Parametric Airfoils and Wings

Helmut Sobieczky

DLR German Aerospace Research Establishment

Bunsenstr. 10, D-37073 Gottingen ,Germany
e-mail: Helmut.Sobieczky@dlr.de

Summary
Explicit mathematical functions are used for 2D curve definition for airfoil design. Flowphe-
nomena-oriented parameters control geometrical and aerodynamic properties. Airfoil shapes
are blended with known analytical section formulae. Generic variable camber wing sections and
multicomponent airfoils are generated. For 3D wing definition all parameters are made func-
tions of a third spanwise coordinate. High lift systems are defined kinematically by modelled
track gear geometries, translation and rotation in 3D space. Examples for parameter variation in
numerical optimization, mechanical adaptation and for unsteady coupling of flow and configu-
ration are presented.

Introduction
Airfoil and wing design methodologies have made large steps forward through the availability
of rapid computational tools which allow for specification of goals in aerodynamic perform-
ance. These goals are mainly to increase a measure of efficiency, like the ratio of lift over drag,
or, in the higher speed regimes, its product with flight Mach number. The need for increased lift
at higher flight speed, with drag kept low, has led to the development of knowledge bases for
aerodynamic design: The art of shaping lift generating devices like aircraft wings is based on
geometric, mechanical and fluid dynamic modelling, carried out with the help of mathematical
tools on rapid computers. Given a designer's refined knowledge about the occurring flow phe-
nomena, his goal may be to obtain certain pressure distributions on wing surfaces: This may be
reached by inverse approaches with a shape resulting from the effort, or by applying optimiza-
tion strategies to drive results toward ideal values.
With such methods we have refined tools available for extending our practical knowledge how
the geometries of airfoils and wings are related to pressure distributions and aerodynamic per-
formance. Certain details of desirable pressure distributions require a modelling of details in the
boundary condition, usually a special feature of the curvature distribution. This is true especial-
ly in the transonic flow regime, where favorable as well as undesirable aerodynamic phenomena
are correctly modelled by certain weak or strong singularities in the local mathematical flow
structure including the flow boundary. Numerical optimization methods iteratively adjusting
the resulting 2D or 3D shapes usually employ smoothing algorithms based on polynomials,
splines and similar algebraic functions. These functions may be ignoring local properties of the
shape being compatible to the inverse input, while they should accomodate the results from an-
alytical inverse methodology using hodograph formulations of the governing equations. Hodo-
graph-type methods, though not practical tools, have led to a deeper understanding about the
relations between surface geometry and the structure of recompression shocks. These methods
are most usefully applied to designing nearly shock-free airfoils and wings with favorable off-

71

K. Fujii et al. (eds.), Recent Development of Aerodynamic Design Methodologies

© Friedr. Vieweg & Sohn Verlagsgesellschaft mbH, Braunschweig/Wiesbaden 1999
design behavior. Understanding the resulting refined shapes and modelling them in a direct ap-
proach with a suitable geometry generator is a continuing challenge for more complex 3D con-
figurations like complete aircraft, turbo machinery components and models for interdisciplinary
design.
The present contribution is aimed at using explicit mathematical functions with a set of free pa-
rameters to define wing surfaces of practical interest for realistic aircraft applications, with a
potential to arrive at optimum values of objective functions like aerodynamic efficiency, with a
minimum of parameters having to be varied, because these parameters are defined from appli-
cation of the fluid mechanic and gasdynamic knowledge base or prescribed by modelling kine-
matic models of a mechanical adaptation device.

