The document discusses the optimization of the design of unmanned aerial vehicle (UAV) wings. It analyzes two types of rectangular wings using vortex lattice CFD software to determine aerodynamic loads and CATIA V5 software for structural analysis. When composite materials are used instead of isotropic materials, a 34% mass reduction is achieved. The optimum design is determined for each wing based on minimum mass, stress, and displacement.
Optimisation of the design of uav wing j.alexander
1. OPTIMISATION OF THE DESIGN OF UAV WING
J.Alexander, S.Prakash, Dr.BSM.Augustine
ABSTRACT
This paper presents the research extensively carried out based on the
static analysis for finding the optimum design of wing structure. The two types of
wings (Rectangular wings) for unmanned airplanes were chosen and designed
at the basis of M=0.21 and at M=0.4 respectively. Further it also covers both the
aerodynamic and structural designs. The aerodynamics study was achieved by
using the vortex lattice method using XFLR-CFD software and the structural
analysis was achieved by using the CATIA V5 package consisting of isotropic
and composite materials.
A wing structure consists of spars, ribs, reinforcements and skin and they
were optimized considering two aspects of weight minimization, and critical load
maximization. The wing carries an elliptically distributed load along the span.
The design variables are based on the positioning of spars and ribs of the
structure. The out come of the research clearly indicates the significant results in
terms of objective functions obtained through the optimization procedures.
W hen composite material is used instead of isotropic material mass
saving of 34 % is achieved. The optimum design for each wing was obtained
according to the mass, stress and displacement. This could be seen in the
presentation subsequent
G: Shear modulus, (pa)
Nomenclature ν: Poisson s ratio
ρ: Density, (kg/m3)
Vn: Normal velocity, (m/s) D.O.F.: Degree of freedom
V∞: free stream velocity, (m/s) M.A.C.: Mean aerodynamic chord (m)
Θ Flight path angle, (deg.) A/C: Aircraft
Wi: Induced flow component, Gr/Ep: Graphite/epoxy
(m/s) H.M Gr/Ep: High modulus
α: Angle of attack, (deg.) graphite/epoxy
Aij: Influence coefficient Sg/Ep: Glass/epoxy
Li: Lift of panel i, (N) Wto - total aircraft weight
Γ: Strength of vortex (m2/s) Ww :weight of wing structure
L: Lift of wing, (N)
Wf :weight of fuel stored in wing
Di: Drag of panel i, (N)
D: Drag of wing, (N) L :length of wing
CLcom: Lift coefficient in Lf :length of fuel tank within wing
compressible flow Co :chord length at wing root
CLincom: Lift coefficient in Ct :chord length at wing tip
incompressible flow Co :width of fuel tank at wing root
M: Mach number
E: Young s modulus, (pa) Ctf width of fuel tank at Lf
structural design of an A/C wing is
INTRODUCTION an important factor in th e
An Unmanned Air Vehicle (UAV) is performance of the airplanes i.e.
an unpiloted aircraft. UAV are obtaining a wing with a high
becoming standard means of stiffness/weight ratio and sustaining
collecting information. The optimum the unexpected loading such as gust
and maneuvering situations. This is
1
2. accomplished by studying the The shape of the drag force vs.
different design parameters required angle of attack is approximately
to specify the wing geometry. The parabolic. It is desirable for the wing
idea of the structural optimization in to have the maximum lift and
the classical sense, has been smallest possible drag.
considered to be the minimization of To develop a more accurate model
(2)
structural mass by varying member for such optimal design studies ,
sizes or shell thickness of a model the Structure of the wing that
in which the geometry remains combines the composite(11) and
unchanged. Therefore many studies isotropic materials and compare the
have been made during the last wing with the wing that design only
years to find the structural from isotropic material in order to
optimization. obtain high strength/weight ratio for
One of the major loading is finding optimum design.
