Document
National Aeronautics and Space Administration
A Dynamic Model for the Evaluation of Aircraft
Engine Icing Detection and Control - Based
Mitigation Strategies
Paper GT2017 - 65128
Donald L. Simon Aidan W. Rinehart Scott M. Jones
NASA Glenn Research Center Vantage Partners, LLC NASA Glenn Research Center
21000 Brookpark Road 3000 Aerospace Parkway 21000 Brookpark Road
Cleveland, OH, 44135 Cleveland, OH, 44135
Brook Park, OH 44142
ASME Turbo Expo 2017 June 26 - 30, 2017 Charlotte, NC www.nasa.gov National Aeronautics and Space Administration
Outline
• Background:
– The ice particle threat to engines in flight
– A control - based approach to icing risk mitigation
• Dynamic engine model overview and features
– Closed - loop control logic
– Heat extraction due to ice particle ingestion
– Flow blockage due to ice buildup in engine compression
system
– Engine actuators
• Comparison of dynamic engine model to engine ice crystal
icing test cell data and manufacturer’s customer deck
• Summary
www.nasa.gov National Aeronautics and Space Administration
The Ice Particle Threat to Engines in Flight
• Since 1990, there have been a number of jet engine powerloss events reported on aircraft operating in ice particle conditions o Temporary or sustained power loss, engine uncontrollability , engine shutdown • Ice crystals enter the engine’s core, melt, and accrete on engine components during flight • Many possible causes of power loss: o Damage due to ice shedding o Flame - out due to combustor ice ingestion o Compressor surge o Sensor icing o Engine rollback • Within the aviation community, research is ongoing to characterize the environmental conditions under which engine icing can occur , understand the mechanisms by which ice particles can accrete on engine components, Images courtesy of NASA and develop mitigation strategies.
www.nasa.gov National Aeronautics and Space Administration
Control - Based Icing Risk Mitigation
• Atmospheric conditions, including ice crystals • Engine operating condition
• Potential mitigations to engine icing
problem include
Aircraft Actuator
– Avoidance of flight through ice crystal
Engine commands
atmospheric conditions
– R e - design of engine hardware
Engine Aircraft & engine Control
– Ice protection systems
sensor measurements Icing detection logic
• Active control icing risk mitigation
architecture
Ice accretion risk Activate calculation
– Includes detection and control - based
control No
mitigation logic
mitigation strategy Icing
• This paper presents a dynamic aircraft risk?
engine model created for the initial
Yes
development and evaluation of aircraft
engine icing detection and control -
Active Control Icing Risk
based mitigation strategies.
Mitigation Architecture
www.nasa.gov National Aeronautics and Space Administration
Dynamic Engine Model Overview
Engine model block diagram
• Dynamic model of Honeywell ALF502R - 5 turbofan engine – Experimental versions of this engine underwent engine icing testing at NASA Glenn in 2013 and 2015 – 0D component level model – Derived from Numerical Propulsion System Simulation (NPSS) model of the ALF502R - 5 – Coded in the Matlab /Simulink environment using a NASA - developed open - source thermodynamic simulation package – Toolbox for the Modeling and Analysis of Thermodynamic Systems (T - MATS) – Includes performance losses caused by heat loss due to ice particle ingestion and ice blockage in the engine’s compression system – Includes engine control logic enabling the simulation of transient engine operation www.nasa.gov National Aeronautics and Space Administration
Heat Extraction Due to Ice Particle Ingestion
• Model includes heat (enthalpy) extraction effects to account for the phase
transition ( ice→water→vapor ) that ingested ice particles undergo as they pass
through the engine’s compression system.
• Heat extraction is modeled to occur both within the LPC and the HPC: T T c w H w T T c w Q extraction heat LPC 25 2 : melt water ice f ice melt ice ice LPC T T c w H w T T c w Q extraction heat HPC 3 25 : boil steam ice v ice boil water ice HPC c = specific heat of water H = heat of fusion of ice w = ice mass flow rate water f ice c = specific heat of ice H = heat of vaporization of water T = ice melting temp ice v melt T = water boiling temp boil www.nasa.gov National Aeronautics and Space Administration
Flow Blockage Due to Ice Buildup in LPC
• Model includes an “LPC ice blockage” input, a lumped parameter that captures LPC
performance changes due to ice accretion
• Captured through a series of modified LPC maps, each representing a different
amount of ice blockage (maps generated from NASA - developed mean line
compressor code (COMDES))
• Results in a series of maps that can be stacked and interpolated between to
simulate changing levels of ice blockage
Stacked series of LPC compressor maps reflecting increasing levels of ice blockage www.nasa.gov National Aeronautics and Space Administration
Engine Control Logic
• User can operate the engine in either open - loop or closed - loop control
mode
– In open - loop operation, user supplies fuel flow input
– In closed - loop operation, user specifies power lever angle (PLA), and engine
operates on core speed (N2) governor droop line
– Closed - loop control logic allows the model to emulate the ALF502R - 5 engine’s
response to ice particle ingestion and ice blockage in the LPC
“Iced” engine “Iced” engine N2 governor operating line operating point droop line N2 governor Increasing droop line Shift in operating line PLA due to ice ingestion and increasing ice blockage Wf Wf P3 P3 Decreasing PLA Nominal engine Nominal engine operating line operating line Nominal engine Nominal engine operating point operating point N2 N2 Movement of N2 governor Movement of engine operating line caused by heat droop line with changing PLA transfer due to ice ingestion and increasing ice blockage www.nasa.gov National Aeronautics and Space Administration
Auxiliary Actuators
• Four auxiliary actuators added to model – enables future studies to assess • Four auxiliary actuators added to model – enables future studies to assess how modulation of these actuators impacts ice accretion risk. how modulation of these actuators impacts ice accretion risk.
