Presented by Niko Manopulo An Introduction to the Finite Element Analysis
Dec 18, 2015
Presented byNiko Manopulo
An Introduction to the Finite Element Analysis
Joint Advanced Student School St.Petersburg 2005
Agenda
PART IIntroduction and Basic Concepts
1.0 Computational Methods1.1 Idealization1.2 Discretization1.3 Solution
2.0 The Finite Elements Method2.1 FEM Notation2.2 Element Types
3.0 Mechanichal Approach3.1 The Problem Setup3.2 Strain Energy3.3 External Energy3.4 The Potential Energy Functional
Joint Advanced Student School St.Petersburg 2005
Agenda
PART IIMathematical Formulation
4.0 The Mathematics Behind the Method 4.1 Weighted Residual Methods
4.2 Approxiamting Functions4.3 The Residual4.4 Galerkin’s Method4.5 The Weak Form4.6 Solution Space4.7 Linear System of Equations4.8 Connection to the physical system
Joint Advanced Student School St.Petersburg 2005
Agenda
PART IIIFinite Element Discretization
5.0 Finite Element Discretization 4.1 The Trial Basis
4.2 Matrix Form of the Problem4.3 Element Stiffness Matrix4.4 Element Mass Matrix4.5 External Work Integral4.6 Assembling4.7 Linear System of Equations
6.0 References7.0 Question and Answers
PART I
Introduction and Basic Concepts
Joint Advanced Student School St.Petersburg 2005
1.0 Computational Methods
Joint Advanced Student School St.Petersburg 2005
1.1 Idealization
• Mathematical Models• “A model is a symbolic device built to
simulate and predict aspects of behavior of a system.”
• Abstraction of physical reality
• Implicit vs. Explicit Modelling• Implicit modelling consists of using existent
pieces of abstraction and fitting them into the particular situation (e.g. Using general purpose FEM programs)
• Explicit modelling consists of building the model from scratch
Joint Advanced Student School St.Petersburg 2005
1.2 Dicretization
1. Finite Difference Discretization• The solution is discretized• Stability Problems• Loss of physical meaning
2. Finite Element Discretization• The problem is discretized• Physical meaning is conserved on elements• Interpretation and Control is easier
Joint Advanced Student School St.Petersburg 2005
1.3 Solution
1. Linear System Solution Algorithms• Gaussian Elimination• Fast Fourier Transform• Relaxation Techniques
2. Error Estimation and Convergence
Analysis
Joint Advanced Student School St.Petersburg 2005
2.0 Finite Element Method
• Two interpretations1. Physical Interpretation:
The continous physical model is divided into finite pieces called elements and laws of nature are applied on the generic element. The results are then recombined to represent the continuum.
2. Mathematical Interpretation:The differetional equation reppresenting the system is converted into a variational form, which is approximated by the linear combination of a finite set of trial functions.
Joint Advanced Student School St.Petersburg 2005
2.1 FEM Notation
Elements are defined by the following properties:1. Dimensionality
2. Nodal Points
3. Geometry
4. Degrees of Freedom
5. Nodal Forces
(Non homogeneous RHS of the DE)
Joint Advanced Student School St.Petersburg 2005
2.2 Element Types
Joint Advanced Student School St.Petersburg 2005
3.0 Mechanical Approach
• Simple mechanical problem• Introduction of basic mechanical concepts• Introduction of governing equations• Mechanical concepts used in mathematical
derivation
Joint Advanced Student School St.Petersburg 2005
3.1 The Problem Setup
Joint Advanced Student School St.Petersburg 2005
• Hooke’s Law:
where
• Strain Energy Density:
3.2 Strain Energy
dx
dux )(
)()( xEx
)()(2
1xx
Joint Advanced Student School St.Petersburg 2005
• Integrating over the Volume of the Bar:
• All quantities may depend on x.
3.2 Strain Energy (cont’d)
dxEAuuU
dxuEAudxpdVU
L
LL
V
0
00
''2
1
')'(2
1
2
1
2
1
Joint Advanced Student School St.Petersburg 2005
3.3 External Energy
• Due to applied external loads
1. The distributed load q(x)
2. The point end load P. This can be included in q.
• External Energy:
L
dxquW0
Joint Advanced Student School St.Petersburg 2005
3.4 The Total Potential Energy Functional
• The unknown strain Function u is found by minimizing the TPE functional described below:
)]([)]([)]([ xuWxuUxu
or
WU
PART II
Mathematical Formulation
Joint Advanced Student School St.Petersburg 2005
4.0 Historical Background
• Hrennikof and McHenry formulated a 2D structural problem as an assembly of bars and beams
• Courant used a variational formulation to approximate PDE’s by linear interpolation over triangular elements
• Turner wrote a seminal paper on how to solve one and two dimensional problems using structural elements or triangular and rectangular elements of continuum.
Joint Advanced Student School St.Petersburg 2005
4.1 Weighted Residual Methods
The class of differential equations containing also the one dimensional bar described above can be described as follows :
.0)1()0(
)1(10),()())((][
uu
xxquxzdx
duxp
dx
duL
Joint Advanced Student School St.Petersburg 2005
It follows that:
Multiplying this by a weight function v and integrating over the whole domain we obtain:
For the inner product to exist v must be “square integrable”
Therefore:
Equation (2) is called variational form
0][ quL
4.1 Weighted Residual Methods
)2(0)][,()][(1
0 quLvdxvquL
)1,0(2Lv
Joint Advanced Student School St.Petersburg 2005
4.2 Approximating Function
• We can replace u and v in the formula with their approximation function i.e.
• The functions jandj are of our choice and are meant to be suitable to the particular problem. For example the choice of sine and cosine functions satisfy boundary conditions hence it could be a good choice.
N
jjj
N
jjj
xdxVxv
xcxUxu
1
1
)()()(
)()()(
Joint Advanced Student School St.Petersburg 2005
4.2 Approximating Function
• U is called trial function and V is called test function• As the differential operator L[u] is second order
• Therefore we can see U as element of a finite-diemnsional subspace of the infinite-dimensional function space C2(0,1)
• The same way
)1,0()1,0( 22 CUCu
)1,0()1,0( 2CSU N
)1,0()1,0(ˆ 2LSV N
Joint Advanced Student School St.Petersburg 2005
4.3 The Residual
• Replacing v and u with respectively V and U (2) becomes
• r(x) is called the residual (as the name of the method suggests)
• The vanishing inner product shows that the residual is orthogonal to all functions V in the test space.
qULxr
SVrV N
][)(
ˆ,0),(
Joint Advanced Student School St.Petersburg 2005
• Substituting into
and exchanging summations and integrals we obtain
• As the inner product equation is satisfied for all choices of V in SN the above equation has to be valid for all choices of dj
which implies that
4.3 The Residual
N
jjj xdxV
1
)()( 0)][,( qULV
NjdqULd j
N
jjj ,...,2,1,0)][,(
1
NjqULj ,...,2,10)][,(
Joint Advanced Student School St.Petersburg 2005
• One obvious choice of would be taking it equal to• This Choice leads to the Galerkin’s Method
• This form of the problem is called the strong form of the problem. Because the so chosen test space has more continuity than necessary.
• Therefore it is worthwile for this and other reasons to convert the problem into a more symmetrical form
• This can be acheived by integrating by parts the initial strong form of the problem.
4.4 Galerkin’s Method
jj
NjqULj ,...,2,10)][,(
Joint Advanced Student School St.Petersburg 2005
• Let us remember the initial form of the problem
• Integrating by parts
4.5 The Weak Form
dxqzupuvquLv 1
0
])''([)][,(
.0)1()0(
10),()())((][
uu
xxquxzdx
duxp
dx
duL
0')''(])''([1
0
1
0
1
0
vpudxvqvzupuvdxqzupuv
Joint Advanced Student School St.Petersburg 2005
4.5 The Weak Form
• The problem can be rewritten as
where
• The integration by parts eliminated the second derivatives from the problem making it possible less continouity than the previous form. This is why this form is called weak form of the problem.
• A(v,u) is called Strain Energy.
0),(),( qvuvA
dxvzupuvuvA )''(),(1
0
Joint Advanced Student School St.Petersburg 2005
4.6 Solution Space
• Now that derivative of v comes into the picture v needs to have more continoutiy than those in L2. As we want to keep symmetry its appropriate to choose functions that produce bounded values of
• As p and z are necessarily smooth functions the following restriction is sufficient
• Functions obeying this rule belong to the so called Sobolev Space and they are denoted by H1. We require v and u to satisfy boundary conditions so we denote the resulting space as
dxzuupuuA ))'((),(1
0
22
10
1
0
22)'(
H
dxuu
Joint Advanced Student School St.Petersburg 2005
• The solution now takes the form
• Substituting the approximate solutions obtained earlier in the more general WRM we obtain
• More explicitly substituting U and V (remember we chose them to have the same base) and swapping summations and integrals we obtain
4.7 Linear System of Equations
10),(),( HvqvuvA
N
N
SVqVUVA
HSVU
0
100
),(),(
,
N
kjkjk NjqAc
1
,...,2,1,),(),(
Joint Advanced Student School St.Petersburg 2005
4.8 Connection to the Physical System
Mechanical Formulation Mathematical Formulation
dxEAuuUL
0
''2
1
L
dxquW0
0 WU
dxvzupuvuvA )''(),(1
0
0),(),( qvuvA
),( qv
PART III
Finite Element Discretization
Joint Advanced Student School St.Petersburg 2005
• Let us take the initial value problem with constant coefficients
• As a first step let us divide the domain in N subintervals with the following mesh
• Each subinterval is called finite element.
5.0 Finite Element Discretization
Njxx jj :1),,( 1
1...0 10 Nxxx
.0)1()0(
0,
10),(''
uu
zp
xxqzupu
Joint Advanced Student School St.Petersburg 2005
• Next we select as a basis the so called “hat function”.
