aerospace

Article
A Fast Aerodynamic Model for Aircraft Multidisciplinary
Design and Optimization Process
Frédéric Moëns

ONERA, Université Paris-Saclay, F92190 Meudon, France; frederic.moens@onera.fr

Abstract: A multidisciplinary design analysis and optimization process is developed at ONERA for
the design of tube and wing and blended wing–body aircraft configurations. This process is composed
of different disciplinary modules (geometry, propulsion, aerodynamics, structure, handling qualities
and flight mission), and the overall process considers different fidelity levels for these modules at
each step of the design process. This article describes the low-fidelity aerodynamic module used
during the preliminary design optimization process. Analytical formulations retained for lift and drag
components are presented in the first part. Then, the performances estimated by the aerodynamic
module on some reference configurations are compared with both numerical and experimental data,
showing a quite good agreement for both tube and wing and blended wing–body configurations not
only for global performance but also for individual drag components.

Keywords: aircraft performance; analytical method; drag evaluation; drag components; blended
wing–body

1. Introduction
Among the different options for more efficient aircraft configurations, the blended
wing–body (BWB) seems to offer a very promising reduction in emissions (CO2 , NOx).
However, such a configuration needs a complete, integrated design process as all the differ-
ent disciplines (cabin arrangement, aerodynamics, structure, propulsion, flight dynamics,
etc.) strongly interact altogether [1]. In order to explore, in depth, the potential benefits of
Citation: Moëns, F. A Fast such configurations, a multidisciplinary design analysis and optimization (MDAO) process
Aerodynamic Model for Aircraft dedicated to blended wing–body aircraft configurations has been developed since 2015 at
Multidisciplinary Design and ONERA [2–5]. The design workflow covers the range of tools from level 0 (L0), considering
Optimization Process. Aerospace 2023, (semi-) empirical methods, over level 1 (L1), taking into account low-level physics-based
10, 7. https://doi.org/10.3390/ methods, to level 2 and 3 (L2, L3), considering high-fidelity methods (Figure 1).
aerospace10010007 The modules implemented are easily exchangeable to adapt the workflow for specific
Academic Editor: Giuseppe Pezzella
configurations or a higher level of fidelity in specific domains if different methods are
needed. This design tool evolves continually due to the enrichment of the disciplinary
Received: 21 November 2022 modules integrated or the addition of new disciplinary modules. The last version is com-
Revised: 13 December 2022 posed of six main disciplinary modules: aircraft geometry, propulsion, aerodynamics,
Accepted: 16 December 2022 structure and weight, mission and performance and, finally, handling qualities. Addition-
Published: 22 December 2022
ally, some acoustic analysis can be performed offline of the multidisciplinary process for
the selected configurations (Figure 2). Those modules are integrated in the NASA OpenM-
DAO [6] framework to constitute the multidisciplinary design analysis and optimization
Copyright: © 2022 by the author.
(MDAO) process.
Licensee MDPI, Basel, Switzerland. When the main characteristics of the aircraft are not fixed, a large number of config-
This article is an open access article urations are evaluated during the exploratory phase of the overall aircraft design (OAD)
distributed under the terms and process with a large variety of architectures and planforms. As an example, Figure 3
conditions of the Creative Commons presents some BWB planforms evaluated for building up a multi-fidelity surrogate model
Attribution (CC BY) license (https:// for the first pre-design optimization loop, showing a large variety of architectures to be
creativecommons.org/licenses/by/ considered (from [4]). A first constraint for the aerodynamic module is to be able to manage
4.0/). 100% of these different architectures.

Aerospace 2023, 10, 7. https://doi.org/10.3390/aerospace10010007 https://www.mdpi.com/journal/aerospace

9, x FOR PEER REVIEW
Aerospace 2023, 10, 7 2 of227
of 27

Figure 1. The different fidelity levels for flow physics modeling of the aerodynamic solvers.

The modules implemented are easily exchangeable to adapt the workflow for spe-
cific configurations or a higher level of fidelity in specific domains if different methods
are needed. This design tool evolves continually due to the enrichment of the discipli-
nary modules integrated or the addition of new disciplinary modules. The last version is
composed of six main disciplinary modules: aircraft geometry, propulsion, aerodynam-
ics, structure and weight, mission and performance and, finally, handling qualities. Ad-
ditionally, some acoustic analysis can be performed offline of the multidisciplinary pro-
cess for the selected configurations (Figure 2). Those modules are integrated in the
NASA OpenMDAO [6] framework to constitute the multidisciplinary design analysis
andFigure 1. The different
optimization Figurefidelity
(MDAO) levels fidelity
process.
1. The different for flow physics
levels modeling
for flow of the aerodynamic
physics modeling solvers.
of the aerodynamic solvers.

The modules implemented are easily exchangeable to adapt the workflow for spe-
cific configurations or a higher level of fidelity in specific domains if different methods
are needed. This design tool evolves continually due to the enrichment of the discipli-
nary modules integrated or the addition of new disciplinary modules. The last version is
composed of six main disciplinary modules: aircraft geometry, propulsion, aerodynam-
ics, structure and weight, mission and performance and, finally, handling qualities. Ad-
ditionally, some acoustic analysis can be performed offline of the multidisciplinary pro-
cess for the selected configurations (Figure 2). Those modules are integrated in the
NASA OpenMDAO [6] framework to constitute the multidisciplinary design analysis
and optimization (MDAO) process.

Figure 2. ONERA MDAO/OAD toolMDAO/OAD

Figure 2. ONERA for BWB applications.
tool for BWB applications.
 figurations are evaluated during the exploratory phase of the overall aircraft design
(OAD) process with a large variety of architectures and planforms. As an example, Fig
ure 3 presents some BWB planforms evaluated for building up a multi-fidelity surrogat
model for the first pre-design optimization loop, showing a large variety of architecture
Aerospace 2023, 10, 7 to be considered (from [4]). A first constraint for the aerodynamic module is3 to
of 27
be able to
manage 100% of these different architectures.

Figure3.3.Extract
Figure Extractofof shapes
shapes considered
considered for afor
DoEa DoE for a design.
for a BWB BWB design.

Then,
Then,forfor
each
eachconfiguration,
configuration,the mission modulemodule
the mission generatesgenerates
an aerodynamic database. data
an aerodynamic
As an order of magnitude, these databases consider about 5000 aerodynamic flow condi-
base. As an order of magnitude, these databases consider about 5000 aerodynamic flow
tions to be computed for each planform, which makes the use of an aerodynamic disci-
conditions to be computed for each planform, which makes the use of an aerodynami
plinary module with a very fast computing time and a reliable performance level predic-
disciplinary
tion mandatory. module with a very fast computing time and a reliable performance leve
prediction mandatory.
Therefore, among the different numerical methods available for aerodynamic mod-
eling, Therefore, among to
the ones belonging thethedifferent numerical
“low-fidelity” familymethods available
(L0 or L1) are oftenfor aerodynamic
used for OAD mod
eling, the
phases. For ones belonging
the present to the a“low-fidelity”
application, family (L0 module
dedicated L0 aerodynamic or L1) are often
based used for OAD
on analyti-
cal formulations
phases. For thederived
present from the theory or
application, a from the dataL0
dedicated analysis of past ormodule
aerodynamic present aircraft
based on ana
is
lytical formulations derived from the theory or from the data analysis “Tube
developed for a fast evaluation of the aerodynamic performance for standard of pastand
or presen
Wing” (T&W), “Flying Wing” (FW) or “Blended Wing Body” (BWB) configurations for
aircraft is developed for a fast evaluation of the aerodynamic performance for standard
typical subsonic cruise flight conditions. This module considers only wing planform data,
“Tube and Wing” (T&W), “Flying Wing” (FW) or “Blended Wing Body” (BWB) configu
and no surface grid is needed.
rations
The for typical
different subsonic cruise
formulations retainedflight conditions.
are described This
in the firstmodule considers
part of the article. only
In wing
planform data, and no surface grid is needed.
the second part, the results issued from the module are compared to the numerical or
The different
experimental formulations
results of retained
different aircraft are described in the first part of the article. In
configurations.
the second part, the results issued from the module are compared to the numerical o
2. Standard Atmosphere
experimental results ofModel
different aircraft configurations.
To characterize the evolution of the ambient static flow conditions throughout the
atmosphere,
2. Standardthe aerodynamicModel
Atmosphere module integrates an analytic formulation of the International
Standard Atmosphere (ISA) [7]. ISA is a mathematical model that divides the atmosphere
To characterize
into several the evolution
layers (troposphere, of the ambient
stratosphere, mesosphere,static flow
etc.) thatconditions throughout th
gives the evolution
atmosphere,
of the ambient the aerodynamic
pressure, module
temperature integrates
and density withan theanalytic
altitude.formulation of the Interna
The flight altitudes
tional Standard
considered Atmosphere
in the module belong(ISA)
to the[7]. ISA is a mathematical
troposphere model Within
and the stratosphere. that divides
each the at
mosphere into several layers (troposphere, stratosphere, mesosphere, etc.) that
layer or sublayer, the temperature is assumed to have a linear evolution with respect to thegives th
altitude,
evolution theofpressure is calculated
the ambient by thetemperature
pressure, hydrostatic balance from Equation
and density with the(1),altitude.
and the air
The fligh
density is calculated assuming the air as a perfect gas from Equation (2):
altitudes considered in the module belong to the troposphere and the stratosphere
Within each layer or sublayer, thedP temperature is assumed to have a linear evolution
= −ρg
with respect to the altitude, the pressure (1)
is calculated by the hydrostatic balance from
dz
Equation (1), and the air density is calculated assuming the air as a perfect gas from
P
Equation (2): ρ= (2)
RT
For the air considered as a perfect gas, there is R = 287.04 J kg−1 K−1 , and the standard
gravity constant is g = 9.80665 m/s2 .
 𝑑𝑃
= −𝜌𝑔 (1)
𝑑𝑧
𝑃
𝜌= (2)
𝑅𝑇
Aerospace 2023, 10, 7 For the air considered as a perfect gas, there is R = 287.04 J kg−1 K−1, and the 4 of 27
stand-
ard gravity constant is g = 9.80665 m/s2.
Knowing the evolution of the temperature with the altitude (Figure 4) and consider-
Knowing
ing some the evolution of
thermodynamic the temperature
assumptions withdifferent
for the the altitude (Figure
layers, the4)integration
and considering
of Equa-
some thermodynamic assumptions for the different layers, the integration
tion (1) between the two altitudes that bound the given layer leads to some analyticof Equation (1) for-
between the two altitudes that
mulations, as presented in Table 1.bound the given layer leads to some analytic formulations,
as presented in Table 1.

Figure 4. Evolution
Figure Evolutionof
ofthe
thetemperature
temperaturewith
withthethe
altitude—ISA model.
altitude—ISA model.

Table 1. Analytic
Analyticstandard
standardatmosphere
atmosphere model
model used.
used.

