Numeric Simulation of Glider Winch Launches Andreas Gäb, Christoph Santel Communicated by Prof. Dr.-Ing. D. Moormann Institute of Flight System Dynamics RWTH Aachen University Jul. 28-Aug. 4, 2010 XXX OSTIV Congress, Szeged, Hungary 1
Contents Simulation tool Overview Main components Simulated procedure Results of analyses Reference case Effect of wind Kavalierstart Summary/Outlook Jul. 28-Aug. 4, 2010 XXX OSTIV Congress, Szeged, Hungary 2
Simulation Tool: Framework Main simulation as block diagram in Simulink Matlab GUI for easy access Outputs page Realized as a student s thesis Several extensions since then Inputs page Jul. 28-Aug. 4, 2010 XXX OSTIV Congress, Szeged, Hungary 3
Simulation Tool: Building Blocks Interaction of six main components: Aircraft Pilot Cable Winch Winch operator Atmosphere Each component as a subsystem Atmosphere included in aircraft and cable Main level of the Simulink model Jul. 28-Aug. 4, 2010 XXX OSTIV Congress, Szeged, Hungary 4
Simulation Tool: Aircraft 6 degree of freedom, rigid body Flat, non-rotating earth Forces and moments arising from Aerodynamics Gravity Cable Ground reaction Aerodynamics formulated as lookup tables, including ground effect Model corresponds to a training two-seater (simplified ASK 21 data); different aircraft possible Jul. 28-Aug. 4, 2010 XXX OSTIV Congress, Szeged, Hungary 5
Simulation Tool: Glider Pilot Approximation of human behavior by linear control theory Three separate channels: Maintain airspeed with elevator Maintain bank angle/ground track with aileron Minimize angle of sideslip with rudder Gains scheduled by scaling with inverse dynamic pressure Human factors (neuro-muscular delay, reaction time) Faded in after reaching safety altitude χ + C - χ Φ + C - k Φχ, T I, T D Φ k ξφ + q C q C q ξ p k ξp Aileron controller channel Jul. 28-Aug. 4, 2010 XXX OSTIV Congress, Szeged, Hungary 6
Simulation Tool: Winch/Winch Operator Winch: Simple drive train model 400 hp Diesel engine simulated Operator controls force with the engine throttle Measurement of cable force assumed Linear control theory, analogous to pilot model Jul. 28-Aug. 4, 2010 XXX OSTIV Congress, Szeged, Hungary 7
Simulation Tool: Cable Cable discretized into finite mass points, connected by massless cylinders Subject to weight, aerodynamic drag, ground reaction forces and internal cable tension Integration of equations of motion for all mass points Different types of cable (steel, synthetic) modeled by mechanic properties (mass, E modulus, ) m j+1 m j m j-1 T j-1 F ext,j T j Finite cable element with acting forces Jul. 28-Aug. 4, 2010 XXX OSTIV Congress, Szeged, Hungary 8
Simulation Tool: Ground Reaction Forces Spring/damper combination at predefined contact points Wheels, skids Exposed structure elements (wing tips) Finite cable elements Spring deflection and deflection speed calculated from equations of motion Vertical reaction force (spring force) produces horizontal friction force s µ F F = c s + d s' Forces acting on contact point Jul. 28-Aug. 4, 2010 XXX OSTIV Congress, Szeged, Hungary 9
Simulated Procedure Launch with constant cable force during main climb phase Pilot controls airspeed, climb angle as result Recent discussion in Germany 1 indicates possible safety benefits with this procedure Winch operator needs to know actual cable force Readily available with electric winches Diesel winch: constant throttle, but operator experience required to find optimal setting Initial force low (~0.5 glider weight), increase to 1.5 weight after liftoff 1 Eppler, R.: "Windenschlepp und optimale Ausklinkhöhe", Draft available from: www.daec.de/se/downfiles/2010/windenstart_prof_eppler20100301.pdf Jul. 28-Aug. 4, 2010 XXX OSTIV Congress, Szeged, Hungary 10
