Arning-Scalabilty of Model Flight Test Results

Microsystems & Electronics; Optronic Systems & Signal Processing

Scalability of Model Flight Test ResultsR. Arning

DGLR-Workshop on System Identification, Parameter Estimation and Optimisation

Ottobrunn, June 9th, 2005

Ingenieurbro Dr. Richard K. Arning


Overview Overview

Motivation of scaled model flight testing

Applicability of method

Quality influences

Results


Applicability of methodApplicability of method

Quality and relevance vs. model size ?

X-38ALFLEX

F-18

X-33NASA

NASA

NASA


Motivation of scaled model flight testingMotivation of scaled model flight testing

Advantages:

Full set (6-DOF) measurement of dynamic derivatives

No wind tunnel / sting mounting interference

Early flight data

Low cost

Low risk



Regions of constructional realisation for dynamically scaled models below 200 N MTOW and 200 N/m2 wing loading

0

20

40

60

80

100

120

140

160

180

200

G[N

]

1 2 3 4 5Spannweite [m]

Segelflzg.

Motorsegler

Raumflzg.

0

20

40

60

80

100

120

140

160

180

200

1 2 3 4 5Spannweite [m]

Segelflzg.

Motorsegler

Sportflzg.+Transportflzg.

Verkehrsflzg.

Glider

SailplaneMotor GliderGeneral Aviation

Spaceplane

Airliner

Span [m]

W

e

i

g

h

t

[

N

]

Limit for airliner models due to extreme high wing loading;

In general: To small Reynolds numbers

Wing loading for fighter aircraft models above limit


Applicable to aircraft:

with relative low relative mass ratio:

at low speeds (uncompressible flow)

with acceptable chord length at scale sized (wing, tail)

Applicable to general aviation and below category aircrafts &

space planes during take-off and landing phase

NOT applicable: In stall region and for performance testing


bSm

=

))original

1

altitude original

altitude model

5,1

original

modelmodel ReRe

=

ll

))


RS 180 Sportsman1:4,8-scale

Tail-Spin testing Tail-Spin testing

Speed Canard1:4-scale

MS 893 Morane1:4-scale

ELAC1:31.3-scale

ASK-211:3-scale

PHOENIX1:7-scale

System identification of winglets influence

System identification(closed and open loop)

Proof-of Concept /Engine failure characteristics

System identification of airbrakes influence

TT-621:5-scale

Applicability of methodApplicability of method IngenieurbroDr. Richard K. Arning


scale center of gravity

dynamically scaled mass m

dynamically scaled moments of inertia I

dynamically scaled control system (if relevant)

quality of sensors and calibration (integrated), redundancy of sensor information

additional external equipment (e.g. nose-boom)

simplification of aerodynamic model

coverage of flight envelope and maneuvers by recorded flight data

Quality influencesQuality influences

*lr

*lplll0ll rCpCCCCCC +++++=


Sensors and data acquisition system

Telemetriebodenstation





150

200

250

300

350

400

K

[

P

a

/

V

]

D

-12.0 -8.0 -4.0 0.0 4.0 8.0 12.0

[]

= -5 = 0 = 5

= 10 = 15 = 20

= 25 = 30




1.E-03

1.E-02

1.E-01

1.E+00

1 . E -03 1. E -02 1 . E -01 1. E +00 1. E +01 1 . E +02 1. E +03 1 . E +04 1. E +05

0.01

0.1

1

0.001

G

y

r

o

A

l

l

a

n

-

D

e

v

i

a

t

i

o

n

[

/

s

]

0.01

0.1

1

0.001

G

y

r

o

A

l

l

a

n

-

D

e

v

i

a

t

i

o

n

[

/

s

]

Gyro

Integration Time [s]

0.001 0.1 1 10 100 1000 104 1050.01

10-4

10-3

10-2

10-5 Ac

c

e

l

.

A

l

l

a

n

-

D

e

v

i

a

t

i

o

n

[

g

=

9

.

8

1

m

/

s

]

Accelerometer

Allan-plot of MEMS accelerometer and gyro


Estimated 1-Sigma Value of measurement systemELAC 1997 2005 possible

rotational rates 1,5 /s < 0,5 /s

acceleration 0,05 g < 0,01 g

, 1 0,5 control surfaces deflection 0,3 0,3

dynamic pressure ? combine with vel. measurement

mass 10 g 2 gIx, Iy, Iz 3% 3%Ixz 2 deviation from main axis 1 Ixy, Iyz set to 0 measure

position of center of gravity 2 mm 1 mm ?

position of aerodynamic 30 mm 30 mm

measurements



Example: Minaturised control system


Autopilot HW with fully integrated MEMS-based flight control sensors, 100 Hz control frequency