Geometry generator
In the series of Notes on Numerical Fluid Mechanics the author has had the chance to present
concepts, tools and examples of shape definition for aerodynamic components, with a strong
emphasis on using mathematical functions which are drawn from analytical modelling of flow
phenomena as they occur in the transonic regime. The need for reduction of shock losses has
sparked an inverse procedure to find shock-free airfoils and wings, with the additional option to
adapt wing geometries to varying operating conditions [1]. The increased need for creating test
cases for numerical flow simulation (CFD), along with the requirements for precise definition
of boundary conditions has then inspired the presentation of a wing within a transonic wind tun-
nel, with all boundaries including the tunnel and the inlet and exit flow conditions given [2], to
be simulated and compared with experiments [3]. Later, the mathematical tools for defining
such boundary conditions were further developed to model real aircraft components: wings, fu-
selages, propulsion components and their integration to complete configurations [4]. Since then,
various applications have been studied and more recent refinements led to several versions of
"geometry preprocessor software tools". These support modem developments in a multidisci-
plinary design environment for aerospace components and not restricted to aerodynamic opti-
mization.
Aircraft wings are the primary subject to optimization efforts, progress in aerodynamic design
methodology is mostly influenced by new ideas to improve the lift-generating devices. Airfoils
are the basic elements of wing geometry, they determine a large share of wing flow phenomena
though they are just two-dimensional (2D) sections of the physical wing surface. Well-known
aspects of wing theory are the reason for options of such idealization, with a large accumulated
knowledge base resulting for 2D airfoil theory. It has, therefore, been well founded to use airfoil
shapes with documented performance results from wind tunnel tests for the design of wing
shapes. These airfoils are usually contained in published or proprietary data bases, we use them
as dense data sets to describe the sections of wings with planform, twist and dihedral given by
analytical model functions. Properties relevant for flow quality, for instance curvature, of these
latter functions are simple and easily controlled by parameters while the airfoil input data are to
be spline-interpolated to obtain a required distribution of surface data. With all the experience
gained by using our shape-generating tools and updating them with recent developments in de-
signing high speed flow examples, an effort is made to generate 2D wing sections in the same
way the 3D shape parameters are already defined. Suitable functions should replace the hitherto
required airfoil data sets. The goal is to propose functions with a minimum set of input param-
eters for shape variation, function structure and their parameters chosen to address special aer-
odynamic or fluid mechanic phenomena. This desirably relatively small number of control
parameters will then effectively support optimization procedures.

72
 Airfoil functions
With airfoil theory and airfoil data bases being well established components of applied aerody-
namics on the ground of lifting wing theory, it is necessary to allow for using such data as a
direct input in any wing geometry definition program. This fact was the motivation to provide
spline interpolation for such given airfoil data in a first version of our geometry code, which has
been described in various papers and publications. Recently these developments have been
summarized in [5] , here we focus on continuing this activity in the area of describing airfoils
with more a sophisticated method than providing a set of spline supports.
Functions to describe airfoil sections are known for many applications, like the NACA 4 and 5
digit airfoils and other standard sections. Aircraft and turbomachinery industry have developed
their own mathematical tools to create specific wing and blade sections, suitably allowing par-
ametric variation within certain boundaries. We define such functions for airfoil coordinates in
coordinates X, Z non-dimensionalized with wing chord therefore quite generally

with P = (Pl, pz, ... , Pk) a parameter vector with k components and Fj a special function using
these parameters in a way determined by a switch j. The goal is to try to keep the number k of
needed parameters as low as possible while controlling the important aerodynamic features ef-
fectively.
z

a x
X=1

Ql0r-~--~----~------r------r-----'

...... NACA 0012

- blend PARSEC I NACA
-PARSEC
- blend PARSEC I Whitcomb

x 1-+ Whitcomb 12%

-0. 1fb.'-::o,----,;oc-;.2:;----;o-:-.4;----:o~.6;::--~-0;;-.-;;-8- --:'1.0

Fig. 1: "PARSEC" airfoil geometry defined by 11 basic parameters: leading edge ra-
dius, upper and lower crest location including curvature there, trailing edge coordi-
nate (at X = 1), thickness, direction and wedge angle, (a) .
Example: Variations of PARSEC airfoil by blending with NACA or Whitcomb airfoil (b)

73
Figure 1 illustrates 11 basic parameters for an airfoil family "PARSEC" which we found quite
useful for applications. There is strong control over curvature by prescribing leading edge radius
and upper and lower crest curvatures. Similar to 4Digit NACA series we choose a polynomial,
though of a higher (6th) order:

6
'" an(p) . Xn -112
Z PARSEC = £..
n=1

for upper and lower surface independently, the coefficients lin determined from the given geo-
metric parameters as illustrated in Fig.l. Comparison with other new or well known airfoil gen-
erator functions is made possible by including those functions in the software, a combination of
individual features is then straightforward:

Blending of different aiifoil generator functions

With an additional blending parameter Pmix, some known airfoils are included in this geometry
tool as basic default functions, like NACA series airfoils as coded by Ladson [6]and Whit-
comb's supercritical wing sections as coded by Eberle [7]. These known sections require input
of a subgroup of the above 11 basic parameters and they can be blended in with the more refined
geometries.
Figure 1b shows an example of an airfoil series whith the 11 basic parameters kept fixed and
using only the blending parameter, resulting in two different variations of the special choice
PARSEC airfoil: Blending with an NACA 4Digit section for -1 < Pmix < 0 and blending with a
12% thick Whitcomb airfoil for 0 < Pmix < 1.

