Professional Documents
Culture Documents
Intro To The Finite Element Method (Lecture Notes) - Y. Liu (1998) WW
Intro To The Finite Element Method (Lecture Notes) - Y. Liu (1998) WW
Chapter 1. Introduction
Chapter 1. Introduction
I. Basic Concepts
The finite element method (FEM), or finite element analysis
(FEA), is based on the idea of building a complicated object with
simple blocks, or, dividing a complicated object into small and
manageable pieces. Application of this simple idea can be found
everywhere in everyday life as well as in engineering.
Examples:
Lego (kidsplay)
Buildings
Approximation of the area of a circle:
Element Si
i
R
S N = Si =
i =1
1 2
2
R N sin R 2 as N
N
2
Chapter 1. Introduction
Examples:
...
1998 Yijun Liu, University of Cincinnati
Chapter 1. Introduction
Chapter 1. Introduction
Example:
Chapter 1. Introduction
Computer Implementations
Preprocessing (build FE model, loads and constraints)
FEA solver (assemble and solve the system of equations)
Postprocessing (sort and display the results)
Chapter 1. Introduction
Chapter 1. Introduction
(1)
a n1 x1 + a n 2 x 2 + ...+ a nn x n = bn
(2)
where
a11
a
21
A = aij =
...
a
n1
[]
x1
x
2
x = {xi }=
:
xn
a12
a22
...
an2
... a1n
... a 2 n
... ...
... ann
b1
b
2
b = {bi }=
:
bn
(3)
Chapter 1. Introduction
v2
v3 ]
w1
w = w 2
w
3
with
cij = a ij + bij
D=A B
with
d ij = a ij bij
Scalar Multiplication
[ ]
A = a ij
Matrix Multiplication
For two matrices A (of size lm) and B (of size mn), the
product of AB is defined by
C = AB
Chapter 1. Introduction
Transpose of a Matrix
If A = [aij], then the transpose of A is
[ ]
A T = a ji
or
a ij = a ji
1
0
I=
...
0
0
0
0
1
...
...
Note that AI = A, Ix = x.
Determinant of a Matrix
The determinant of square matrix A is a scalar number
denoted by det A or |A|. For 22 and 33 matrices, their
determinants are given by
a b
det
= ad bc
c d
1998 Yijun Liu, University of Cincinnati
Chapter 1. Introduction
and
a11 a12
det a21 a22
a31 a32
a13
a23 = a11a22a33 + a12a23a31 + a21a32a13
a33
Singular Matrix
A square matrix A is singular if det A = 0, which indicates
problems in the systems (nonunique solutions, degeneracy, etc.)
Matrix Inversion
For a square and nonsingular matrix A (det A 0), its
inverse A-1 is constructed in such a way that
AA 1 = A 1 A = I
1
CT
det A
10
Chapter 1. Introduction
Examples:
1
1
d b
a b
(1)
=
(ad bc)
c a
c d
Checking,
1
1
d b a b 1 0
a b a b
=
c d c d (ad bc) c a c d = 0 1
1 1 0
3 2 1
3 2 1
1
2 2 1 = 2 2 1
(2) 1 2 1 =
(4 2 1)
0 1 2
1 1 1
1 1 1
Checking,
1 1 0 3 2 1 1 0 0
1 2 12 2 1 = 0 1 0
0 1 2
1 1 1
0 0 1
11
Chapter 1. Introduction
[ ]
A( t ) = a ij ( t )
da (t )
d
A(t ) = ij
dt
dt
and the integration by
12
Chapter 1. Introduction
13
Chapter 1. Introduction
important is simple.
i
fi
ui
j
uj
fj
Two nodes:
i, j
Nodal displacements:
Nodal forces:
F = k
with = u j ui
Linear
Nonlinear
F
k
14
Chapter 1. Introduction
and at node j,
f j = F = k ( u j ui ) = kui + ku j
In matrix form,
k
k
k ui f i
=
k
u j f j
or,
ku = f
where
k = (element) stiffness matrix
u = (element nodal) displacement vector
f = (element nodal) force vector
Note that k is symmetric. Is k singular or nonsingular? That is,
can we solve the equation? If not, why?
15
Chapter 1. Introduction
Spring System
k1
k2
u1, F1
u2, F2
u3, F3
For element 1,
k1
k
1
k1 u1 f 11
= 1
k1
u2 f 2
element 2,
k2
k
2
k 2 u2 f 12
= 2
k2
u3 f 2
at node 2,
F2 = f 21 + f 12
and node 3,
F3 = f 22
1998 Yijun Liu, University of Cincinnati
16
Chapter 1. Introduction
That is,
F1 = k1u1 k1u2
F2 = k1u1 + ( k1 + k 2 )u2 k 2 u3
F3 = k 2 u2 + k 2 u3
In matrix form,
k1
k
1
k1
k1 + k 2
k2
0 u1 F1
k 2 u2 = F2
k2
u3 F3
or
KU = F
K is the stiffness matrix (structure matrix) for the spring system.
k1
k
1
k1
k1
0
0
0
0 k
2
0 k 2
0u1 f 11
0u2 = f 21
0
u3 0
0 u1 0
k 2 u2 = f 12
2
k2
u3 f 2
17
Chapter 1. Introduction
k1
k1
k
1
k1 + k 2
k2
0 u1 f 11
k 2 u2 = f 21 + f 12
k2
u3 f 2
u1 = 0
and
F2 = F3 = P
we have
k1
k
1
k1
k1 + k 2
k2
0 0 F1
k 2 u2 = P
k2
u3 P
which reduces to
k1 + k 2
k
k 2 u2 P
=
k2
u3 P
and
F1 = k1u2
Unknowns are
u2
U=
u3
1998 Yijun Liu, University of Cincinnati
18
Chapter 1. Introduction
2 P / k1
u2
=
/
/
u
P
k
+
P
k
2
3
1
2
and the reaction force
F1 = 2 P
19
Chapter 1. Introduction
Example 1.1
k1
1
Given:
k2
2
k3
Find:
Solution:
(a) The element stiffness matrices are
100 100
k1 =
(N/mm)
100
100
(1)
200 200
(N/mm)
k2 =
200 200
(2)
100 100
k3 =
(N/mm)
100 100
(3)
20
Chapter 1. Introduction
u2
u3
u4
0
0
100
100
100 100 + 200
0
200
K=
200 + 100 100
200
0
0
0
100
100
or
0
0
100 100
100 300 200
0
K=
200 300 100
0
0
0
100 100
0
0 u1 F1
100 100
100 300 200
0
0
u2
=
200 300 100u3 P
0
0
0
100 100
F4
u4
(4)
21
300 200u2 0
200 300 u = P
Chapter 1. Introduction
(5)
u2 P / 250 2
=
= ( mm)
u
P
/
3
500
3
3
(6)
(c) From the 1st and 4th equations in (4), we get the reaction forces
j j
22
Chapter 1. Introduction
Example 1.2
4
k4
1
k1
2
4
F1
1
x
k2
F2
k3
Node i (1)
4
2
3
2
Node j (2)
2
3
5
1
23
u4
k1
k1 =
k1
u2
u2
k1
k1
u3
k3
k3 =
k 3
Chapter 1. Introduction
k2
k2 =
k 2
u3
k2
k2
u2
u5
k3
k3
k4
k4 =
k 4
u1
k4
k4
u1
k4
k
4
K= 0
0
u2
u3
u4
u5
k4
k1 + k 2 + k 4
k2
k1
k2
k1
k2 + k 3
0
0
k1
k3
0
0
k3
0
k3
24
25
ui
fi
A,E
fj
L
L
length
cross-sectional area
elastic modulus
u = u( x )
displacement
= ( x)
strain
= ( x)
stress
Strain-displacement relation:
=
du
dx
(1)
Stress-strain relation:
= E
(2)
26
x
x
u( x ) =
1 ui + u j
L
L
(3)
we have
u j ui
L
= E =
( = elongation)
E
L
(4)
(5)
We also have
=
F
A
(F = force in bar)
(6)
where k =
EA
= k
L
(7)
EA
is the stiffness of the bar.
