervised neural network technique’ SPIE Defense & Security Symposium, Orlando, Florida, April 2004.
11.
MenonS.UluyolO.KimK.NwadiogbuE., ‘Incipient fault detection and daignosis in turbine engines using hidden markov models’ ASME Turbo Expo, Atlanta, Georgia, June 2003.
12.
ZeddaM.SinghR., ‘Fault diagnosis of a turbofan engine using neural networks — a quantitative approach’ 34th AIAA/ASME Joint Propulsion Conference and Exhibit, Cleveland, Ohio, July 1998, paper No: AIAA 98-3602.
13.
KanelopoulosK.StamatisA.MathioudakisK., ‘Incorporating neural networks into gas turbine performance diagnostics’ (1997): ASME paper No. 97-GT-35.
14.
VolponiA. J.de PoldH.GanguliR.DaguangC., ‘The use of Kalman filter and neural network methodologies in gas turbine performance diagnostics: a comprehensive study’ (2000): ASME paper No. 200-GT-547.
15.
KimK.MylaraswamyD., ‘Fault diagnosis and prognosis of gas turbine engines based on qualitative modeling’ ASME Turbo Expo, Barcelona, Spain, May 2006.
16.
SiuC.ShenQ.MilneR., ‘TMDOCTOR: a fuzzy rule and case based expert system for turbomachinery diagnosis’ IFAC Symposium: SAFEPROCESS'97, Kingston Upon Hull, England, August 1997.
17.
BreeseJ. S.HorvitzE. J.PeotM. A.GayR.QuentinG. H., ‘Automated decision-analytic diagnosis of thermal performance in gas turbines’ (1992): ASME paper No. 92-GT-399.
18.
TumerI. Y.BajwaA., ‘A survey of aircraft engine health monitoring systems’ 35th AIAA/ASME Joint Propulsion Conference and Exhibit, Los Angeles, California, June 1999, paper No: AIAA 99-2528.
19.
SurenderV.GanguliR., ‘Adaptive myriad filter for improved gas turbine condition monitoring using transient data’ ASME J. Eng. Gas Turbines Power1272 (2005): 329–339.
20.
LiY. G., ‘A gas turbine diagnostic approach with transient measurements’ Proc. Instn Mech. Engrs, Part A: J. Power and Energy217 A2 (2003): 169–177 DOI: 10.1243/09576500360611317.
21.
UluyolO.KimK.NwadiogbuE., ‘Synergistic use of soft computing technologies for fault detection in gas turbine engines’ IEEE Trans. Syst., Man Cybern.364 (2006): 476–484.
22.
BishopC., Neural networks for pattern recognition (Oxford: Clarendon Press1995).
