Numerical Analysis
BVP
Dr. Nasir M Mirza
Email: [email protected]
Applications of BVP
1. Extensive application and use. Boundary Value Problems
occur
all physical problems where space variations are
sought and are seen particularly in structural and solid
mechanics.
2.
Typical transport problems have BVP types. For example,
thermal processing of food and biomaterials (drying,
cooling, baking, microwaving, freezing,).
Movement of water and chemicals in soil and other
porous media.
Applications of BVP
2. The material properties in adjacent elements do not
have to be the same. This allows application of BVP
to composite materials.
3. Irregularly shaped boundaries can be approximated
using elements with straight sides or matched exactly
using elements with curved boundaries. The finite
element or finite difference method is used and it is
not limited to nice shapes with easily defined
boundaries (like many analytical solutions or some
other numerical methods).
Applications of BVP
4. The size of the elements can be varied. This allows
the element grid or mesh to be expanded or refined
as the need arises.
5. Boundary conditions such as discontinuous surface
loadings present no difficulties for the method.
Mixed boundary conditions can be easily handled.
8. BVP are an integral part of CAD/CAM (computer
aided design/computer aided manufacturing) and
CFD (computational fluid dynamic) systems.
9. Applications of BVP are also done using high
performance computing and parallel algorithms.
10. A a few test cases can include crash testing,
explosion, combustion, fluid-structure interaction,
large CFD problems, plasma fusion, biomedical
research, constructive surgery,.
10. Steady state and time-dependent problems are
also part of BV- problems.
Applications of BVP
Mixed Boundary Value Problems occur, in a natural way, in varieties of
branches of Physics and Engineering and several mathematical
methods have been developed to solve this class of problems of
Applied Mathematics.
While understanding applications of such boundary value problems
are of immense value to Physicists and Engineers, analyzing these
problems mathematically and determining their solutions by utilizing
the most appropriate analytical or numerical
methods are the
concerns of Applied Mathematicians.
Of the various analytical methods, which are useful to solve certain
mixed boundary value problems arising in the theory of Scattering of
Surface Water Waves, the methods involving complex function theory
and singular integral equations will be examined in detail along with
some recent developments of such methods.
6
Systems of Ordinary Differential
Equations - BVP
Boundary Value Problems
Shooting Methods
Linear
Nonlinear
Finite Difference
Linear
Nonlinear
Ordinary differential Equations Boundary Value Problems
Approximation of the solution to boundary value
problems, differential equations with conditions
imposed at different points.
y f x, y, y,
a xb
with the boundary conditions
y a ya and y b yb
Boundary Conditions
There are three types of boundary conditions:
Dirichlet boundary conditions
Newman boundary conditions
ya ,
yb
Mixed conditions
yb
ya ,
ya c1 ya ,
yb c2 yb
Linear ODE -Boundary Value
Problems
For simple equation with Dirichlet boundary
conditions between a < x < b
y px y qx y r x ,
y a ya and y b yb
where, the p(x), q(x) and r(x) are functions of the
independent variable.
Linear ODE -Boundary Value
Problems
Rewrite the equation in terms of two initial value
problems
u px u qx u r x ,
v px v qx v,
u a ya and ua 0
va 0 and va 1
if the slope at the end, b, is not zero we can write
equation as:
yx u x Avx
Linear ODE -Boundary Value
Problems
At the boundary y(b) = yb,
yb u b
y b yb u b Av b A
vb
Substitute back into the equations.
yb u b
yx u x
v x
vb
Boundary Value Problems
Rewrite the equations as a set of first order
differential equations and solve:
w1 u
w1 px w1 qx u r x
w2 v
w2 px w2 qx v
The differential equations
The first order equations can be solved using the
techniques we developed in one dimensional initial
value problems.
Taylor Series
Euler / Modified Euler
Runge-Kutta
Adam Bashforth
Higher-order differential Example
Problem
Consider a simple second order differential
equation.
d x
t
t 1 x
2
dt
5
2
x t 1 x
5
The initial conditions are x(1) =2, x(3) =-1 and
Dt = 0.2.
