Biomedical Engineering Reference
In-Depth Information
Being q
I
an interest point discretizing the natural boundary curve, Q
Ct
ð
C
Þ
, or the
natural boundary surface, Q
Ct
ð
S
Þ
, as represented in Fig.
3.18
.
A closer look on Eq. (
3.40
) permit find some resemblances between the force
integrals and the Lagrange multipliers integrals. In fact the Lagrange multipliers
can be seen as ''smart forces'' forcing u
u ¼ 0. In order to obtain the discretized
formulation of the last two terms in Eq. (
3.40
), the Lagrange multipliers have to
be approximated using the respective nodal values and the shape functions for the
field nodes along the essential boundaries, sets X
Cu
ð
C
Þ
and X
Cu
ð
S
Þ
. Thus, since
the Lagrange multipliers are unknown functions of the coordinates variables,
the Lagrange multipliers approximation on an interest point q
I
, belonging to the
essential boundary (q
I
2
Q
Cu
ð
C
Þ
or q
I
2
Q
Cu
ð
S
Þ
), can be determined with,
dk
ð
q
I
Þ
¼
X
n
k
U
i
ð
q
I
Þ
dk
ð
x
i
Þ
¼
N
local
dk
ð
x
Þ
|{z }
½3n
k
1
;
x
2
C
u
ð
3
:
47
Þ
I
|{z}
½3
3n
k
i¼1
being n
k
the number of field nodes used for this interpolation. The vector of the
virtual Lagrange multipliers nodal values, considering only the nodes on the
essential boundary C
u
, can be represented for the classical three-dimensional
deformation theory as,
dk
ð
x
Þ
¼
dk
ð
x
1
Þ
x
dk
ð
x
1
Þ
y
dk
ð
x
1
Þ
z
dk
ð
x
n
k
Þ
x
dk
ð
x
n
k
Þ
y
dk
ð
x
n
k
Þ
z
ð
3
:
48
Þ
The local diagonal interpolation matrix N
local
I
is defined by,
ð
3
:
49
Þ
The interpolation function value U
i
ð
q
I
Þ
is not obtained using the MLS
approximation, alternatively U
i
ð
q
I
Þ
it is obtained using, for example, the Lagrange
interpolants.
This is the point where the Lagrange multipliers methodology becomes more
delicate. Each integration point q
I
2
Q
Cu
possesses two influence-domains: one
with n field nodes belonging to C
u
used to construct the MLS approximation shape
functions u
ð
q
I
Þ
, Eq. (
3.18
), and another with n
k
field nodes, also belonging to C
u
,
used to construct the interpolation functionU
ð
q
I
Þ
,