Example: Transonic aiifoils

The 11 basic parameters in Fig. 1b are selected to re-model an efficient wing section from a pre-
vious study using the above mentioned spline support airfoil definition technique. A 300swept
wing with a 12% thick main section had been designed for Mach = 0.85 and Re = 40 Mill. First
favorable results of CFD analysis suggested a more detailed development of this wing and its
main section. The PARSEC routine is applied here by choosing the 11 basic parameters directly
from analysis of the spline support section. Application of swept wing theory requires a thick-
ening by a factor of 1.1547 resulting in the airfoil to be investigated in Mach = 0.74. Here and
in the following the Drela-Zores airfoil expert system software [8] is used for fast viscous tran-
sonic flow analysis, with pressure distributions, dragrise and polars resulting, Fig. 2.

0.8rL...-_ _ _-::;!.-....
c,
0.7

0.6

0.5

0.00~.7 0.72 0.74 0.76

M_
Fig. 2: Dragrise and lift/drag polar for Re = 40 Mill. transonic flow past PARSEC airfoil
(symbol 0) and Whitcomb airfoil (symbol +) using 6 of the 11 basic parameters. De-
sign conditions at Mach =0.74, = 0.6. c,
74
Comparative Whitcomb section results are shown, too, indicating that for the relatively low
transonic design Mach number 0.74 the PARSEC airfoil seems well suited, (ratio lift/drag =80),
while the use in higher design Mach numbers a blending with a Whitcomb section - or a varia-
tion of the basic PARSEC parameters guided by the Whitcomb airfoil geometry - will be a way
to optimize the wing section with a relatively modest effort.

Parameter variations
The parametric airfoil generator PARSEC allows for control of curvature at the nose, at the up-
per and at the lower crest. With these additional degrees of freedom - compared to airfoil func-
tions without curvature control - we may vary aerodynamic performance and shift the optimum
conditions to desired operation conditions. The example illustrated in Fig. 3 shows a variation
of the above PARSEC airfoil by, first, only increasing the leading edge radius and, second, also
decreasing the upper crest curvature, which is suggested by the analyzed curvature values for
the Whitcomb airfoil. We see a shift of the drag rise toward higher Mach numbers. Other pa-
rameter variations give similar substantial changes in performance. Here we stress the observed
fact that for the PARSEC airfoil model function some single parameter changes may already
improve a given section for selected operating conditions.

O.025~--~--~--~--~
G-El basic PARSEC (Fig. 1b)
I3-£J modified 'Ie
~ plus modified Zxx,up

+-r basic Whitcomb (Fig. 1b)

M=
Fig, 3: Shifting drag rise to higher Mach numbers by changing single parameters

Trailing edge (TE) variations

Refined control of viscous flow parameters near the wing trailing edge may influence circula-
tion and hence aerodynamic efficiency quite remarkably. In the past this has led to specially de-
signed airfoil and wing TE shapes: Special solutions to the outer in viscid flow model equations
were proposed to create a flow field in the vicinity of the TE which has a favorable pressure
gradient on the airfoil surface. One little known theoretical base for the shaping of high perform-
ance airfoils has been presented by Garabedian [9]. Based on a complex hodograph analysis,
the principle can be modelled by increased curvature only quite closely at the TE, to counteract
the boundary layer's de-cambering effect.
This occurs on the upper surface more locally than on the lower surface. The practical conse-
quence for physically relevant airfoils which are not having negative thickness or too thin TE's,
is a blunt TE base, a convex upper surface contour shaping with curvature increasing toward the
TE and a more evenly distributed curvature on the concave lower surface, resulting in a mini-
mum thickness of the airfoil a few percent upstream of the TE. Such TE refinements have been

75
studied on practical wing sections and have been termed 'Divergent Trailing Edge - DTE' air-
foils [10],[11] . Modifications based on the hodograph analysis are added to the basic PARSEC
shape: In the simplest case a single additional parameter L\a controls the functions added to air-
foil upper and lower surface to become a DTE wing section, see Figure 4. Based on our hitherto
quite limited experience with case studies, modification lengths L j ,2 range between 20 and 50
% of airfoil chord, for the exponents values we use n =3 and 1.8> 11 > 1.3 (Garabedian's ho-
dograph solutions suggest 11 = 4/3).

Lz.
:' 4
~2 :1 ~
X1
Fig. 4: Local airfoil geometry modifications to model a divergent trailing edge.

