INVESTIGATING THE PIAGGIO AVANTI DESIGN USING

CEASIOM
Zhang, M.∗ , Cristofaro, M.∗∗ , Wang, Y.∗∗∗ , Da Ronch, A.∗∗ , Rizzi, A.∗
∗ Royal Institute of Technology, Sweden
∗∗ University of Southampton, Southampton, S017 1BJ, United Kingdom,
∗∗∗ China Academy of Aerospace Aerodynamics, Beijing, 100074, China

Keywords: Computational fluid dynamics, aircraft design, stability and control

Abstract duce a refinement in defining the minimum goals

of a candidate project. At the end of the concep-
In early steps of aircraft design the unification tual design phase all the design layouts will have
of configuration definition is important to avoid been analysed, and the “best” one, or possibly
user–input errors. Also coupling with each other two, designs will be down–selected to the prelim-
can strengthen different tools with different spec- inary design phase. During the preliminary defi-
ifications simultaneously provided that the geom- nition, project design is still undergoing a some-
etry definition is transferred with minimum data what fluid process and indeed warrants some el-
loss. This is vitally useful especially when geom- ement of generalist–type thinking, but the mini-
etry data is transferred in order to perform high– mum goals of the project have already been es-
fidelity analysis. This paper reports the analy- tablished during the conceptual definition phase
sis for the pitch control of a three–lifting–surface and the aim is to meet these targets using methods
aircraft Piaggio Avanti using C EASIOM, a tool– with higher order than those used during the con-
chain software for aircraft preliminary design, ceptual definition phase. Furthermore, the partic-
with the baseline configuration coming from the ipants in this working group are mostly genuine
conceptual design code A AA, linked by a com- specialists in each respective discipline. Figure 1
mon name–space C PACS for the means of data indicates the way in which data, or information,
collaboration. is passed between specialist groups during the de-
sign process. The specialist groups must consider
1 Introduction the level of advanced technology to be adopted
together with all of the other active constraints
Figure 1 spells out the details in the early steps on the design. The data flow lines indicate how
of aircraft design for the definition of the config- the technology areas influence the aircraft config-
uration. The figure illustrates two design loops in uration through its performance. The specialist
the conceptual design phase that follow the first– departments/offices provide the input data to the
guess sizing (usually done by a spread–sheet) to project designers who then coordinate a system-
obtain the initial layout of the configuration. The atic search to find the “optimum” configuration
first one, the pre–design loop, is aimed at estab- and settle disputes between conflicting specialist
lishing a very quick (time–scale can be from one opinions. There exists today a good deal of ineffi-
to a few weeks) yet technically consistent sized ciencies in interactions between all these various
configuration with a predicted performance. The groups.
second one, the concept–design loop, is a pro- This paper shows the application of the high
tracted and requires intensive effort involving fidelity aircraft design code C EASIOM [1], the
more advanced first–order trade studies to pro- Computerised Environment for Aircraft Synthe-

1
 ZHANG, CRISTOFARO, WANG, DA RONCH, RIZZI

sis and Integrated Optimization Methods for Pi- A AA has been improved and upgraded several
aggio Avanti configuration which comes from times.
Advanced Aircraft Analysis A AA [2] by inves- A AA enables a fully functioning three–
tigating its longitudinal stability and control. The dimensional aircraft drafting tool Shark/AP [3].
goal is to model the known three–channel con- More information about A AA geometry format
trol surfaces and to show how the three–lifting– and description can be found in [2] or on the web-
surface for pitch control gives lower trim drag site 1 .
than conventional two–lifting–surface configura-
tions. 3 Preliminary Design Toolset C EASIOM

The Computerized Environment for Aircraft

Synthesis and Integrated Optimization Meth-
ods, C EASIOM, developed within the European
6th Framework Programme SimSAC (Simulat-
ing Aircraft Stability And Control Characteris-
tics for Use in Conceptual Design), is a frame-
work tool for conceptual aircraft design that in-
tegrates discipline–specific tools like: CAD and
mesh generation, computational fluid dynamics
(CFD), structures, stability and control analysis,
etc., all for the purpose of early preliminary de-
sign [1]. It is an ad hoc framework that offers
possible ways to increase the concurrency and
Fig. 1 The two design loops in the conceptual de- agility of the classical conceptual–preliminary
sign phase process and the down–select to project process outlined in Fig. 1. C EASIOMsoftware has
study in preliminary design four core functions: geometry and mesh genera-
tion, CFD, aeroelastic analysis, and stability and
control (flight dynamics). Significant features de-
2 Conceptual Design Tool A AA veloped and integrated in C EASIOM as modules
are:
Advanced Aircraft Analysis (A AA) provides a • Geometry module CPACScreator–
powerful framework to support the iterative and sumo [4, 5]. A customized geometry
non–unique process of aircraft preliminary de- construction system coupled to automated
sign. The A AA program allows design engi- surface and volume grid generators,
neers and preliminary design engineers to take resulting model exported to Computer
an aircraft configuration from early weight siz- Aided Design (CAD) via Initial Graphics
ing through open loop and closed loop dynamic Exchange Specification (IGES) standard.
stability and sensitivity analysis, while working
within regulatory and cost constraints. • Aerodynamic Model Builder AMB–
The current version of A AA is based on the CFD [6]. A complete toolbox of aerody-
methods of Airplane Design Parts I–VIII by Jan namic analysis methods ranging from the
Roskam, Airplane Flight Dynamics Parts I–II by empirically based DATCOM to physics-
Jan Roskam, Airplane Aerodynamics and Perfor- based linear and non–linear CFD (Euler &
mance by Jan Roskam and Eddie Lan and meth- RANS) offering broad choice in fidelity:
ods developed for airplane design by DARcorpo-
– Digital DATCOM.
ration engineers. Since 1991, when DARcorpo-
ration acquired the rights for A AA and contin- 1 http://www.darcorp.com/Software/AAA/

ued the development as a commercial venture, [retrieved 14 July, 2014]

2
 Investigating the Piaggio Avanti Design using C EASIOM

– Steady/unsteady vortex–lattice code and thus needs to collaborate with a tool like
(VLM) TORNADO for low–speed A AA, which was described in Section 2.
(linear) aerodynamics and aeroelas-
ticity. 4 Interfaces and Wrappers
– CFD solvers in EDGE code. Euler
If an analysis module is not developed to explic-
solver (EDGE code in Euler mode)
itly serve a central data model it is unlikely that
for inviscid flow cases where total
the module and the central model share the same
pressure and vorticity fields are too
parameterization. Hence conversions need to be
complex to model with isentropic
made. The first step in such a conversion is the
equations e.g. at high speed or
filtering of data. By applying mapping rules only
swirling flow. Examples of these
the data relevant for the analysis module is trans-
are shock waves and propeller slip-
ferred. In a second step the tool wrappers do the
streams. RANS (Reynolds–Averaged
conversion of the data.
Navier–Stokes) flow simulator (e.g.
Figure 2 shows that all the related software
EDGE CFD code) for high fidelity
tools for aircraft concept–design are linked to the
viscous flow analysis at extreme
central model approach C PACS [11] (visualized
flight conditions.
via CPACScreator), then the data are sent to the
• Stability and Control module S & C (e.g. higher order physics-based analysis tools C EA -
SDSA [7]). A simulation and dynamic SIOM . The baseline geometry studied in this pa-

stability and control analyser and flying– per is obtained from A AA–C PACS interface [3].
quality assessor. Includes:

– Performance prediction.
– Test flights by six Degrees of Free-
dom flight simulation.
– Stability Augmentation System
(SAS).

• Aero–elastic module NeoCASS [8].

Quasi–analytical structural analysis meth-
ods that support aero–elastic problem
formulation and solution. Fig. 2 Different conceptual design tools linked
to preliminary design tool–chain C EASIOM by
C EASIOM is intended to support engineers in
C PACS
the conceptual/preliminary design process of the
aircraft, with emphasis on the improved predic-
tion of stability and control properties of elas- 5 C EASIOM Down–Select Configuration &
tic aircraft achieved by higher–fidelity methods Pitch Control Study
than found in contemporary aircraft design tools.
Moreover C EASIOM integrates into one appli- Figure 3 shows the A AA 3–view drawing of the
cation the main design disciplines, e.g. aero- final conceptual design for the Piaggio Avanti.
dynamics, structures, and flight dynamics, im- This is the configuration that is down–selected
pacting on the aircraft performance. It is thus from conceptual design and is now ready for pre-
a multi–disciplinary analysis toolbox brought to liminary design, as illustrated in Fig. 1.
bear on the design of the aero–servo–elastic air- A primary goal of preliminary design is to ob-
craft [9, 10]. C EASIOM however does not carry tain the final wing design with optimized perfor-
out the initial sizing of a baseline configuration, mance: e.g. maximized aerodynamic efficiency

3
 ZHANG, CRISTOFARO, WANG, DA RONCH, RIZZI

Table 1 Structure of the aerodynamic database constructed in C EASIOM for use in the flight simulation
SDSA module.
α M β δele δrud δail p q r CL CD Cm CY C` Cn
x x x – – – – – – x x x x x x
x x – x – – – – – x x x x x x
x x – – x – – – – x x x x x x
x x – – – x – – – x x x x x x
x x – – – – x – – x x x x x x
x x – – – – – x – x x x x x x
x x – – – – – – x x x x x x x

or minimized drag coefficient (CD ), usually start- the vehicle performances and handling qualities.
ing with the cruise point. Table 1 indicate how the aerodynamic database
The wing aerfoils are chosen by a skilful en- computed in C EASIOM is organized. It shows
gineer as the initial design, which may not be op- the static and quasi–static stability coefficients
timum, but can be used as a good starting point. and the control coefficients. In Table 1 are pre-
Then it should be put in an optimization loop to sented the lift, drag and lateral force coefficients
determine the optimized airfoil shapes (thickness (CL , CD , CY respectively), pitching, rolling and
and cambers), twist distributions according to the yawing moment coefficients (Cm , C` , Cn respec-
limits of lift coefficient (CL ), pitching moment tively). The angle of attack is presented as α,
coefficient (Cm ), bending moment, span loading, M is the Mach number and β the side slip an-
fuel tank volumes, etc. for corresponding flight gle while q, p and r are the three rotations in
conditions. Take the Piaggio Avanti for exam- pitch, roll and yaw. The three control surfaces
ple, the aerfoils of the wing are well–designed that can be deflected are the elevator (δe ), the
with cambers and twists to give sufficient lift in rudder (δr ) and the aileron (δa ). The dynamic
order to balance the weight during cruise. Addi- derivatives (Cmα̇ ,CZα̇ ,CXα̇ ,CYβ˙ ,C`β˙ ,Cnβ˙ ) are in-
tionally to get a well designed wing a multi–level stead computed only for different Mach numbers.
optimization [12] should be considered, namely, The coefficients must be obtained for each of
optimizations for cruise, take–off and landing. these parameters throughout the flight envelope,
hence the data is voluminous. In this paper we
only focus on longitudinal control analysis in or-
der to validate the advantage of this three–lifting–
surface configuration. Thus only the second and
sixth rows in Table 1 are filled with CFD Euler
computations.

6 Piaggio Avanti

Piaggio Avanti is a three–lifting–surface twin–

engine turboprop aircraft that has a small forward
wing (canard) to produce extra lift and a T–tail for
Fig. 3 3–view drawing of Piaggio–Avanti in A AA longitudinal and lateral control. It is claimed that
it can save up to around 30% fuel compared with
Another goal is to determine a database of similar aircraft due to the non–traditional config-
aerodynamic forces and moments that cover suf- uration. The main wing is designed to have lami-
ficiently the flight envelope so that it is appro- nar flow over a very high percentage of the wing
priate input to a flight simulator for the study of chord, and the fixed canard is designed to stall

4
 Investigating the Piaggio Avanti Design using C EASIOM

before the main wing, resulting in a nose–down

effect improving the airplane good performance
at high angles of attack. The Piaggio Avanti is
designed to cruise at Mach number of 0.62 at
an altitude of 39,000 ft (economy cruise). More
about the cruise speed and altitude can be re-
ferred in [13] and summarized in Table 2 accord-
ing to Instrument Flight Rules IFR Range & Pay-
load graph.

Table 2 Cruise speed at maximum cruise power [13]

Description Speed [kts] Altitude [ft]
Service ceiling at OEI – 24,000
Maximum speed2 395 30,000
Cruise 370 37,000
∗ Economy cruise 356 39,000
Service ceiling 320 41,000

The typical mid–cruise weight is estimated

by:

W = Operating weight + 4PAX + 1/2Full fuel (1)

at ISA condition and IFR reserves. The lift is

produced to balance the weight, which is around
450 kN estimated by taking maximum payload
and half full fuel [13]. The CFD solvers operate
inside C EASIOM and then all the data are sent to
Stability and Control Analysis module SDSA in Fig. 4 Control surface illustrated on 3–view
C EASIOM to model/simulate the pitch controls. drawing [13]
Figure 4 shows its 3–channel standard control
surfaces illustrated on the 3–view drawing. The
moderate number (often less than 30) spline sur-
canard is a fixed lifting surface, and elevators on
faces. This description is used to generate input
the horizontal tail control the pitch.
for CFD solutions based on the Euler equations.
Tornado VLM and Edge Euler computations
Horizontal trim is studied for both full and canard
are carried out at Mach number 0.62 (economy
off configurations.
cruise 356 kts) in order to build a complete aero–
database to verify the advertised the superior
flight qualities. The VLM method is fast but with 6.1 Aerodynamics using Vortex Lattice
lower fidelity. It is used to quickly generate a Method
complete data–set as Table 1 is shown that is sent
into SDSA for stability and control analysis. In VLM method the compressibility Prandtl–
At the next higher level of detail, the graph- Glauert correction was applied for high subsonic
ical surface modelling tool sumo can be used Mach number (0.62). Figure 5 are the VLM mesh
to define a more detailed geometry based on a and solution for economy cruise at Mach number
of 0.62.
2 At ISA conditions

5
 ZHANG, CRISTOFARO, WANG, DA RONCH, RIZZI

poor fidelity on non–linear aerodynamics such as

the wing in transonic flow. All of these above
motivate us to turn to Euler solver to get more
accurate solutions.

(a) Tornado VLM panels

(a) Euler mesh generated by sumo and TetGen [14]

(b) Euler computation for cruise at CL = 0.565 and

Mach=0.62 at 39,000 ft
(b) Tornado VLM pressure coefficient distribution for
steady flight at M=0.62 and CL =0.65
Fig. 6 Euler solutions from C EASIOM
Fig. 5 Tornado VLM solutions from C EASIOM
Figure 7 shows the forces and moments pre-
dicted for both the full and the canard–off con-
6.2 Euler solutions figurations from Euler solver in C EASIOM. The
canard is very small that only produces slight
Figure 6(a) shows an Euler mesh used in this pa- lift, so the total lift from both configurations are
per generated in sumo with 9.4 million nodes. quite close. We see significant differences on CD
The results for movable control surfaces are com- and Cm for both configurations. Note that the er-
puted not by physically deflected, but by transpi- ror bars showing maximum deviations of the last
ration boundary condition in Edge. Figure 6(b) 500 iterations indicated very poor convergence
shows the Mach contour on the same condition for steady flow computed for drag coefficient and
predicted by C EASIOM Euler solver. We see that pitching moment for negative angles of attack.
a weak shock is formed at the mid–chord of the Hence, the simulations below zero degree angle
wing due to high lift. Note that the VLM model of attack is computed in unsteady model. This
only includes lifting surfaces such as wing, verti- means that the deviation bars for the results of
cal tail and stabilizer, while the sumo geomet- negative angle of attack cases indicates the actual
rical representation also includes aerfoil thick- unsteadiness predicted by simulation.
ness and non–lifting surfaces such as the fuse- Figure 9 shows the pressure distributions C p
lage. Again the VLM only treats the flow fields from Euler solutions at trimmed flight conditions
around the lifting surfaces linearly, which has for both full & canard–off configurations. For the

6
 Investigating the Piaggio Avanti Design using C EASIOM

full configuration, the horizontal trim is achieved

by deflecting the elevator up (negative) 3.33 deg
at α=0.56 deg to maintain the desired lift. For the
alternative geometry without canard, the horizon-
tal trim can be achieved at α=0.62 deg with ele-
vator deflection at δe =-4.25 deg. The elevator de-
flection angles are small for both configurations
at trimmed flight, however the canard–off aircraft
is too stable compared with its fully configured
counterpart. During the presented study the cen-
ter of gravity position was assumed unchanged
between canard ON & OFF cases at 23.41 ft
(a) Lift coefficient from the nose. As figure 7(c) shows, the static
margin for the full configuration is around 7.5%
MAC, while for the canard–off one is around
42% MAC. Adding the canard moves the aerody-
namic center forward that reduces the static mar-
gin accordingly.

(b) Drag coefficient

Fig. 8 C p from Euler solutions for trimmed flight

at Mach number of 0.62 (U = 356 kts at 39,000
ft), full & canard–off configurations

Then we would like to ask why both configu-

rations have very close trim angles. If we go back
to figure 7(c), we could see that, round α = 0
(computed cruise lift coefficient CL ≈ 0.56), the
pitching moments Cm for both configurations are
quite close. This means to get the pitching mo-
(c) Pitching moment coefficient, reference point ments balanced for both configurations require
at 7.41m from the nose similar efforts, namely, similar nose–up moments
provided from the horizontal stabilizer. How-
Fig. 7 Euler solutions at Mach number of 0.62, ever, if we expand the speed and altitude from
canard ON & OFF cases; error bars represent the economy cruise point according to Table 2,
maximum deviations during the last 500 CFD it- we found that although the differences of the
erations trimmed angles of attack between the full and
canard–off configurations are within 0.5 deg for
concerned speeds and altitudes, the elevator de-

7
 ZHANG, CRISTOFARO, WANG, DA RONCH, RIZZI

flections vary significantly for both configura-

tions. From figures 9 and 10 we can see the
full canard configuration superiors in the hori-
zontal trim flight: the elevator deflected angles
are almost kept constantly at small negative val-
ues; while for the canard–off one the behavior is
on a common level, the elevator deflects less (in
absolute value) as the speed increases.

(a) Trim elevator deflection

(a) Trim angle of attack

(b) Trim angle of attack

(b) Trim elevator deflection

(c) Trim drag

Fig. 9 Longitudinal trim calculated and interpo-
lated/extrapolated from Euler solutions for full
Fig. 10 Economy cruise trimmed conditions at
(solid line) & canard–off (dashed line) configu-
altitude of 39,000 ft, calculated from C EASIOM–
rations
Euler

8
 Investigating the Piaggio Avanti Design using C EASIOM

6.2.1 Innovative design omy cruise condition. The freestream velocity

is theoretically zero (but set to very small value,
The forward canard contributes to lift, since it is
e.g. 1 m/s to assure numerical convergence in
a fixed surface, and the pitch angle of the forward
CFD computation) and the aircraft is evoked by
wing is configured so that it always stalls before
the angular velocity. Canard provides additional
the main wing. The resulting automatic nose–
lift, also notice that the flow starts to accelerate
down effect assures excellent in flight behavior at
at the lower side of root of the canard. This ma-
high angles of attack. These aerodynamic advan-
neuver can be achieved with incremental control
tages resulting from the aircraft innovative design
force and elevator deflection, in order to climb to
and construction, cause the airflow to be laminar
a higher altitude, for instance, to the service ceil-
over a very high percentage of the aircraft wing
ing altitude.
chord.
Euler simulations for both full and canard–
6.2.2 Steady pull–up off configurations for 3g steady pull–up maneu-
ver are obtained, for flight at 39,000 ft. The ele-
To judge the maneuverability of the Piaggio vator per g for the full configuration is -1.58 deg,
Avanti, the elevator per g is calculated using Eu- whereas the value for the canard–off configura-
ler solver, by maintaining a steady pull–up with tion is around -4.8 deg, indicating that the full
a constant angular velocity as if the aircraft were configuration has better maneuverability.
attached to the end of a whirling arm provided
very far away. Figure 11 shows the Mach plot for 7 Conclusions

The aircraft design stages, conceptual and pre-

liminary, are necessarily collaborative by their
very nature. An example design exercise was
carried out to illustrate the collaborative aspects
of design using the tools A AA and C EASIOM,
working respectively on conceptual and prelim-
inary design. The chosen example is the Piag-
gio Avanti that has three lifting surfaces and has
(a) Upper surface an advantage in horizontal flight performance.
The exercise brought out some of the details in-
volved when exporting the configuration geome-
try from conceptual design, where the model is
usually not water–tight and meshable, to prelim-
inary design where a meshed model is the nec-
essary starting point for further design work. In
the example a small computer routine was writ-
ten to convert the configuration data from A AA to
input data for C EASIOM–sumo so that the con-
figuration was water–tight and meshable [3]. The
(b) Lower surface
common–language used to minimize the re–work
is a standard for the data describing the aircraft,
Fig. 11 Mach contour from C EASIOM–Euler so- i.e., the CPACS standard proposed by DLR (see
lutions for Piaggio–avanti 3g steady pull–up ma- Ref [11]).
neuver Euler simulations for both full and canard–
off configurations at steady and level flight con-
the full configuration aircraft with 3g pull–up ma- ditions were computed. The trim analysis is car-
neuver predicted from C EASIOM–Euler, at econ- ried out a number of different cruise conditions,

9
 ZHANG, CRISTOFARO, WANG, DA RONCH, RIZZI

showing that the full configuration with a canard CFD grids. An automated approach for con-
has advantages with minimized elevators angle ceptual design,” Progress in Aerospace Sci-
changes while no (significant) drag was added. ences, Vol. 47, No. 11, 2011, pp. 589–596, doi:
The elevator per g for the full configuration is 10.1016/j.paerosci.2011.08.005.
much less than the configuration without a ca- [5] Ciampa, P. D., Nagel, B., Meng, P., Zhang,
nard, when the aircraft is under a steady pull–up M., and A., R., “Modeling for Physics Based
maneuver, provided that the neutral point is fixed. Aircraft Pre–design in a Collaborative Envi-
All of these above validates that the three–lifting– ronment,” 4th CEAS Air & Space Conference,
surface configuration has some aerodynamic ad- Linkoping, Sweden, September 2013.
vantages than conventional configurations. [6] Da Ronch, A., Ghoreyshi, M., and Badcock,
In the future a more realistic model with pro- K. J., “On the generation of flight dynam-
ics aerodynamic tables by computational fluid
pellers can be made in C EASIOM for Euler sim-
dynamics,” Progress in Aerospace Sciences,
ulations. The effects are well modelled for in-
Vol. 47, No. 11, 2011, pp. 597–620, doi:
viscid flow and these should be considered in the 10.1016/j.paerosci.2011.09.001.
design process, one good example is Lötstedt’s
[7] Goetzendorf—-Grabowski, T., Mieszalski, D.,
work [15].
and Marcinkiewicz, E., “Stability analysis us-
ing SDSA tool,” Progress in Aerospace Sci-
8 Acknowledgments ences, Vol. 47, No. 11, 2011, pp. 636–646, doi:
10.1016/j.paerosci.2011.08.007.
The computations were performed on resources
[8] Cavagna, L., Ricci, S., and Travaglini, L., “Neo-
provided by the Swedish National Infrastructure CASS: An integrated tool for structural siz-
for Computing (SNIC) at PDC Centre for High ing, aeroelastic analysis and MDO at concep-
Performance Computing (PDC–HPC). Marco tual design level,” Progress in Aerospace Sci-
Cristofaro thanks Politecnico di Torino for the ences, Vol. 47, No. 11, 2011, pp. 621–635, doi:
exchange period at the University of Southamp- 10.1016/j.paerosci.2011.08.006.
ton. Yongzhi Wang acknowledges the financial [9] Rizzi, A., Eliasson, P., Goetzendorf-Grabowski,
support provided by China Scholarship Council T., Vos, J. B., Zhang, M., and Richardson, T. S.,
(CSC). “Design of a canard configured TransCruiser
using CEASIOM,” Progress in Aerospace Sci-
References ences, Vol. 47, No. 11, 2011, pp. 695–705, doi:
10.1016/j.paerosci.2011.08.011.
[1] Rizzi, A., “Modeling and simulating air- [10] Richardson, T. S., McFarlane, C., Isikveren, A.,
craft stability and control–The SimSAC Badcock, K. J., and Da Ronch, A., “Analysis of
project,” Progress in Aerospace Sciences, conventional and asymmetric aircraft configura-
Vol. 47, No. 11, 2011, pp. 573–588, doi: tions using CEASIOM,” Progress in Aerospace
10.1016/j.paerosci.2011.08.004. Sciences, Vol. 47, No. 11, 2011, pp. 647–659,
[2] Anemaat, W. A. and Kaushik, B., “Geome- doi: 10.1016/j.paerosci.2011.08.008.
try Design Assistant for Airplane Preliminary [11] Rizzi, A., Zhang, M., Nagel, B., Boehnke,
Design,” 49th AIAA Aerospace Sciences Meet- D., and Saquet, P., “Towards a Unified Frame-
ing including the New Horizons Forum and work using CPACS for Geometry Management
Aerospace Exposition, AIAA-2011–162, Or- in Aircraft Design,” 50th AIAA Aerospace Sci-
lando, Florida, USA, 4–7 January 2011. ences Meeting, Nashville, Tennessee, 09–12
[3] Rizzi, A., Meng, P., Nagel, B., Boehnke, D., January 2012.
Anemaat, W. A. J., and Carroll, J., “Collab- [12] Lyu, Z., Kenway, G. K. W., and Martins,
orative Aircraft Design using AAA and CEA- J. R. R. A., “RANS-based Aerodynamic Shape
SIOM linked by CPACS Namespace,” 4th Optimization Investigation of the Common Re-
CEAS Air & Space Conference, Linkoping, search Model Wing,” 52nd Aerospace Sciences
Sweden, September 2013. Meeting, AIAA 2014–0567, National Harbor,
[4] Tomac, M. and Eller, D., “From geometry to

10
 Investigating the Piaggio Avanti Design using C EASIOM

Maryland, USA, 13–17 January 2014, doi:

10.2514/6.2014-0567.
[13] Williams, J. E. and Vukelich, S. R., “Avanti
P180 II, specification and Description,” Tech.
rep., Piaggio Aero Industries SpA, Via Cibrario
4, 165154 Genova, Italy, January 2005, File
P180 Avanti II Spec & Option-1-R2.doc, Revi-
sion 5.0.
[14] Si, H., “TetGen: a quality tetrahedral mesh gen-
erator and 3D delaunay triangulator,” Tech. rep.,
User’s Manual, WIAS technical Report No. 13,
2013.
[15] Lötstedt, P., “Accuracy of a Propeller Model
in Inviscid Flow,” Journal of Aircraft, Vol. 32,
No. 6, 1995, doi: 10.2514/3.46880.

9 Contact Author Email Address

Zhang, M. [email protected]
Cristofaro, M. [email protected]
Wang, Y. [email protected]
Da Ronch, A. [email protected]
Rizzi, Z. [email protected]

Copyright Statement

The authors confirm that they, and/or their company or or-

ganization, hold copyright on all of the original material
included in this paper. The authors also confirm that they
have obtained permission, from the copyright holder of any
third party material included in this paper, to publish it as
part of their paper. The authors confirm that they give per-
mission, or have obtained permission from the copyright
holder of this paper, for the publication and distribution of
this paper as part of the ICAS 2014 proceedings or as indi-
vidual off–prints from the proceedings.

11