aerodynamic loading. Aerodynamic
characteristics of the UAV (7) vary Types of load acting on the
with certain parameters like the aircraft wing
angle of attack and others. 1.Aerodynamic loading
Experimental works on UAVs have 2.Structural loading
been conducted in many places with 3.Fuel W eight
various aerofoil profiles but enough The optimization chain consists
work with the Computational Fluid of:
Dynamics (CFD) analysis is not
1 Aerodynamic grid
available that much till now. The
present work contains mainly CFD(8) generation
analysis to determine the flow 2 CFD Analysis by using
pattern a n d th e aerodynamic XFLR software
characteristics of an UAV. The 3 Parametric CFD model
shape of an aircraft is designed to generation
make the airflow through the surface
4 Structural grid generation
to produce a lifting force in most
efficient manner. In addition to the 5 FEM Analysis and
lift, a force directly opposing the structure sizing
motion of the wing through the air is
always present, which is called a AIM AND SCOPE OF THE
drag force. The angle between the PRESENT INVESTIGATION
relative wind and the chord line is
the angle of attack of the aerofoil. A light rectangular wing of an UAV
The lift and drag forces developed with NACA0012 airfoil is developed
by an aircraft will vary with the using CATIA software and the static
change of angle of attack. The cross analysis of the wing by CATIA V5
sectional shape obtained by the package has been presented and
intersection of the wing with the used to determine the optimum
perpendicular plane is called an design for the wing. This operation
aerofoil. Here NACA 0012(10) involves using the vortex lattice
symmetric aerofoil profiles have method to obtain the aerodynamic
been used for the present research load(lift) for rectangle wing. This
work. The lift force increases aerodynamic result (lift) is
almost linearly with the angle of combined(4) with the fuel weight and
attack until a maximum value is structural weight of the wing, and it is
reached where upon the wing is said used to simulate the wing loading on
to stall. the wings during the static analysis.
2
3. Some of the parameters of aerofoil The whole wing is divided into many
and properties of air have been kept numbers of panels and each panel is
constant and some have been considered as a bound vortex. The
varied. The flow of air over the required strength of the bound vortex
aerofoil is varied as per requirement. on each panel is to be calculated by
The chord length of the aerofoil is applying a surface flow boundary
450mm. (10)The free stream airflow condition. The equation used is the
has been kept 60 m/s and the effect usual condition of zero flow normal
of the temperature in the study has to the surface. For each panel the
been neglected. The density of air condition is applied at the 3/4 chord
(ρo)=1.22 kg/m3 , operating pressure position along the centre line of the
(Po)=0.101 MPa (1.01 bar) and panel(9). The normal velocity is made
absolute viscosity up of a free stream component and
(μ)=1.789x10-5kg/m-s.The Reynolds an induced flow component, as
Number has been considered as shown in Fig.(2).
variable.
METHOD OF ANALYSIS
AERODYNAMICS
The aircraft motion through the
atmosphere produces forces called
the aerodynamic forces mainly the
lift and drag forces and the wing is
the major source of this aerodynamic
α V Panel i
force. The wing is that surface of
aircraft which supports the aircraft by
Fig 2 : Velocity component of the
means of the dynamic reaction on
panel i on the wing
the air producing pressure
distribution. This pressure
distribution on the wing changes with Vn = 0 = Sinθ+W i --------------------(1)
different wing angles of attack and
flight condition. The induced component is a function
of strength of all vortex panels on the
Methods of calculating aerodynamic
load (Lift) wing. Assuming small angles
1.Prandtl’s classical lifting line theory
Sinθ=sin(α-β)= (α-β)=-dz/dx--------(2)
2.Vortex lattice numerical method
3.Panel method Wi= Гj ---------------------(3)
In this research work vortex lattice
method is used Гj= -Vsin(θ) -----------------(4)
Vortex lattice method(8) Where θ is flight path angle
Aij is the influence coefficient which
represents the induced flow on panel
I due to the vortex on panel j
A solution for the strength of the
vortex lines on each panel is found
by solving the matrix of equations
the lift coefficient for the wing at a
given angle of attack will be obtained
by integrating the panel lift
Fig1: Wing panels distribution. The lift on a particular
panel can be found using the Kutta
law.
3
4. Li = ρV∞ Гj 2k Lift of panel i x = position along wing
ka = lift profile
L= Lift of wing ---------(4) coefficient
The down wash velocity induced on
a panel can be calculated once the
strength of the wing loading is
known. The variation between local
flow angles of the panel and the free
stream velocity can be found. Figure 3: Derivation of Lift
consequence of this down wash flow
is that the direction of action of each
panels lift vector is rotated relative to We can determine the total lift by
the free stream direction. The local integrating across the length of the
lift vectors are rotated backward and wing:
hence give rise to a lift induced drag. Within the notebook interface we
By integrating the component of define ql(x) and calculate its integral
panel lift coefficient that acts parallel (Figure 5). W e incorporate math
to the free stream across the span equations, descriptive text, and
then the induced drag coefficient can images into our calculations to
be found. clearly document our work.