Weight < 25 grams

Micro UAV: DO-MAV

500 grams

0,42 m span


ELAC-space plane configuration with additional external installations(e.g. noseboom, skids)wind tunnel model scale factor 1:65

ELAC-space plane configuration free flyingmodel (rudderactuation)



-0,20

-0,15

-0,10

-0,05

0,00C [-]l

6,0 8,0 10,0 12,0 14,0 16,0 [-]

Freiflug

Windkanal

-0,8

-0,6

-0,4

-0,2

0,0C [-]l

6,0 8,0 10,0 12,0 14,0 16,0 [-]

Freiflug

Windkanal

0,00

0,02

0,04

0,06

0,08

0,10C [-]l

6,0 8,0 10,0 12,0 14,0 16,0 [-]

Freiflug

Windkanal

-0,40

-0,30

-0,20

-0,10

0,00C [-]lp

6,0 8,0 10,0 12,0 14,0 16,0 [-]

Freiflug

Windkanal

0,0

0,2

0,4

0,6

0,8

1,0C [-]lr

6,0 8,0 10,0 12,0 14,0 16,0 [-]

Freiflug

Vortex-Lattice

Comparison of roll moment derivatis btweenflight test and wind tunnelmeasurements

Results: Comparison wind tunnel / free flightResults: Comparison wind tunnel / free flight


From differences in wind tunnel and flight tests derived parameter uncertainties of rolling damping moment coefficient for space plane X-33

(Cobleigh: Development of the X-33 Aerodynamic Uncertainty ModelNASA TP-1998-206544, April 1998)

Results: Comparison with literature about deviation of wind tunnel and free flight dataResults: Comparison with literature about deviation of wind tunnel and free flight data


Results: Comparison with literature about typical deviation of wind tunnel and free flight dataResults: Comparison with literature about typical deviation of wind tunnel and free flight data

FALKE-Orbiter Space-Shuttle Orbiter

Study 1 Study 2 ELAC model

AC +0.08 +/- 0.050 +0.008 +0.036 AC +12.5% +0.8% +15.7%

WC -0.04 +0.07 +/- 0.0125 -0.010 +0.036mC +0.008 +/- 0.022 +/- 0.008 -0.009 -0.001 mC -5% -20% +40% +/- 28% -6.4% +0.3%

mqC +/- 80% -33.3% -19.2%mqC VL +13.0% +19.1%mqC DC -10.0% -1.2% QC -15% +/- 25% -0.140 +0.200 0.067 0.104 QC +/- 0.028 0.043 0.063 QC +/- 0.086 -0.046 -0.030 lC +0.024 +/- 0.030 +/- 0.045 0.001 0.042 lC -10% -25% +40% +/- 25% -10.8% 4.2% lC -6% -30% +60% +/- 0.011 -0.040 -0.200

lpC -16 +57% -40% +70% -1.2% 16.6%lrC VL -200% +150% 36.0% 152.7% nC +0.009 +/- 0.030 -0.065 +0.046 -0.070 -0.007 nC +/- 0.024 +/- 0.017 -0.056 -0.025 nC -17% -30% +60% +/- 26% -41.3% -25.2%

npC VL -200% +100% -42.4% -13.9%nrC +/-4% -35% +40% -27.5% 0.5%

VL: Vortex-Lattice DC: DATCOM-Methode


Comparison of ELAC result (deviation modelflight testing / wind tunnel) with parameteruncetainties from literature for X-33

Results: Comparison with literature about typical deviation of wind tunnel and free flight data

Results: Comparison with literature about typical deviation of wind tunnel and free flight data

-150

-100

-50

0

50

100

150

200

250

300[

%

]

C

A

C

W

C

m

C

m

C

W

K

m

q

C

V

L

m

q

C

D

C

m

q

-400

-300

-200

-100

0

100

200

300

[

%

]

C

Q

C

Q

C

Q

C

l

C

l

C

l

C

l

p

C

V

L

l

r

C

n

C

n

C

n

C

V

L

n

p

C

n

r


Method is applicable to subsonic/low speed flight regime and configurations with specific lower mass density (e.g. GA)

Quality of miniaturised (low-cost) sensors (drift & sensitivity) has been improved significantly over the recent years; aerodynamic sensors and nose-boom integration remain challenging

Even miniaturised autopilots are available today to investigate configurations with closed control loop and realistic position of center of gravity

For applications to determine the quasi-unsteady aerodynamics of these configuration the method is reliable enough today

SummarySummary

Arning-Scalabilty of Model Flight Test Results

Documents

scaled models

model size

control system

fighter aircraft models

airliner models

spannweite msegelflzg

scale sized wing

low speeds uncompressible