Reference Case 1000 m synthetic cable No wind Target airspeed 30 m/s (110 km/h) Glider weight ~5 kn Cable force as above F [N] 8000 7000 6000 5000 4000 3000 2000 1000 Reaches 431 m after Force on glider Commanded winch force 0 35 s of winching 0 5 10 15 20 25 30 35 t [s] Time history of winch forces during reference launch Jul. 28-Aug. 4, 2010 XXX OSTIV Congress, Szeged, Hungary 11
Reference Case: Flight Path -400-350 Flight path FEM cable at t=20 s Secant cable at t=20 s -300-250 z [m] -200-150 -100-50 0 0 100 200 300 400 500 600 700 800 900 1000 x [m] Cable sag illustrated by difference between FEM and secant cables Jul. 28-Aug. 4, 2010 XXX OSTIV Congress, Szeged, Hungary 12
Reference Case: Airspeed Airspeed safety margin is at least 25% during initial phase Drops to 18% shortly before force reduction Airspeed controller (pilot) gains authority at ca. t=10 s Safety margin defined as VTAS Vstall S = 100% V stall Speed [m/s] 40 35 30 25 20 15 10 5 True airspeed Stalling speed 0 0 5 10 15 20 25 30 35 t [s] Jul. 28-Aug. 4, 2010 XXX OSTIV Congress, Szeged, Hungary 13
Reference Case: Load Factors Aerodynamic load factor (lift/weight) well above 2 But canceled by cable force: vertical acceleration perceived by pilot is close to 1g Stalling speed increases with square root of aerodynamic load factor! Load factor [-] 2.5 2 1.5 1 0.5 n aero n Cable n Pilot 0 0 5 10 15 20 25 30 35 t [s] Jul. 28-Aug. 4, 2010 XXX OSTIV Congress, Szeged, Hungary 14
Wind Effects Constant longitudinal wind About 5 m altitude gain per km/h headwind Controller maintains speed margin from reference case even with tailwinds Real life pilot has to consciously disregard groundspeed when controlling airspeed z [m] -600-500 -400-300 -200-100 0 Headwind 10 m/s Headwind 5 m/s Headwind 2.5 m/s Reference case Tailwind 2.5 m/s Tailwind 5 m/s 0 100 200 300 400 500 600 700 800 x [m] Flight paths for head-/tailwind cases Jul. 28-Aug. 4, 2010 XXX OSTIV Congress, Szeged, Hungary 15
Kavalierstart Kavalierstart: excessively steep initial launch attitude May be caused by pilot (pulling too early) or winch operator (initial winch force too high for given glider) z [m] Altitude gain with Kavalier winch operator (more energy added to system), but reduced safety Kavalier pilot alone can only marginally increase altitude -500-400 -300-200 -100 0 Reference case Kavalier pilot Kavalier winch Both Kavalier 0 100 200 300 400 500 600 700 800 x [m] Flight paths for Kavalierstart cases Jul. 28-Aug. 4, 2010 XXX OSTIV Congress, Szeged, Hungary 16
Kavalierstart: Safety Safety margin to stalling speed significantly reduced in all cases Kavalier winch operator causes overshoot of allowable speed Pilot + operator both have responsibility for safe launch procedure True airspeed [m/s] Stalling speed [m/s] Safety margin [%] 50 40 30 20 10 0 0 5 10 15 20 25 30 35 40 50 40 30 20 10 0 0 5 10 15 20 25 30 35 40 40 30 20 Reference case Kavalier pilot Kavalier winch Both Kavalier 10 0 5 10 15 20 25 30 35 40 t [s] Jul. 28-Aug. 4, 2010 XXX OSTIV Congress, Szeged, Hungary 17
Summary Simulation framework realized for analysis of winch launches Comprises aircraft, winch, pilot and winch operator as linear control theory models, cable as FEM model Reference launch with constant cable force Safety of this procedure illustrated Safety reduction during Kavalierstart shown Winch operator and pilot both take part in ensuring a safe launch Jul. 28-Aug. 4, 2010 XXX OSTIV Congress, Szeged, Hungary 18
Outlook Blade Element aerodynamics model being realized to allow varying airflow in spanwise direction Background: simulate roll-over accidents during early launch phase More sophisticated winch models to allow comparison of Diesel and electric winches Jul. 28-Aug. 4, 2010 XXX OSTIV Congress, Szeged, Hungary 19
Thank you for your attention! Jul. 28-Aug. 4, 2010 XXX OSTIV Congress, Szeged, Hungary 20