Unformatted text preview: renner and Scott 3], Section 3.6). With p = 2, we have
n2 = 10 monomial terms and we can determine Lagrangian shape functions by placing
nodes at the four vertices and at the midpoints of the six edges (Figure 4.5.3). With
p = 3, we have n3 = 20 and we can specify shape functions by placing a node at each of
the four vertices, two nodes on each of the six edges, and one node on each of the four
faces (Figure 4.5.3). Higher degree polynomials also have nodes in the element's interior.
In general there is 1 node at each vertex, p ; 1 nodes on each edge, (p ; 1)(p ; 2)=2 nodes
on each face, and (p ; 1)(p ; 2)(p ; 3)=6 nodes in the interior.
4
1
0
1
0
1
0 8 1
0
1
0
1
0 1 1
0
1
0
1
0 1
0
1
0
1
0 1
0
1
09
1
0
1
0
1
0
1
0
5 1
010
1
0
1
0 1
0
1
0
1
0 1
0
1
06
1
0 1
03
1
0
1
0 2 1
0
11
00
1
0
11
00
1
0
11
00
1
0
11
00
11
00
11
00
11 11
00 00
11
00
1
0
11 11
00 00
11
00
11
00
1
0
1
0
1
0
1
0
1
0
1
0
1
0
11
00
1 11
0 00
1
0
1 11 11
0 00 00
11
00
1
0
11
00
1
0 Figure 4.5.3: Node placement for quadratic (left) and cubic (right) interpolants on tetrahedra.
Example 4.5.1. The quadratic shape function N12 associated with vertex Node 1 of a
tetrahedron (Figure 4.5.3, left) is required to vanish at all nodes but Node 1. The plane
1 = 0 passes through face A234 and, hence, Nodes 2, 3, 4, 6, 9, 10. Likewise, the plane
2
1 = 1=2 passes through Nodes 5, 7 (not shown), and 8. Thus, N1 must have the form N12 ( 1 2 3 4) = 1 ( 1 ; 1=2): Since N12 = 1 at Node 1 ( 1 = 1), we nd = 2 and N12 ( 1 2 3 4 ) = 2 1 ( 1 ; 1=2): Similarly, the shape function N52 associated with edge Node 5 (Figure 4.5.3, left) is
required to vanish on the planes 1 = 0 (Nodes 2, 3, 4, 6, 9, 10) and 2 = 0 (Nodes 1, 3,
4, 7, 8, 10) and have unit value at Node 5 ( 1 = 2 = 1=2). Thus, it must be N52 ( 1 2 3 4) = 4 1 2: 26 Finite Element Approximation
1,1,2
2,1,2 1
0
1
0
1
0 1
0
1
0
1
0 ζ 1
01,2,2
1
0
1
0
1
0
1
02,2,2
1
0
111
000η 111
000
1
0
111
000
1
0
111
000
ξ 000
1
0
111
1,1,1
1
0
1
0
1
0
2,1,1 1
0
1
0
1
0
1
0
1
0
1
0 1,2,1 2,2,1 1
0
1
0
11
00
1
0
1
0
1
0
1
0
11
00
1
0
1
0
11
00
11
00
1 11
0 00
1
0
1
0 1
0
1
0
1
0
1
0
1
0
1
0
1
0
1
0
1
0
1
0
1
0
1
0
1
0
1
0
1
0
1
0
1
0 1
0
1
0
11
00
1
0
11
00
1
0
1
0
1
0
1
0
11
00
1
0
11
00
1
0
1
0
1
0
1
0
11
00
1
0
11
00
1
0
1
0 1
0
1
0
11
00
1
0
1
0
1
0
1
0
11
00
1
0
1
0
1
0
11
00
1
0
1
0 Figure 4.5.4: Node placement for a trilinear (left) and triquadratic (right) polynomial
interpolants on a cube. 4.5.2 Lagrangian Shape Functions on Cubes
In order to construct a trilinear approximation on the canonical cube f
j;1
1g, we place eight nodes numbered (i j k), i j k = 1 2, at its vertices (Figure
4.5.4). The shape function associated with Node (i j k) is taken as
Ni j k(
) = Ni( )Nj ( )Nk ( )
(4.5.7a)
where Ni( ), i = 1 2, are the hat function (4.3.1d,e). The restriction of U to this element
has the form
222
XXX
U(
)=
ci j k Ni j k(
)
(4.5.7b)
i=1 j =1 k=1 Once again, ci j k = Ui j k = U ( i j k ).
The placement of nodes at the vertices produces bilinear shape functions on each...
View
Full
Document
This document was uploaded on 03/16/2014 for the course CSCI 6860 at Rensselaer Polytechnic Institute.
 Spring '14
 JosephE.Flaherty
 The Land

Click to edit the document details