Example Problem
The equation can be written as a set of two initial
value problems.
t
x1 t 1 x1 ,
x1 1 2, x1 1 0
t
x2 1 x2 ,
x2 1 0, x2 1 1
Example Problem
The equation can be written as a set of two initial
value problems.
w1 x1
t
w1 t 1 x1 ,
5
w2 x2
x1 1 2, x1 1 0
w2 1 x2 ,
5
x2 1 0, x2 1 1
Example Problem
The two segments are determined:
t
1
1.2
1.4
1.6
1.8
2
2.2
2.4
2.6
2.8
3
x[1]
2
2.052
2.21359
2.49414
2.90571
3.46283
4.18217
5.08228
6.18316
7.50579
9.07154
v[1]
0
0.532
1.1034
1.7283
2.4196
3.189
4.0464
5.0001
6.0563
7.2185
8.4867
dv/dt[1]
2.6
2.7595
2.9938
3.296
3.6597
4.0777
4.542
5.0428
5.5679
6.1025
6.6286
x*[1]
2
2.1584
2.4343
2.8398
3.3896
4.1006
4.9914
6.0823
7.3944
8.9495
10.769
v*[1]
0.52
1.0839
1.7021
2.3875
3.1515
4.0045
4.9548
6.0087
7.1699
8.439
9.8125
dv*/dt[1]
2.72
2.954
3.2553
3.6175
4.0338
4.4963
4.9956
5.5195
6.0536
6.5798
7.0768
x[2]
0
0.2
0.4061
0.6238
0.8584
1.1149
1.398
1.7123
2.062
2.4512
2.8833
v[2]
1
1.0152
1.0594
1.1307
1.2275
1.3487
1.4931
1.6596
1.8468
2.0527
2.2751
dv/dt[2]
0
0.152
0.2924
0.4242
0.5494
0.669
0.7829
0.8904
0.9898
1.0785
1.1533
x*[2]
0.2
0.403
0.618
0.8499
1.1039
1.3847
1.6967
2.0442
2.4314
2.8617
3.3383
v*[2]
1
1.0456
1.1179
1.2155
1.3374
1.4825
1.6497
1.8377
2.0447
2.2684
2.5057
dv*/dt[2]
0.152
0.29019
0.42022
0.54397
0.66236
0.7754
0.88226
0.98123
1.06982
1.1447
1.2018
x[3]
2
1.3534
0.7951
0.3151
-0.093
-0.432
-0.701
-0.899
-1.02
-1.056
-1
v[3]
-3.493
-3.014
-2.597
-2.221
-1.868
-1.522
-1.169
-0.797
-0.395
0.0482
0.5398
Two Dimensional Steady-State..
Consider:
T T
0
2
2
x
y
2
Laplace Equation
Two Dimensional Steady-State..
Analytical:
4To
1
y
sin( 2n 1)
a
n 0 2n 1
b
sinh( b 1)( 2n 1)
cos ec ( 2n 1)
a
a
Two Dimensional Steady-State..
Consider:
System of grid-points (i,j)
and its four immediate neighbors
i,j+1
i,j
I-1,j
Dy
i+1,j
Dx
i,j-1
Two Dimensional Steady-State..
T T
0
2
2
x
y
2
Difference equation:
Ti 1, j 2Ti , j Ti 1, j
2
Dx
Ti , j 1 2Ti , j Ti , j 1
0
2
Dy
Two Dimensional Steady-State..
If Dx = Dy:
Ti , j
Ti 1, j Ti 1, j Ti , j 1 Ti , j 1
4
Two Dimensional Steady-State..
Example:
What is the temperature at P
P= (100 + 0 + 0 + 0)/4
= 25
0
100
100
100
not very accurategrid is too coarse!
Two Dimensional Steady-State..
Apply the equation to each interior node:
Ta = (Tb + 0 + 100 + Tc)/4
Tb = (0 + 0 + Ta + Td)/4
Tc = (Td + Ta + 0 + 100)/4
Td = (0 + Tb + Tc + 0)/4
100
100
100
0
0
100
Four equations --> four unknown!
Two Dimensional Steady-State..
Gauss-Seidel Iteration:
a. replace all = by <--
b. make starting guesses
c. compute new approximations
for Ta, Tb, Tc, Td
d. repeat step c until converge.
