Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd
Upload
230 views

Triangular Element

The document discusses the formulation of a 3-noded linear triangular finite element for structural analysis. It describes: 1) Developing the FEM formulation in the global coordinate system for ease of integration rather than in the local coordinate system. 2) Deriving the shape functions and derivatives for the triangular element, which are linear and constant, respectively. 3) Assembling the element stiffness matrix and calculating the nodal load vector due to applied body loads and surface tractions on the element.

Uploaded by

katakgoreng
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
230 views

Triangular Element

The document discusses the formulation of a 3-noded linear triangular finite element for structural analysis. It describes: 1) Developing the FEM formulation in the global coordinate system for ease of integration rather than in the local coordinate system. 2) Deriving the shape functions and derivatives for the triangular element, which are linear and constant, respectively. 3) Assembling the element stiffness matrix and calculating the nodal load vector due to applied body loads and surface tractions on the element.

Uploaded by

katakgoreng
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 24

Triangular element

Assistant Prof. Hang Thu Vu


hang@civil.uwa.edu.au
3-noded linear triangular element
3-noded linear triangular element and 6-
noded quadratic triangular element





We develop the FEM formulation for
triangular element. Linear triangular is used
for an illustration.

Element in global and local coordinate
systems
The element is mapped between the global
coordinate system and the local coordinate
system
In local coordinate system
The 2D integration does not integrate
into the element triangle area
The triangular geometry is not symmetric
about local coordinate axis. It requires a
special technique to derive the Gauss points
and Gauss weights, which involves a
triangular coordinate system
Difficult to map among three coordinate
systems
Integration in local coordinate is hence
difficult

} }

1
1
1
1
q d d
In global coordinate system
It is observed that derivatives of shape
functions in the global coordinate systems
are constants with values depending on the
nodal coordinates.
The integration is easy to evaluate
FEM formulation for the triangular elements is
the special case where formulation at
element level is preferably performed in
global coordinate system
3-noded linear triangular element
Computation at element level conducted in
global coordinates. For each element






| |
6 2 2 3 6 3
)] , ( [ ] [ ) , (

= y x N L y x B
| | | | | || | | | | || | dxdy t y x B D y x B dV y x B D y x B K
Area
T
V
T
) , ( ) , ( ) , ( ) , (
6 6
} }
= =

Node sequence
Note: When formulating at the element level, the
element nodal sequence must be in the anti-clockwise
direction for every element in the domain for the finite
element computation to work.
Trial solution
The trial solution is a linear function of form




Where unknown coefficients is to be
determined.
To be FEM trial solution, must yield nodal
solutions at three nodal points, i.e. if substituting
x and y values at each nodal point into the function,
we will have

y a x a a y x u
2 1 0
) , ( + + =
| |

=
2
1
0
1 ) , (
a
a
a
y x y x u
i
a
) , ( y x u
i
u
Compute coefficients



In matrix form
3 3 2 3 1 0
2 2 2 2 1 0
1 1 2 1 1 0
u y a x a a
u y a x a a
u y a x a a
= + +
= + +
= + +
| | } { } {
1
1
1
3
2
1
2
1
0
3 3
2 2
1 1
u a A
u
u
u
a
a
a
y x
y x
y x
=

(
(
(

Compute coefficients
Hence


is area of the linear triangular element



The element node numbering must be in
sequence to make the determinant a positive
number




| |

(
(
(

=

3
2
1
1 2 3 1 2 3
2 1 1 3 3 2
1 2 2 1 3 1 1 3 2 3 3 2
2
1
0
1
2
1
} { } {
u
u
u
x x x x x x
y y y y y y
y x y x y x y x y x y x
A
a
a
a
u A a
area
area
A
3 3
2 2
1 1
1
1
1
2
1
y x
y x
y x
A
area
=
Trial solution

Substitute found coefficient vector {a} into
function for trial solution and rearrange the
terms to describe in the form


is shape function at node i. It has a
value of 1 at node i and a value of 0 at two
other nodes. Its formula is
) , ( y x u
3 3 2 2 1 1
) , ( ) , ( ) , ( ) , ( u y x N u y x N u y x N y x u + + =
) , ( y x N
i
Shape functions





We see that the shape function is in the form

Its derivatives are

y b x b b y x N
i i i
i
) (
2
) (
1
) (
0
) , ( + + =
( ) ( ) ( ) | | y x x x y y y x y x
A
y x N
area
2 3 3 2 2 3 3 2 1
2
1
) , ( + + =
( ) ( ) ( ) | | y x x x y y y x y x
A
y x N
area
3 1 1 3 3 1 1 3 2
2
1
) , ( + + =
( ) ( ) ( ) | | y x x x y y y x y x
A
y x N
area
1 2 2 1 1 2 2 1 3
2
1
) , ( + + =
) (
1
i
i
b
x
N
=
c
c
) (
2
i
i
b
y
N
=
c
c
For each element
Matrix [B(x,y)]

| |
(
(
(

=
(
(
(
(
(
(
(

c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
=
) 3 (
1
) 3 (
2
) 2 (
1
) 3 (
2
) 2 (
2
) 3 (
1
) 2 (
2
) 1 (
1
) 1 (
2
) 1 (
2
) 2 (
1
) 1 (
1
3 3 2
3 2
1
2 1 1
1
1 1
0
0 0
0 0
0
0
0 0
0 0
0
) , (
b b b
b b
b
b b b
b
b b
x
N
y
N
x
N
y
N
y
N
x
N
y
N
x
N
y
N
y
N
x
N
x
N
y x B
Element stiffness matrix