Di=ρV∞Гi2kSin(αi)
Through integration, we find that
Drag of panel i
Lift=πL2ka/4 ---------------(7)
D= Drag of the wing--(5) W e determine ka by equating the
lift expression that we just calculated
with the lift expressed in terms of the
The induced flow angle (αi)
aircraft’s load factor. In aircraft
represents the amount of rotation of
design, load factor is the ratio of lift
the lift vector backward and must be
calculated from the velocities to total aircraft weight:
induced on the bound vortex of the n=Lift/W to ----------------(8)
panel by other panels and the free Load factor equals 1 during straight
stream. and level flight, and is greater than 1
Pitching moment about the wing root when an aircraft is climbing or during
loading edge can be calculated by other maneuvers where lift exceeds
summing the panel lift multiplied by a aircraft weight.
moment arm which extends in the x- Equating two lift expressions,
direction from the loading edge of
Wton/2= πL2ka/4 -------------------(9)
the wing to the centre of the bound
vortex for the panel. and solve for the unknown ka
ANALYTICAL MODELING OF term. Our analysis assumes that lift
AIRCRAFT WING LOADS forces are concentrated on the two
wings of the aircraft, which is why
Lift the left-hand side of the equation is
We assume an elliptical distribution divided by 2. W e do not consider lift
for lift across the length of the wing, on the fuselage or other surfaces
resulting in the following expression
forlift profile(4)
---------(6)
where L = length of wing L i ft= ∫ dx-----10)
4
5. Plugging ka into our original ql(x)
expression, we obtain an expression
for lift:
An analytical model like this helps
us understand how various
parameters affect lift. For example,
we see that lift is directly proportional
to load factor (n) and that for a load
factor of 1, the maximum lift , occurs
at the wing root (x=0).
Weight of Fuel Stored in Wing
Weight of Wing Structure
We assume that the load caused by We define the load from the weight
the weight of the wing structure is of the fuel stored in the wing as a
proportional to chord length (the piecewise function where load is
width of the wing), which is highest zero when x > Lf. We assume that
at the wing base (Co) and tapers off this load is proportional to the width
toward the wing tip (Ct). Therefore, of the fuel tank, which is at its
the load profile can be expressed as maximum at the base of the wing
. W e define qw(x) and integrate it (Cof) and tapers off as we approach
across the length of the wing to the tip of the fuel storage tank (Ctf).
calculate the total load from the wing We derive qf(x) in the same way that
structure: we derived qw(x), resulting in an
We then equate this structural load equation of the same form:
equation with the structural load
expressed in terms of load factor
and weight of the wing structure
(Wws)
and solve for kw.
Plugging kw into our original qw(x) ---------(12)
expression, we obtain an analytical
expression for load due to weight of
the wing structure.
Figure 5: showing the weight of
Figure 4: showing the weight of fuel
wing structure
Total Load
We calculate total load by adding the
three individual load components.
This analytical model gives a clear
view of how aircraft weight and
geometry parameters affect total
load.
5
6. Results of the theoretical
aerodynamic analysis by CFD-
XFLR. Table (1) shows the main
characteristics of the wings. The
aerodynamic results (lift) are used to
simulate the wing loading on the
MODELING AND ANALYSIS wings during the static stress
analysis.
Table1: Main characteristics of the
Aerodynamic model is developed
wings
and analyzed by using CFD-XFLR
Characteristic Rectangle
software and the structural model is
wing
generated by CATIA V5 and is
Wing span (m) 3.35
analysed by using ansys.
Gross area (m2) 1.53
CFD ANALYSIS Aspect ratio 7.33
Taper ratio 1
The wing is designed with single Tip chord (m) 0.45
aerofoil profile. The aerofoil number Root chord (m) 0.45
is NACA 0012 which is a symmetric M. A. C. (m) 0.45
aerofoil. Sweep angle of 0
leading edge (deg.)
