Product Life Cycle

NEEDS

FINITEELEMENTMETHOD
PHASE-
OUT
DESIGN

(MetodeElemenHingga)

OPERATE MANU-
& MAINT. FACTURE

1
1.1. Introduction A. Formal Definition of FEA:
The Finite Element Method (FEM) is a versatile An approximate mathematical analysis tool to study
and powerful mathematical (numerical) tool that the behavior of a continua (or a system) to an external
has wide applications in a multitude of physical influence such as stress or strain, heat, pressure,
problems such as stress analysis, fluid flow, heat temperature, fluid velocity, magnetic field, etc.
transfer, acoustics, aero-elasticity, micro-
This involves generating a mathematical formulation
fluidics, MEMS (Micro-Electro-Mechanical of the physical process followed by a numerical
Systems), electrical and magnetic fields, solution of the mathematics model.
electrostatic coupling and many others.

2
 Engineers Mathematicians
B. History of FEA: Trial functions Finite differences
Richardson 1910
Hyper-static structure Navier 1819 Variational Weighted Liebman 1918
methods residuals Southwell 1940
Energy theorem Maxwell 1864 Rayleigh 1870 Gauss 1795
Ritz 1909 Galerkin 1915
Castigliano 1878 Biezeno-Koch 1923

Approximation method Ritz 1908

Galerkin 1915 Structural analogue Piecewise continous
substitution trial function
Approximation by finite elements Courant 1940 Hrenikoff 1941 Courant 1943
McHenry 1943 Prager-Synge 1947
Matrix method: Newmark 1949
Levy & Garvey 1953 Direct continuum Variational finite
Force method in aircraft industry elements differences
Modern FEM Argyris 1955
Turner et al1. 1956
Varga 1962

Force method Argyrys-Denke 1955

Displacement method Argyris-Turner 1956 Modern FEM
Figure 1-1(a) Historical background to modern FEM, after J.F. Imbert [2] Figure 1-1(b) Historical background to modern FEM, after O.C. Zienkiewics [3]

C. Basic Concept: 1.2. Basic Illustration

A. Circumference:
Division of a given domain into a set of simple sub- Se
He Q
domains called finite elements accompanied with Q R
R R
polynomial approximations of solution over each
element in terms of nodal values.
1. FE Discretization
Assembly of element equation with inter-element Each line segment is an element, He
Collection of these line segments is called a mesh
continuity of solution and balance of force is Element are connected at nodes
considered. 2. Element equations
H e 2 R sin( 2 )

3
3. Assembly of equations and solution
n
P He
B. Frame Structure:
e 1

2
For , H e 2 R sin( ), P 2nR sin( )
n n n

4. Exact solution
As n , P 2R
1 sin(x )
If x P 2 R
n x
n x0
sin(x) cos(x) n P nEe
lim 2R 2R 1 2R
x 0 x lim
x 0 1 2,5E-16 6,283185307

5. Error Estimation 10 6,18034 0,10284542

(a) Real structure (b) Discretized structure

Error, Ee Se H e 2 R sin
100 6,28215 0,001033492

n n 1000 6,28317 1,03354E-05

Total Error nEe 2R P 10000 6,28319 1,03354E-07

Figure 1-2 Example of discretization of a frame structure by FEM

C. Continuous problem: 1.3. General Step in the FEM

Step 1 Discretize and Select Dividing the body into an equivalent system of
Element Types finite elements with associated nodes and
choosing the most appropriate element type.

Step 2 Select a displacement Choosing a displacement function

function (approximation function) within each element.

Step 3 Define the strain- Both relationships are necessary for deriving the
displacement and stress- equations for each element.
strain relationship
(b) Discrete model
Step 4 Derive the element stiffness Based on the concept of stiffness influence
matrix and equations coefficients (direct equilibrium method, work or
(a) Continuous problem energy method, weighted residual method.)

Figure 1-3 Descritization of an elasticity 2D continuous problem by FEM

4
 Level of Nodal Defor. Work of Stiffness Nodal
Step 5 Assemble the element Individual element equations generated in step 4 Formulation This image cannot currently be display ed.
Displ. Energy Ext. forces Matrix Forces
equations to obtain the is added together using a method of
global equations and superposition (called the direct stiffness method). Element di
vj
introduce boundary d d j e 1 eT e e
U 2 d K d e d e F e Ke Fe
T
i e
conditions j uj d
k
k
Step 6 Solve for the unknown Global equations obtained from step 5 is a set of
degrees of freedom (or simultaneous algebric equations. These Assemblage
generalized displacements) equations can be solved by using an elimination d1
U 12 d T Kd dT F K F
method (Gausss method) or an iterative method Global
u j
(Gauss-Seidel, etc.) i e j d j v j di
Virtual Work Principle Linear Equation

Step 7 Solve for the element strains For the structural stress-analysis problem, strains k d 0 System

and stresses and stress (or moment and force) can be d d j
obtained.
U KdF
d T K d d T F
Step 8 Interpret the results The final goal is to interpret and analyse the d k

results for use in the design/analysis process.
n : total number of nodes d n Solution d

Class Name Geometry Class Name Geometry

Point
0D Elasticity 2D
element
(tin) shell
Truss 2D
1D element (Plan Bending
(Line Ele- plate
Ele- ment)
ment) Frame
element Coque

Figure 1-4 (a) Different type of elements Figure 1-4 (b) Different type of elements

5
 Class Name Geometry
Class Name Geometry
3D Volume
Torus (Volu
axisymetric me
Axi-
symet Ele-
Thick Coque
ric ment)
Coque
axisymetric
Gap element that have stiffness
Special Element
only for compression direction.
Figure 1-4 (c) Different type of elements
Figure 1-4 (d) Different type of elements

1.4. Analysis Type 1.5. Computer Code

Assemblage
Restraints
Analysis Linear Non-linear Start
Linear static Static Non-linear Sub program for
K, F
Kq F Input Data
matrix calculation

Static
Initial stability Non-linear stability Solution LES
Elements
K K G X F Library
FE modeling
q
Modal Non-linear dynamic Element
K M X 0, 2 Characteristics Elements stress
calculation
Dynamic response Direct integration step by step Figure 1-7
Dynamic
Cq Kq F(t )
Mq Ke, Fe Print Result Simplified flowchart
for static analysis
Modal superposition
(displacement method)
Direct integration step by step End

6
1.6. Application
Structural areas:
Stress analysis, including truss and frame analysis both for
structural and non-structural concentration problems typically
associated with holes, fillets, or other changes in geometry in
a body.
Buckling problem
Vibration analysis

Non-structural problems:
Heat transfer
Fluid flow, including seepage through porous media
Distribution of electric or magnetic potential

7
8
References:
1. Logan, D.L., 1992, A First Course in the Finite Element
Method, PWS-KENT Publishing Co., Boston.
2. Imbert, J.F.,1984, Analyse des Structures par
Elements Finis, 2nd Ed., Cepadues.
3. Zienkiewics, O.C., 1977, The Finite Eelement Method,
3rd ed., McGraw-Hill, London.

Pegas:

Pendekatan mekanik: