Classification of EM Problems

The classification of the EM problem is important in helping to find the

best method for solution.
1) The solution region of the problem,
2) Equations describing the problem,
3) Boundary Conditions.

1) The Solution Regions

Solution regions may be classified as, closed region, bounded, or open

region, unbounded. i.e. Wave propagation in a waveguide is a closed
region problem, where radiation from a dipole antenna is an open
region problem.

A problem is also classified in terms of the electrical, constitutive

properties. We shall be concerned with simple materials here.

2) Classification of the Equations Describing the Problem

EM problems may be stated by using:

1) Differential Equations
2) Integral Equations
3) Integro-Differential Equations

In general, EM problems have the following form:

L g

L: Operator (integral, differential, integro-differential),

g : Excitation or source,
: Unknown function.
Example: The Poissons Equation in the differential form (differential
operator).

v
2V

Here,
L 2
v
g

V
Example: The Poissons Equation in the integral form (integral
operator).

v r '
V r dv '
v
4 o r r '
Where
dv '
L
v
4 r r '
g V r
v r '

o

This formulation is used in the MoM to find the charges on

conductors when the potential distribution is known.
Example: Electric field intensity of a wire antenna integro-
differential operator).

1 2G z, z '
I z ' dz '
/2

j o
Ez k 2
G z , z '
z
2

/2

e jkr
Where, G z, z '
4 r .

EM problems satisfy second order partial differential equations

(PDE).

i.e. Wave equation, Laplaces equation.

A two dimensional second order PDE may have the following

form:

2 2 2
a 2 b c 2 d e f g
x xy y x y

The operator is:

2 2 2
L a 2 b c 2 d e f
x xy y x y
Any linear second order PDE can be classified as follows:

If , b 2 4ac 0 Elliptic
If , b 2 4ac 0 Parabolic
If , b 2 4ac 0 Hyperbolic

The differential equation is said to be non-linear if the coefficients (a, b,

c, d, e, f) is a function of x and y .

If, g ( x, y) 0 , then PDE is homogeneous.

If, g ( x, y) 0 , then PDE is inhomogeneous.

Examples: Elliptic PDE, Poissons and Laplaces Equations

2 2
2 g ( x, y ) Poisson ' s Eqn.
x 2
y
2 2
2 0 Laplace ' s Eqn
x 2
y
For both cases a=c=1, b=0.

b2 4ac 0

An elliptic PDE usually models the closed region problems.

Example: Hyperbolic PDEs, the Wave Equation in one dimension.

2 1 2
0
x 2 u 2 t 2

A Hyperbolic PDE usually models the Propagation Problems (Open

region problems).

a u 2 , b 0, c 1
b2 4ac 0
Example: Parabolic PDE, Diffusion Equation for an RC transmission line.

2V V
RC 0
x 2 t
1 2V V
0
RC x 2
t
1
a ,bc0
RC
b2 4ac 0
A Parabolic PDE usually arises in propagation problems.
 3) Classification of Boundary Conditions:

Consider again,

L g

The solution of this equation is to find the unknown function , within a

solution region R , subject to certain boundary conditions on surface S
bounding R .

These boundary conditions are Dirichlet and Neumann type boundary

conditions .
1) Dirichlet B.C.

(r ) 0, R on S

i.e , vanishes on S

2) Neumann B.C
(r )
0, R on S .
n
i.e. the normal derivative of vanishes on S.

3) Mixed Boundary Conditions

A PDE in general can have both:

1) Boundary Values
2) Initial values
Example:

Consider the 1D wave equation:

2 1 2
2 2 0
x 2
u t
The solution of this second-order wave equation requires the following
initial conditions:

x,0 ( x)

x, t
( x)
t t 0