Local surface bumps

Transonic airfoils of high efficiency differ from classical low speed airfoils mainly due to their
delicate curvature distribution on the upper (lifting) surface where supersonic flow conditions
occur. In the case of exactly shock-free flow we observe distinct curvature maxima close to
where local Mach number unity flow is found ; these shape details can be understood from lo-
cally valid model solutions to the inviscid basic equations which have led to systematic design
methods based on operational CFD analysis codes; a review of this concept with a more detailed
discussion of the interaction of transonic gasdynamics with geometrical boundaries is given by
the author in [12]. It is shown that the addition of two suitable bumps to a given conventional
airfoil can convert the flow to be shock-free in high subsonic Mach numbers. This is done by a
first bump near the leading edge which triggers a cluster of expansion waves, and a second one
absorbs recompression waves which coalesce near the sonic recompression.
Based on this established knowledge for the design of transonic airfoils with high aerodynamic
efficiency it has been found useful to influence local curvature of given airfoils in critical re-
gions by surface bumps of varying shape and size, which can be built in an aircraft wing as a
flow control device. Even reducing this effort to only adding a very small bump, extending to 2
- 3 % of chord near the location of a recompression shock, has been claimed to improve aero-
dynamic performance by dispersing the shock at the foot point on the airfoil, thus favorably in-
fluencing shock - boundary layer interaction [13].

76
 C!.2 = 2m· S i n(f(~)l(l;)

f(~) = a~+be
g(~) (P-QI;)· (I-(I-c) · sin 1;)

Z
Zm, ZXX,m

Z -IX-X2/f

..
X1 Xm X2 X
I
0 1t ~

Fig. 5: Local airfoil geometry modifications to model a bump with strong shape control.

A function for arbitrary bumps has been added therefore to the airfoil generator program which
generates bumps (Fig. 5) with strong local curvature control. Chordwise extent (Xl, X z) is de-
fined by choice of the local variable ~ . Possible requirement of an unsymmetrical bump crest
location (X m, Zm) will be taken care for by coefficients a and b, crest curvature (ZXXm) and cur-
vature control at the bump ramps (e, f) are controlled by the coefficients P, Q and c in the equa-
tion for g(x) as can easily be verified.

Variable camber models

Flexible parts for a wing geometry may be used for widening the range of optimum efficiency
in variable operating conditions. Structural constraints restrict the use of such parts to the areas
of trailing and leading edges. For ensuring acceptable flow quality, sealed flaps at the TE and
sealed slats at the leading edge (LE) maintain the smooth contour without gaps or comers. Set-
ting up a geometrical model for realistic wings or wing sections of course depends on knowl-
edge of the mechanical device putting the concept to reality. For rapid predesign studies, the
task to geometrically model such contour variations is to define only the deflection angle ffi as
a single input parameter, for a given kinematics (in the simplest case the hinge point H coordi-
nates) and fixed domain (.1X) of some elastic surface modification. The sketch Figure 6 illus-
trates this for both a sealed slat and a sealed flap.
Without further specification of the elastic surface mechanics a simple analytic function pro-
vides a smooth connection between the original airfoil and its rotated nose or tail portion. For a
specified mechanism solving the problem of connecting the solid parts with an elastic and
sealed contour the model function needs to be adapted to the hardware data.

77
 t::.Xs
ZI ~

~---------------

~s~-- _ _ _ _ _ _ _ _ _- - - - -

Fig. 6: Local geometry variation at leading and trailing edges by sealed slat and flap
model

Multicomponent airfoils
A more complex task of generic parameterized modelling is the geometry of high lift compo-
nents like Fowler flaps and slats. Here we start also from a given airfoil, but we need to carve
out separated lift-generating airfoils from the nose area and one or more of such sections at the
rear portion of the basic airfoil. Figure 7 shows the added geometry details for a given airfoil
modified to include a single slat which can be moved by a combined translation and rotation.
Choice of coordinates for CO, C 1 and C2 and curvatures there define curve functions similar to
the above PARSEC approach for the remaining fixed airfoil portion (or similarly at the flap nose
portion).

Fig. 7: Selecting a portion of airfoil contour (C 1C:z) to carve a slat geometry cs. Added
functions for carved surface, coupled translation (0 0 0) and rotation angle (0.