– Fan customer bleed
– Core customer bleed
– Anti - ice bleed
– Horsepower extraction
www.nasa.gov National Aeronautics and Space Administration
Aircraft Engine Ice Crystal Icing Testing in
NASA Glenn Propulsion Systems Laboratory (PSL)
• The NASA Glenn PSL is an altitude simulation facility for experimental
research on air - breathing propulsion systems
• A PSL test cell has been upgraded to include a water spray nozzle array
system to produce simulated ice crystal cloud conditions
• Experimental versions of Honeywell ALF502R - 5 engine underwent ice
crystal icing testing in PSL in 2013 (LF01) and 2015 (LF11).
Water injection Experimental ALF502R - 5 spray bars installed engine installed in PSL in PSL test cell test cell Normalized measurement parameters recorded during uncommanded engine rollback event caused by ice crystal icing (LF01 Run 193) www.nasa.gov National Aeronautics and Space Administration
Comparison of Dynamic Engine Model
to LF01 Engine Experimental Data
• Model was run under both open - loop and closed - loop control mode
– Recorded parameters of altitude, Mach, dTamb , Wf (open - loop only), and PLA (closed - loop only) were supplied as model inputs – Additional model input parameters of ice particle concentration and LPC ice blockage were determined based on experimental data • Ice particle concentration profile is calculated as a function of measured spray bar water flow rate and engine volumetric flow rate, with scale factor adjustment to produce a comparable temperature drop as that observed in recorded HPC exit temperature (T3) • The percentage of LPC ice blockage was not measurable during the test. Model input of this parameter was selected to match measured engine core speed (N2) response.
Calculated ice particle concentration and T3 Calculated LPC ice blockage and N2 response response during LF01 Run 193 rollback event during LF01 Run 193 rollback event www.nasa.gov National Aeronautics and Space Administration
Modeling of LF01 Run 193 Engine Rollback Event
• Run 193 rollback event was
simulated by running the dynamic
engine model in both open - loop and
closed - loop control mode
• Flight condition for Run 193 was 28K
feet, 0.5 Mach, and ISA +28˚F
Model Input Parameters (in addition to Normalized Engine and Model Output Parameters Alt, Mach, and dTamb ) • Fan speed (N1) • Core speed (N2) • Ice concentration • HPC exit pressure (P3) • HPC exit temp (T3) • % ice blockage • Exhaust gas temp (T45) • Fuel flow ( Wf ) • Fuel flow ( Wf ): open - loop only • LPC exit temp (T25) • Power lever angle (PLA): closed - loop only www.nasa.gov National Aeronautics and Space Administration
Modeling of Additional LF01 Engine
Rollback Events
• In addition to Run 193, four additional LF01 test runs (Run 199, Run 206, Run 570 and Run 940) were selected for comparison against the model • Input parameters for these runs are shown below
Model Input Parameters
www.nasa.gov National Aeronautics and Space Administration
Modeling of Additional LF01 Engine
Rollback Events (cont.)
Model Output Parameters
www.nasa.gov National Aeronautics and Space Administration
Comparison of Engine Model to Customer Deck
• In follow - on studies, the developed engine model will be used to evaluate the feasibility of control - based strategies for mitigating the risk of engine icing. This will entail modulation of the model’s auxiliary actuators and assessing the corresponding impact on icing risk .
• The manufacturer’s steady - state customer deck was used to assess correct implementation of auxiliary actuators within the model.
Comparison of Model and Customer Deck Steady - State
Response to Actuator Modulation
www.nasa.gov National Aeronautics and Space Administration
Summary
• A dynamic model of the ALF502 - 5R turbofan engine has been
developed and evaluated
• Model was shown to emulate engine system - level behavior during ice
crystal icing test cell evaluations as well as the steady - state outputs
produced by the manufacturer’s customer deck
• Key features of the model include
– Closed - loop controller allowing the simulation of engine transients
– Heat extraction effects reflecting the heat loss the engine experiences as
ingested ice crystals melt and vaporize in its compression system
– Flow blockage effects reflecting the buildup of ice in the engine’s low
pressure compressor
– Auxiliary actuators enabling the modulation of engine performance
• Potential follow - on work
– The model can be used in follow - on studies to develop and evaluate
potential icing risk detection and control - based mitigation strategies
www.nasa.gov National Aeronautics and Space Administration
Acknowledgments
• This work was conducted under the NASA Advanced Air
Vehicles Program, Advanced Air Transportation Technologies
Project. The authors graciously acknowledge Honeywell
Engines and the Ice Crystal Consortium for their support of
the engine testing that enabled this work.
www.nasa.gov