Next: Fourier series
Up: Electrostatics
Previous: Method of Images and
We shall now consider a method of solving the boundary value problem in
electrostatics where one uses a basis of orthogonal functions. The basic idea is as
follows. For a class of differential operators (
) defined over a region of
space, there is a set of solutions
The solutions
are called the eigenfunctions and
the
eigenvalues. Here a point in the region of space is denoted by
. Further, the
eigenfunctions
are orthogonal, i.e.
if
. Here it is understood that the integration is performed over the
region of space over which the operator
is defined. By suitable scaling,
we can normalise these functions. Thus, they satisfy orthonormality relation
In addition, these function form
a complete set. That is, they satisfy the relation
The fact that the set of functions
form a complete and orthonormal set
over a region of space means that any function
, defined in the region of the
set of functions can be expressed in the form
 |
|
|
(60) |
The coefficients
appear to be arbitrary but actually, they are defined uniquely
by minimising
with respect to
the variation of
's.
The relation is
 |
|
|
(61) |
Subsections
Next: Fourier series
Up: Electrostatics
Previous: Method of Images and
S.C.Phatak
2007-02-20