Section profile NACA 0012
Table 2: Calculated polar through
CFD analysis
XFLR5_v4.17 polar for: NACA 0012
Mach = 0.2 Re = 1.000 e 6
A OA Cl Cd Cm
1 0.11 0.0051 0.001
Fig6 :Cl Vs Alpha Curve At 3 0.32 0.0058 0.0022
various Reynolds number 5 0.54 0.0078 0.0054
7 0.82 0.01 0.0084
9 0.992 0.013 -0.0084
11 1.15 0.0177 0.0012
13 1.29 0.023 0.0102
15 1.33 0.035 0.021
17 1.1801 0.086 0.0093
The aerodynamic results are taken
for a range of wing incidences
Figs.(7) and and only the maximum
wing loadings are applied to the
wings to serve wing stress
conditions
Fig 7: Lift Distribution over the wing
6
7. load
distance (m) N Wto = 108 kg
0.25 73.39 Ww = 14.17kg
0.5 36.352 Wf = 32 kg
0.75 17.885 L = 3.35 m
1 17.471 Lf = 2.3 m
1.5 16.235 Co = 0 .45 m
2 14.349
Ct = 0.45 m
2.5 4.59
Co = 0.275 m
3 -0.594
Ctf = 0.275 m
3.35 -8.763
Table 3: Load acting at various
portion of the w ing
Distance from leading edge in m
α Fig 9 Load distribution over the
wing
Fig 8) Alpha Vs Cl Curves at STRUCTURAL STATIC ANALYSIS
Mach no 0.2
Structural analysis is probably the
CALCULATION OF WING most common application of the
LOADING finite element method. The static
analysis is used to determine
Plot of individual load displacements, stresses, etc. under
components and total load along static loading conditions. In this work
the length of the wing. the structural static analysis was
we see that lift is the largest achieved by using the CATIA V5
contributor to total load and that the package in order to obtain stress
maximum load occurs at the end of and displacement distribution in the
the fuel tank. Fuel load also wings by using isotropic and
contributes significantly to the total composite material which are used
load, while the weight of the wing is commonly these days especially for
the smallest contributor. unmanned A/C. The CATIA V5
While it is useful to visualize wing program has many finite element
loads, what really concerns us are analysis capabilities, ranging from a
the shear force and bending simple, linear, static analysis to a
moments resulting from these loads. complex, non linear, transient
We need to determine whether dynamic analysis. A typical CATIA
worst- case bending moments ex- V5 analysis has following distinct
perienced by the wing are within steps:
design limits i) Build the model.
ii) Apply loads
7
8. iii) Obtain the solution.
iv) Review the results.
Materials
Historically, aluminum materials
have been the primary material for
aircraft and space craft construction.
Today, structural weight and
stiffness requirements have
exceeded th e capability of
conventional aluminum, and high-
performance payloads have
Fig 10.CATIA WING MODEL
demanded extreme thermo-elastic
stability in the aircraft design
Optimization Procedure
environment. During the past
The wing structural design problem
decades, advanced composite
is composed into two levels in a
materials have been increasingly
hierarchical structure at the first
accepted for aircraft and aerospace
level, the wing configuration is
structural materials by numerous
completely made up of isotropic
developments and flight applications.
material and the following design
Composite materials are those
parameters were investigated such
containing more than one bonded
as number of stiffeners, skin
material, each with different
thickness, cross sectional area of
structural properties.
stiffener and the type of material.
For the sake of simplicity two types
The second level wing substructure
of materials have been used for the
(spars, stiffener) is made of isotropic
optimization procedure.One is
material and the skin panel is made
isotropic material that wing
of composite material. Various types
configuration is completely made up
of composite materials are also
of Aluminum or titanium (7075-T6,
considered. Then from the results
adv. Aluminum, Ti6A 14V). Another
we take the optimum design .
one is composite material that wing
configuration is Graphite/Epoxy or S-
RESULTS AND DISCUSSIONS
glass/Epoxy.The material properties
are shown in Table (4)
The structural analysis was achieved
(12) by using the CATIA V5 package in
Table 4: Material properties
Isotropic 7075-T6 Adv Ti6A14V order to obtain stress and
material Aluminium displacement distributions in the
E(N/m2) 71E+9 82.7E+9 110.32E+9 wings by using isotropic and
γ 0.33 0.318 0.29
Yield 470E+6 424E+6 598E+6
composite material to find the
stress(N/m2) optimum design. The work involves
2800 2906.4 4429 the investigation effects of changing
Ρ(Kg/m3) the skin shell thickness (0.001-
0.003m), the stiffener beam area
Composite H M Gr/Ep Gr/Ep Sg/Ep (44mm2-81mm2), adding ribs span
material wisely (3-5) adding stringers chord
E1(Gpa) 137.9 145 55
wisely (0-5),. The optimum design
E2=E3 (Gpa) 14.48 10 16
γ 12= γ 12= γ 12 0.21 0.25 0.28 parameter was achieved, where the
G12=G23=G13 5.86 4 .8 7 .6 skin thickness (0.001m) for rectangle
(Gpa) wing and the stiffener (3×5) . The
Ρ(Kg/m3) 1743.84 1580 1593.42 optimum isotropic material is (7075-
T6) . Also the optimum cross section
.
area of stiffener equal to (44 mm2)
8
9. Figures(12), (13) show the Von-
Table (5) shows the Von-Mises Mises stress distribution in case
stress and (Uy) results for rectangle of using isotropic and composite
wing when adding the ribs span materials for rectangular wing, in
wisely, increasing the number of ribs both cases the root Von-Mises
span wisely gives small reduction variation increases through out
effect on the Von-Mises stress and the (0.2-0.35) chord zone.
displacement distribution.
