0% found this document useful (0 votes)

31 views

Finite Element 2D Wing

This document describes a multi-disciplinary design optimization of a 2D wing using computational mechanics. It presents the finite element discretization and solution of the governing equations to model incompressible, non-viscous flow over an airfoil. Streamlines, pressure distribution, lift and other aerodynamic quantities are computed by solving for uniform potential and circular flows and combining them using the Kutta condition. The example demonstrates applying these methods to optimize the shape of a 2D wing section.

Uploaded by

issamoune

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

0% found this document useful (0 votes)

31 views

Finite Element 2D Wing

Uploaded by

issamoune

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

You are on page 1/ 17

Muldisciplinary

design opmizaon
in computaonal mechanics

Applicaon case 2D wing
Piotr Breitkopf
26.11.2014

Angle of aack
Air velocity

Drag
Aerodynamic center

Chord
Reference line

Hypothesis

Simplifying assumpons
incompressible, non-viscous (non-rotaonal) ow

@v
@u @v
+
= 0,
@x @y
@x

@u
=0
@y

velocity eld determined by stream funcon

Wortmain airfoil FX60.126 - computing domain

1.5

0.5

@
,v =
u=
@y

@
@x

pressure given by Bernoulli's principle

-0.5

-1

-1.5

-2

2
P (x, y) = P1 + (Vwind V 2 (x, y))
2
far eld pressure P
1
, ow velocity Vwind

-2

-1.5

-1

-0.5

0.5

1.5

2.5

Streamlines

Curves that are instantaneously tangent to the velocity vector of the ow

0
1

Integral (weak formulaon)

dierenal formulaon

(x, y) = 0, (x, y) 2

residual weighted by test funcon

W =

(x, y) d = 0

aer integrang by parts and using Greens theorem

W =

(r , r )d

(r , n)d(@) = W W@ = 0
6

Finite element discrezaon

Integral terms

W =
W@ =

ne Z
X

(r , r )d =

(r , n)d(@) =
@

e=1 e
ne I
X
e=1

(r , r )de
(r , n)d(@e )

are computed over surface and boundary mesh

2
1.5
1
0.5
0
-0.5
-1
-1.5
-2

7
-2

-1.5

-1

-0.5

0.5

1.5

2.5

Finite element approximaon

(x1 , y1 )

l1
l2

(x2 , y2 )

A3
(x, y)

(x, y)

(x1 , y1 )

(x2 , y2 )
Ni (l) = li /L
2
X
(x, y) =
Ni (l) (xi , yi )
r (x, y) =

i=1
2
X
i=1

rNi (l) (xi , yi )

(x3 , y3 )

Ni (x, y) = Ai /A, i = 1...3

3
X
(x, y) =
Ni (x, y) (xi , yi )
r (x, y) =

i=1
3
X
i=1

rNi (x, y) (xi , yi )

Explicit form of shape funcons and derivaves for a triangle

1
A = ((x2 x1 )(y3 y1 ) (x3 x1 )(y2 y1 ))
2
1
((x2 x)(y3 y) (x3 x)(y2 y))
N1 =
2A
1
((x x1 )(y3 y1 ) (x3 x1 )(y y1 ))
N2 =
2A
1
((x2 x1 )(y y1 ) (x x1 )(y2 y1 ))
N3 =
2A

1
@N1
=
(y3
@x
2A
1
@N2
=
(y3
@x
2A
1
@N3
=
(y1
@x
2A

y2 )

1
@N1
=
(x3
@y
2A

x2 )

y1 )

1
@N2
=
(x1
@y
2A

x3 )

y2 )

1
@N3
=
(x2
@y
2A

x1 )
9

Surface term

r (x, y) =

3
X
i=1

= N1 (x, y)

rNi (x, y)

N2 (x, y)

L y2 y3
=
6 x3 x2

1 y2
Be =
2A x3

We =

y3
x2

y3
x1

N3 (x, y) 4

y3 y1
x1 x3

y1
x3

T
e Ke e , Ke

y1
x2

1
2
3

3
5

y1 y2 4
x2 x1

y2
x1

1
2
3

5 = Be

= AB T B

Boundary term

(r , n) = N1 (s) N2 (s) (ny un nx vn )

1
e = N1 (s) N2 (s)
2
W@e =

T
n
e Me (ny u

nx v), u
n =

u
1
u
2

v
, vn = 1
v2

L 2 1
Me =
6 1 2

Finite element linear system

W =

e
X

We =

e=1

(K F ) = 0
)K

boundary condions
Neumann at the external boundary intergrated in the RHS
Dirichlet

K11
T
K12
1

K12
K22

= K111 (F

F1
0

K12 2 )

@
,
(
@y

@
) = (
u, v)
@x

non-rotaonal ow

-2

-1

@
,
(
@y
2

@
) = (
u, v)
13
@x
3

Circular ow

=0
=1

Kua (Joukowski) condion

Kua condion: no circulaon around the trailing edge

resultant velocity follows reference line
y

0.25

0.2

0.15

0.1

0.05

-0.05

-0.1

-0.15

-0.2

-0.25
0.75

0.8

0.85

0.9

0.95

1.05

1.1

1.15

1.2

-0.25
0.75

0.8

0.85

0.9

0.95

1.05

0.25

1.1

1.15

1.2

1.25

-0.25

0.8

0.9

1.1

1.2

V = V1 + V2

weighted sum of uniform and circular ows

Li and li coecient

L = V ,

L
=
(r, n), cL = 1 2
@
2 V l
15

Summed ows
p

1.5

0.5

-0.5

-1

-1.5

-2

16
-2

-1.5

-1

-0.5

0.5

1.5

2.5