Layer Altitude
Altitude (km) Temperature (K) Pressure (Pa)
Layer Temperature (K) Pressure (Pa)
Troposphere (km)
Z ≤ 11 T = 288.15 − 6.5 × Z

T
5.2561
P = 101325.0 × 288.15
.
𝑇
Troposphere Z
11≤≤11
Z ≤ 20 𝑇 = 288.15 − 6.5 × 𝑍
T = 216.65 𝑃 = 101325.0 ×
P = 22630.6 × 10−(288.15
z−11
Stratosphere 14.596 )
11 20
≤Z ≤ ≤Z 20
≤ 32 𝑇 216.65
T= = 216.65
− ( Z − 20)
Stratosphere 𝑃 = 22630.6 × 10 .
20 ≤ Z ≤ 32 𝑇 = 216.65 − 𝑍 − 20
Finally, the air dynamic viscosity is computed using Sutherland’s relation:
Finally, the air dynamic viscosity
 iscomputed
3 
using
 Sutherland’s relation:
T 2 T0 + 110.4
µ = µ0 (3)
T0 𝑇 T +𝑇 110.4
+ 110.4 (3)
𝜇=𝜇
−5
𝑇 𝑇 + 110.4
where µ0 = 1.711 10 kg/(ms) and T0 = 273.15 K.
where μ0 = 1.711 10−5 kg/(ms) and T0 = 273.15 K.
3. Aerodynamic Module
3. Aerodynamic Module
The aerodynamic module is dedicated to T&W, FW or BWB configurations in subsonic,
attached
The flow conditions.
aerodynamic It doesisnot
module considertoflow
dedicated conditions
T&W, with configurations
FW or BWB separation, buffet,
in sub-
high-lift configurations, supersonic flight and unconventional aircraft configurations,
sonic, attached flow conditions. It does not consider flow conditions with separation, such
as multiplane or boxed wing.
buffet, high-lift configurations, supersonic flight and unconventional aircraft configura-
In the first stage, a reference wing is built based on geometrical characteristics such as
tions, such as multiplane or boxed wing.
the airfoil relative thickness, the leading-edge sweep angles, the local chord, etc., and refer-
In the first stage, a reference wing is built based on geometrical characteristics such
ence geometrical parameters are then derived (reference surface, mean aerodynamic chord).
as
For theairfoil
the relative
performance thickness,this
evaluation, thewing
leading-edge sweep
is considered angles, the local
“aerodynamically chord, with
optimized” etc., and
an elliptical span loading in order to have an estimation of the local lift coefficients.
In the second stage, the influence of other elements, such as the fuselage, winglets
or nacelles, are considered as drag increments. Note that the span loading of the wing
considering these elements is not computed by the module.
 coefficients.
In the second stage, the influence of other elements, such as the fuselage, winglets
or nacelles, are considered as drag increments. Note that the span loading of the wing
considering these elements is not computed by the module.
Aerospace 2023, 10, 7 The module is suited to compare different wing planforms within a preliminary de-5 of 27
sign loop. By construction, identical results will be obtained for two configurations with
the same wing planform and the same airfoil thickness spanwise evolution. The optimi-
zation of geometrical details,is suited
The module such as the airfoil
to compare shape,
different camber
wing or twist,
planforms withinis not possible
a preliminary design
with the use loop.
of thisBymodule
construction, identical
and has, results to
therefore, willbebeconsidered
obtained forin
two
theconfigurations
next step of with
the the
design processsame wing
using planform
more and the
advanced L1same
or L2airfoil thickness
numerical spanwise evolution. The optimization
methods.
of geometrical details, such as the airfoil shape, camber or twist, is not possible with the use
The following chapters present the geometrical inputs used and the analytical for-
of this module and has, therefore, to be considered in the next step of the design process
mulations considered
using more for the estimation
advanced L1 or L2of the aerodynamic
numerical methods. performance (lift and drag).
The following chapters present the geometrical inputs used and the analytical formu-
3.1. Geometrical Inputs
lations considered for the estimation of the aerodynamic performance (lift and drag).
3.1.1. Wing
3.1. Geometrical Inputs
The wing3.1.1.
planform
Wing is defined by the use of NAIRF airfoil sections of NWSEG segments
(NAIRF = NWSEG+1). TheBy construction,
wing planformthe first airfoil
is defined by theisuse
located
of NAIRF at airfoil
the symmetry
sections ofplane
NWSEGY(1) =
segments
0, and the last(Nsection Y(N ) corresponds to the wing tip. For each wing segment,
AIRF = NWSEG +1). By construction, the first airfoil is located at the symmetry plane
AIRF the
leading-edge Y(1)
sweep= 0, angle,
and theanlastindex
section(XTra)AIRF
Y(N for) laminar
correspondscomputation andFor
to the wing tip. theeach
airfoil
wingtech-
segment,
nology factorthe leading-edge
used for wave sweep angle, an index
drag calculations (XTra ) factor”,
(“Korn for laminar
see computation
chapter 3.5)andarethe
pro-airfoil
vided. Additionally, for each airfoil, the relative thickness (t/c) and an estimation of the are
technology factor used for wave drag calculations (“Korn factor”, see chapter 3.5)
maximum liftprovided. Additionally, for each airfoil, the relative thickness (t/c) and an estimation of the
(Cl max) are provided. Figure 5 gives an example of data for a wing defined
maximum lift (Cl max ) are provided. Figure 5 gives an example of data for a wing defined
by five airfoilby
sections (or four wing segments).
five airfoil sections (or four wing segments).

(a) (b)
Figure 5. Geometrical
Figure 5.data used fordata
Geometrical wing definition.
used (a) Geometrical
for wing definition. data related
(a) Geometrical to wing
data related to seg-
wing segments.
ments. (b) Geometrical data related
(b) Geometrical to airfoil
data related sections.
to airfoil sections.

It should be Itnoted
should be no
that noted that nogeometrical
detailed detailed geometrical
inputs for inputs for airfoils
airfoils are usedare used (shape,
(shape,
twist, camber).
twist, camber).
Then, the wing span is divided into NDY subsections. For a given subsection, the
Then, thedifferent
wing span is divided into NDY subsections. For a given subsection, the
geometrical characteristics are obtained by a linear interpolation of the given
different geometrical characteristics
parameter between two given are obtained
wing sections by aoflinear interpolation
the segment. of the
This allows thegiven
evolution
parameter between two given wing
of the leading-edge sections
lines X LE (Y) and of Ythe
LE segment.
(Y), the chordThis
law allows
C(Y) the
and evolution
the different of
sweep
the leading-edge lines with
evolutions XLE(Y)
wingand YLE(ϕ
span (Y),
0 the
(Y), ϕ 25chord
(Y), ϕ 50law
(Y)) C(Y)
to be and the
obtained. different sweep
The reference
evolutions with wing areaφconsidered
span (φ0(Y), 25(Y), φ50(Y)) fortothebeaerodynamic
obtained. coefficient calculations is the
geometrical wing area obtained by
The reference area considered for the aerodynamic the sum of thecoefficient
different trapezoid segments:
calculations is the ge-
ometrical wing area obtained by the sum iof =N
the different trapezoid segments:
[Y (i + 1) − Y (i )] × [C (i + 1) + C (i )]
SREF = SWing = ∑i=0 WSEG 2
(4)

The wing geometrical aspect ratio λ corresponds to:

(2 Y ( NAIRF ))2 b2
λ= = (5)
SWing SWing
 The wing geometrical aspect ratio λ corresponds to:
2𝑌 𝑁 𝑏
𝜆= = (5)
𝑆 𝑆
Aerospace 2023, 10, 7 6 of 27
The wing taper ratio ε is:
𝐶 𝑁
𝜀= (6)
The wing taper ratio ε is: 𝐶 1
C ( NAIRF )
Finally, the value of the aerodynamic mean
ε = chord and its location on the half-wing (6)
can be obtained: C (1)
Finally, the value of the aerodynamic
/ mean chord and its location on the half-wing
can be obtained: 2
𝐴𝑀𝐶 = 𝐶 𝑦Zb/2𝑑𝑦 (7)
𝑆 2 2
AMC = C (y)dy (7)
SRe f
/ 0
2
𝑋 = 𝑋 Zb/2
𝑦 × 𝐶 𝑦 𝑑𝑦 (8)
𝑆 2
X AMC = X LE (y) × C (y)dy (8)
SRe f
0
2 b
(9)
𝑌 = 𝑌2 𝑦 Z2 × 𝐶 𝑦 𝑑𝑦
𝑆
YAMC = YLE (y) × C (y)dy (9)
SRe f
0

3.1.2. Fuselage
3.1.2. Fuselage
A fuselage is modeled
A fuselage as a slender
is modeled circularcircular
as a slender cylinder of length
cylinder LFUS and
of length a diameter
LFUS and a diameter
DFUS. The aerodynamic module considers the fuselage to be set on the symmetry
DFUS . The aerodynamic module considers the fuselage to be set on the symmetry plane
plane but
but with nono
with consideration about
consideration aboutitsitsrelative
relativeposition
position to the wing.
to the wing. Therefore,
Therefore,similar
similarperformance
per-
formancewillwill be evaluated
be evaluated bymodule
by the the module fordifferent
for the the different configurations
configurations presented
presented in 6. The
in Figure
Figure wing
6. Thedihedral
wing dihedral angle is not taken into
angle is not taken into account. account.

(a) (b)
Figure 6.Figure
Fuselage arrangements
6. Fuselage with similar
arrangements aerodynamic
with similar performance.
aerodynamic (a) Front
performance. view. view.
(a) Front (b) Top
(b) Top view.
view.
3.1.3. Nacelles
3.1.3. Nacelles
The effect of airframe installation on performance is considered for the drag contribu-
Thetion onlyof(there
effect are no
airframe propulsion
installation oneffects considered).
performance The evaluation
is considered is therefore
for the drag contri-similar
bution to a “through-flow
only nacelle” consideration.
(there are no propulsion NacelleThe
effects considered). elements are considered
evaluation is thereforeas cylinders.
sim-
Single flux (NAC1 elements only) or double flux nacelles (NAC1 and NAC2 elements) can
be considered (Figure 7), but it is supposed that all the nacelles are identical in the present
version of the module.
Aerospace 2022, 9, x FOR PEER REVIEW 7 of 27

ilar to a “through-flow nacelle” consideration. Nacelle elements are considered as cylin-

ders. Single flux (NAC1 elements only) or double flux nacelles (NAC1 and NAC2 ele-
Aerospace 2023, 10, 7 7 of 27
ments) can be considered (Figure 7), but it is supposed that all the nacelles are identical
in the present version of the module.

Nacelletypes
Figure7.7.Nacelle
Figure typesconsidered
consideredininthe
themodule.
module.

3.2. Formulation for Aircraft Lift

3.2. Formulation
3.2.1. Estimationforof
Aircraft Lift
Wing Maximum Lift
3.2.1. Estimation of Wing Maximum Lift
By construction, the aerodynamic module considers an ideal elliptic span loading,
By for
KZ(Y), construction, thewing,
the reference aerodynamic module
which is fully considers
defined by the an
loadideal elliptic
at the spanplane
symmetry loading,
KZ0 :
KZ(Y), for the reference wing, which is fully defined
v by the load at the symmetry plane
KZ0: u !2
u Y
KZ (Y ) = KZ0 1 − b
t (10)
𝑌2
𝐾𝑍 𝑌 = 𝐾𝑍 1− (10)
where: 𝑏
2SW I NG 2
KZ0 =   × CL (11)
where: π 2b

For a given Y section, the local 𝐾𝑍

airfoil 2𝑆
= lift coefficient
× 𝐶 corresponds to:
𝑏 (11)
𝜋
2 (Y )
KZ
Cl (Y ) = (12)
C (Y )
For a given Y section, the local airfoil lift coefficient corresponds to:
Wing maximum lift is considered to be𝐾𝑍 𝑌
reached when Cl (Y) = Cl max (Y) at one
𝐶 𝑌 = (12)
Y section. 𝐶 𝑌
Wing
3.2.2. Lift maximum
Slope lift is considered to be reached when Cl (Y) = Cl max (Y) at one Y sec-
tion.
The aircraft ∂C L
∂α slope (α in rad) considered is the Polhamus formulation [8], with the
effects
3.2.2. LiftofSlope
fuselage taken into account [9]:

The aircraft slope (α in rad) considered DFUSis2the Polhamus formulation [8], with
 h i 
πλ 1.07 1 +
into account [9]: b
 
the effects of fuselage∂C taken
L DFUS
∂α
= r  1− b (13)
1+tan ϕ50 2 − M2

1 + 1 + λ2 𝐷4
𝜕𝐶 𝜋𝜆 1.07 1 + 𝐷
𝑏
= 1− (13)
where ϕ50 is the mean wing𝜕𝛼 1 + 𝑡𝑎𝑛 𝜑 −𝑀 𝑏
1 +sweep
1 + 𝜆angle at mid-chord. For FW or BWB configurations,
D is set to zero. 4
FUS
whereSome φ50 ismodules
the mean of the
wingMDAO
sweepprocess
angle require the CL (α)
at mid-chord. Forcurve
FW asor the
BWBinput, which means
configurations,
Dthat
FUS isitset
is to
necessary
zero. to know the zero-lift incidence of the aircraft α0 . The aerodynamic
module
Some is modules
not able toofdetermine
the MDAO thisprocess
parameter, but itthe
require canC
beL(α)
estimated based
curve as the on preliminary
input, which
validation exercises on similar configurations. Otherwise, a value of α = 0 ◦ can be used.
means that it is necessary to know the zero-lift incidence of the aircraft 0 α0. The aerody-
namic module is not able to determine this parameter, but it can be estimated based on
3.3. Formulations for Aircraft Drag Evaluation
preliminary validation exercises on similar configurations. Otherwise, a value of α0 = 0°
can beAmong
used. the different existing drag formulations, the one retained for the aerodynamic
module is derived from the one presented by Gur [10]:

CDTotal = CD Induced + CDFriction + CD Add + CDWave + CDParasitic (14)

The following sections describe the formulations used for these different terms.
Aerospace 2023, 10, 7 8 of 27

3.3.1. Lift-Induced Drag

Only the wing contribution to lift-induced drag is considered with the influence of the
fuselage or winglet. The contribution of the tail surfaces to the total lift-induced drag is not
considered in the module. The standard formulation is considered for the lift-induced drag
coefficient of the reference wing:

CL2
CD Induced = (15)
π λ Osw
where λ is the wing geometric aspect ratio defined in Equation (5), and Osw is the Oswald
factor that characterizes the deviation from the ideal elliptic span loading. For a given
wing planform, several methods are available for the determination of the Oswald factor
(see [11] for instance). Most of them consider an Oswald factor such as:

1
Osw = (16)
1+δ
where δ is a parameter based on wing geometrical characteristics. The one proposed in the
aerodynamic module is based on the formulation used by Anderson [11,12]:
h i h p  i
δAnderson = 0.0015 + 0.016(ε − 0.4)2 × λ 1 − M2 − 4.5 (17)

leading to:
1
Osw Anderson = (18)
1 + δAnderson
However, this formulation is valid for unswept wings only, which is obviously not
suitable for BWB configurations. To take the wing sweep angle into account, Hörner [12]
proposes a simple correction:

Osw ϕ = Osw( ϕ=0) × cos ϕ (19)

Some validations carried out on several CFD results analyzed by a far-field drag
decomposition tool have shown that the lift-induced drag component computed for swept
wings was surrounded by the evaluations made considering OswAnderson or Oswϕ and that
a mean value between these two formulations leads to an excellent agreement (Figure 8).
Therefore, the formulation retained in the code is simply:

1 + cos ϕ
OswWing = Osw Anderson (20)
2
When a fuselage or winglets are considered, the Oswald factor is modified, as de-
scribed by Nita [11]:

OswWing+ Fuselage+W LT = OswWing × K Fus × KW LT (21)

The correction factor for the fuselage is:

 2 !
DFUS
K Fus = 1−2 (22)
b

DFus is the fuselage diameter, and b is the total wing span.

The correction factor for the winglets is:

2 HW LT 2
 
1
KW LT = 1+ (23)
Coe f W LT b
Aerospace 2023, 10, 7 9 of 27

where CoefWLT is a correction factor derived from works from Bourdin [13] or Delavenne
Aerospace 2022, 9, x FOR PEER REVIEW
[14]
9 of 27
to take into account the effect of the winglet cant angle on lift-induced drag:
Figure 8. Lift-induced drag formulation
−4
for swept wings.
−5 2
Coe f W LT = 1.0 + 4.10 δW LT + 1.10 δW LT − 3.10−8 δW
3
LT − 5.10
−10 4
δW LT (24)
When a fuselage or winglets are considered, the Oswald factor is modified, as de-
scribed by Nita [11]:
𝑂𝑠𝑤 = 𝑂𝑠𝑤 ×𝐾 ×𝐾 (21)
The correction factor for the fuselage is:
𝐷
𝐾 = 1−2 (22)
𝑏
DFus is the fuselage diameter, and b is the total wing span.
The correction factor for the winglets is:
1 2𝐻
𝐾 = 1+ (23)
𝐶𝑜𝑒𝑓 𝑏
where CoefWLT is a correction factor derived from works from Bourdin [13] or Delavenne
[14] to take into account the effect of the winglet cant angle on lift-induced drag:
Figure 8.
Figure Lift-induceddrag
8. Lift-induced dragformulation
formulationfor
for swept
swept wings.
wings.
𝐶𝑜𝑒𝑓 = 1.0 + 4. 10 𝛿 + 1. 10 𝛿 − 3. 10 𝛿 − 5. 10 𝛿 (24)
HHWLT
When is
WLT a
the winglet
is fuselage height,
or winglets
the winglet b is
height, b◦arethe total reference
considered,
is the wing
the Oswald
total reference span without
factor
wing span winglets,
is modified,
without and
as de-
winglets, and
δ WLT is
scribed the
δWLT is by winglet
theNita cant angle (in ) (see definitions in Figure 9).
[11]:cant angle (in °) (see definitions in Figure 9).
winglet
𝑂𝑠𝑤 = 𝑂𝑠𝑤 ×𝐾 ×𝐾 (21)
The correction factor for the fuselage is:
𝐷
𝐾 = 1−2 (22)
𝑏
DFus is the fuselage diameter, and b is the total wing span.
The correction factor for the winglets is:
Figure9.9.Definitions
Figure Definitionsofofthe
thewinglet
wingletparameters.
parameters.
1 2𝐻
𝐾 = 1+ (23)
3.3.2.
3.3.2.Friction/Form
Friction/Form DragDrag 𝐶𝑜𝑒𝑓 𝑏
whereAccording
CoefWLT isto
According the methodology
atocorrection
the methodology described in [9,10],
described
factor derived from in the
[9,10],
works skin
from friction
theBourdin
skin and or
friction
[13] pressure
and drag
pressure
Delavenne
ofdrag
the
[14] different
to of theinto
take components
different
account are calculated
components
the effect are
of using cant
calculated
the winglet the following
using relation:
the following
angle relation:
on lift-induced drag:
𝐶𝑜𝑒𝑓 = 1.0 + 4. 10 C𝛿 𝐶 + 1. 𝛿 SWET
10 · FF −𝑆3. 10 𝛿 − 5. 10 𝛿 (24)
DFriction = C=F 𝐶 ·. 𝐹𝐹. 𝑆 (25)
(25)
SREF
HWLT is the winglet height, b is the total reference wing span without winglets, and
δwhere
WLT is C
the winglet
F is cant skin
a flat plate angle (in °) (see
friction definitions
coefficient, FF isinthe
Figure
form 9).
factor of the component, and
SWET is the wetted area of this component.
• Friction coefficient CF
For a flat plate of length L, the friction coefficient is obtained considering the definition
of the boundary layer momentum thickness. At the end of the plate, the integrated friction
coefficient is:
Θ
CF = 2 L (26)
C (Y )
FigureIf9.the flow is laminar,
Definitions the use
of the winglet of the Blasius relationships leads to:
parameters.

3.3.2. Friction/Form Drag 1.328

CFLam = √ (27)
Re L
According to the methodology described in [9,10], the skin friction and pressure
drag of the different components are calculated using the following relation:
𝑆
𝐶 = 𝐶 . 𝐹𝐹. (25)
𝑆
 If the flow is laminar, the use of the Blasius relationships leads to:
1.328
𝐶 = (27)
𝑅𝑒
Aerospace 2023, 10,For
a turbulent flow, the friction coefficient is obtained by the compressible10 of 27
7
Schlichting relation [9]:
0.455
For a𝐶turbulent
= flow, the friction coefficient is obtained by the compressible Schlichting
(28)
relation [9]: 𝑙𝑜𝑔 𝑅𝑒 . 1 + 0.144𝑀 .
0.455
For wings considering natural laminar CFTurb = flow (NLF) 2.58 or hybrid 2 laminar
0.65 flow (HLF) (28)
(log10 Re L ) (1 + 0.144M )
technologies, it is possible to have an estimation of the maximum laminar extent X
For wings considering natural laminar flow (NLF) or hybrid laminar flow (HLF)
for the wing sections as a function of the local Reynolds number and the φ25 sweep angle
technologies, it is possible to have an estimation of the maximum laminar extent XLammax
(Figure 10, from [15]).
for theThis
wingmaximum
sections as a laminar
function ofextent occurs
the local for number
Reynolds a design
andlift
thecoefficient
ϕ25 sweep angle
(CL_Adapt), and the(Figure
change 10,in transition
from location
[15]). This maximum due to theextent
laminar modification ofdesign
occurs for a the pressure
lift coefficient
(C
gradients over the airfoil ), and the change in transition location due to the modification
LAdapt when the angle of attack changes has to be taken into account. of the pressure
gradients over the airfoil when the angle of attack changes has to be taken into account.

(a) (b)

Figure 10. MaximumFigure 10. Maximum

possible laminarpossible laminar
flow extent asflow extent as of
a function a function
Reynoldsof Reynolds
numbernumber and wing sweep
and wing
angle. (a)
sweep angle. (a) NLF. (b) HLF.NLF. (b) HLF.

Though the module does not consider any airfoil shape, it is therefore necessary to
Though
Aerospace 2022, 9, x FOR the module
PEER REVIEW consider does not consider
a “suction” any airfoil
and a “pressure” side shape, it is therefore
for the wing surfaces in necessary 11to
order to take ofthese
27
consider a “suction” andinatransition
changes “pressure” side
location for
into the wing
account. surfaces
The module usesin
theorder to take
simplified modelthese
presented
in Figure 11.
changes in transition location into account. The module uses the simplified model pre-
sented in Figure 11.

(a) (b)
Figure
Figure 11. 11. Simplified
Simplified modelmodel
usedused for transition
for transition location
location on wing
on wing surfaces.
surfaces. (a) Model
(a) Model for transition
for transition
change on on
change wing suction
wing side.
suction (b) (b)
side. Model for for
Model transition change
transition on on
change wing pressure
wing side.
pressure side.

Once Oncethethe portion

portion ofoflaminar
laminarflowflow(X
(XLam ) isestimated
) is
Lam estimatedfor
for the
the considered
considered wing
wing section,
sec-
tion, the global friction coefficient is obtained via the computation of the boundary layermo-
the global friction coefficient is obtained via the computation of the boundary layer
mentum thickness
momentum thicknessatatthe thetrailing-edge
trailing-edge through
through thethe
useuse
of aof
“fictitious turbulent
a “fictitious length”
turbulent
(L
length” in Figure 12).
Fict (LFict in Figure 12).
 Figure 11. Simplified model used for transition location on wing surfaces. (a) Model for transition
change on wing suction side. (b) Model for transition change on wing pressure side.

Once the portion of laminar flow (XLam) is estimated for the considered wing sec-
tion, the global friction coefficient is obtained via the computation of the boundary layer
Aerospace 2023, 10, 7 momentum thickness at the trailing-edge through the use of a “fictitious turbulent 11 of 27
length” (LFict in Figure 12).

Figure12.
Figure 12.Estimation ofΘΘatatairfoil
Estimationof airfoiltrailing-edge
trailing-edgefor
fortransitional
transitionalflow.
flow.

Firstly, the momentum thickness at the transition point (Θ ) is computed using the
Firstly, the momentum thickness at the transition point (ΘTrTr) is computed using the
Blasius relationships.
Blasius relationships.
0.664
0.664
Θ Tr𝛩 = = 𝑋 ××
( X Lam C (𝐶Y 𝑌
)) q (29)
(29)
𝑅𝑒( L=XLam ××C(Y ))
Re

Then, using the Michel relationships [16], the length LFict of a fully turbulent bound-
Then, using the Michel relationships [16], the length LFict of a fully turbulent boundary
ary layer with the same ΘTr momentum thickness is estimated as:
layer with the same ΘTr momentum thickness is estimated as:
/ .

𝐿 =
" 1
𝛩  𝑅𝑒 × 1/6 #1.2 (30)
1 Re( L= XLam ×C(Y )
L Fict = 0.02208 Θ Tr 𝑋 ×𝐶 𝑌 (30)
0.02208 X Lam × C (Y )
Finally, ΘTE is computed as the momentum thickness of a turbulent flow that devel-
ops over a length
Finally, ofcomputed
ΘTE is LTur: as the momentum thickness of a turbulent flow that develops
over a length of LTur : 𝐿 = 𝐿 + 1−𝑋 ×𝐶 𝑌 (31)
L Tur = L Fict + (1 − X Lam ) × C (Y ) (31)
𝐿
𝛩 = 0.02208 L (32)
Θ TE = 0.02208 𝑅𝑒 Tur
1/6/ (32)
Re LTur
The friction coefficient for transitional flow on one surface is finally obtained by in-
The friction coefficient for transitional flow on one surface is finally obtained by
troducing ΘTE from Equation (32) into Equation (26).
introducing ΘTE from Equation (32) into Equation (26).
• Form Factors FF
• Form Factors FF
There are several models in the literature available for form factors (see for instance [10]
for a comparison of different formulations for FFWING or for body of revolutions). Table 2
presents the ones retained in the aerodynamic module and the corresponding SWET .

Table 2. Form factors and wetted area considered in the aerodynamic module.

Element Form Factor SWet

h
t

2
3 i
Wing [17] − 0.4578 ct + 13.0119 ct cos2 ϕ50 + 1 2 SWING


FFW I NG = 3.4004 c
 
L FUS
Fuselage [9] FFFUS = 1 +  L 60 3 + 0.0025 D FUS
π LFUS DFUS
FUS
D FUS

t


Winglets [17] FFW LT = 1 + 3.52 c W LT cos ϕW LT 2 SWLT
t


Tail surfaces [17] FFTAIL = 1 + 3.52 c TAIL cos ϕ TAIL 2 STAIL
 
Nacelles [18] D N AC 2π DNAC LNAC
FFN AC = 1 + 0.35 L N AC
 𝐷
Nacelles [18] 𝐹𝐹 = 1 + 0.35 2π DNAC LNAC
𝐿

3.3.3. Interaction Effects for Nacelles

Aerospace 2023, 10, 7 12 of 27

In addition to the simple geometrical effect on the friction drag, some interaction ef-
fects of the nacelles 3.3.3.
withInteraction
a surface (wing or fuselage) are considered through the dis-
Effects for Nacelles
tance parameter ZNAC (Figure 13). The
In addition to thepositive values ofeffect
simple geometrical ZNAConarethe for standard
friction drag, somenacelle ar-
interaction
rangements. The negativeeffects ofvalues of Zwith
the nacelles NAC aare for (wing
surface buried nacelles.
or fuselage) areOn the basis
considered of these
through the dis-
geometrical characteristics, the interaction coefficient QN is defined according to thenacelle
tance parameter Z NAC (Figure 13). The positive values of Z NAC are for standard fol-
arrangements. The negative values of ZNAC are for buried nacelles. On the basis of these
lowing statement given in [9]: “For a nacelle or external store mounted directly on the fuselage
geometrical characteristics, the interaction coefficient QN is defined according to the follow-
or wing, the interference
ingfactor
statement is about
QN given 1.5.
in [9]: If athe
“For nacelle
nacelle or store
or external store is mounted
mounted lessonthan
directly aboutor
the fuselage
wing, the interference factor
one diameter away, the QN factor is about 1.3. If Q N it is mounted much beyond one diameter, the Qone
is about 1.5. If the nacelle or store is mounted less than about N
diameter away, the QN factor is about 1.3. If it is mounted much beyond one diameter, the QN factor
factor approaches 1.0.” approaches 1.0.”

(a) (b)

Figure 13. Definition of Figure NACDefinition

the Z13. parameter.of the(a)
ZNAC parameter. (a)(b)
“Standard”. “Standard”.
“Buried”.(b) “Buried”.

For the positive values of ZNAC , a linear evolution of QN is considered in the module:
For the positive values of ZNAC, a linear evolution of QN is considered in the module:
ZN AC
If ZNAC > 0; Q N = max (1, 1.5 − 0.25 ) (33)
If ZNAC > 0; 𝑄 = 𝑚𝑎𝑥 1, 1.5 − 0.25 DFan
(33)
For the negative values of ZNAC , the interference factor is considered at its maximum
For the negativevalue
values
(i.e.,of Zbut
1.5), , the
NACwe takeinterference
into account thefactor is considered
modification of the wettedat its maximum
surface of the nacelle
by imposing a minimum value of 1.0 to the factor. It leads
value (i.e., 1.5), but we take into account the modification of the wetted surface of theto the following relationship:
  
acos 1 + 2 ZDNFan
 
AC

If ZNAC < 0; Q N = max 1.0, 1.5 × 1 −  (34)

π

Once this coefficient is evaluated, the global friction drag coefficient for the NENG
nacelles is obtained by the following relation:


CDENG = NENG × Q N CDN AC1 + CDN AC2 (35)

where the individual friction drag coefficients for NAC1 and NAC2 are computed according
to Equation (25) with the form factor and wetted areas given in Table 2.
Aerospace 2023, 10, 7 13 of 27

3.4. Additional Drag Due to Lift

An additional profile of drag due to lift is considered for the wing. For each wing
segment, the additive drag is estimated according to the relation from [18]:
2 q
CL − CL0 SSegment

CD ADD = 0.75 CD ADD 1 − ( Mcosϕ25 )2 (36)
Re f CL Max − CL0 SREF

where:
"    4 !#
t t
CD ADD = 0.010 CL Max − 0.0046 1 + 2.75 + 100 cos3 ϕ25 (37)
Re f c c

CLMax is the maximum lift coefficient of the wing estimated from Equation (12), and
CL0 is the minimum drag lift coefficient provided in the data file.

3.5. Wave Drag

For aircraft missions at transonic flight conditions, the drag increase due to compress-
ibility effects has to be considered. In the module, the estimation of this drag component
is based on the Korn equation [10]. In the original work, Korn gives an estimation of the
divergence Mach number for an airfoil as:
 
1 t
MDD = K A − Cl − (38)
10 c

The KA coefficient, referred to as the “Korn factor”, is an airfoil technology coefficient,

depending on the nature of the airfoil. It is proposed to use KA = 0.95 for “modern” super-
critical airfoils and KA = 0.87 for “conventional” airfoils. For wing segments corresponding
to a transition area with the fuselage on a BWB configuration, KA = 0.90 can be used.
However, Equation (38) is valid for 2D airfoils only. In order to have an estimation of MDD
for a wing section, the simple swept wing theory, considering a normal to the leading-edge
flow assumption, can be used:
M2D = M3D cos ϕ (39)
1
Cl = CL3D (40)
cos2 ϕ
   
t t 1
= (41)
c 2D c 3D cos ϕ
Introducing Equations (39)–(41) into Equation (38), the following relation for the 3D
swept wing section is obtained:
 
KA 1 Cl 1 t
MDD = − − (42)
cos ϕ 10 cos3 ϕ cos2 ϕ c

Then, the local critical Mach number MCr is estimated according to:
 1
0.1 3
MCr = MDD − = MDD − 0.108 (43)
80

Finally, for a given wing subsection, the wave drag contribution is obtained:

Selement
I f M > MCr ; CD WAVE = 20 ( M − MCr )4 (44)
SREF

The total wave drag of the wing can be found by adding the contributions of all the
NDY wing subsections.
 The total wave drag of the wing can be found by adding the contributions of
NDY wing subsections.

3.6. Parasitic Drag

Aerospace 2023, 10, 7 14 of 27
Finally, an additional parasitic drag (CD PARASITIC) (due to protuberances, an
probes, paint, etc.) is considered. A general formulation used in the pre-design p
to consider this contribution as a percentage of the total friction drag:
3.6. Parasitic Drag
Finally, an additional parasitic𝐶 drag (CDPARASITIC
= 𝑋 ) (due× to𝐶 protuberances, antenna,
probes, paint, etc.) is considered. A general formulation used in the pre-design phase is to
Forthis
consider ancontribution
aircraft performance
as a percentageestimation, XPARAdrag:
of the total friction = 0.025 is generally used. Ho
for a validation purpose with comparisons with the CFD results or experimental
C = XPARA × CDFRICTION (45)
a wind tunnel, XPARA = 0.0DPARASITIC
has to be used.
For an aircraft performance estimation, XPARA = 0.025 is generally used. However,
4. Module
for Validation
a validation purpose with comparisons with the CFD results or experimental data in a
wind tunnel, XPARA = 0.0 has to be used.
The following chapters present the validation of the aerodynamic model on
4.ard tube Validation
Module and wing or blended wing–body configurations. The results obtained
module are compared
The following chapters to the CFD
present results oforthe
the validation experimental data on
aerodynamic model in standard
terms of overa
formance and drag components when possible.
tube and wing or blended wing–body configurations. The results obtained by the module
are compared to the CFD results or experimental data in terms of overall performance
Note that all the drag coefficients presented are expressed in drag counts. and
drag components when possible.
Note that all the drag coefficients presented are expressed in drag counts.
4.1. Standard Tube and Wing Configurations
4.1. Standard
4.1.1. ONERA Tube NOVA
and WingConfiguration
Configurations
4.1.1. ONERA NOVA Configuration
The first configuration considered is the Nextgen ONERA Versatile Aircraf
The first configuration considered is the Nextgen ONERA Versatile Aircraft
VA) [19]
(NOVA) [19](Figure 14),which
(Figure 14), which is designed
is designed with awith a downward
downward winglet
winglet element andelement
different and di
engineconfigurations.
engine configurations.

Figure14.14.
Figure ONERA
ONERA NOVA
NOVA Aircraft.
Aircraft.

The aerodynamic flow conditions for this test case are a Mach number of M = 0.82 at
The aerodynamic flow conditions for this test case are a Mach number of M =
an altitude of Z = 11,300 m. For this configuration, drag decompositions using the ffd72
tools [20] of a of
an altitude Z = 11,300
different m. Forare
configuration this configuration,
available, dragit decompositions
which makes possible to compareusing th
tools
the [20] of adrag
lift-induced different
for the configuration
clean wing and thearewing
available,
equippedwhich makes
with the it possible
winglet with a to co
negative cant angle drag −18the
δWLT =for ◦ . clean wing and the wing equipped with the winglet
the lift-induced
negative cant angle δWLT = −18°. from the aerodynamic module (Figure 15) shows
The comparison with the prediction
a quite good prediction of this drag component, even though only four wing segments are
The comparison with the prediction from the aerodynamic module (Figu
used for the wing definition.
showsNOVA a quite good prediction
is a platform of thisthe
used to investigate drag
effectcomponent, evenofthough
of the integration only fou
ultra-high
segments
bypass ratio are usedengines.
(UHBR) for theDifferent
wing definition.
engine arrangements are available (Figure 16) that
make possible a first validation of the different nacelle interference ratios on drag.
Table 3 compares the drag increase for different engine installations for the NOVA
aircraft between the CFD results and the estimation from the module. A quite good agree-
ment can be found for the different cases proving that the formulation of the interference
factor for the nacelle is realistic.
 NOVA is a platform used to investigate the effect of the integration of ultra-high
bypass ratio (UHBR) engines. Different engine arrangements are available (Figure 16)
NOVA is a platform used to investigate the effect of the integration of ultra-
that make possible a first validation of the different nacelle interference ratios on drag.
Aerospace 2023, 10, 7 bypass ratio (UHBR) engines. Different engine arrangements are available 15 of 27 (Figure
that make possible a first validation of the different nacelle interference ratios on dra

Figure 15. NOVA configuration—winglet effect (δWLT = −18°). Lines—aerodynamic module; sym-
◦ ). Lines—aerodynamic module;
= −=18−18°).
bols—CFD. Figure 15.
Figure NOVA configuration—winglet
15. NOVA configuration—winglet effect (δWLT
effect (δWLT Lines—aerodynamic module; sy
symbols—CFD.
bols—CFD.

Figure 16. NOVA

Figure 16. NOVA configurations configurations
used used for integration.
for UHBR engine UHBR engine integration.

Figure 16. NOVA configurations used for UHBR engine integration.

Table 3. NOVA configuration—drag increase (drag counts) for different engine installations.
Table 3 compares the drag increase for different engine installations for the NOVA
aircraft between the CFD Table 3 compares
results andZNACthe
the drag increase
estimation
(m) from forthe
different
Aerodynamic module.
Module engine installations
A quite good
CFD for the NO
aircraft
agreement can be found = between
CLfor the CFD
the different
0.50 casesresults
provingand D thethe
Cthat estimation
∆CDfrom
formulation of the
the module.
∆CD
inter- A quite g
agreement
ference factor for theRef.
nacelle can be found
is realistic. –
(No Engine) for the different
210 cases proving– that the formulation
– of the i
ference factor for the +10.00
Baseline nacelle is realistic.
232 +22 +28
Table 3. NOVA configuration—drag
Podded increase+0.75
(drag counts) for244
different engine+34
installations. +34
Table 3. NOVA configuration—drag increase (drag counts) for different engine installations.
BLI (m)
ZNAC −0.50
Aerodynamic226
Module +16
CFD +18
CL = 0.50 ZCNAC
D
(m) Aerodynamic
ΔC D
Module
ΔCD CFD
4.1.2. NASA
CL -- Common
= 0.50 Research Model (CRM) C ΔC ΔCD
Ref. (No Engine) 210 --D -- D
The second configuration considered for the validation exercise is the NASA CRM
Baseline Ref. +10.00
(No Engine) -- 210 -- --
model (Figure 17), for which232 +22
both experimental and numerical data are+28
available [21–23].
Podded Baseline
+0.75 +10.00
244 232
+34 +22
+34 +28
BLI Podded
−0.50 +0.75
226 244
+16 +34
+18 +34
BLI −0.50 226 +16 +18
4.1.2. NASA Common Research Model (CRM)
4.1.2. NASA Common Research Model (CRM)
The second configuration considered for the validation exercise is the NASA CRM
The second
model (Figure 17), for which both configuration
experimental considered for the
and numerical validation
data exercise
are available [21–is the NASA C
23]. model (Figure 17), for which both experimental and numerical data are available
23].
Aerospace 2022, 9, x FOR PEER REVIEW 16 of 27
Aerospace 2023, 10, 7 16 of 27

Figure 17. Common research model (CRM) reference configuration.

Figure 17. Common research model (CRM) reference configuration.
For the evaluation
Figure of the performance,
17. Common the
research model module
(CRM) considered
reference only six wing sec-
configuration.
For the evaluation of the performance, the module considered only six wing sections
tions (Figure 18). (Figure 18).
For the evaluation of the performance, the module considered only six wing sec-
tions (Figure 18).

Figure 18. CRM wing—data used by the aerodynamic module.

Figure 18.
Figure 18. CRM wing—data CRMby
used wing—data used by the
the aerodynamic aerodynamic module.
module.
The flow
The flowconditions
conditionsconsidered
consideredare
arethe
theones
ones used
used forfor
thethe different
different AIAA
AIAA Drag
Drag Pre-
Predic-
The flow conditions
diction
tion considered
Workshop
Workshop (DPW) are
(DPW) the
exercises:ones
exercises: used
modelmodel for
scale, the1/10;
scale,
1/10; different
Mach Mach AIAA
number
number Drag
= 0.850;
= 0.850; Pre-
and and Reyn-
Reynolds
olds
diction Workshop number number
(DPW) =exercises: = 5.36 × 10 6.
5.36 × 10 .model scale, 1/10; Mach number = 0.850; and Reyn-
6
Note
olds number = 5.36 × 10Note
6. thatfor
that forthe
theCRM CRM configuration,
configuration, thethe reference
reference areaarea
usedused forforce
for the the force coeffi-
coefficients
cients is not the geometrical wing one, as considered by the module (S Wing = 411.806
2 m2a),
Note that for is
thenotCRM
the geometrical
configuration, wing one, theasreference
consideredarea by the module
used for 2(S
the = 411.806
force
Wing m ), but
coeffi-
but a modified
modified trapezoidal
trapezoidal one (“Wimpress”
oneas(“Wimpress” area =thearea =m383.69
383.69 2 ). Form the). For the present
present compari-
comparisons, the
cients is not the geometrical
sons, the wing
force one,
coefficients considered
considered bythe
for CRMmodule
database (SWing
(from= 411.806
CFD or m2),
experiments)
force coefficients considered for the CRM database2(from CFD or experiments) are therefore
but a modified trapezoidal
are thereforeone corrected
(“Wimpress” area =deal383.69 m ). For the presentin SREFcompari-
corrected in order to deal in withorder
this to
difference with
in Sthis difference
REF with a factor of 0.93172.
with a factor of
sons, the force coefficients
0.93172. considered for the CRM database (from CFD or experiments)
The first validation exercise considered the effect of the fuselage on the CL (α) curve.
are therefore corrected
FigureThe 19first
in validation
order
compares to the
deal Cexercise
with considered
this differencethe effectinofStheREF fuselage
with a on the CLof
factor (α) curve.
L (α) curves from the aerodynamic module, with or without a
0.93172. Figure 19
fuselage compares
taken the CL(α)
into account, withcurves
the CFDfrom the aerodynamic
results from ONERAmodule, obtained with or without
within the DPWa
fuselage
framework. taken into account,
Theconsidered
module considers with the CFD
a zero-lift results from ONERA obtained
= −C1.55 ◦ within the
The first validation exercise the effect of theincidence
fuselageofon α0the L(α) to improve the
curve.
DPW framework. The module considers a zero-lift incidence of α 0 = −1.55° to improve
Figure 19 comparescomparison.
the CL(α)Acurves
the wing–fuselage
quite good
comparison. Aarrangement
from agreement can be observed
the aerodynamic
quite good agreement can with
for thewith
module,
be observed
estimation of the liftaslope of
or without
the by the module the CFD for data.the estimation of the lift
fuselage taken intoslope
account, with the CFD
of the wing–fuselage results from
arrangement by theONERA
module with obtained
the CFDwithindata. the
DPW framework. The module considers a zero-lift incidence of α0 = −1.55° to improve
the comparison. A quite good agreement can be observed for the estimation of the lift
slope of the wing–fuselage arrangement by the module with the CFD data.
Aerospace 2022, 9, x FOR PEER REVIEW 17 of 27

Aerospace 2022, 9, x FOR PEER REVIEW 17 of 27

Aerospace 2023, 10, 7 17 of 27

Figure 19. CRM configuration—fuselage effect on the CL(α) curve. Lines—aerodynamic module;
symbols—CFD.

Regarding
Figure
Figure 19. CRM
19. the overall drag estimation,
CRM configuration—fuselage
configuration—fuselage effect onFigure
effecton the
theCC 20curve.
L (α)
L(α) compares
curve. the CL(CD) curves
Lines—aerodynamic
Lines—aerodynamic com-
module;
module;
puted by the module with the CFD results from ONERA [23]. The drag coefficients from
symbols—CFD.
symbols—CFD.
the CFD results were obtained by the ffd72 far-field analysis tool [20] that allows the
Regarding
elimination
Regarding the
of thethe overall
artificial drag
dragestimation,
overallspurious drag from
estimation, Figure
the20
Figure compares
numerical
20 the the
CL (CC
solution
compares DL)(C
and curves
gives com-
the
D) curves drag
com-
puted by the module with the CFD results from ONERA [23]. The drag coefficients from the
breakdown
puted by the ofmodule
the physical components
with the CFD results(lift-induced,
from ONERA viscous
[23]. and
The wave). It can be seen
drag coefficients from
CFD results were obtained by the ffd72 far-field analysis tool [20] that allows the elimination
that
the the
CFDagreement
results wereis quite good by
obtained (ΔC D = ffd72
the 1 d.c.far-field
maximum) between
analysis tool the
[20]module evalua-
that allows the
of the artificial spurious drag from the numerical solution and gives the drag breakdown of
tion and the of
elimination CFDtheresults upspurious
artificial to lift coefficients
drag fromofthe around 0.50 solution
numerical (using SREFand
the physical components (lift-induced, viscous and wave). It can be seen that the agreement
= Sgives
WING) the
as well
drag
as for the
breakdown wing–fuselage configuration and for the wing–fuselage–horizontal
is quite good (∆CD = 1 d.c. maximum) between the module evaluation and the CFD results seen
of the physical components (lift-induced, viscous and wave). It tail
can planes
be
case.
that
up tothe
lift agreement
coefficients is quite good
of around (ΔCD =S1REFd.c.
0.50 (using = Smaximum) between the module evalua-
WING ) as well as for the wing–fuselage
tion and the CFD
configuration results
and for up to lift coefficients of tail
the wing–fuselage–horizontal around
planes0.50 (using SREF = SWING) as well
case.
as for the wing–fuselage configuration and for the wing–fuselage–horizontal tail planes
case.

CRM configuration—C
configuration—CLL(C = 5=×510 6
Figure20.
Figure 20.CRM (CDD))curves
curves(M
(M= 0.85, ReRe
= 0.85, ). 6Lines—aerodynamic
× 10 module;
). Lines—aerodynamic mod-
symbols—CFD.
ule; symbols—CFD.

Figure 20. CRM configuration—CL(CD) curves (M = 0.85, Re = 5 × 106). Lines—aerodynamic mod-

ule; symbols—CFD.
 Aerospace 2022, 9, x FOR PEER REVIEW 18 of 27
x FOR PEER REVIEW 18 of 27

Aerospace 2023, 10, 7 18 of 27

When considering the different drag components, namely, lift-induced, viscous and
When considering
wave the different
drag, drag components,
a very good agreement cannamely,
also be lift-induced,
found (Figure viscous and discrepancies
21). Some
wave drag, a verycan When
good
be considering
agreement
observed CLthe
for can different
also
values be dragthan
found
higher components,
(Figure namely,
21).viscous
0.50 for Some andlift-induced,
discrepancies viscous
wave drag, butand
in these
wave drag,
conditions,
can be observed for a
CL values very good
a separationagreement
higher than can can
be for
0.50 also
found be found
on theand
viscous (Figure
wingwave 21).
by CFD, Some
drag, which discrepancies
but inisthese can be in
not considered
observed for C
the aerodynamic L values higher
module than 0.50 for viscous and wave drag, but in these conditions,
conditions, a separation can be found on formulations.
the wing by CFD, which is not considered in
a separation can be found on the wing by CFD, which is not considered in the aerodynamic
the aerodynamic module formulations.
module formulations.

(a) (b)
Figure 21. CRM configuration—drag components. Lines—aerodynamic module; symbols—CFD.
(a) (a)
Lift-induced. (b) Wave and viscous. (b)
Figure 21. CRM configuration—drag components. Lines—aerodynamic
Figure 21. CRM configuration—drag module; symbols—CFD.
components. Lines—aerodynamic module; symbols—CFD.
(a) Lift-induced. (b) (a)
Wave Finally, some
and viscous. experimental data are provided on the CRM website for a wing
Lift-induced. (b) Wave and viscous.
equipped with or without through-flow nacelles. Due to their relative position with the
Finally, somewing surface,
Finally,
experimental nacelles
some data are
areconsidered
experimental as
onout
data are provided
provided ofon
the interaction
CRMthe CRM (QN =for
website
website 1). Itacan
fora beequipped
wing
wing seen that the
with
drag or without through-flow
increment found nacelles. Due
experimentally is to their relative
correctly position
predicted by with
the the wing surface,
aerodynamic module
equipped with or without through-flow nacelles. Due to their relative position with the
nacelles 22).
(Figure are considered as out of interaction (QN = 1). It can be seen that the drag increment
wing surface, nacelles are considered as out of interaction (QN = 1). It can be seen that the
found experimentally is correctly predicted by the aerodynamic module (Figure 22).
drag increment found experimentally is correctly predicted by the aerodynamic module
(Figure 22).

Figure 22. CRM

Figure 22. CRMconfiguration—Nacelle
configuration—Nacelle(TFN)(TFN)
effect. effect. Lines—aerodynamic
Lines—aerodynamic module; symbols—
module; symbols—experiments.
experiments.
4.1.3. CRM-NLF Configuration
4.1.3.The next validation
CRM-NLF exercise considers the CRM model with a modified wing with a
Configuration
Figure 22. CRM configuration—Nacelle
natural laminar flow (TFN)
on the effect. Lines—aerodynamic
upper surface. module;
This CRM-NLF symbols—
wing has been designed for tests
The next validation exercise considers the CRM model with a modified wing with a
experiments. in the NASA NTF wind tunnel [24–26], and some experimental results are provided on
natural laminar flow on the upper surface. This CRM-NLF wing has been designed for
tests in the NASA NTF wind tunnel [24–26], and some experimental results are provided
4.1.3. CRM-NLF Configuration
The next validation exercise considers the CRM model with a modified wing with a
natural laminar flow on the upper surface. This CRM-NLF wing has been designed for
 Aerospace 2022, 9, x FOR PEER REVIEW 19 of 27
Aerospace 2022, 9, x FOR PEER REVIEW 19 of 27

Aerospace 2023, 10, 7 19 of 27

on the NASA CRM website. The data used for the CRM reference plane are modified for
thethe
on inboard
NASAwingCRMsection
website.inThe
order to used
data take the wing
for the planform
CRM change
reference plane(Figure 23) intofor
are modified ac-
count.
the NASASimilar
theinboard towebsite.
wing
CRM the CRM
section inreference
The order case,
to take
data used theCRM
for the experimental
wing data
planformplane
reference are
are corrected
change (Figure 23)
modified in into
for order
the ac-to
take
count. the
inboard change
Similar in the
to the CRM
wing section reference wing
reference
in order area
to take case, into account.
the planform
the wing experimental data
change are corrected
(Figure in order to
23) into account.
Similar
take the to the CRM
change reference
in the case,wing
reference the experimental data are corrected in order to take the
area into account.
change in the reference wing area into account.

Figure 23. CRM-NLF geometry used by the aerodynamic module.

Figure
Figure23.
23.CRM-NLF
CRM-NLF geometry usedby
geometry used bythe
theaerodynamic
aerodynamic module.
module.
The flow conditions used for the validation are similar to the NTF ones: a Mach
numberThe of
The flow M =conditions
flow 0.8565 and
conditions a Reynolds
used
used for
forthe number
thevalidation
validationof
are15×10 6. The
similar
are similar to NLF
the capabilities
NTF
to the NTF ones: ofathe
a Mach
ones: aer-
Mach
number of M = 0.8565 and a Reynolds number of 15 × 10 6 . The NLF capabilities of the
odynamic
number of M module
= 0.8565 are considered
and a Reynolds fornumber
this validation
of 15×106exercise,
. The NLF but wing lower
capabilities surfaces
of the aer-
aerodynamic
are considered module are
turbulent, considered
as during forthethis validation
wind tunnel exercise,
tests. The
odynamic module are considered for this validation exercise, but wing lower surfacesbut wing
estimated lower surfaces
maximum loca-
are considered
tion of the turbulent,point
transition as during
on the wind wing
different tunnelsections
tests. Theisestimated
presented maximum
in Figurelocation
24 at CL =
are considered turbulent, as during the wind tunnel tests. The estimated maximum loca-
of the
0.50of transition
forthe point
thetransition
wind tunnel on different
flow wing sections is presented in Figure 24 at C = 0.50
tion point on conditions.
different wing It can be seenisthat
sections an extended
presented natural
in Figure L
24 laminar
at CL =
for the
flow is wind tunnel
possible flow conditions.
(around 50%) for a Itlarge
can be seen that
portion of an extended
the wing, natural
except for laminar
some flow of
portions
0.50 for the wind tunnel flow conditions. It can be seen that an extended natural laminar
is possible (around 50%) for a large portion of the wing, except for some portions of the
the inboard
flow is possiblewing, which50%)
(around are more limited
for a large in terms
portion of NLF
of the wing, capabilities.
except for some portions of
inboard wing, which are more limited in terms of NLF capabilities.
the inboard wing, which are more limited in terms of NLF capabilities.

CRM-NLFconfiguration—estimated
Figure 24. CRM-NLF
Figure configuration—estimated maximum
maximum transition
transition location
locationby
bythe
themodule
module(M =
6× 6 C = 0.50).
0.8565,
(M = Re =
0.8565, 15
Re ×
= 10
15 , C L =, 0.50).
10 L
Figure 24. CRM-NLF configuration—estimated maximum transition location by the module (M =
0.8565, Re = 15 × 106, CL = 0.50).
Then,
Then, these
thesevalues
valuesareareconsidered
considered as as
thethe
XLam
X max parameter,
parameter,as shown
as shownin Figure 11, 11,
in Figure
the location of the transition point XLam is computed for each CL considering CLAdapt = 0.50.
the location
The Then,
comparison
of the
these transition
values
of
point XLam
are considered
the transition as the
line estimated X the module
is computed
by
for each Cas
parameter, L considering
shown
at CL = 0.425,
CL_Adapt
withininfrared
Figure 11,=
0.50.
the The
location
pictures
comparison
of the
available
of the
ontransition
transition
point XisLam
the CRM website,
line estimated
is computed
presented
by the
for25.
in Figure each module
CLbe
It can
at C
considering
L = 0.425, with
CL_Adapt =
seen that, though
infrared
0.50. The pictures available
comparison of theon the CRMline
transition website, is presented
estimated in Figure
by the module at 25.
CL =It 0.425,
can bewith
seen
infrared pictures available on the CRM website, is presented in Figure 25. It can be seen
 that, though not exact, the laminar flow extent estimated by the module is quite realistic,
especially on the outboard wing.
Aerospace 2022, 9, x FOR PEER REVIEW 20 of 27
Aerospace 2023, 10, 7 20 of 27

that, though not exact, the laminar flow extent estimated by the module is quite realistic,
not exact, the laminar flow extent estimated by the module is quite realistic, especially on
especially on the wing.
the outboard outboard wing.

Figure 25. CRM-NLF configuration—transition line estimated by the module at CL = 0.425—

comparison with experimental infrared image.

Moreover,Figure
of most 25.importance
CRM-NLFis the evaluation of the effect
configuration—transition of this laminar
line estimated by the flow moduleex- at
Figure
C 25.
= CRM-NLF configuration—transition
0.425—comparison with experimental line estimated
infrared image. by the module at C L = 0.425—
tent on thecomparison
overall
L performance. Figure 26 compares the CL(α) and the CL(CD) curves
with experimental infrared image.
from the aerodynamic module
Moreover, withimportance
of most the experimental data. A of
is the evaluation zero-lift angle
the effect of laminar
of this attack of flow
α0 = −0.90° is considered
extent on the
Moreover, to improve
ofoverall the comparison
performance.
most importance is Figure for
26 the of
compares
the evaluation CLthe(α) curve.
theeffect For
CL (α)ofand comparison,
thisthe CL (CD )flow
laminar curvesex-
the performances from
tent on the the aerodynamic
estimated module
by the module
overall performance. with the experimental
in fully
Figure turbulent
26 compares data. A
the mode zero-lift
are the
CL(α) and angle
presentedof attack
in of
CL(CD) curves
α = − 0.90 ◦ is considered to improve the comparison for the C (α) curve. For comparison,
dashed lines.
fromIt can
0
the be seen thatmodule
aerodynamic the agreement
with the on the drag evaluation
experimental is quite
data. A Lzero-lift anglegood for of
of attack
the performances estimated by the module in fully turbulent mode are presented in dashed
α0 = −0.90° isaround
NLF flow conditions considered the todifferent
improve C the comparison
L values for the(between
available CL(α) curve. 1.5Ford.c.comparison,
and 3
lines. It can be seen that the agreement on the drag evaluation is quite good for NLF flow
d.c.). the performances estimated by the module in fully turbulent mode are presented in
conditions around the different CL values available (between 1.5 d.c. and 3 d.c.).
dashed lines. It can be seen that the agreement on the drag evaluation is quite good for
NLF flow conditions around the different CL values available (between 1.5 d.c. and 3
d.c.).

(a) (b)
(a) (b) 6).106 ).
Figure 26. CRM-NLF
Figureconfiguration—global aerodynamic
26. CRM-NLF configuration—global performances
aerodynamic (M = 0.8565,
performances Re = Re
(M = 0.8565, 10×
15 =× 15
Lines—aerodynamic
Lines—aerodynamic
Figure 26.module;
CRM-NLF module; symbols—experiments.
symbols—experiments.
configuration—global (a) CL(α).(a)
aerodynamic (b)CLC(α).
L(C(b)
D). CL (C
performances (MD ).= 0.8565, Re = 15 × 106).
Lines—aerodynamic module;
4.1.4. NLF Regional symbols—experiments.
Aircraft Configuration (a) CL(α). (b) CL(CD).
4.1.4. NLF RegionalThe
Aircraft Configuration
CRM-NLF configuration considered laminar flow on the upper wing only. In order
4.1.4. NLF Regional Aircraft Configuration
The CRM-NLF configuration
to evaluate considered
the performance predictionlaminar flow
capabilities onaerodynamic
of the the upper model
wing for
only. In in
a wing
order to evaluate the performance prediction capabilities of the aerodynamic model for a In
The CRM-NLF configuration considered laminar flow on the upper wing only.
order to evaluate the performance prediction capabilities of the aerodynamic model for a
wing in which laminar flow extents on both surfaces, we considered the AG2-NLF wing
wing in which laminar flow extents on both surfaces, we considered the AG2-NLF wing
 design by ONERA in the framework of the Clean Sky 2 AIRGREEN2 program. The ref-
Aerospace erence
Aerospace2022,
2023, 10,x 7FORaircraft
9, considered is a 90-pax turboprop configuration with wing airfoils rede- 2121ofof2727
PEER REVIEW
signed by ONERA at cruise conditions for natural laminar flow capabilities [27–29]. This
configuration is referred to as AG2-NLF. Figure 27 presents the complete configuration,
which considers design some laminar
which winglets,
by ONERA flowan under-carriage
inextents
the fairing
on both surfaces,
framework of the and
Sky a2 AIRGREEN2
we considered
Clean wing–body
the AG2-NLF junction
wing design
program. The ref-by
Karman that are not ONERA
erence in the in
aircraft
modeled framework
considered of
is athe
the aerodynamic Clean
90-pax Sky 2 AIRGREEN2
turboprop
module. order toprogram.
Inconfiguration takewith The
into reference
wing airfoilsaircraft
account rede-
the effect of the Karman considered
signed byon is
ONERAa 90-pax
the viscous turboprop
at cruise
drag, configuration
conditions
the forwing
first with
natural wing
laminar
section is airfoils
flow redesigned
capabilities
considered by ONERA
[27–29].
a thick This
at cruise conditions
configuration is for natural
referred to as laminar flow
AG2-NLF. capabilities
Figure 27 presents [27–29].
the This configuration
complete configuration, is
airfoil (t/c = 40%) with a leading-edge sweep of 30° at the symmetry plane.
referred
which to as AG2-NLF.
considers Figure 27an
some winglets, presents the complete
under-carriage configuration,
fairing which considers
and a wing–body junction
some winglets, an under-carriage fairing and a wing–body junction
Karman that are not modeled in the aerodynamic module. In order to take into Karman thataccount
are not
modeled
the effect in
of the
the aerodynamic
Karman on the module.
viscousIn drag,
order the
to take
firstinto
wing account theiseffect
section of the Karman
considered a thick
on the(t/c
airfoil viscous
= 40%) drag,
withthe first wing section
a leading-edge sweepisof
considered
30° at the asymmetry
thick airfoil (t/c = 40%) with a
plane.
leading-edge sweep of 30◦ at the symmetry plane.

Figure 27. AG2-NLF configuration—data used by the aerodynamic module (4 wing sections).

The flow conditions

Figure 27. are
Figure 27. AG2-NLF a Mach
AG2-NLF number of M
configuration—data
configuration—data = 0.52
used
used by theat an altitude
by the aerodynamic
aerodynamic of Z(4(4=wing
module
module 6100
wing m
sections).
sections).
(20,000 ft). The NLF capabilities of the module are evaluated by comparison with 3D
The flow conditions are a Mach number of M = 0.52 at an altitude of Z = 6100 m
RANS results obtained Theby theconditions
flow ONERA elsA are a software
Mach number withofintegrated
M = 0.52 at transition
an altitude
(20,000 ft). The NLF capabilities of the module are evaluated by comparison with 3D
predic-
of Z = 6100 m
(20,000
tion capabilities [30].
RANS Theft). The
results
results NLF capabilities
are presented
obtained of the
in Figure
by the ONERA module are evaluated
28. The with
elsA software zero-lift by comparison
anglestransition
integrated with 3D
of attackprediction
RANS results
CL(α) curves
considered for thecapabilities obtained
[30].are
The by the
α0results ONERA
= −1.7°are elsA
forpresented software
turbulentinconditions with integrated
Figure 28. and transition
α0 = −1.96°
The zero-lift for ofpredic-
angles attack
the NLF aircraft. tion
It cancapabilities
be seen
considered [30].
that
for the CLThe
(α) results
the are
are presented
agreement
curves α0on=−the infor
1.7◦dragFigure 28. The
evaluation
turbulent zero-lift angles
is quiteand
conditions goodα0 of
=−attack
1.96◦
considered
for theand
NLFNLFfor the
aircraft.C L(α) curves are α0 = −1.7° for turbulent conditions and α0 = −1.96° for
It can be seen that
for both the turbulent flow conditions at the agreement
around on =
CL_Adapt the0.50,
dragwith
evaluation is quite good
a differ-
the
for NLF
both aircraft.
the It canand
turbulent be NLF
seen flow
that conditions
the agreement
at on theC drag evaluation
around = 0.50, is quite
with good
a difference
ence of less than 1for
point in LoD. L_Adapt
both the turbulent and NLF flow conditions at around CL_Adapt = 0.50, with a differ-
of less than 1 point in LoD.
ence of less than 1 point in LoD.

(a) (b)
(a) (b)
Figure 28. AG2-NLF configuration—global aerodynamic performances (M = 0.52, Re = 17.5 × 106).
Figure 28. AG2-NLF configuration—global
Figure 28. AG2-NLFmodule;
Lines—aerodynamic aerodynamic
configuration—global
symbols—CFD.performances
aerodynamic (M = 0.52, Re(M
(a) CL(α). (b)performances
CL(LoD). = 17.5
= 0.52, Re6).= 17.5 × 106 ).
× 10
Lines—aerodynamicLines—aerodynamic
module; symbols—CFD.module; (a) CL(α). (b) C(a)
symbols—CFD. L(LoD).
CL (α). (b) CL (LoD).
4.2. BWB Configurations
4.2. BWB Configurations
4.2. BWB Configurations As the module is used for an MDAO design process of BWB configurations, it is
As the module is used for an MDAO design process of BWB configurations, it is
important
As the module is used to for
check
anififMDAO
the formulations used areof
design process valid
BWB for configurations,
such configurations it iswith high
important to check the formulations used are valid for such configurations with high
values
important to check of
if the
values leading-edge
of formulations sweep angle.
used angle.
leading-edge sweep are valid for such configurations with high
values of leading-edge sweep angle.
4.2.1. AVECA

4.2.1. AVECA
 The first BWB configuration considered is one of the optimized planforms of the
AVECA project [31,32]. AVECA is a long-range BWB configuration designed by Airbus.
Aerospace
Several
Aerospace 2022,
2023, 9,
wing planforms, considering some changes in the geometrical characteristics or 22
10,x 7FOR PEER REVIEW 22 of
of 27
27
different volume constraints for the passenger cabin or cargo, have been optimized by
ONERA using the adjoint method. One of the optimized configurations is selected for
the validation purpose 4.2.1.Theconsidering
AVECA only fourconsidered
first BWB configuration wing segments
is one of or
thefive wing planforms
optimized sections of the
(Figure 29), whichAVECA is very crude.
Theproject The flow
[31,32].
first BWB AVECAconditions
configuration are aisMach
is considered
a long-range BWB number
configuration
one of of M
the optimized =planforms
0.850byatAirbus.
designed of the
an altitude of Z = 11,000Several m.
wing The results
planforms, of the aerodynamic
considering some module
changes in the are compared
geometrical
AVECA project [31,32]. AVECA is a long-range BWB configuration designed to the
characteristics
by Airbus. or
different
CFD results post-processed volume
Several wing with constraints
the ffd72
planforms, for the passenger
tool in order
considering cabin
sometochanges or
obtaininthe cargo,
thedrag have been
breakdown
geometrical optimized
be-
characteristics by
or
ONERA
different using
volume theconstraints
adjoint method.
for theOne of the optimized
passenger cabin or configurations
cargo, have been isoptimized
selected for by
tween the different drag components. Note that the engines were not considered for
the
ONERAvalidation purpose
using the adjointconsidering
method. One only four
of the wing segments
optimized or five
configurations wing sections
is selected for the
both CFD and the (Figure module. 29), which is very crude. The flow conditions are a Mach number of M = 0.850 at
validation purpose considering only four wing segments or five wing sections (Figure 29),
an altitude
which of Z
is very = 11,000
crude. Them. Theconditions
flow results of are
the aaerodynamic
Mach number module
of M =are compared
0.850 to the
at an altitude
CFD results post-processed with the ffd72 tool in order to obtain the drag breakdown
of Z = 11,000 m. The results of the aerodynamic module are compared to the CFD results be-
tween the different
post-processed withdrag components.
the ffd72 Notetothat
tool in order the the
obtain engines
drag were not considered
breakdown between thefor
both CFDdrag
different and the module. Note that the engines were not considered for both CFD and
components.
the module.

Figure 29. AVECA BWB configuration—data used by the aerodynamic module (5 wing sections).

Figure 29.
Figure 30 compares
Figure AVECA
29.the
AVECA BWB
CL(α),
BWB configuration—data
with a zero-lift angle
configuration—data used by
used by the
ofthe aerodynamic
attack module
of α0 =module
aerodynamic +0.175° (5
(5 wing sections).
to im-
wing sections).
prove the comparison,Figure
and thecompares
CL(CD) curves from the aerodynamic module (continuous ◦ to improve
Figure 3030 comparesthe theCCL (α), with a zero-lift angle of attack of α of= α
L(α), with a zero-lift angle of attack 0
+0.175
0 = +0.175° to im-
lines) with the CFD results (symbols).
the comparison, and the It can
CLthe
(CDC be seen
) Lcurves that
from thethe agreement
aerodynamic on the
module lift slope
(continuous lines)
prove the comparison, and (CD) curves from the aerodynamic module (continuous
and on the total drag
withevaluation
the CFD resultsis quite good.
(symbols). It Similarly,
can be seen athatvery
the good correlation
agreement on the lift
lines) with the CFD results (symbols). It can be seen that the agreement on the lift slope
can be
slope and on
the total drag
found on the different evaluation is (lift-induced,
quite good. Similarly, a very
andagood correlation can in
be found
and ondrag components
the total drag evaluation is quite good. viscous
Similarly, wave) presented
very good correlation can be
on the different drag components
Figure 31, with a small (lift-induced,
dragviscous
by theand wave) presented in Figure 31,
foundunderestimation
on the different drag of the viscous
components (lift-induced, module.
viscous and wave) presented in
with a small underestimation of the viscous drag by the module.
Figure 31, with a small underestimation of the viscous drag by the module.

(a) (b)
(a) 30. AVECA BWB configuration—global aerodynamic performances
Figure (b) (M = 0.85, Re = 180 ×
106). Lines—aerodynamic module; symbols—CFD. (a) CL(α). (b) CL(CD).
Figure 30. AVECA BWB configuration—global
Figure aerodynamicaerodynamic
30. AVECA BWB configuration—global performances (M = 0.85,
performances (MRe = 180
= 0.85, Re×= 180 × 106 ).
10 ). Lines—aerodynamic
6 module; symbols—CFD.
Lines—aerodynamic (a) CL(α).(a)
module; symbols—CFD. (b)CLC(α).
L(CD ). CL (CD ).
(b)
FOR PEER REVIEW 23 of 27
Aerospace 2023, 10, 7 23 of 27

(a) (b)
Figure 31. AVECA BWB configuration—drag components. Lines—aerodynamic module; sym-
bols—CFD. (a) Lift-induced. (b) Wave and viscous.

(a)Configuration
4.2.2. NACOR-SMILE (b)
FigureThe
31. AVECA
second BWBBWB configuration—drag
Figureconfiguration
31. AVECA BWB
components.
considered for Lines—aerodynamic
the validation
configuration—drag
module;
components.exercise
sym-opti- module;
is the
Lines—aerodynamic
bols—CFD. (a) Lift-induced.
mized SMILE planform (b) Wave
from the
symbols—CFD. and viscous.
CS2 NACOR
(a) Lift-induced. program.
(b) Wave This configuration is designed
and viscous.
for a short–medium range (SMR) mission, similar to the A320 aircraft. Therefore, com-
4.2.2. 4.2.2.Configuration
NACOR-SMILE Configuration
paredNACOR-SMILE
to the AVECA configuration, its span is much smaller (36 m for SMILE, 80 m for
The second BWB configuration considered for the validation exercise is the optimized
The second
AVECA). BWB
It should be configuration
noted that the considered
aerodynamicfor the validation
module is used exercise
withinisthe
SMILE planform from the CS2 NACOR program. This configuration is designed for a
theOAD-
opti-
mized SMILE planform
MDAO definition phase, from the
but therange
short–medium CS2 NACOR
final(SMR) program.
optimization This
detailstoare
mission, similar configuration
the carried out using
A320 aircraft. is designed
the CFD
Therefore, compared to
for a short–medium
methods [33]. therange
AVECA (SMR) mission,itssimilar
configuration, to the
span is much A320(36
smaller aircraft. Therefore,
m for SMILE, com-
80 m for AVECA). It
paredFigure
to the32
AVECA configuration,
should
compares be
a noted
CAD that its
thespan is much
aerodynamic
rendering of smaller
module
the final (36 within
is used
optimized m shape
for the
SMILE, 80
themdata
OAD-MDAO
with for
definition
AVECA).
used by theIt shouldphase, but
be noted
aerodynamic the final optimization
that the
module. details
Onlyaerodynamic are carried
module
six wing sections out using
areisconsidered the
used within CFD methods
the OAD-
to model [33].
the
Figure 32 compares a CAD rendering of the final optimized shape with the data used
MDAO definitionWinglet
BWB planform. phase, but
andthe final optimization
nacelles details are
are also considered forcarried out using the
the performance CFD
evalua-
by the aerodynamic module. Only six wing sections are considered to model the BWB
methods [33].
tions by the module.planform. Winglet and nacelles are also considered for the performance evaluations by
Figure 32 compares a CAD rendering of the final optimized shape with the data
the module.
used by the aerodynamic module. Only six wing sections are considered to model the
BWB planform. Winglet and nacelles are also considered for the performance evalua-
tions by the module.

Figure 32. SMILE BWB configuration—data used by the aerodynamic module (6 wing sections).
Figure 32. SMILE BWB configuration—data used by the aerodynamic module (6 wing sections).
Figure 33 compares the CL (α), with a zero-lift angle of attack of α0 = −0.225◦ to
Figure 33 compares
improvethe theCcomparison,
L(α), with anda zero-lift
the CL (Cangle of attack
D ) curves of module
from the α0 = −0.225° to CFD
with the im- results.
prove the comparison, and the CL(CD) curves from the module with the CFD results. Itis quite
It can be seen that the agreement on the lift slope and on the drag evaluation
can be32.
Figure seen thegood.
that BWB
SMILE
Similarly,
agreement ona the
configuration—data
verylift
good correlation
slope
used oncanthebedrag
andaerodynamic
by the
found on the different
evaluation
module
drag
is quite
(6 wing
components
good.
sections).
(lift-induced, viscous and wave) presented in Figure 34.
Similarly, a very good correlation can be found on the different drag components (lift-
Figure
induced, 33 compares
viscous thepresented
and wave) CL(α), with
in aFigure
zero-lift
34. angle of attack of α0 = −0.225° to im-
prove the comparison, and the CL(CD) curves from the module with the CFD results. It
can be seen that the agreement on the lift slope and on the drag evaluation is quite good.
Similarly, a very good correlation can be found on the different drag components (lift-
induced, viscous and wave) presented in Figure 34.
FOR PEER REVIEW 24 of 27
FOR PEER REVIEW 24 of 27
Aerospace 2023, 10, 7 24 of 27

(a) (b)
(a) (b)
Figure 33. SMILE BWB configuration—global aerodynamic performances (M = 0.78, Alt. = 41,000
Figure Figure
33. SMILE BWB 33. SMILE BWB configuration—global
configuration—global aerodynamic aerodynamic performances
performances (M = 0.78,(M = 0.78,
Alt. Alt. = 41,000 ft).
= 41,000
ft). Lines—aerodynamic module; symbols—CFD. (a) CL(α). (b) CL(CD).
Lines—aerodynamic
ft). Lines—aerodynamic module; symbols—CFD.
module; symbols—CFD. (a) C
(a) CL(α). (b) CL(C
(α).D).(b) CL (CD ).

(a) (b)
(a) (b)
Figure 34. SMILE BWB configuration—drag
Figure 34. SMILE BWBcomponents. Lines—aerodynamic
configuration—drag components. module; symbols— module;
Lines—aerodynamic
Figure(a)34.
CFD. SMILE BWB(b)
Lift-induced. configuration—drag
Wave and(a)viscous. components. Lines—aerodynamic module; symbols—
symbols—CFD. Lift-induced. (b) Wave and viscous.
CFD. (a) Lift-induced. (b) Wave and viscous.
5. Computing Performances
5. Computing Performances
5. Computing Performances The use of analytical formula allows quite fast computing times for this module, which
The use of analytical formula
makes it quite usefulallows quite
for its use fasta computing
within timesoptimization/design
multi-disciplinary for this module, process
The use of analytical formula allows quite fast computing times for this module,
which makes it quite useful for its use within a multi-disciplinary optimization/design
where a large number of configurations are considered. As an example, for a complete
which makes it quite useful
aircraft for its use
configuration, as within
the CRMa case
multi-disciplinary optimization/design
process where a large number of configurations arepresented, the data
considered. As anfile example,
considers about
for a90 lines
process where a large only (including comments), and the computing time to obtainan
number of configurations are considered. As oneexample,
complete C for
L (Ca
D ) curve
complete aircraft configuration, as the CRM case presented, the data file considers about
(~120 values) is in
complete aircraft configuration, asthe
theorder
CRM of case
1 sec presented,
on a workingthe
station
data(DELL Precision 3630).
file considers about
90 lines only (including comments), and the computing time to obtain one complete
90 lines only (including comments), and the computing time to obtain one complete
CL(CD) curve (~1206.values) is in the order of 1 sec on a working station (DELL Precision
Conclusions
CL(CD) curve (~120 values) is in the order of 1 sec on a working station (DELL Precision
3630). In the framework of a multi-disciplinary design analysis and optimization process
3630). for BWB configurations, an analytic aerodynamic module was developed in order to
6. Conclusions evaluate the aerodynamic lift and drag components of different wing planforms and
6. Conclusions aircraft architectures with a fast restitution time.
In the framework of a multi-disciplinary design analysis and optimization process
In the framework of a multi-disciplinary design analysis and optimization process
for BWB configurations, an analytic aerodynamic module was developed in order to
for BWB configurations, an analytic aerodynamic module was developed in order to
evaluate the aerodynamic lift and drag components of different wing planforms and air-
evaluate the aerodynamic lift and drag components of different wing planforms and air-
craft architectures with a fast restitution time.
Aerospace 2023, 10, 7 25 of 27

The module considered analytic formulations derived from the theory, literature or
from statistical data. Some validation exercises showed that it could be used to estimate the
aerodynamic performances of T&W and BWB configurations within an MDAO pre-design
process for subsonic cruise flight conditions with a quite satisfactory level of accuracy.
The lift slope, the total and individual drag components (lift-induced, viscous and wave)
estimated by the module were in quite good agreement with the reference data issued from
the numerical and experimental data. In addition, the effects of different elements on the
performance, such as fuselage, winglets, nacelles and tail surfaces, were well captured by
the module. Finally, the performance of the wings using laminar flow technologies (NLF or
HLF) could be estimated with a quite good level of accuracy for use in a pre-design phase.
However, it is important to note that this module cannot be used for detailed optimiza-
tion (airfoil shape, wing twist or camber, nacelle positions). For instance, the module will
estimate the same performances for two wings with similar planform and airfoil thickness
evolution with span. These fine optimization steps should be considered in the next stage
of the aircraft design process, using more advanced numerical methods once the general
architecture is defined in the preliminary phase.

Funding: The work presented in this paper was funded by the internal research project CICAV
funded by ONERA from 2015 to 2019 and by the EU NACOR Project, which has received funding
from the Clean Sky 2 Joint Undertaking under the European Union’s Horizon 2020 research and
innovation program under grant agreement N◦ . CS2-AIR-GAM-2018-2019-01. Part of the results
used in this paper were carried out in the framework of AIRGREEN2 Project, which received funding
from the Clean Sky 2 Joint Undertaking under the European Union’s Horizon 2020 research and
innovation program, grant agreement N◦ . 807089—REG GAM 2018—H2020-IBA-CS2-GAMS-2017.
Institutional Review Board Statement: Not Applicable.
Informed Consent Statement: Not Applicable.
Data Availability Statement: Experimental and reference CFD data related to the CRM configura-
tions are available at CRM (https://commonresearchmodel.larc.nasa.gov/ (accessed on 21 November
2022)) or at the Drag Prediction Workshop (https://aiaa-dpw.larc.nasa.gov/ (accessed on 21 Novem-
ber 2022)) websites.
Conflicts of Interest: The author declares no conflict of interest.

Nomenclature
b Wing span (m)
c Airfoil chord (m)
d.c. Drag count (CD × 104 )
t Airfoil thickness
z Altitude (m or km)
P Pressure (Pa)
T Temperature (K)
V Velocity (m/s)
X Coordinate in the streamwise direction (m)
Y Coordinate in the spanwise direction (m)
LoD Lift over drag ratio
M Mach number
MDD Divergence Mach number
MCr Critical Mach number
N Number of components
Re Reynolds number
FF Form factor
QN Interference factor for nacelles
Osw Oswald factor
SWING Wing area (m2 )
SREF Reference area (m2 )
Aerospace 2023, 10, 7 26 of 27

Cl Local (airfoil) lift coefficient

CL Lift coefficient
CD Drag coefficient
KA Airfoil Korn factor
XTra Index for turbulent or laminar flow computations
XLam Laminar flow extent (%)
NDY Number of wing section subdivisions
AMC Aerodynamic mean chord (m)
Greek Symbols
α Angle of attack (◦ )
δ Parameter for Oswald factor
δWLT Winglet cant angle (◦ )
λ Wing aspect ratio
ρ Air density (kg/m3 )
µ Dynamic viscosity (kg/(ms))
ϕ Sweep angle (◦ )
ε Wing taper ratio
Subscripts
LE Leading-edge
TE Trailing-edge
FUS Fuselage
WLT Winglet
ENG Engines
NAC1 Nacelle or Fan
NAC2 Turbine
TAIL Tail surfaces

References
1. Liebeck, R.H. Design of the Blended Wing Body Subsonic Transport. AIAA J. Aircr. 2004, 41, 10–25. [CrossRef]
2. Gauvrit-Ledogar, J.; Defoort, S.; Tremolet, A.; Morel, F. Multidisciplinary Overall Aircraft Design Process Dedicated to Blended
Wing Body Configurations. In Proceedings of the 2018 Aviation Technology, Integration, and Operations Conference, Atlanta,
GA, USA, 25–29 June 2018; p. 3025.
3. Tremolet, A.; Gauvrit-Ledogar, J.; Brevault, L.; Defoort, S.; Morel, F. Multidisciplinary Overall Aircraft Design and Optimization
of Blended Wing Body Configurations. In Proceedings of the 2019 European Conference for Aeronautics and Space Sciences,
Madrid, Spain, 1–4 July 2019.
4. Iwanizki, M.; Wöhler, S.; Fröhler, B.; Zill, T.; Méheut, M.; Defoort, S.; Carini, M.; Gauvrit-Ledogar, J.; Liaboeuf, R.; Tremolet, A.;
et al. Conceptual Design Studies of Unconventional Configurations. In Proceedings of the 3AF Aerospace Europe Conference
2020, Bordeaux, France, 25–28 February 2020.
5. Gauvrit-Ledogar, J.; Tremolet, A.; Moens, F.; Meheut, M.; Defoort, S.; Liaboeuf, R.; Morel, F.; Paluch, B. Multidisciplinary Design
Analysis and Optimization Process dedicated to Blended Wing Body Configurations. In Proceedings of the 33rd Congress of
ICAS, Stockholm, Sweden, 4–9 September 2022.
6. Gray, J.; Moore, K.; Naylor, B. OpenMDAO: An Open Source Framework for Multidisciplinary Analysis and Optimization.
In Proceedings of the 13th AIAA/ISSMO Multidisciplinary Analysis Optimization Conference, Fort Worth, TX, USA, 13–15
September 2010; p. 9101.
7. US Standard Atmosphere 1976, NASA-TM-X 74335, NOAA-NASA-USAF, Washington, DC, USA. 1976. Available online:
https://ntrs.nasa.gov/citations/19770009539 (accessed on 21 November 2022).
8. Lowry, J.; Polhamus, E. A Method for Predicting Lift Increments Due to Flap Deflection at Low Angles of Attack in Incompressible Flow;
NASA TN-3911; NACA: Washington, DC, USA, 1957.
9. Raymer, D.P. Aircraft Design—A Conceptual Approach. In AIAA Education Series, 2nd ed.; American Institute of Aeronautics and
Astronautics Inc.: Washington, DC, USA, 1992.
10. Gur, O.; Mason, W.H.; Schetz, J.A. Full Configuration Drag Estimation. In Proceedings of the AIAA Applied Aerodynamics Conference,
San Antonio, TX, USA, 23–25 June 2009; AIAA: Reston, VA, USA, 2009; pp. 2009–4109.
11. Nita, M.; Scholz, D. Estimating the Oswald Factor from Basic Aircraft Geometrical Parameters. In Proceedings of the Publikationen
zum DLRK 2012 (Deutscher Luft-und Raumfahrtkongress, Berlin, Germany, 10–12 September 2012.
12. Hörner, S.F. Fluid Dynamic Drag; Otterbein Press: Dayton, OH, USA, 1951.
13. Bourdin, P. Numerical Predictions of Wing-Tip Effects on Lift Induced Drag. In Proceedings of the 23rd ICAS Congress, Toronto,
ON, Canada, 8–13 September 2002.
14. Delavenne, M.; Barriety, B.; Vetrano, F.; Ferrand, V.; Salaun, M. Parametric Analysis of an Active Winglet Concept for High Aspect
Ratio Wing using CFD/CSM Computations. In Proceedings of the AIAA Aviation Forum, Virtual, 15–19 June 2020.
Aerospace 2023, 10, 7 27 of 27

15. Reneaux, J. Réduction de la traînée des avions de transport par laminarité étendue. In Proceedings of the AAAF 24e Colloque
Aérodynamique Appliquée, Poitiers, France, 26–28 October 1987.
16. Cousteix, J. Turbulence et Couche Limite; Cepadues Editions: Toulouse, France, 1989; ISBN 2854282108.
17. Haftman, B.; Debbeler, F.J.; Gielen, H. Takeoff Drag Prediction for an Airbus A300–600 and A310 Compared with Flight Test
Results. AIAA J. Aircr. 1988, 25, 1088–1096. [CrossRef]
18. Torenbeek, E. Synthesis of Subsonic Airplane Design; Delft University Press: Delft, The Netherlands, 1986.
19. Wiart, L.; Atinault, O.; Paluch, B.; Hue, D.; Grenon, R. Development of NOVA Aircraft Configurations for Large Engine Integration
Studies. In Proceedings of the AIAA Paper, Dallas, TX, USA, 22–26 June 2015; pp. 2015–2254.
20. Destarac, D. Drag Extraction from Numerical Solutions to the Equations of Fluid Dynamics: The Far-Field Philosophy. In
Proceedings of the 43ème Colloque d’Aérodynamique Appliquée AAAF, Poitiers, France, 10–12 March 2008.
21. Vassberg, J.C.; DeHaan, M.A.; Rivers, S.K.; Wahls, R.A. Development of a Common Research Model for Applied CFD Validations
Studies. In Proceedings of the AIAA Paper, Honolulu, HI, USA, 18–21 August 2008; pp. 2008–6919.
22. Rivers, M.B.; Dittberner, A. Experimental Investigations of the NASA Common Research Model in the NASA Langley NTF and
the NASA Ames 11-ft Transonic Wind Tunnel, AIAA 2011-1126. In Proceedings of the 49th AIAA Aerospace Sciences Meeting
including the New Horizons Forum and Aerospace Exposition, Orlando, FL, USA, 4–7 January 2011.
23. Hue, D.; Esquieu, S. Computational Drag Prediction of the DPW4 Configuration using the Far-Field Approach. J. Aircr. 2011, 48,
1658–1670. [CrossRef]
24. Lynde, M.; Campell, R. Computational Design and Analysis of a Transonic Natural Laminar Flow Wing for a Wind Tunnel
Model. In Proceedings of the AIAA Aviation Forum 2017, 35th AIAA Applied Aerodynamics Conference, Denver, CO, USA,
5–7 June 2017. [CrossRef]
25. Rivers, M.; Lynde, M.; Campbell, R.; Viken, S.; Chan, D.; Watkins, A.N.; Goodliff, S. Experimental Investigation of the NASA
Common Research Model with a Natural Laminar Flow Wing in the NASA Langley National Transonic Facility. In Proceedings
of the AIAA SciTech 2019 Forum, San Diego, CA, USA, 7–11 January 2019. [CrossRef]
26. Lynde, M.; Campbell, R.; Viken, S. Additional Findings from the Common Research Model Natural Laminar Flow Wind Tunnel
Test. In Proceedings of the AIAA AVIATION Forum 2019, Dallas, TX, USA, 17–21 June 2019. [CrossRef]
27. de Gaspari, A.; Moens, F. Aerodynamic Shape Design and Validation of an Advanced High-Lift Device for a Regional Aircraft
with Morphing Droop Nose. Hindawi Int. J. Aerosp. Eng. 2019, 7982168. [CrossRef]
28. Moens, F. Augmented Aircraft Performance by the use of Morphing Technology for a Turboprop Regional Aircraft Wing.
Biomimetics 2019, 4, 64. [CrossRef] [PubMed]
29. Dimino, I.; Andreutti, G.; Moens, F.; Fonte, F.; Pecora, R.; Concilio, A. Integrated Design of a Morphing Winglet for Active Load
Control and Alleviation of Turboprop Regional Aircraft. Appl. Sci. 2021, 11, 2439. [CrossRef]
30. Moens, F.; Perraud, J.; Krumbein, A.; Toulorge, T.; Iannelli, P.; Hanifi, A. Transition Prediction and Impact on 3D High-Lift Wing
Configuration. AIAA J. Aircr. 2008, 45, 1751–1766. [CrossRef]
31. Meheut, M.; Grenon, R.; Carrier, G.; Defos, M.; Duffau, M. Aerodynamic Design of Transonic Flying Wing Configurations. In
Proceedings of the CEAS/KAT net II Conference on Key Technologies, Bremen, Germany, 11–12 May 2009.
32. Meheut, M.; Arntz, A.; Carrier, G. Aerodynamic Shape Optimizations of a Blended Wing Body Configuration for Several Wing
Planforms. In Proceedings of the 30th Applied Aerodynamics Conference, New Orleans, LA, USA, 25–28 June 2012; AIAA: Reston, VA,
USA, 2012; pp. 2012–3122.
33. Bennehard, Q. Aerodynamic Shape Optimization of a Short Medium Range Blended Wing Body Aircraft. In Proceedings of the
56th 3AF International Conference on Applied Aerodynamics, Toulouse, France, 28–29 March 2022.

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.