L
27
k
k=
k
EA
EA
k L
L
=
k
EA EA
L
L
or
k=
EA 1 1
L
1 1
(8)
L 1 1 u j f j
(9)
28
N j () =
(10)
where
=
x
,
L
0 1
(11)
or
ui
N j = Nu
u j
u = Ni
(12)
du d
= Nu = Bu
dx dx
(13)
B=
i.e.,
d
N i ( )
dx
B = [ 1 / L 1 / L]
N j () =
d
N i ()
d
d
N j ( )
dx
(14)
29
= E = EBu
(15)
1
1
T dV =
2
2
V
(u
B T EBu )dV
1 T
T
= u (B EB)dV u
2
(16)
1
1
1
f i ui + f j u j = u T f
2
2
2
(17)
U =W
(18)
which gives
1 T
1
T
u (B EB)dV u = u T f
2
2
T
(B EB)dV u = f
or
1998 Yijun Liu, University of Cincinnati
30
ku = f
(19)
where
k=
(B
EB)dV
(20)
k=
1 / L
EA 1 1
E[ 1 / L 1 / L]Adx =
1
/
L
L
1 1
(21)
31
Example 2.1
1 2A,E
1
2 A,E
2
u2
2 EA 1 1
k1 =
L
1 1
Element 2,
u2
u3
EA 1 1
k2 =
L
1 1
0 1 1
u3 F3
1998 Yijun Liu, University of Cincinnati
32
u1 = u 3 = 0,
F2 = P
FE equation becomes,
2 2 0 0 F1
EA
2 3 1u2 = P
0 F3
0 1 1
Deleting the 1st row and column, and the 3rd row and column,
we obtain,
EA
3 {u2 }= {P}
[]
L
Thus,
u2 =
PL
3EA
and
u1
0
PL
u2 =
1
u 3EA 0
3
Stress in element 1 is
u1
1 = E1 = EB1u1 = E[ 1 / L 1 / L]
u2
=E
u2 u1 E PL
P
=
0
=
L
L 3EA 3 A
33
u3 u2 E
PL
P
= 0
=
L
L 3EA
3A
Notes:
In this case, the calculated stresses in elements 1 and 2
are exact within the linear theory for 1-D bar structures.
It will not help if we further divide element 1 or 2 into
smaller finite elements.
For tapered bars, averaged values of the cross-sectional
areas should be used for the elements.
We need to find the displacements first in order to find
the stresses, since we are using the displacement based
FEM.
34
Example 2.2
1
1
A,E
2
2
3
E = 2.0 10 4 N / mm2 ,
L = 150 mm, =1.2 mm
Solution:
We first check to see if or not the contact of the bar with
the wall on the right will occur. To do this, we imagine the wall
on the right is removed and calculate the displacement at the
right end,
PL (6.0 104 )(150)
0 =
=
= 18
. mm > = 12
. mm
4
EA (2.0 10 )(250)
0 1 1
u3 F3
1998 Yijun Liu, University of Cincinnati
35
F2 = P = 6.0 10 4 N
u1 = 0,
u3 = = 12
. mm
FE equation becomes,
1 1 0 0 F1
EA
1 2 1u2 = P
F3
0 1 1
u2
EA
[2 1] = {P}
L
that is,
EA
EA
[2]{u2 }= P +
L
L
1 PL
. mm
u2 =
+ = 15
2 EA
and
u1 0
. ( mm)
u2 = 15
u 12
3 .
36
37
Distributed Load
q
i
qL/2
qL/2
Wq =
qL
1
1
uqdx =
u( )q ( Ld ) =
u( )d
2
2
2
0
qL
=
2
qL
=
2
[N ( )
i
0
1
[1
0
ui
N j ( ) d
uj
ui
]d
uj
1 qL
=
2 2
qL ui
2 u j
qL / 2
uj
qL
/
2
1
ui
2
38
that is,
qL / 2
with f q =
qL
/
2
1 T
u fq
2
Wq =
(22)
(23)
which yields
ku = f + f q
(24)
+
f
qL
/
2
j
(25)
In an assembly of bars,
q
1
qL/2
1
qL
2
qL/2
3
39
ui
vi
ui
X
Local
Global
x, y
X, Y
ui' , vi'
ui , vi
1 dof at node
2 dofs at node
u
ui' = ui cos + vi sin = [l m] i
vi
u
vi' = ui sin + vi cos = [ m l ] i
vi
where l = cos , m = sin .
1998 Yijun Liu, University of Cincinnati
40
In matrix form,
m ui
ui' l
' =
vi m l vi
(26)
or,
~
u i' = Tu i
m
~ l
T=
m l
~
~
is orthogonal, that is, T 1 = T T .
(27)
m
ui' l
'
vi m l
' =
uj 0 0
v 'j 0 0
0 ui
0 0 v i
l
m u j
m l v j
(28)
~
T 0
with T =
~
0 T
(29)
or,
u ' = Tu
(30)
41
L 1 1 uj f j
EA 0
L 1
0
0 1 0 ui' f i '
0 0 0 vi' 0
' =
0 1 0 u j f j'
0 0 0 v 'j 0
or,
k 'u ' = f '
(31)
(32)
42
Explicit form,
ui
vi
uj
l2
lm
m2
EA lm
k=
L l 2 lm
2
lm m
vj
l 2 lm
lm m2
l2
lm
lm
m2
(33)
l = cos =
X j Xi
L
m = sin =
Yj Yi
L
(34)
Element Stress
ui'
1
= E = EB ' = E
L
uj
ui
1 l m 0 0 vi
L 0 0 l m u j
v j
That is,
ui
v
E
i
= [ l m l m]
L
uj
v j
1998 Yijun Liu, University of Cincinnati
(35)
43
Example 2.3
45o
2
P2
2
P1
45o
Solution:
k 1' =
EA 1 1
'
k
=
2
1
1
L
= 45o , l = m =
2
2
44
u1
v1
u2
v2
1 1 1
1
1
1
1
EA
k 1 = T1T k 1' T1 =
1
2 L 1 1 1
1 1 1
1
Element 2:
= 135o , l =
2
2
, m=
2
2
We have,
u2
v2
u3
v3
1 1 1 1
1 1
EA 1 1
T '
k 2 = T2 k 2 T2 =
1 1
2 L 1 1
1 1 1 1
v1
u2
v2
u3
v3
1 1 1 0
0 u1 F1 X
1
1
1 1 1 0
0 v1 F1Y
0 1 1 u2 F2 X
EA 1 1 2
2
1 1v2 F2Y
2 L 1 1 0
0
0 1 1
1 1u3 F3 X
0
0
1
1
1
v
F
3 3Y
45
u1 = v1 = u3 = v3 = 0,
F2 X = P1 , F2Y = P2
Condensed FE equation,
EA 2 0u2 P1
=
2 L 0 2v2 P2
Solving this, we obtain the displacement of node 2,
u2 L P1
=
v
EA
P2
2
Using formula (35), we calculate the stresses in the two bars,
0
E 2
L 0
2
=
1
1
1
1
1 =
( P1 + P2 )
[
] P
L 2
EA 1 2 A
P2
P1
E 2
L P2
2
2 =
[1 1 1 1] 0 = ( P1 P2 )
L 2
EA 2 A
46
Example 2.4
(Multipoint Constraint)
y
3
P
2
2
L
1 Y
3
1
45o
X
P = 1000 kN,
A = 6.0 10 4 m 2
L = 1 m,
E = 210 GPa ,
A = 6 2 10 4 m 2 for element 3.
Determine the displacements and reaction forces.
Solution:
We have an inclined roller at node 3, which needs special
attention in the FE solution. We first assemble the global FE
equation for the truss.
Element 1:
= 90 o , l = 0, m = 1
47
u1
v1 u2 v2
0 0
0
0 1
( N / m)
0 0
0 1
Element 2:
= 0o ,
l = 1, m = 0
u2
1
(210 10 9 )(6.0 10 4 ) 0
k2 =
1
1
0
v 2 u3 v 3
0 1 0
0 0 0
( N / m)
0 1 0
0 0 0
Element 3:
= 45o , l =
1
1
, m=
2
2
u1
v1
u3
v3
k3 =
0.5
0.5 0.5 0.5
2
0.5 0.5 0.5
0.5
( N / m)
48
1
1 0
0 u2 F2 X
5
1260 10
v = F
1
0
0
2 2Y
.
15
0.5 u3 F3 X
Sym.
.
0
5
v3 F3Y
u1 = v1 = v2 = 0, and v3' = 0,
F2 X = P , F3 x ' = 0.
From the transformation relation and the BC, we have
2
v3' =
2
2 u3
2
( u3 + v3 ) = 0,
=
2 v3 2
that is,
u3 v3 = 0
This is a multipoint constraint (MPC).
Similarly, we have a relation for the force at node 3,
2
F3 x ' =
2
2 F3 X
2
=
( F3 X + F3Y ) = 0,
F
2
2
3Y
that is,
F3 X + F3Y = 0
1998 Yijun Liu, University of Cincinnati
49
0 0.5 0.5
v3 F3Y
Further, from the MPC and the force relation at node 3, the
equation becomes,
1 1 0 u2 P
. 0.5u3 = F3 X
1260 105 1 15
0 0.5 0.5
u3 F3 X
which is
P
1 1
u
1260 105 1 2 2 = F3 X
u3
F
1
3X
0
1 1u2 P
1260 105
=
1 3 u3 0
Solving this, we obtain the displacements,
50
u 2
1
3 P 0.01191
=
=
( m)
5
u
P
0
.
003968
3 2520 10
1Y
2
5
0
0 u3 = 0.0 ( kN )
F2Y = 1260 10 0
v 500
1 15
F
.
.
0
5
3X
3
0.5
0.5
F3Y
500
Aj u j = 0
51
3-D Case
y
Y
i
X
Z
Local
Global
x, y, z
X, Y, Z
ui , vi , wi
1 dof at node
3 dofs at node
52
y
vi, Fi
vj, Fj
j
i
i, Mi
E,I
L
j, Mj
length
elastic modulus
v = v( x )
dv
dx
F = F ( x)
shear force
M = M ( x)
d 2v
EI 2 = M ( x)
dx
(36)
My
I
(37)
53
Direct Method
Using the results from elementary beam theory to compute
each column of the stiffness matrix.
vi
vj
6 L 12 6 L vi Fi
12
M
2
i
6 L 2 L2
EI 6 L 4 L
=
L3 12 6 L 12 6 L v j F j
6 L 2 L2 6 L 4 L2 M
j j
(38)
54
Formal Approach
Apply the formula,
L
k=
B T EIBdx
(39)
N 1 ( x ) = 1 3x 2 / L2 + 2 x 3 / L3
N 2 ( x) = x 2 x 2 / L + x 3 / L2
(40)
N 3 ( x ) = 3x / L 2 x / L
2
N 4 ( x) = x 2 / L + x 3 / L2
Then, we can represent the deflection as,
v( x ) = Nu
= [N 1 ( x )
N 2 ( x)
N 3 ( x)
vi
i
N 4 ( x )]
v j
(41)
55
d 2v d 2
= 2 Nu = Bu
2
dx
dx
(42)
d2
B = 2 N = N 1" ( x ) N 2" ( x) N 3" ( x) N 4" ( x )
dx
6 12 x
4 6 x 6 12 x
2 6x
=
+ 2
2
3
2
2
3
L
L
L
L
L
L
L
L
(43)
1
1
T dV =
2
2
V
1
=
2
My 1 My
dAdx
I E I
0 A
d 2 v
1 d 2 v
T 1
M
Mdx =
EI 2 dx
dx
EI
2 dx 2
1
(Bu)T EI (Bu)dx
2
0
1 T
T
= u B EIBdx
u
k=
B T EIBdx
56
ui
vi
EA
0
L
12 EI
0
L3
6 EI
0
2
L
k=
EA
0
L
12 EI
0
L3
6 EI
0
L2
0
6 EI
L2
4 EI
L
0
6 EI
L2
2 EI
L
uj
vj
EA
L
0
0
EA
L
0
0
0
12 EI
L3
6 EI
2
L
0
12 EI
L3
6 EI
2
L
j
0
6 EI
L2
2 EI
L
0
6 EI
2
L
4 EI
57
Example 2.5
Y
P
M
1
1
E,I
2
3
Given:
Find:
v2
6 L 12 6 L
12
2
2
EI 6 L 4 L 6 L 2 L
k1 = 3
L 12 6 L 12 6 L
6 L 2 L2 6 L 4 L2
v2
v3
6 L 12 6 L
12
2
2
EI 6 L 4 L 6 L 2 L
k2 = 3
L 12 6 L 12 6 L
6 L 2 L2 6 L 4 L2
58
v2
v3
6 L 12 6 L
0
0 v1 F1Y
12
6 L 4 L2 6 L 2 L2
0
0 1 M 1
0
12 6 L v2 F2Y
EI 12 6 L 24
=
2
2
2
L
L
L
L
L
6
2
0
8
6
2
L3
2 M 2
0
0
12 6 L 12 6 L v3 F3Y
2
2
L
L
L
L
0
0
6
2
6
4
3 M 3
Reduced FE equation,
EI
L3
24 0 v2 P
0 8 L2 = M
24
EI
3
M
2
59
F1Y
12 6 L
2
M
1 EI 6 L 2 L v2
= 3
=
F
L
12
6
2
3Y
6L
2 L2
M3
2 P + 3 M / L
1 PL + M
4 2 P 3 M / L
PL + M
Stresses in the beam at the two ends can be calculated using the
formula,
= x =
My
I
d 2v
EI 2 = M ( x)
dx
and
dM
= V (V - shear force in the beam)
dx
dV
= q (q - distributed load on the beam)
dx
Thus,
d 4v
EI 4 = q( x )
dx
If q(x)=0, then exact solution for the deflection v is a cubic
function of x, which is what described by our shape functions.
60
qL/2
qL/2
qL2/12
qL2/12
j
L
qL/2
qL
qL2/12
L
61
Example 2.6
E,I
L
Given:
Find:
f
m
E,I
L
where
f = pL / 2,
m = pL2 / 12
62
6 L 12 6 L v1 F1Y
12
2
M
6 L 2 L2
1
EI 6 L 4 L
1
=
L3 12 6 L 12 6 L v2 F2 Y
6 L 2 L2 6 L 4 L2 M
2 2
M2 = m
EI
L3
12 6 L v2 f
6 L 4 L2 = m
=
3
EI
6
/
Lf
m
pL
EI
3
6
6
2
(A)
6
EI
3
Lf
pL
/
4
EI
2
(B)
63
The errors in (B) will decrease if more elements are used. The
equivalent moment m is often ignored in the FEM applications.
The FE solutions still converge as more elements are applied.
From the FE equation, we can calculate the reaction force
and moment as,
F1Y L3 12 6 L v2 pL / 2
= 2
=
2
M
6
L
2
L
5
pL
12
EI
1
where the result in (A) is used. This force vector gives the total
effective nodal forces which include the equivalent nodal forces
for the distributed lateral load p given by,
pL / 2
2
pL / 12
The correct reaction forces can be obtained as follows,
F1Y pL / 2
= 2
/
M
pL
5
12
pL / 2 pL
= 2
2
/
pL
12
pL / 2
64
Example 2.7
Y
P
E,I
1
2
1
L
Given:
2
3
L
Find:
Solution:
The beam has a roller (or hinge) support at node 2 and a
spring support at node 3. We use two beam elements and one
spring element to solve this problem.
The spring stiffness matrix is given by,
v3
k
ks =
k
v4
k
k
65
v1
v2
12 6 L 12 6 L
4 L2 6 L 2 L2
24
0
EI
2
8
L
3
L
Symmetry
v3
v4
0 v1 F1Y
0 1 M 1
12
6L
0 v2 F2Y
2
6L 2L
0 2 = M 2
12 + k ' 6 L k 'v3 F3Y
4 L2
0 3 M 3
k '
v4
F4Y
0
0
0
0
in which
L3
k
k '=
EI
is used to simply the notation.
We now apply the boundary conditions,
v1 = 1 = v2 = v4 = 0,
M 2 = M 3 = 0,
F3Y = P
8 L2
6 L 2 L2 2 0
EI
6
L
12
+
k
6
L
'
v3 = P
L3 2
2
2
L
6
L
4
L
3 0
Solving this equation, we obtain the deflection and rotations at
node 2 and node 3,
66
2
3
2
PL
v3 =
7 L
EI (12 + 7 k ')
3
9
v
0
01744
=
.
m
3
0.007475 rad
3
F1Y
M
1
=
F2Y
F4Y
69.78 kN
69.78 kN m
116.2 kN
3.488 kN
1
69.78 kNm
50 kN
2
116.2 kN
3
3.488 kN
67
for stresses,
x , y , z , xy , yz , zx
for strains.
and
y
yz
xy
zx
z
x
z
Under contain conditions, the state of stresses and strains
can be simplified. A general 3-D structure analysis can,
therefore, be reduced to a 2-D analysis.
75
(z 0)
(1)
Plane strain:
z = yz = zx = 0
(z 0)
(2)
76
x 1 / E
y = / E
0
xy
/E
1/ E
0
0 x
0 y +
1 / G
xy
x 0
y0
xy 0
(3)
or,
= E 1 + 0
G=
E
2(1 + )
(4)
=
y
2
1
xy
0
1
x
1
0
0 0 (1 ) / 2
xy
x 0
y 0
xy 0
(5)
or,
= E + 0
where 0 = E0 is the initial stress.
1998 Yijun Liu, University of Cincinnati
77
The above relations are valid for plane stress case. For
plane strain case, we need to replace the material constants in
the above equations in the following fashion,
E
1 2
1
G G
E
(6)
0
1
x
E
1
0
y =
y
1
1
2
(
+
)(
)
0
(1 2) / 2
0
xy
xy
x 0
y 0
xy 0
x 0 T
T
y0
xy 0
(7)
78
x =
u
v
u v
, y = , xy =
+
x
y
y x
In matrix form,
0
x / x
u
/ y , or = Du
y = 0
/ y / x v
xy
(8)
Equilibrium Equations
In elasticity theory, the stresses in the structure must satisfy
the following equilibrium equations,
x
+
x
xy
+
x
xy
y
y
y
+ fx = 0
(9)
+ fy = 0
where fx and fy are body forces (such as gravity forces) per unit
volume. In FEM, these equilibrium conditions are satisfied in
an approximate sense.
79
Boundary Conditions
ty
p
tx
y
St
Su
x
u = u, v = v,
tx = tx , ty = ty ,
on S u
on S t
(10)
80
Example 3.1
A plate is supported and loaded with distributed force p as
shown in the figure. The material constants are E and .
y
p
x
The exact solution for this simple problem can be found
easily as follows,
Displacement:
p
y
E
p
x,
E
v=
p
,
E
y =
x = p,
y = 0,
u=
Strain:
x =
p
,
E
xy = 0
Stress:
xy = 0
81
u N 1
=
v 0
0
N1
N2
0
0
N2
u1
v
1
L
2
L
v
2
or
u = Nd (11)
or
= Bd
(12)
82
1
1
T dV =
2
2
V
1
2
( +
x
+ xy xy )dV
(E) dV
T
1 T
E dV
2
V
1
= d T B T EB dV d
2
V
1
= d T kd
2
From this, we obtain the general formula for the element
stiffness matrix,
k=
B T EB dV
(13)
Note that unlike the 1-D cases, E here is a matrix which is given
by the stress-strain relation (e.g., Eq.(5) for plane stress).
The stiffness matrix k defined by (13) is symmetric since E
is symmetric. Also note that given the material property, the
behavior of k depends on the B matrix only, which in turn on
the shape functions. Thus, the quality of finite elements in
representing the behavior of a structure is entirely determined by
the choice of shape functions.
Most commonly employed 2-D elements are linear or
quadratic triangles and quadrilaterals.
83
u3
y
v
v1
1
(x1, y1)
v2
u
(x, y)
u2
(x2, y2)
u1
x
u = b1 + b2 x + b3 y , v = b4 + b5 x + b6 y
(14)
xy = b3 + b5
(15)
84
u1 = b1 + b2 x1 + b3 y1
u2 = b1 + b2 x2 + b3 y 2
M
v3 = b4 + b5 x3 + b6 y3
Solving these equations, we can find the coefficients b1, b2, ...,
and b6 in terms of nodal displacements and coordinates.
Substituting these coefficients into (14) and rearranging the
terms, we obtain,
u N1
=
v 0
N2
N3
N1
N2
u1
v
1
0 u2
N3
v2
u3
v3
(16)
(17)
and
1998 Yijun Liu, University of Cincinnati
85
1 x1
1
A = det 1 x 2
2
1 x3
y1
y2
y3
(18)
x
y 23
1
0
Bd
=
=
y
2A
x32
xy
0
x32
y31
0
0
x13
y12
0
y23
x13
y31
x21
u1
v
0 1
u
x 21 2 (19)
v2
y12
u3
v3
k=
B T EB dV = tA( B T EB )
(20)
86
=0
=b
=a
=1
=1
(a, b)
2
N1 = , N 2 = , N 3 = 1
(21)
Notice that,
N1 + N 2 + N 3 = 1
(22)
1,
Ni =
0,
at node i;
at the other nodes
(23)
and varies linearly within the element. The plot for shape
function N1 is shown in the following figure. N2 and N3 have
similar features.
87
=0
3
N1
=1
1
2
1
(24)
or,
x = x13 + x 23 + x3
y = y13 + y23 + y3
(25)
u = x
y u
u
=
J
u
u
y
(26)
88
x
J = 13
x23
y13
,
y 23
J 1 =
1 y 23
2A
x 23
y13
x13
(27)
where det J = x13 y23 x 23 y13 = 2 A has been used (A is the area of
the triangular element. Prove this!).
From (26), (27), (16) and (21) we have,
x
1 y23
u =
x
A
2
23
u
y13
x13
1 y23
=
2A
x23
(28)
y13 u1 u3
x13
u2 u3
Similarly,
v
x
1 y 23
v =
x
2
A
23
y13 v1 v3
x13
v 2 v3
(29)
2A
x32
0
x32
y31
0
0
x13
y12
0
y23
x13
y31
x 21
0
x 21
y12
(30)
89
90
v5
5
v6
y
u6
u5
v2
v1
1
u1
u4
u2
v4
x
(31)
x = b2 + 2b4 x + b5 y
y = b9 + b11 x + 2b12 y
(32)
91
N 1 = ( 2 1)
N 2 = ( 2 1)
N 3 = ( 2 1)
(33)
N 4 = 4
N 5 = 4
N 6 = 4
in which = 1 . Each of these six shape functions
represents a quadratic form on the element as shown in the
figure.
=0
3
=1/2
6
=1
1
N1
4
u = N i ui ,
i =1
v = N i vi
(34)
i =1
92
v3
v4
=1
u3
v2
v1
1
y
u4
u2
u1
=1
= 1
=1
Note that
1
N 2 = (1 + )(1 )
4
1
N 4 = (1 )(1 + )
4
(35)
i=1
u = N i ui ,
i =1
v = N i vi
(36)
i =1
93
=1
6
8
5
1
y
=1
= 1
=1
1
N 1 = (1 )( 1)( + + 1)
4
1
N 2 = (1 + )( 1)( + 1)
4
1
N 3 = (1 + )(1 + )( + 1)
4
1
N 4 = ( 1)( + 1)( + 1)
4
(37)
94
1
N 5 = (1
2
1
N 6 = (1 +
2
1
N 7 = (1 +
2
1
N 8 = (1
2
)(1 2 )
)(1 2 )
)(1 2 )
)(1 2 )
8
Again, we have
i=1
u = N i ui ,
i =1
v = N i vi
(38)
i =1
95
Example 3.2
A square plate with a hole at the center and under pressure
in one direction.
y
p
A
FE Analysis:
From the knowledge of stress concentrations, we should
expect the maximum stresses occur at points A and B on the
edge of the hole. Value of this stress should be around 3p (=
300 psi) which is the exact solution for an infinitely large plate
with a hole.
96
No. Elem.
DOF
Max. (psi)
T6
966
4056
310.1
Q4
493
1082
286.0
Q8
493
3150
327.1
...
...
...
...
Q8
2727
16,826
322.3
Discussions:
Check the deformed shape of the plate
Check convergence (use a finer mesh, if possible)
Less elements (~ 100) should be enough to achieve the
same accuracy with a better or smarter mesh
Well redo this example in next chapter employing the
symmetry conditions.
1998 Yijun Liu, University of Cincinnati
97
98
Transformation of Loads
Concentrated load (point forces), surface traction (pressure
loads) and body force (weight) are the main types of loads
applied to a structure. Both traction and body forces need to be
converted to nodal forces in the FEA, since they cannot be
applied to the FE model directly. The conversions of these
loads are based on the same idea (the equivalent-work concept)
which we have used for the cases of bar and beam elements.
qB
q
fB
qA
fA
s
B
L
B
A
Traction on a Q4 element
Suppose, for example, we have a linearly varying traction q
on a Q4 element edge, as shown in the figure. The traction is
normal to the boundary. Using the local (tangential) coordinate
s, we can write the work done by the traction q as,
L
Wq = t un ( s )q( s )ds
0
99
un ( s ) = (1 s / L )unA + ( s / L )unB
The traction q(s), which is also linear, is given in a similar way,
q( s ) = (1 s / L )q A + ( s / L )qB
Thus, we have,
Wq = t [unA
0
L
= [unA
= [unA
1 s / L
q A
1 s / L s / L] ds
unB ]
[
s
/
L
q B
(1 s / L ) 2
( s / L )(1 s / L ) q A
unB ]t
ds q
2
(s / L)
0 ( s / L )(1 s / L )
B
tL 2 1q A
unB ]
6 1 2
q B
L
f A tL 2 1q A
=
f B 6 1 2q B
Note, for constant q, we have,
f A qtL 1
=
f B 2 1
For quadratic elements (either triangular or quadrilateral),
the traction is converted to forces at three nodes along the edge,
instead of two nodes.
Traction tangent to the boundary, as well as body forces,
are converted to nodal forces in a similar way.
1998 Yijun Liu, University of Cincinnati
100
Stress Calculation
The stress in an element is determined by the following
relation,
x
x
=
E
y
y = EBd
xy
xy
(39)
e Y
where e is the von Mises stress and Y the yield stress of the
material. This is a generalization of the 1-D (experimental)
result to 2-D and 3-D situations.
101
1
(1 2 ) 2 + (2 3 ) 2 + (3 1 ) 2
2
(40)
2 =
P
x + y
2
x + y
2
x y
2
+ xy
x y
2
+ xy
2
(41)
(42)
Averaged Stresses:
Stresses are usually averaged at nodes in FEA software
packages to provide more accurate stress values. This option
should be turned off at nodes between two materials or other
geometry discontinuity locations where stress discontinuity does
exist.
102
Discussions
103
Readings:
Sections 3.1-3.5 and 3.8-3.12 of Cooks book.
104
I. Symmetry
A structure possesses symmetry if its components are
arranged in a periodic or reflective manner.
Types of Symmetry:
Reflective (mirror, bilateral) symmetry
Rotational (cyclic) symmetry
Axisymmetry
Translational symmetry
...
Examples:
105
Examples:
Cautions:
In vibration and buckling analyses, symmetry concepts, in
general, should not be used in FE solutions (works fine in
modeling), since symmetric structures often have antisymmetric
vibration or buckling modes.
106
Physical Meaning:
A finite element model of a portion of structure.
Mathematical Meaning:
Boundary matrices which are load and stiffness matrices
reduced (condensed) from the interior points to the exterior or
boundary points.
107
Disadvantages:
Increased overhead for file management
Matrix condensation for dynamic problems introduce
new approximations
...
108
Iterative Methods:
Solution time is unknown beforehand
Reduced storage requirement
Suitable for large problems, or bulky structures (large
bandwidth, converge faster)
Need solving again for different load cases
109
x 3 3
0 3 3
or Ax = b .
Forward Elimination:
Form
(1) 8 2 0
(2) 2 4 3
(3)
0 3 3
2
1;
(3)
3
0 3
(2) +
2
2;
14
(3) (3):
3
(1) 8 2
0
(2) 0 14 12
(3)
2
0 0
2
2;
12
Back Substitution:
x 3 = 12 / 2 = 6
x 2 = ( 2 + 12 x 3 ) / 14 = 5
x1 = (2 + 2 x 2 ) / 8 = 1.5
or
1.5
x = 5 .
6
110
(A is symmetric)
or
aij x j = bi ,
i = 1, 2, ..., N .
j =1
( k + 1)
bi
1
=
a ii
i 1
a ij x j
( k + 1)
j =1
a ij x j
for i = 1, 2, ..., N .
(k )
j=i+ 1
In vector form,
1
x ( k + 1) = A D b A L x ( k + 1) A L x ( k ) ,
T
where
x ( k + 1) x ( k )
x(k )
111
u = N u
=1
Stiffening Effect:
FE Model is stiffer than the real structure.
In general, displacement results are smaller in
magnitudes than the exact values.
112
Exact Solution
FEM Solutions
No. of DOFs
113
V. Numerical Error
Error Mistakes in FEM (modeling or solution).
Types of Error:
Modeling Error (beam, plate theories)
Discretization Error (finite, piecewise )
Numerical Error ( in solving FE equations)
u2
P
1
k1
k2
FE Equations:
k1
k
1
and
k1 u1 P
=
k1 + k 2
u 2 0
Det K = k1 k 2 .
114
P
k1
u2 = u1
u2
u2 =
k1
u1
k1 + k 2
u1
P/k1
u2 = u1
u2
P
k1
u2 =
k1
u1
k1 + k 2
P/k1
u1
115
Types of Refinement:
h-refinement:
p-refinement:
r-refinement:
Examples:
116
U = Ui ,
i =1
U = U i* ,
*
i =1
M
U E = U Ei ,
i =1
Ui =
1 T 1
2 s E s dV ;
V
i
U* =
i
1 *T 1 *
2 s E s dV ;
V
i
U Ei =
1 T 1
2 s E E s E dV ;
V
i
117
= E .
U + U E
(0 1)
Examples:
118
Applications:
Shear walls
Floor panels
Shelves
119
y
y
My
q(x,y)
Mxy
Qy
t
Mx
Qx
Mid surface
Mxy
Stresses:
z
yz
y
xy
xz
xy
120
t/2
( N m / m)
(1)
t/2
( N m / m)
(2)
( N m / m)
(3)
( N / m)
(4)
( N / m)
(5)
M y = t / 2 y zdz ,
Twisting moment (per unit length):
t/2
M xy = t / 2 xy zdz ,
Shear Forces (per unit length):
t/2
Q x = t / 2 xz dz ,
t/2
Q y = t / 2 yz dz ,
Maximum bending stresses:
( x ) max
6M
= 2x,
t
( y ) max =
6M y
t2
(6)
121
xz = yz = 0 .
Displacement:
z
w
x
w
x
w = w( x, y ),
w
u = z
,
x
w
v = z
.
y
( deflection)
(7)
122
Strains:
2w
x = z 2 ,
x
2w
y = z 2 ,
y
xy
(8)
2w
.
= 2 z
x y
=
y
2
1
xy
0
1
x
1
,
0
y
0 0 (1 ) / 2 xy
or,
2w
2
1
0
x
x2
E
w .
z
1
0
=
y
2
y 2
1
0 0 (1 ) 2 w
xy
xy
(9)
123
Governing Equation:
D 4 w = q ( x , y ) ,
(10)
where
4
4
4
( 4 + 2 2 2 + 4 ),
x
x y y
4
Et 3
D=
(the bending rigidity of the plate),
2
12(1 )
q = lateral distributed load (force/area).
Compare the 1-D equation for straight beam:
d 4w
EI
= q( x ) .
4
dx
Note: Equation (10) represents the equilibrium condition
in the z-direction. To see this, refer to the previous figure
showing all the forces on a plate element. Summing the forces
in the z-direction, we have,
Q x y + Q y x + qxy = 0,
which yields,
Q x Q y
+ q( x, y ) = 0 .
+
y
x
Substituting the following relations into the above equation, we
obtain Eq. (10).
1999 Yijun Liu, University of Cincinnati
124
M x M xy
,
+
y
x
Qy =
2w
2w
M x = D 2 + 2 ,
y
x
M xy
x
M y
y
2w
2w
M y = D 2 + 2 .
x
y
Boundary Conditions:
Clamped:
w = 0,
w
= 0;
n
(11)
Simply supported:
w = 0,
M n = 0;
(12)
Free:
Q n = 0,
M n = 0;
(13)
boundary
1999 Yijun Liu, University of Cincinnati
125
Examples:
A square plate with four edges clamped or hinged, and
under a uniform load q or a concentrated force P at the center C.
C
L
x
y
L
Simply supported
0.00126 qL4/D
0.00406 qL4/D
0.00560 PL2/D
0.0116 PL2/D
in which: D= Et3/(12(1-v2)).
These values can be used to verify the FEA solutions.
126
xz 0,
yz 0 .
This means that a line which is normal to the mid surface before
the deformation will not be so after the deformation.
w
y
x
w
w
x
127
New relations:
u = z y ,
v = z x ;
(14)
y
,
x
y = z x ,
y
xy = z ( y x ),
y
x
w
xz =
+ y ,
x
w
yz =
x.
y
x = z
(15)
xz =
w
+ y = 0,
x
yz =
w
x = 0,
y
128
3
x
2
1
w w
w1 , ,
x 1 y 1
w w
w2 , ,
x 2 y 2
w,
w w
,
.
y y
w
w
w( x, y ) = N i wi + N xi ( ) i + N yi ( ) i ,
x
y
i =1
4
129
8-Node Quadrilateral
z
y
3
x
2
w, x and y.
On each element:
n
w( x, y ) = N i wi ,
i =1
n
x ( x, y ) = N i xi ,
i =1
n
y ( x, y ) = N i yi .
i =1
130
3
6
4
1
w w
, , x , y ;
x y
w w
At each node: w, x =
.
, y =
x y
Total DOF = 9 (DKT Element).
Incompatible w(x,y); convergence is faster (w is cubic
along each edge) and it is efficient.
131
Test Problem:
z
P
C
y
L
wc ( PL2/D)
0.00593
0.00598
0.00574
0.00565
:
0.00560
132
Example:
Sea shell, egg shell (the wonder of the nature);
Containers, pipes, tanks;
Car bodies;
Roofs, buildings (the Superdome), etc.
Forces in shells:
Membrane forces + Bending Moments
(cf. plates: bending only)
133
internal forces:
p
membrane stresses
dominate
Shell Theory:
Thin shell theory
Thick shell theory
Shell theories are the most complicated ones to formulate
and analyze in mechanics (Russians contributions).
Engineering Craftsmanship
Demand strong analytical skill
134
Shell Elements:
v
u
Q4 or Q8 shell element.
135
v
i
136
Test Cases:
q
L/2
L/2
F
A
R
80o
F
Roof
Pinched Cylinder
F2
F
b
A
A
L
F1
F
Pinched Hemisphere
Difficulties in Application:
Non uniform thickness (turbo blades, vessels with
stiffeners, thin layered structures, etc.);
137
x
z
y,v
yx
yz
xy
zy
x
z
zx
xz
x, u
z, w
1999 Yijun Liu, University of Cincinnati
138
y
z
= { }=
xy
yz
zx
or
[ ]
(1)
ij
Strains:
x
y
= { }= z
xy
yz
zx
or
[ ]
(2)
ij
Stress-strain relation:
x
y
z
xy
yz
zx
v
v
0
1 v
v
v
1 v
0
v
1 v
0
v
1 2v
0
E
0
0
=
2
v
v
(
1
)(
1
2
)
+
0
0
0
0
0
0
0
or
1999 Yijun Liu, University of Cincinnati
= E
0
0
0
0
1 2v
2
0
0
0 x
0 y
0 z
xy
0 yz
1 2v zx
2
(3)
139
Displacement:
u( x, y , z ) u1
u = v ( x, y , z ) = u2
w( x, y , z ) u
( 4)
Strain-Displacement Relation:
u
v
w
, y =
, z =
,
x
y
z
w v
u w
v u
, yz =
, xz =
+
= +
+
y z
z x
x y
x =
xy
(5)
or
1 ui u j
ij =
+
,
2 x j xi
(i, j = 1, 2, 3)
or simply,
ij =
1
(ui , j + u j ,i )
2
( tensor notation)
140
Equilibrium Equations:
x xy xz
+
+ fx = 0 ,
+
y
z
x
yx y yz
+ fy = 0 ,
+
+
z
y
x
zx zy z
+ fz = 0 ,
+
+
z
y
x
( 6)
or
ij , j + f i = 0
on u ( specified displacement )
ti = ti ,
on ( specified traction )
( 7)
( traction ti = ij n j )
p
n
( = u + )
u
Stress Analysis:
Solving equations in (6) under the BCs in (7).
141
u=
N i ui
i =1
v=
N i vi
(8)
i =1
w=
N i wi
i =1
Nodal values
In matrix form:
N1
u
= 0
v
w
0
( 31)
0
N1
0
0
N2
0
0
N2
0
0
N1
N2
u1
v1
L
w1
(9)
L
u2
L
( 33 N ) v2
w2
M ( 3N 1)
u=Nd
or
Using relations (5) and (8), we can derive the strain vector
=B d
(61) (63N)(3N1)
142
Stiffness Matrix:
k = B T E B dv
(10)
(3N) (3N6)(66)(63N)
143
linear (4 nodes)
Hexahedron (brick):
linear (8 nodes)
Penta:
linear (6 nodes)
144
Element Formulation:
Linear Hexahedron Element
6
5
7
2
y 8
1
4
3 mapping (x)
(-1 ,, 1)
x
z
(-1,1,-1) 4
(-1,1,1) 8
3 (1,1,-1)
7 (1,1,1)
o
(-1,-1,-1) 1
(-1,-1,1) 5
2 (1,-1,-1)
6 (1,-1,1)
u = N i ui ,
i =1
v = N i vi , w = N i wi
i =1
(11)
1i =1
145
Shape functions:
1
N 1 ( , , ) = (1 ) (1 ) (1 )
8
1
N 2 ( , , ) = (1 + ) (1 ) (1 )
8
1
N 3 ( , , ) = (1 + ) (1 + ) (1 )
8
M
M
1
N 8 ( , , ) = (1 ) (1 + ) (1 + )
8
,
,
,
(12)
N i ( , , ) = 1.
i =1
x = N i xi , y = N i yi , z = N i zi .
i =1
i =1
(13)
i =1
Isoparametric element.
146
Jacobian Matrix:
u
u
=
u
x
x
z
z
u
x
u
y
u
z
(14)
Jacobian matrix
u
u
x
u
1 u
=J ,
y
u
u
z
u
=
N i
u
,
etc
.
i
i =1
and
v
v
x
v
1 v
=J ,
y
v
v
z
(15)
also for w.
147
u
x
v
x
y
y
w
z z
= = x u = L use (15) = B d
xy +
yz x y
w v
zx +
y z
u
w
+
z x
where d is the nodal displacement vector,
i.e.,
= Bd
(16)
(61) (624)(241)
148
Strain energy,
1 T
1
( E ) T dV
dV
2V
2V
1
= T E dV
2V
U=
1 T T
d B E B dV d
2
(17)
(18)
(2424) (246)(66)(624)
In coordinates:
dV = (det J ) d d d
(19)
1 1 1
k = B T E B (det J ) d d d
( 20)
1 1 1
( Numerical integration)
149
Loads:
Distributed loads Nodal forces
pA/3
pA/12
Area =A
Stresses:
= E = EB d
Principal stresses:
1 , 2 , 3 .
von Mises stress:
e = VM =
1
( 1 2 ) 2 + ( 2 3 ) 2 + ( 3 1 ) 2 .
