• DocumentCode
    3205746
  • Title

    Advanced test cell diagnostics for gas turbine engines

  • Author

    Roemer, Michael J. ; Kacprzynski, Gregory J. ; Schoeller, Michael ; Howe, Ron ; Friend, Richard

  • Author_Institution
    Impact Technol., Rochester, NY, USA
  • Volume
    6
  • fYear
    2001
  • fDate
    2001
  • Firstpage
    2915
  • Abstract
    Improved test cell diagnostics capable of detecting and classifying engine mechanical and performance faults as well as instrumentation problems is critical to reducing engine operating and maintenance costs while optimizing test cell effectiveness. Proven anomaly detection and fault classification techniques utilizing engine Gas Path Analysis (GPA) and statistical/empirical models of structural and performance related engine areas can now be implemented for real-time and post-test diagnostic assessments. Integration and implementation of these proven technologies into existing USAF engine test cells presents a great opportunity to significantly improve existing engine test cell capabilities to better meet today´s challenges. A suite of advanced diagnostic and troubleshooting tools have been developed and implemented for gas turbine engine test cells as part of the Automated Jet Engine Test Strategy (AJETS) program. AJETS is an innovative USAF program for improving existing engine test cells by providing more efficient and advanced monitoring, diagnostic and troubleshooting capabilities. This paper describes the basic design features of the AJETS system; including the associated data network, sensor validation and anomaly detection/diagnostic software that was implemented in both a real-time and post-test analysis mode. These advanced design features of AJETS are currently being evaluated and advanced utilizing data from TF39 test cell installations at Travis AFB and Dover AFB
  • Keywords
    aerospace computing; aerospace engines; aircraft testing; automatic test equipment; automatic testing; belief networks; fault diagnosis; feature extraction; gas turbines; military aircraft; vibration measurement; Bayesian network; Dover AFB; TF39 test cell; Travis AFB; USAF engine test cells; anomaly detection; automated jet engine test; engine gas path analysis; engine operating costs; engine vibration; fault classification; feature extraction; gas turbine engines; maintenance costs; mechanical faults; performance faults; real-time diagnostic; sensor validation; statistical models; test cell diagnostics; troubleshooting tools; Automatic testing; Cost function; Fault detection; Instruments; Jet engines; Monitoring; Performance analysis; Sensor phenomena and characterization; Sensor systems; Turbines;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Aerospace Conference, 2001, IEEE Proceedings.
  • Conference_Location
    Big Sky, MT
  • Print_ISBN
    0-7803-6599-2
  • Type

    conf

  • DOI
    10.1109/AERO.2001.931313
  • Filename
    931313