5.1 The Trial Basis
otherwise
xxxxx
xx
xxxxx
xx
x jjjj
j
jjjj
j
j
0
)( 11
1
11
1
Joint Advanced Student School St.Petersburg 2005
• With the basis in the previous slide we construct our approximate solution U(x)
• It is interesting to note that the coefficients correspond to the values of U at the interior nodes
5.1 The Trial Basis
Joint Advanced Student School St.Petersburg 2005
• The problem at this point can be easily solved using the previously derived Galerkin’s Method
• A little more work is needed to convert this problem into matrix notation
5.1 The Trial Basis
N
kjkjk NjqAc
1
,...,2,1,),(),(
Joint Advanced Student School St.Petersburg 2005
• Restricting U over the typical finite element we can write
• Which in turn can be written as
in the same way
5.2 Matrix Form of the Problem
],[)]()([)(
)(][)( 1
11
11 jj
j
jjj
j
jjj xxx
c
cxx
x
xccxU
],[)()()( 111 jjjjjj xxxxcxcxU
],[)]()([)(
)(][)( 1
11
11 jj
j
jjj
j
jjj xxx
d
dxx
x
xddxV
Joint Advanced Student School St.Petersburg 2005
• Taking the derivative
• Derivative of V is analogus
5.2 Matrix Form of the Problem
1
1
1
1 ],[]/1/1[/1
/1][)('
jjj
jjj
j
jjj
j
jj
xxh
xxxc
chh
h
hccxU
],[]/1/1[/1
/1][)(' 1
11 jj
j
jjj
j
jjj xxx
d
dhh
h
hddxV
Joint Advanced Student School St.Petersburg 2005
• The variational formula can be elementwise defined as follows:
5.2 Matrix Form of the Problem
j
j
j
j
j
j
x
x
x
x
Mj
x
x
Sj
Mj
Sjj
N
jjj
dxVqqV
dxzVUUVA
dxUpVUVA
UVAUVAUVA
qVUVA
1
1
1
),(
),(
''),(
),(),(),(
0]),(),([1
Joint Advanced Student School St.Petersburg 2005
• Substituting U,V,U’ and V’ into these formulae we obtain
4.3 The Element Stiffnes Matrix
11
11
][11
11][),(
]/1/1[/1
/1][),(
1
1
1
21
11
1
1
jj
j
j
jjjj
jx
xj
jjSj
j
jx
x jjj
jjj
Sj
h
pK
c
cKdd
c
cdx
h
pddUVA
c
cdxhh
h
hpddUVA
j
j
j
j
Joint Advanced Student School St.Petersburg 2005
4.4 The Element Mass Matrix
• The same way
21
12
6
][),(
][][),(
11
11
11
1
jj
j
jjjj
Sj
j
jx
x jjj
jjj
Mj
zhM
c
cMddUVA
c
cdxzddUVA
j
j
Joint Advanced Student School St.Petersburg 2005
• The external work integral cannot be evaluated for every function q(x)
• We can consider a linear interpolant of q(x) for simplicity.
• Substituting and evaluating the integral
4.5 External Work Integral
j
j
x
xdxVqqV
1
),(
],[),()()( 111 jjjjjj xxxxqxqxq
jj
jjjj
jjjj
jjj
x
xj
jjjj
qqhl
ldddxq
qddqV
j
j
2
2
6
],[],[],[),(
1
1
11
11
11
Element load vector
Joint Advanced Student School St.Petersburg 2005
• Now the task is to assemble the elements into the whole system in fact we have to sum each integral over all the elements
• For doing so we can extend the dimension of each element matrix to N and then put the 2x2 matrix at the appropriate position inside it
• With all matrices and vectors having the same dimension the summation looks like
4.6 Assembling
N
j
SjA
1
KcdT
21
121
.........
121
121
12
h
pK
1
2
1
1
2
1
......
NN d
d
d
d
c
c
c
c
Joint Advanced Student School St.Petersburg 2005
• Doing the same for the Mass Matrix and for the Load Vector
4.6 Assembling
N
j
MjA
1
McdT
41
141
.........
141
141
14
6
zhM
N
jjqV
1
),( ldT
NNN qqq
qqq
qqq
h
12
321
210
4
...
4
4
6l
Joint Advanced Student School St.Petersburg 2005
• Substituting this Matrix form of the expressions in
we obtain the following set of linear equations
• This has to be satisfied for all choices of d therefore
4.7 Linear System of Equations
0lcMKdT ])[(
N
jjj qVUVA
1
0]),(),([
0lcMK )(
Joint Advanced Student School St.Petersburg 2005
References
• Carlos Felippa
http://caswww.colorado.edu/courses.d/IFEM.d/IFEM.Ch06.d/IFEM.Ch06.pdf
• Joseph E Flaherty,Amos Eaton Professorhttp://www.cs.rpi.edu/~flaherje/FEM/fem1.ps
• Gilbert Strang, George J. Fix
An Analysis of the Finite Element Method
Prentice-Hall,1973