Data-Driven Modeling of Aircraft Engine Performance
Yashovardhan Sushil Chati

Citation
Yashovardhan Sushil Chati. "Data-Driven Modeling of Aircraft Engine Performance". Talk or presentation, 28, May, 2015.

Abstract
Knowledge of aircraft engine performance, and especially fuel consumption is important for various stakeholders, like the pilots, airlines, engine manufacturers, environmentalists and researchers in the field of air traffic management. Current practice in engine performance analysis involves the use of either gas turbine performance simulation softwares or data-driven models built on non-operational data. These methods cannot capture the variability in engine performance arising from a variability in operations. The objective of our research is to develop models of engine performance using operational flight data from Flight Data Recorders (FDRs) in combination with insights from the physical principles governing the operation of an aircraft engine. Regression trees (CART) are used to model the aircraft engine fuel flow rate as a function of the aircraft pressure altitude, ground speed, vertical speed and takeoff mass in the climb out and approach phases of flight. Prediction intervals from regression give an estimate of the variability observed in engine fuel flow rates for the same aircraft/engine type. The fuel flow rate results from the CART models are compared with those given by the ICAO Engine Exhaust Emissions Databank, a databank which gives point estimates of fuel flow rates for an engine from manufacturer conducted ground tests and which is used for developing fuel burn and emission inventories. For the different aircraft types in our FDR dataset, the CART models give a lower prediction error on the test dataset as compared to the ICAO Engine Exhaust Emissions Databank. The presentation ends with a discussion on the activities envisaged for the future towards fulfilling our research objective of developing operational data-driven models of engine performance.

Electronic downloads


Internal. This publication has been marked by the author for FORCES-only distribution, so electronic downloads are not available without logging in.
Citation formats  
  • HTML
    Yashovardhan Sushil Chati. <a
    href="http://www.cps-forces.org/pubs/72.html"
    ><i>Data-Driven Modeling of Aircraft Engine
    Performance</i></a>, Talk or presentation,  28,
    May, 2015.
  • Plain text
    Yashovardhan Sushil Chati. "Data-Driven Modeling of
    Aircraft Engine Performance". Talk or presentation, 
    28, May, 2015.
  • BibTeX
    @presentation{Chati15_DataDrivenModelingOfAircraftEnginePerformance,
        author = {Yashovardhan Sushil Chati},
        title = {Data-Driven Modeling of Aircraft Engine Performance},
        day = {28},
        month = {May},
        year = {2015},
        abstract = {Knowledge of aircraft engine performance, and
                  especially fuel consumption is important for
                  various stakeholders, like the pilots, airlines,
                  engine manufacturers, environmentalists and
                  researchers in the field of air traffic
                  management. Current practice in engine performance
                  analysis involves the use of either gas turbine
                  performance simulation softwares or data-driven
                  models built on non-operational data. These
                  methods cannot capture the variability in engine
                  performance arising from a variability in
                  operations. The objective of our research is to
                  develop models of engine performance using
                  operational flight data from Flight Data Recorders
                  (FDRs) in combination with insights from the
                  physical principles governing the operation of an
                  aircraft engine. Regression trees (CART) are used
                  to model the aircraft engine fuel flow rate as a
                  function of the aircraft pressure altitude, ground
                  speed, vertical speed and takeoff mass in the
                  climb out and approach phases of flight.
                  Prediction intervals from regression give an
                  estimate of the variability observed in engine
                  fuel flow rates for the same aircraft/engine type.
                  The fuel flow rate results from the CART models
                  are compared with those given by the ICAO Engine
                  Exhaust Emissions Databank, a databank which gives
                  point estimates of fuel flow rates for an engine
                  from manufacturer conducted ground tests and which
                  is used for developing fuel burn and emission
                  inventories. For the different aircraft types in
                  our FDR dataset, the CART models give a lower
                  prediction error on the test dataset as compared
                  to the ICAO Engine Exhaust Emissions Databank. The
                  presentation ends with a discussion on the
                  activities envisaged for the future towards
                  fulfilling our research objective of developing
                  operational data-driven models of engine
                  performance.},
        URL = {http://cps-forces.org/pubs/72.html}
    }
    

Posted by Carolyn Winter on 10 Jun 2015.
Groups: forces
For additional information, see the Publications FAQ or contact webmaster at cps-forces org.

Notice: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright.