Numerical simulation of a flow around an unmanned aerial vehicle/Bepilociu lektuvu skraidymo skaitmeninis modeliavimas.
Meftah, S.M.A. ; Imine, B. ; Imine, O. 等
1. Introduction
The unmanned aerial vehicle (UAV) is a small plane which flies in
an autonomous way with only one virtual pilot on board. That does not
mean that the plane is completely out of control once it leaves the
track. The presence of a human being in the loop of control is
essential. But the operator is on a station on the ground, either on
ground, or at sea or in a control center. He can intervene at any time,
examine the situation and the rules of engagement, and stop the mission
if necessary. The concept of an UAV resounds in several research
laboratories of any developed country. The first instigators of this
type of the projects were the soldiers. Indeed, the armies of many
countries prefer to rather send UAV's for the recognition in the
high-risk zones that man. This type of flying machine can take much
various missions, extending from the control of the scientific
experiments to the intelligence collecting the monitoring during the day
or the night.
The numerical simulation in the field of aerodynamics is relatively
recent research tools. The specialists raise the question of knowing if
the field of aerodynamics is purely theoretical or numerical, they
estimate however that it is theoretical, because of the many numerical
tests necessary for the stage to insufficiencies of the knowledge of the
calculation methods [1]. Since a score of years, one can say that
numerical aerodynamics lost much of its empirical nature due to the
convergent efforts of mathematics to solve the nonlinear problems
involved in the calculation methods in aerodynamics and also thanks to
the invention of the very fast machines of calculation and the
development of data processing.
[FIGURE 1 OMITTED]
In this present work, a numerical simulation of a flow around an
UAV is presented. The first phase of this work consists in using of
construction of this UAV to create a solid model which specifies its
external geometry in Solid Works (Fig. 1). This solid model is then
exported in to a programme of the grid generation in order to return its
geometry adapted to aerodynamic calculations and tested in various
flying conditions [2]. A computer code of computational fluid dynamics
(CFD) [3] is used to obtain estimates of the coefficients of lift and
drag to 20 meters per second. The model of turbulence Spalart-Allmaras
[4] is used to leap the transport equations of Navier-Stokes in order to
correctly predict the complex flow around the plane. A study of the
independence of the grid is then carried out to determine the precision
of the grid of calculation used for analysis CFD.
2. Description of the UAV
This UAV as shown in Fig. 2 is 2 meters long and provided with a
pendular wing of the scale of 2.9 m. the profile of this wing is of type
Clark yh with a cord of 0.236 m, stalled on the fuselage of 4[degrees]
with a dihedral of 4[degrees]. The empennage group of the shape of V
reversed is profiled with Naca 0012 and attached to the wings by two
beams which give also the name of an UAV to twin-boom. These UAV's
aspire of a remarkable stability to the no desirable movements of roll.
This UAV is designed to have a pusher engine which will be embarked
behind the fuselage.
[FIGURE 2 OMITTED]
3. Transport equations and turbulence model
The flow around the plane is considered turbulent asymmetrical. The
general forms of the transport equations can be written in Cartesian
coordinates [5] as:
* continuity equation
[partial derivative]/[partial
derivative][x.sub.j]([bar.[rho]][[??].sub.j]) = 0 (1)
* momentum conservation equation
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (2)
* turbulence model
The model of turbulence used in this present work is a model with a
transport equation for the v quantity suggested by Spalart and Allmaras
[4]
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (3)
On the terms of production and destruction are defined as follows
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (4)
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (5)
with
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (6)
and
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (7)
and the coefficients of closing are given by the following values:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (8)
4. Results
Several compilations were carried out for various angles of
incidence to determine the lift and drag. Each case was carried out at
the same speed: 20 m/s. The angle of attack extended from 3 to
18[degrees].
The evolution of the drag coefficient according to the incidence is
presented in Fig. 3. It is a parabolic type tendency similar to those
found on the planes in general. In particular, one notes the increase in
the drag with the increase in the angle of incidence. This curve is
presented with drag of the isolated wing which equips the UAV in
question and which was obtained by wind tunnel test. The two curves
follow the same tendency separately which is normal because variation of
the drag between the two curves corresponds to the drag of the remainder
of the UAV which varies very slightly with the incidence.
[FIGURE 3 OMITTED]
Fig. 4 treats the evolution of the lift according to the incidence
for the UAV and the isolated wing. An agreement of the two curves is
noted in the linear zone which is explained by the fact what it is the
wing which contributes in a great proportion to the total lift produced
on the plane. The other parts of the plane, in particular the
stabilizer; contribute slightly to weak and moderate incidence. However
the lift of the isolated wing exceeds that of the plane for the bigger
angles of incidence. This is dues with the fact that the Reynolds number
of the tests of the isolated wing is much higher than that of the plane
tested numerically which leads to higher maximum Cz for the isolated
wing. Numerical calculations show well the phenomenon of unhooking of
the UAV which is located at approximately 12[degrees]. It should be
noted that this incidence which appears is weak only seemingly because
it should be mentioned that the reference of measurement of the angle of
attack is the base of the fuselage and that the wing is fixed with
4[degrees].
[FIGURE 4 OMITTED]
The map of shear stress along X is shown in Figs. 5 and 6 for
respectively an incidence of 0[degrees] and 15[degrees]. A flow return
is noted on almost the totality of the wing, in particular, it is much
accentuated in the central part. This effect is expected because the
wing is not twisted and the great deflections of the air are perceptible
in this area. The separation of the flow recorded on the nose of the
fuselage still persists with this angle of incidence.
[FIGURE 5 OMITTED]
[FIGURE 6 OMITTED]
Fig. 7 shows the chart of the parietal flow recorded on the level
of the wing root with the fuselage with 15[degrees] of incidence. It is
a representation of the streamlines of the flow inside the internal zone
of the boundary layer around the UAV which is completely turbulent. It
is a very useful chart for the draftsman of the aerodynamic forms
because it reveals the design defects characterized by the appearance of
the zones of flow separated colour blue as shown in the Fig. 8 on the
wing and on the level from the bases from fastener from the beams.
[FIGURE 7 OMITTED]
[FIGURE 8 OMITTED]
Fig. 9 shows the two marginal swirls which release from salmons of
the two wings of the UAV to the passage of the flow for the incidence
15[degrees], the size of the marginal swirl for such an incidence is
considered to be considerable. Moreover, one notes the complexity of the
flow around the plane for an incidence judged like the beginning of
unhooking.
[FIGURE 9 OMITTED]
[FIGURE 10 OMITTED]
5. Conclusion
The present study was undertaken to observe the aerodynamic
performances of an UAV by using a CFD computer code. This concept of
design of flying machines is judged like an alternative to the wind
tunnel tests for a possible reduction of the design cost. As shown in
Fig. 10, this work brought moreover, certain important aerodynamic
corrections around this UAV such as the aerodynamic careenage of the
nose of the fuselage and the setting of the bases of fastener of the
beams on the level of the under-surface of the wings
Received June 02, 2010
Accepted April 05, 2011
References
[1.] Ronzheimer, A. 2005. Post-parameterization of CAD-geometries
using freeform deformation and grid generation techniques, Notes on
Numerical Fluid Mechanics and Multidisciplinary Design, vol.87: 382-389.
[2.] Flores, J.; Reznick, S.G.; Hol, T.; Gundy, K. 1988. Transonic
Navier-Stokes solutions for a fighter-like configuration, Journal of
Aircraft 25, October 1988: 875-881.
[3.] Bhaskaran, R. FLUENT Short Course, http://instruct1.
cit.cornell.edu/courses/fluent. Includes step-by-step tutorials and
problem sets. Chapters: Introduction to Computational Fluid Dynamics;
Laminar pipe flow; Turbulent pipe flow; Compressible nozzle flow;
Airfoil flow.
[4.] Li, J.; Li, F.; Qin, E. 2000. Numerical simulation of
transonic flow over wing-mounted twin-engine transport aircraft, Journal
of Aircraft, vol.37, No.3: 469-478.
[5.] Benhamza, M.E.; Belaid, F. 2009. Computation of turbulent
channel flow with variable spacing riblets, Mechanika 5(79): 36-41.
S.M.A. Meftah, Mechanical Engineering Faculty, USTO Oran, B.P 1505
El Mnaouer U.S.T, Oran, Algeria, E-mail: hmeftahdz@yahoo.fr
B. Imine, Mechanical Engineering Faculty, USTO Oran, B.P 1505 El
Mnaouer U.S.T, Oran, Algeria, E-mail: imine_b@yahoo.fr
O. Imine, Mechanical Engineering Faculty, USTO Oran, B.P 1505 El
Mnaouer U.S.T, Oran, Algeria, E-mail: imine_o@yahoo.fr
L. Adjlout, Mechanical Engineering Faculty, USTO Oran, B.P 1505 El
Mnaouer U.S.T, Oran, Algeria, E-mail: adjloutl@yahoo.fr