As terms in matrix B are constant values, the
matrix can be taken out of the integration sign.
The element stiffness matrix becomes



Next, we establish the element nodal load vector
resulted from the body load and the traction.
| | | | | | | | | | | | | | | | | | | |
area
T
Area
T
Area
T
tA B D B dA t B D B dxdy t y x B D y x B K
} }
= = = ) , ( ) , (
Nodal load vector due to body load
In 2D, the body load may be in kN/m2. The
integration is hence performed over the element
area.
If body load is constant over the element domain,
we have
{ } | |
}

=
)
`

=
Area
by
bx
by
bx
by
bx
y
x T
b
f
f
f
f
f
f
dA
q
q
y x N f
3
3
2
2
1
1
) , (
Nodal load vector due to body load
If the body load varies over the element
domain, we can use the shape functions to
interpolate between the nodal values.
It is directly given by isoparametric feature of
the element.

=
)
`

3
3
2
2
1
1
3
3
2
2
1
1
) , ( 0
0 ) , (
) , ( 0
0 ) , (
) , ( 0
0 ) , (
) , (
) , (
y
x
y
x
y
x
y
x
q
q
q
q
q
q
y x N
y x N
y x N
y x N
y x N
y x N
y x q
y x q
Nodal load vector due to body load
Hence





Gauss points and weights are not derived for
element in global coordinates. How do we do
the integration?
{ } | | | |

|
|
.
|

\
|
=
}
3
3
2
2
1
1
) , ( ) , (
y
x
y
x
y
x
Area
T
b
q
q
q
q
q
q
dA y x N y x N f
Nodal load vector due to body load
To overcome the problem with integration, we use the
element average body load multiplying with element area
and distribute equally between the nodes, i.e.

{ }

=
3
3
3
3
3
3
,
,
,
,
,
,
area ave y
area ave x
area ave y
area ave x
area ave y
area ave x
b
A q
A q
A q
A q
A q
A q
f
Nodal load vector due to traction
Next step, we consider the nodal load vector
due to load distributed on the element side
faces.
Nodal load vector due to traction
The traction distributes along the line.
We hence use one-dimensional shape
functions to interpolate the nodal values.
If the traction is constant along the element
edge
{ } | |
} }

=
)
`

=
)
`

=
s s
l
ty
tx
ty
tx
y
x
T
l
y
x T
t
f
f
f
f
ds
t
t
s N
s N
s N
s N
ds
t
t
s N f
2
2
1
1
2
2
1
1
) ( 0
0 ) (
) ( 0
0 ) (
) (
Nodal load vector due to traction
If the traction varies over the element side
face, we use the shape functions to
interpolate between the nodal values.
Having

=
)
`

2
2
1
1
2
2
1
1
) ( 0
0 ) (
) ( 0
0 ) (
) (
) (
y
x
y
x
y
x
t
t
t
t
s N
s N
s N
s N
s t
s t
Nodal load vector due to traction
Substitute the isoparametric expression and
multiply out the terms, the nodal load vector
due to traction becomes

{ } | | | |

(
(
(
(
(
(
(
(

|
|
.
|

\
|
=
}
2
2
1
1
2
2
1
1
3
0
6
0
0
3
0
6
6
0
3
0
0
6
0
3
) ( ) (
y
x
y
x
s s
s s
s s
s s
y
x
y
x
l
T
t
t
t
t
t
l l
l l
l l
l l
t
t
t
t
ds s N s N f
s
Compute matrix terms
I compute one entry of matrix
for illustration.
All other matrix entries are computed in a
similar manner.
Consider the entry located at column 1 and
row 1 intersection.
| | | |
|
|
.
|

\
|
}
s
l
T
ds s N s N ) ( ) (
| | | | | | | |
} }

=
1
1
1 1 1 1
) ( ) ( ) ( ) ( d J N N ds s N s N
T
l
T
s
Compute matrix terms
Where




Substitute these terms in the integral, we
have


2
1
) (
1


= N
2 2
1 2 s
l s s
J =

=
| | | |
3 3
1
4
1
2 2
1
2
) ( ) (
1
1
2 3
1
1
2
1 1
s s
l
s
T
l l
d
l
ds s N s N
s
=
|
.
|

\
|
+ =
|
.
|

\
|

=

} }

You might also like