Table (5) Variation of von
misses stress due to variation
of no of ribs
No. of ribs 3 4 5
properties
V.M. stress 129 129 130
(M Pa)
UY(mm) 0.022 0.022 0.022
Stress ratio 29.57 29.57 30.00
% Fig11: Effect of von mises stress
Mass (Kg) 1.991 2.01 2.11 in changing the composite
Yield/weight 8.02 7.86 7.72 material
* 106( 1/m2 )
Weight/Area 38.27 38.98 39.75
( N/m2 )
Table (6) Result of isotropic
And composite materials
Material Aluminu Graphite/
properties m allay Epoxy
7075 -T6
V.M. stress 103 109
(MPa) Fig:12 Contours of Von misses
UY(m) 0.026 0.027 stress(Pa) distribution for
Stress ratio 21.91 23.19 isotropic w ing
%
Mass (Kg) 5.58 2.01
Yield/weight 5.96 7.76
*106(1/m2 )
Weight/Area 51.48 39.56
(N/m2 )
Table (6) shows comparison of the
results between the isotropic and
composite material. The goal
comparison is to emphasize that the
wing total mass is reduced and Von-
Mises stress increases when
composite material is used instead
of isotropic material.
9
10. Fig:17 Contours of Von misses isotropic and composite materials in
stress(Pa) distribution for design the parts of wing. Therefore
composite wing the major and general observations
and conclusions for this work can be
Figures shows the displacement listed below:
distribution of Isotropic and 1. From the Aerodynamic results it
Composite wings can be noticed that lift coefficient
increases with the increase of Mach
number at the same angle of
attacks due to the compressibility.
2. From the structural results it is
noticed that increasing the skin
thickness is considered as an
important factor in reducing the
stress and displacement levels
compared to the increase of, cross
section area of beam and number of
layers.
3. Increasing the ribs span wisely
does not change noticeable the Von-
Fig13 : Contours of Displacement Misses stress distribution.
distribution of isotropic wing 4. Also it is found that the mass
saves (34%) when composite
material is used instead of isotropic
material wing.
5. It is desirable to use composite
materials in design of the skin of a
wing
References.
1. N. G. R., Iyengar, and S. P.,
Joshi, “Optimal Design of
Antisymmetric Laminated
Composite Plates”, Journal of
Aircraft, Vol. 23, No. 5, May 1986.
Fig :14: Contours of
2. R. A., Confield, R. V., Granhi, and
Displacement distribution of
V. B., Venkayya, “Optimum Design
Composite wing
of Structures with Multiple
Constraints”, AIAA Journal, Vol. 26,
5.SUMMARY AND No. 1, January 1988, p. 78.
CONCLUSSION 3. An example of preliminary
The static analysis of the wing by design of Jet plane E .G
CATIA V5 package has been .Tulaburkara,A.Venkatraman,
presented and used to determine the V.Ganesh 2007
optimum design for the wing. This 4. Dan Doharty, Analytical
operation involves using the vortex modeling of aircraft wing using
lattice method to obtain the matlab, Matlab digest
aerodynamic result at (M=0.2) for 5.Thomas Chamberlain, Methods of
rectangle wing. This aerodynamic calculating aerodynamic load on
result (lift) is used to simulate the aircraft structures
wing loading on the wings during the 6Tobias F.W underlich .”
static analysis and also using Multidisciplinary w ing design and
10
11. optimization for transport aircraft” 10” Design analysis of UAV using
DRL Institute of aeronautics and flow NACA 0012 Airfoil’ Alimul Rajib1,
Technology Sep 2008. Bhuiyan Shameem Mahmud Ebna
7. Shawn E.Gano,John E.Renaud ” Hai1 and Md Abdus Salam2
Optimized unmanned aerial 1Department of Mechanical
vehicle with wing Morphing”, , Engineering, Military Institute of
AIAA Journal 2002“. Science and Technology, Dhaka-
8.”Aerodynamics of 3-D lifting 1216, Bangladesh..
surfaces using vortex lattice 11.A.Kao Mechanics of composite
method” NASA Journal Nov 1998. materials .
9. J. J., Bertin, and M. S., Smith, 12. P. J., Rol, D. N., Marris, and D.
“Aerdynamics for Engineering P.,Schrage,“Combined Aerodynamic
Students”, Department of and Structural Optimization of a
Aeronautical Engineering, University High-Speed Civil Transport Wing”,
of Sydney, 2000. School of Aerospace Engineering,
American Institute of Aeronautic and
Astronautic, Inc., 1995.
11