01_General-Introduction
01_General-Introduction
GENERAL INTRODUCTION
Read: CONTENTS
Chapters 1 and 2 • Engineering and analysis
• Simulation of a physical process
• Examples mathematical model
development
• Approximate solutions and methods
of approximation
• The basic features of the finite element
method
• Examples
• Finite element discretization
• Terminology
• Steps involved in the finite element
JN Reddy model development
INTRODUCTORY REMARKS
What we do as engineers?
• develop mathematical models,
• conduct physical experiments,
• carry out numerical simulations to help
designer, and
• design and build systems to achieve a (1)
functionality in (2) most economical way.
JN Reddy Introduction: 3
Analysis
Analysis is an aid to design and manufacturing,
and not an end in itself.
Analysis involves the following steps:
Physical
System
Assumptions Laws of physics
concerning (conservation
the system principles)
Mathematical
Model
(BVP, IVP)
BVP – Boundary value problems FEM – Finite Element
(equilibrium problems) Method
IVP – Initial value problems FDM – Finite Difference
Numerical
(time-dependent problems) Method
Simulations
BEM – Boundary Element
Method
Numerical method Computational
(FEM,FDM,BEM,etc.) device
JN Reddy Introduction: 5
EXAMPLES OF MATHEMATICAL MODEL
DEVELOPMENT
Objective: Determine heat flow in a heat exchanger fin
Lateral surface and right end are
exposed to ambient temperature, T∞
(a)
Rectangular fins
L (b) 3D to 2D
a
h
Body from which
heat has to be extracted
y Convection, β P (T - T¥ ) , P = perimeter
z
x
l
2D to 1D
(qA )x (qA )x +Δx
a
æ A + Ax + Δx ö÷
( qA )x - ( qA )x +Δx - β P Δx (T - T¥ ) + ρ rh çç x ÷÷ Δx = 0
çè 2 ø
d dT
JN Reddy
- ( Aq) - β P (T - T¥ ) + ρ rh A = 0, q = -k
dx dx Basic Concepts: 6
EXAMPLE OF ENGINEERING
MODEL DEVELOPMENT (continued)
d æç dT ö÷
- çç Ak ÷÷ + β P (T - T¥ ) = ρ rh A
dx è dx ø
u = T - T¥ , a = Ak, c = β P, ρ rh A = f
d æç du ö÷
- çça ÷÷ + cu = f
dx è dx ø
Model Problem
d du
− a( x ) + c( x )u − f ( x ) = 0 in Ω = ( 0, L )
dx dx
du
a + b(u − u0 ) = P at a boundary point
dx
u0 uninsulated bar
a = EA
a = kA
E, A f (x ) P
c = βP f (x )
u(L)
L b=k L
c x
x, u u = T - T¥ , u0 = T¥
Elastic deformation of a bar Heat transfer in a fin
JN Reddy Basic Concepts: 8
ENGINEERING EXAMPLES OF THE
MODEL PROBLEM IN 1-D
Flow of viscous fluid through a channel
d dvx
− μ − f = 0 in Ω = ( 0,b)
dy dy
y U ³0
u( y ) = vx ( y ), horizontal velocity
f = pressure gradient, dp / dx (constant) b
2
μ ( y ) = fluid viscosity b
Q = shear stress x
U = velocity of the top surface
Poiseuille flow
Couette flow
∂J ∂R
L
or = 2 R dx = 0
∂ci 0
∂ci
This method is known as the least-Squares method. We
obtain N algebraic equations in N unknown C’s
∂R
L
0 R ∂ci dx = 0
JN Reddy Basic Concepts: 13
Methods of Approximation
(Continued)
3. Yet, another approach in which the residual, R(x), can be
made zero is in a weighted-residual sense
L
0 = ò y R dx , i = 1,2, , N
i
0
• Collocation method yi ( x) = d( x - xi )
¶R
• Least-squares method yi ( x ) =
¶ci
♦ Petrov-Galerkin method yi ( x) ¹ fi ( x)
♦ Galerkin Method yi ( x) = fi ( x)
JN Reddy Introduction: 16
WORKING EXAMPLE
For a weighted-residual method:
f0 satisfy the actual specified BCs
fi satisfy the homogeneous form of the actual
specified BCs
For the Ritz method:
f0 satisfy the actual specified essential BCs
fi satisfy the homogeneous form of the actual
specified essential BCs
d æç du ö÷
- ça ÷
÷ - f0 = 0, 0 < x < L
dx çè dx ø
du
u(0) = u0 , a =P
JN Reddy dx x =L Introduction: 17
WORKING EXAMPLE
(CONTINUED)
dfi
fi ( 0) = 0, a = 0 f1 = x ( 2L - x )
dx x =L
f0 ( 0) = u0 , fi ( 0) = 0
f0 = u0 , fi ( x ) = x i
JN Reddy Introduction: 18
BASIC FEATURES OF THE
FINITE ELEMENT METHOD (FEM)
m x i i m y i i m z i i
X = i =1
N
, Y = i =1
N
, Z= i =1
N
JN Reddy m
i =1
i m
i =1
i m
i =1
i
Basic Concepts: 20
Example 2: Determine the integral of a function
I = F ( x ) dx
a
x
a b
a1 + b1 x , a ≤ x ≤ x2
Ii = F ( x) dx = (a + b x) dx
i i i
xi xi
F ( x ) ≈ a 2 + b2 x , x2 ≤ x ≤ x3
a + b x, x ≤ x ≤ b I ≈ I1 + I 2 + I 3
3 3 3
Actual function
F(x) Approximation
I1 I2 I3
x
a = x1 x2 x3 b = x4
Basic Concepts:
JN Reddy 22
EXAMPLE 2 (continued) – Refined mesh
Actual function
F(x) Approximation
x
a = x1 b = x7
(temperature profile)
(represents the solution)
Elements
Nodes
•
• •
•
• • •
Domain, Ω • Ωe • •
• • • •
•
°°°° •
° •
Boundary, Γ •
Ωe Domain,Ω h
Boundary flux
JN Reddy Introduction: 26
FEM Terminology
Element A geometric sub-domain of the region being
simulated, with the property that it allows a unique (1)
representation of its geometry and (2) derivation of the
approximation (interpolation) functions.
Node A geometric location in the element which plays a
role in the derivation of the interpolation functions and it
is the point at which solution is sought.
Mesh A collection of elements (or nodes) that replaces
the actual domain.
Weak Form An integral statement equivalent to the
governing equations and natural boundary conditions.
More to come.