78
Figure 8 shows an airfoil with a single flap, various positions depicted. Drela's code for multi-
component airfoils [8] can be used for a rapid manual optimization of the flap track to obtain
high lift coefficients. An estimation of the separation bubble displacement within the flap bay,
to be modelled for each flap (or slat) position, is quite helpful for these pre-design studies: with
the same approach it is straightforward to model also a viscous displacement contour cy to re-
place the concave surface parts. More refined analysis using a NavierlStokes solver is needed
to calibrate cy for the faster analysis methods, but the parameters to do so in a flexible way may
be available already.
It should be stressed that these 2D mUlticomponent airfoils are to be used in the cruising (re-
tracted) configuration only for 3D applications: Swept wings will require a 3D definition of flap
tracks and a shifting and rotation of the whole 3D flap or slat, which cannot be modelled for
each section in a 2D fashion . The modelled retracted 2D components are the baseline for the
real 3D high lift system.

5
detail (flap pos. 3)

Fig. 8: Modelling flap geometry cf and viscous flow replacement contour Cv for fast 20
flow high lift computation. Pressure distributions for original airfoil (1) and two flap po-
sitions (3, 5) in incompressible, in viscid flow, estimated separation in flap bay mod-
elled.

79
 Wing geometries with spanwise section variation
So far our shaping of aerodynamic components is restricted to a 2D space (X, Z), which is non-
dimensionalized with airfoil chord. In the following, this chord will be a function of the span-
wise coordinate, Y =Yo' of a 3D wing, which is the independent variable to scale, shift and ro-
tate each wing section in 3D space (x, y, z).
The flexible geometry generator for 3D wings, also laid out for curve and surface definition
based on suitable parameter input [5], so far makes use of a number of airfoils as 'support sec-
tions' at given spanwise positions; blending functions defined within the resulting intervals give
a section geometry at every spanwise station. For an already very precisely given wing with
many support sections and small intervals, this section blending is used merely for a linear in-
terpolation to obtain a redistributed or refined surface grid. Such approach, in principle, may
lead to inaccuracies in spanwise smoothness, which does not occur from a definition with only
few support sections.
Application of the analytically defined airfoils as wing sections with a smooth variation of the
parameters (here, applying the PARSEC functions, the 11 basic and optionally a few parameters
for surface bumps and trailing edge variations) guarantees surface smoothness to a desired de-
gree with also just few input data and still an option for strong surface variation along span.

Periodic airfoil deformation

Not yet a wing configuration, but introducing time as a third dimension defines a 3D boundary
condition, see the illustration Figure 7: We applied the new parameterized shapes for the gen-
eration of airfoil systems with periodic geometry changes. So far we have studied applications
to new helicopter rotor blades with shape adaptation for suppression of dynamic stall: a periodic
nose drooping to a given airfoil using the geometry manipulation as illustrated for the sealed
slat, Fig. 6, was applied, in certain phase with a periodically changing angle of attack. Results
are obtained indicating a very favorable delay of unsteady separation in the low speed phase of
the retreating rotor blade [14]. An unsteady N/S code is used with a grid conforming and mov-
ing with the varying airfoil geometry. This concept of using periodically adaptive airfoils is also
applied with a refined nose curvature variation to control shock-boundary layer interaction in
high angle of attack airfoil flow [15]. The study revealed the dramatic role of viscous transonic
phenomena occurring in low speed aerodynamics: a small supersonic bubble forms at the lead-
ing edge at high angle of attack. Because of high curvature the recompression shock terminating
this supersonic domain is quite strong. Our earlier approach to design shock-free supercritical
airfoils has taught us to apply bumps for shock suppression; it seems that such approach may be
successfully used very locally with small adaptive bumps at the leading edge in low speed con-
figurations, which might be realized without excessive mechanical effort.
Software tools developed for the comparative visualization of 3D CFD results are suitably ap-
plied here for 2D unsteady flows: Color or 'zebra' isofringes for surface pressure show the onset
of separation, Fig. 9. Iso-surfaces like the sonic bubble M = 1, are displayed as shaded surfaces
and give an impression of the extent of the observed phenomena. Creating this 3D visualization
of an unsteady 2D flow, as well as the use of video animation of such case studies and their 4D
extension if a 3D unsteady process is being investigated, have a high educational value for iden-
tifying the relative importance of single parameters to be varied.

80
 Nose drooped
within time interval
o. T1 < T < T3

-. 05~~
o. .05 . 10 .... . 15
xl"
Unsteady surface pressure
distribution indicating
delayed dynamic stall

supersonic bubble
visualized

Fig. 9: Results for unsteady airfoil flow with periodically drooped nose (Ref. 15): Color
isofringes for surface pressure visualization, M~ = 0.3. Control of locally occurring su-
personic flow and viscous interaction triggering downstream boundary layer separa-
tion.

Aiifoil parameters for wing geometry definition

In a new modification of our wing generator software, the previously used support airfoils at
selected stations along wing span may now be replaced by defining a set of airfoil parameters
like those explained above for the PARSEC family, as functions along span, just like the already
operational way of defining functions for leading and trailing edge coordinates and local wing
section twist. Compared to the amount and flexibility of input data for a set of special support
sections, the new approach seems to open a better use especially for optimization strategies, us-
ing the well proven concept of composing arbitrary spanwise curves for geometry and distribu-
tions with only a few key parameters. Figure 10 illustrates both options to define wing sections
at any spanwise station in the wing coordinate system (x, Yo).

81
 xl x

:tl l l l l Il l l l l l l~ ~
···~ ' ~
....
y =Yo Y=Yo
·· .
i=2,

~
i= 1, i=3
Z = m(Y)ZlX) + (1-m(Y))~(X)
Z= Z(p(Y),X)
a b
Fig. 10: Two methods to define wing sections: Blending support airfoils data (a), and
varying generating parameters (b) along wing span. Sketch shows wing with basic
section over large portion of wing, root and tip sections.

Example: Optimization of a Flying Wing in Supersonic Flow

A Flying Wing aircraft has several advantages both for aerodynamic and structural efficiency.
Such a configuration therefore is an ideal test case for new design and optimization strategies.
The application of a Flying Wing for a new concept of high speed transport aircraft recently has
been studied as an alternative to conventional supersonic configurations [16]. Using variable
sweep angles for the whole configuration results in a wide range of possible optimum aerody-
namic efficiency. Oblique Flying Wing (OFW) examples have been generated by our geometry
preprocessor, both using the previous support section blending technique and also the new par-
ametric 'PARSEC' airfoils defined along span of the OFW.
Several aerodynamic phenomena suggest usage of well-known design methods like applying
transonic (supercritical) airfoil theory, applied to a sufficiently thick 2D basic airfoil in the sub-
sonic Mach number component normal to the leading edge of this large aspect ratio wing. Re-
qirements of an elliptic load distribution suggests the classic elliptic planform, a linear variation
of the wing twist along span and blending the basic airfoil with two modified sections at both
tips yields a wing with the desired load distribution. With computed lift over drag ratio (LID)
of this first case study [17]slightly above the values of known conventional wing-body type con-
figurations, we learned that a more refined approach than designing one 2D basic section would
allow for a better control of crossflow shocks coalescing on the upper wing surface. With a
slight modification of the planform geometry and the new spanwise parametric section defni-
tion this goal has been accomplished already in a first manual approach of optimizing the OFW
[19].

82
 z
wing sections

twist axis
wing sections front view
z leading edge

- - --- -l.:= p
0. 10

~
- l-
- r- -
--
I- ~
I"-- ~

--
I'-
-1-
- -- - -- -
I- /' V

-
0.00 I- - V
'--......
l..--

trailing edge
--- --I--
- I-- V

y
-0.10 O. 2. 4. 6. 8. 10.

4. 6. 8. y 10.

A~~
Fig. 11: Oblique Flying Wing optimized for supersonic flow M~ = 1.4: Example for
spanwise variation of wing sections (above), dihedral, thickness and twist, leading
and trailing edge geometry (center and below). (Note enlarged scale for vertical co-
ordinate z). Aerodynamic performance optimized in swept flow M~ = 1.4, A = 6rfJ, with
constraints for span wise wing section thickness, area and aerodynamic load distribu-
tion.

83
Figure 11 shows the resulting geometry in a threeview: wing sections vary considerably within
the elliptic chord distribution, all II PARSEC parameters were made functions of span. A mod-
ification of the basic elliptic planform to the unsymmetrical shape with stronger sweep at the
trailing tip (y = 10) than at the leading tip (y = 0) is suggested by the observed stronger crossflow
shocks in the trailing area. Simple aerodynamic theories suggest higher normal Mach number
components in the trailing part and therefore lead us to shape the local sections for supercritical
flow in higher Mach numbers. This was a goal in the example Fig. 3, a use of PARSEC func-
tions for span wise section definition therefore was promising.
Constraints based on application of one of the classical aerodynamic theories (the supersonic
area rule) to improve the design, calls for tuning the spanwise section area distribution accord-
ing to the Sears-Haack body of minimum drag for given volume. The simple polynomial struc-
ture of the PARSEC function yields the integral easily for each set of parameters.
An automated optimization procedure for this and similar configurations will perform the de-
sign of a better OFW much more economically than the manual approach done so far, but the
value of learning the role of the individual parameters in the process of a practical design cannot
be estimated high enough.

Wings with multicomponent high lift system

In the same way as for airfoil parameters, the newly developed key points for sealed flaps and
slats as well as the input data for multicomponent airfoil shapes may be made functions along
the third physical dimension in space, the spanwise direction of a wing. 3D space is defined in
the 'wing system', with planform in an (x, Yo)-plane, and the wing shape defined prior to shift-
ing and rotation in general 3D space with aircraft coordinates.
Using airfoil and high lift sections data in the retracted (cruise conditions) position defining
clean wing boundary conditions, wing input data rescale each section plus its components to
physical chord and provides twist as a function of span. Each component (flap, slat, solid re-
maining wing) will then be available for a movement in 3D space.
These surfaces may be defined along most of the wing span (except at fat root fillets or at thin
tips), choice of sections to begin and end slats and flaps is then a matter of constraints and ad-
ditional flexibility. Kinematic requirements, however, in the case of tapered and twisted wings
demand that the sectioning between component ends must be redefined to allow an unobstruct-
ed sliding of the flaps and slats along the fixed part of the wing. In the general case this requires
an intersection of the wing surface with a sphere, its center located at the vertex of a cone tan-
gent to the local wing surface panels.

Example: An extended DLR-F5 test wing configuration

A decade ago the 'OLR-F5 wing' was presented and communicated as a test case for the devel-
opment of CFD methods [2], [3]. The data for this wing have been used by various developers
to tune Navier-Stokes codes for viscous transonic flow. The case is still a difficult task to solve
if the experiment is to be simulated: transonic flow with laminar shock - boundary layer inter-
action remains a problem for CFD so far.
Nevertheless, with the example well known in the CFD community it seems to be a suitable ex-
ample also for other than experimental operating conditions, especially if the wing shape is de-
fined in a parametric way allowing variation of the shape and this way testing design and
optimization strategies.
In a first approach to revisit the OLR-F5 case its wing sections are redefined by PARSEC pa-
rameters. The original wing has a symmetrical basic section which was designed to be nearly

84
shock-free at Moo = 0.78. A set of surface data points was provided and blended with a thick
NACA 0036 section to fonn a prominent wing root fillet. With knowing the nose and crest cur-
vatures, applying the trailing edge modification parameters and blending with the NACA 4digit
generating function, the input set of airfoil data can be replaced altogether by the new analytical
definition. The new parameters are proposed along with mathematical modelling of some ex-
perimental pressure distributions to complete a new test case for direct/inverse CFD and for op-
timization [19].
In addition, the DLR-F5 wing was used for definition of a multicomponent wing with slat and
flap. Figure 12 (a) shows the choice for carving the basic section to shape a slat and a flap, these
section components subsequently are scaled to the DLR-F5 planfonn and additional input for
the 3D flap and slat tracks and rotation angles is provided. Choice of spanwise extent for flap
and slat, a refined sliding sections definition and closure of the components at the sliding sec-
tions completes the preprocessing of this extended test wing for CFD analysis.

Fig. 12: Basic OLR-F5 section, redefined by PARSEC parameters and carved to in-
clude a slat and flap component (a). Selection of 3D flap and slat extending along
span, in clean wing (b) and high lift (c) position.

Conclusions
An effort is made in using basic algebraic and analytic relations to generate realistic airfoil
shapes which are specified from a set of parameters. These are defined by only a few character-
istic dimensions used already in classical airfoil catalogs like NACA airfoil families, but also
allow for a refined shape definition as it results from systematic design processes in the transon-
ic flight regime. Airfoils detennined this way by a minimum set of parameters are subsequently
used as wing sections, with their generating parameters made functions along span, This has

85
been proven earlier for basic wing geometries which have used given data sets for support air-
foils. Some unsteady airfoil flow applications lead the way to fully threedimensional wings
which may be subject to input for manual or automated aerodynamic optimization techniques.
This method to describe all shapes analytically has been extended to high lift systems and adap-
tive devices.
The approach is intended to provide 2D, 3D and, with unsteady, adaptive or evolutionary con-
figurations, also 4D boundary conditions for CFD and CAD.

