239x Filetype PDF File size 0.20 MB Source: www.iitg.ac.in
Chapter-4
Legendre’s Polynomials
4.1 Introduction
The following second order linear differential equation with variable
coefficients is known as Legendre’s differential equation, named after
Adrien Marie Legendre (1752-1833), a French mathematician, who is
best known for his work in the field of elliptic integrals and theory of
numbers :
…(1)
where n is a non-negative integer.
Legendre’s differential equation occurs in many physical and
engineering problems involving spherical geometry and gravitation.
4.2 Legendre’s Differential Equation
We know that the differential equation of the form
…(1)
is called Legendre’s differential equation (or simply Legendre’s equation),
where n is a non-negative integer.
This equation can also be put in the following form:
Clearly, the only singular points of (1) are x = 1, x = − 1 and x = ,
which are regular. Therefore, the Legendre’s differential equation is a
Fuchsian differential equation.
92
The points other than singular points e.g., x = 0, x = 2, etc. behave like
ordinary points of (1).
Let the series solution of (1) be of the form
…(2)
"
!
"
! …(3)
!
!
$
and # "
! "
! …(4)
Putting the above values of y, y and y# in (1), we have
"
! "
!
!
! "
! "
! !
! "
! "
!
$%
"
!
$
or "
! "
!
& '
"
! "
!
"
!
"
!
or !
! "
! "
! & '
"
!
"
!
"
!
or !
! "
! "
!
& '
"
!
"
!
"
!
"
!
or !
! "
! "
!
"
! "
!
…(5)
which is an identity in x and, therefore, the coefficients of various powers
of x in it should be zero.
Thus, equating to zero, the coefficient of the smallest power of ,
namely $ in (5), we get the following indicial equation
( *
" "
or " "
)
93
which gives two indicial roots k = k = 1 and k = k = 0.
1 2
Note that the roots of indicial equation are unequal and differ by an
integer.
Now, to get the recurrence relation, we equate to zero, the coefficient
of $in (5). Thus, we have
! "
! "
! ! "
! "
!
or ! "
! "
!
! …(6)
"
!"
! $%
Next, equating to zero, the coefficient of in (5), we obtain
"
"
…(7)
For k = 0, we note from (7) that C1 is indeterminate.
Thus, putting k = 0 in (6), we get ! ! !
! …(8)
!!
We now express C , C , C …. in terms of C and C , C C …. in terms
2 4 6 0 3 5 7
of C1 by assuming that C1 is finite.
Putting m = 2 in (8), we have
…(9)
+
Putting m = 4 in (8) and using (9), we obtain
, -
-
…(10)
,- ,+
and so on.
Next, putting m = 3 in (8), we obtain
-
…(11)
- -+
Again, putting m = 5 in (8) and using (11), we have
94
. - ,
- -
, …(12)
., .+
and so on.
Now, the solution (2) can be re-written as:
/
0
1
2, where k = 0
% / 0 1
0 / 1
or
2
2 …(13)
0 % / 1
Using the values of C , C , C , C ,…... in the above equation, we get
2 3 4 5
3 44%
4$ 4 4$% 4/0 25
+ 0+
3 4$% 4 /
4$/ 4$% 4 401
25 …(14)
% /+ 1+
which is the required general series solution, C0 and C1 being arbitrary
constants.
4.3. Solution of Legendre’s Differential Equation in
Descending Powers
Consider Legendre’s differential equation of the type
…(1)
where n is a non-negative integer.
It is possible to obtain the solution of (1) in terms of descending
powers of x. Due to its applications to physical problems, this form of
solution of Legendre’s differential equation is more important.
For such a solution, let us assume that the Legendre’s differential
equation (1) has a series solution of the form
$
…(2)
Then, by Frobenius method, we can find two linearly independent
solutions of (1) in descending powers of x as:
no reviews yet
Please Login to review.