2
Examples:
1999 Yijun Liu, University of Cincinnati
150
Baseball bat
shaft
r, u
z, w
r, u
r
rz
r
151
Displacement field:
u = u ( r , z ) , w = w( r , z )
Strains:
u
u
w
,
,
= ,
z =
r
r
z
w u
, ( r = z = 0)
+
rz =
r z
r =
( 21)
u
(r+u)d
r
d
rd
Stresses:
r
E
=
z (1 + v ) (1 2v )
rz
0
v
v
1 v
v 1 v
0
v
0
v 1 v
v
1 2v
0
0
0
r
( 22)
z
rz
152
Axisymmetric Elements:
2
r, u
r, u
3
1
3-node element (ring)
k = B T E B rdr d dz
( 23)
or
2 1 1
k=
B T E B r (det J ) d d d
0 1 1
1 1
= 2
B T E B r (det J ) d d
( 24)
1 1
153
Applications:
Rotating Flywheel:
z
Body forces:
fr = r 2
fz = g
( gravitational force)
154
r0
q = ( p ) 2 r0
Press Fit:
r0
ri
ri +
ring ( Sleeve)
shaft
at r = ri :
uo ui =
MPC
i o
155
156
F(t)
I. Basic Equations
A. Single DOF System
k
m
f=f(t)
m - mass
k - stiffness
c - damping
f (t ) - force
ku
c u&
f(t)
x, u
(1)
157
Free Vibration:
mu&&+k u =0 .
(2)
i.e.,
m+k U = 0.
m+k = 0,
which yields
=
k
.
m
(3)
(4)
158
u = U s in w t
U
t
U
T=1/f
Undamped Free Vibration
(5)
(6)
c
(damping ratio).
cc
(7)
159
(8)
in which:
,A,L
u1
2 m = AL
2
u2
AL
m= 2
AL
0
1442424
3
diagonal matrix
160
m=
N T NdV
(9)
1 T
1
u& m u&
(cf. mv 2 )
2
2
1
1
T
= u& 2 dV = (u& ) u& dV
2 V
2 V
1
T
= (N u& ) (N u& )dV
2 V
1
= u& T N T N dV u&
V 42 43
2
1
1
m =
[1 ]ALd
V
1 / 3 1 / 6 u&&1
= AL
1 / 6 1 / 3 u&&2
(10)
161
v1
1
, A, L
m = NT NdV
V
22 L
54
13L v&&1
156
4 L2
13L
3L2 &&1
AL 22 L
=
13L
156 22 L v&&2
420 54
2
2 &&
13L 3L 22 L 4 L 2
(11)
Choice I
s
m
kg
m/s2
N
kg/m3
Choice II
s
mm
Mg
mm/s2
N
Mg/mm3
162
(12)
[K
M u = 0
(13)
Solutions?
163
(14)
[K
2
i
M ui = 0 .
u iT K u
= 0,
u iT M u j = 0 ,
for i j,
(15)
164
u iT M u i = 1,
u iT K u i = i2 .
(16)
Note:
Magnitudes of displacements (modes) or stresses in normal
mode analysis have no physical meaning.
For normal mode analysis, no support of the structure is
necessary.
i = 0 there are rigid body motions of the whole or a
part of the structure.
apply this to check the FEA model (check for
mechanism or free elements in the models).
Lower modes are more accurate than higher modes in the
FE calculations (less spatial variations in the lower modes
fewer elements/wave length are needed).
165
Example:
v2
2
, A, EI
1
2
L
v2 0
K M = ,
2 0
K=
EI 12 6L
,
3
2
L 6L 4L
M=
AL 156 22L
.
2
420 22L 4L
12 156 6L + 22L
= 0,
2
2
6L + 22L 4L 4L
EVP:
2
4
in which = AL / 420 EI .
#3
#2
#1
EI 2 v 2 1
1 = 3.533
, = 1.38 ,
4
AL
2 1
L
1
1
EI 2 v 2
2 = 34 .81
=
,
7.62 .
4
AL
2 2
L
Exact solutions:
1
EI
1 = 3.516
,
4
AL
2
EI 2
2 = 22.03
.
4
AL
166
III. Damping
Two commonly used models for viscous damping.
A. Proportional Damping (Rayleigh Damping)
C = M + K
(17)
1 =
,
+
2
21
2 =
,
+
2
2 2
Damping ratio
B. Modal Damping
Incorporate the viscous damping in modal equations.
167
[K
i M ui = 0 ,
2
i = 1,2,..., n
(18)
u iT K u
T
ui M u
j
j
= 0,
= 0,
for i j,
and
u iT M u i = 1 ,
T
2
u
K
u
=
i
i
i ,
for i = 1, 2, , n.
(nn )
[u 1
L u
(19)
2
0
M
2
(Spectralmatrix),
T K = =
M
O 0
2
0
0
L
n
T M = I.
(20)
168
u = z1 u 1 + z 2 u 2 + L + z n u n = z ,
(21)
where
z1 ( t )
z (t )
z= 2
M
z n ( t )
where C = I +
p =
(22)
(proportional damping),
f (t) .
2 1
0
=
M
0
2 2
O
L
M
. (23)
2 n n
0
169
Notes:
Only the first few modes may be needed in constructing
the modal matrix (i.e., could be an nm rectangular
matrix with m<n). Thus, significant reduction in the
size of the system can be achieved.
Modal equations are best suited for problems in which
higher modes are not important (i.e., structural
vibrations, but not shock loading).
170
&& + Cu& + Ku = F
Mu
t
1sin
23
Harmonicloading
(25)
(26)
zi
zi ( t ) =
pi i2
(1 ) + ( 2 ii )
2 2
i
sin( t i ),
(27)
where
2 i i
, phase angle
arctan
=
i
2
1 i
i = i ,
ci
ci
, damping ratio
=
=
i c
m
2
c
i
/i
[K
+ i C 2 M u = F .
171
u(t)
t0 t1 t2
t n t n+1
172
&&
u
1
( u n + 1 u n 1 ),
2 t
1
=
( u n +1 2 u n + u
( t)2
=
n 1
1
1
+
(
2
)
( u n +1 u n 1 ) + Ku n = fn ,
M
u
u
u
C
n
1
n
n
1
+
2 t
( t )
which yields,
Au n +1 = F(t )
where
1
1
A =
M+
C,
2
2t
( t )
F ( t ) = f n K 2 2 M u n 1 2 M 1 C u n 1.
2t
( t )
( t )
173
= F (t)
n +1
where
1
C +
M ,
t
( t)2
&& n ).
F ( t ) = f ( f n + 1 , , , t , C , M , u n , u& n , u
A = K +
e . g .,
1
.
2
1
,
=
2
1
4
174
B. Modal Method
First, do the transformation of the dynamic equations using
the modal matrix before the time marching:
u =
i =1
u i zi (t ) = z ,
&z&i + 2 i i z& i + i z i = p i ( t ),
i = 1,2,, m.
Modal Method
Small model
Large model
Single loading
Multiple loading
Shock loading
Periodic loading
175
Examples
Impact, drop test, etc.
176