References
[1] Sobieczky, H.: Computational Methods for the Design of Adaptive Airfoils and Wings.
Notes on Numerical Fluid Mechanics, Vol. 2, ed. E. H. Hirschel, Vieweg (1979) pp. 269
- 278.
[2] Sobieczky, H.: DFVLR-F5 Test Wing Configuration for Computational and Experimenta
Experimental Aerodynamics, Wing Surface Generator Code, Control Surface and
Boundary Conditions. Notes on Num. Fluid Mechanics, Vol. 22, ed. W. KordulIa,
Vieweg (1988), pp. 27 - 37.
[3] Sobieczky, H., Hefer, G., Tusche, S.: DFVLR-F5 Test Wing Experiment for Computa-
tional Aerodynamics. AlAA 5. Appl. Aerodynamics Conf. Proc., AlAA 87-2485CP,
(1987), Notes on Numerical Fluid Mechanics, Vol. 22, ed. W. KordulIa,Vieweg, (1988),
pp.4 - 22.
[4] Sobieczky, H.: Aircraft Surface Generation. Results of EC BritelEuram Project
'Euromesh' 1990-92, Notes on Numerical Fluid Mechanics, Vol. 44, ed. N. Weatherill et
ai, Vieweg (1993), pp. 71 - 76.
[5] Sobieczky, H.: Geometry Generator for CFD and Applied Aerodynamics. in: New
Design Concepts for High Speed Air Transport. CISM Courses and Lectures No. 366,
Springer (Wien, New York), (1997) pp 137 - 157.
[6] Ladson, c., Brooks, c.: Development of a Computer Program to obtain Ordinates for
NACA 4-Digit, 4-Digit Modified, 5-Digit, and 16-Series Airfoils. NASA TM X-3284
(1975).
[7] Eberle, A.: Berichtigung der Whitcomb-Offenlegungsschrift und Profildefinitionspro-
gramm. MBB-UFE-AERO-MT-298 (1977).
[8] Zores, R.: Transonic Airfoil Design with Expert Systems. AIAA 95-1818CP, Proc. 13th
AlAA Applied Aerodynamics Conf., San Diego, CA, (1995).
[9] Garabedian, P. R.: On the Design of Airfoils Having no Boundary Layer Separation.
Advances in Mathematics 15, (1975) pp 164-168.
[10] Henne, P. A.: Innovation with Computational Aerodynamics: The Divergent Trailing
Edge Airfoil", Applied Computational Aerodynamics, ed. P. A. Henne, AlAA Education
Series, AIAA, Washington, D.C., (1990), pp. 221 - 262.
[11] Thompson, B. E., Lotz, R. D.: Divergent-Trailing-Edge Airfoil Flow. J. Aircraft Vol. 33
(1996), pp. 950 - 955.
[12] Sobieczky, H.: Gasdynamic Knowledge Base for High Speed Flow Modelling. in: New
Design Concepts for High Speed Air Transport. CISM Courses and Lectures No. 366,

86
 Springer (Wien, New York), (1997) pp 105 - 119.
[13] Ashill, P. R., Fulker, 1. L., Simmons, M. J., Gaudet, I. M.: A Review of Research atDRA
on Active and Passive Control of Shock Waves. 20th ICAS Congress Conf. Proc. ICAS-
96-2.1.4, (1996) pp. 76 - 87.
[14] Geissler, W., Sobieczky, H.: Unsteady Flow Control on Rotor Airfoils. AIAA 95-
1890CP, Proc. 13th AIAA Applied Aerodynamics Conference, San Diego, CA, (1995).
[15] Sobieczky, H., Geissler, w., Hannemann, M.: Numerical Tools for Unsteady Viscous
Flow Control". Proc. 15th lnt. Conf. on Num. Meth. in Fluid Dynamics. Lecture Notes in
Physics, ed. P. Kutler, J. Flores, J.-J. Chattot, Springer (Berlin, Heidelberg) (1997).
[16] Van der Velden, A.: The Oblique Flying Wing Transport. in: New Design Concepts for
High Speed Air Transport. CISM Courses and Lectures No. 366, Springer (Wien,
New York), (1997) pp 291 - 316.
[17] Seebass, R.: Oblique Flying Wing Studies. in: New Design Concepts for High Speed Air
Transport. CISM Courses and Lectures No. 366, Springer (Wien, New York), (1997) pp
317 - 336.
[18] Li, P., Seebass, R., Sobieczky, H.: Manual Optimization of an Oblique Flying Wing.
AIAA 98-0598 (1998).
[19] Sobieczky, H.: Configuration Test Cases for Aircraft Wing Root Design and Optimiza-
tion. Proc. ISIP'98, Nagano, Japan, (1998).

87