13.3 The Frobenius method 373
EXAMPLE 13.3Indicial equations
(i)
We haveb
0
1 = 1 − 122 ,c
0
1 = 1122 , and the indicial equation isr
2
1 − 13 r 221 + 11221 = 10 with
rootsr
1
1 = 11 andr
2
1 = 1122.
(ii)x
2
y′′ 1 + 1 xy′ 1 + 1 (x
2
1 − 112 4)y 1 = 10
This is the Bessel equation (see Table 13.1) forn 1 = 1122. We haveb(x) 1 = 111 = 1 b
0
and
c(x) 1 = 1 x
2
1 − 1124 so thatc
0
1 = 1 − 124. Thenr
2
1 − 11241 = 1 (r 1 − 112 2)(r 1 + 112 2) 1 = 10 and the
indicial parameters arer 1 = 1 ± 12 2.
(iii)xy′′ 1 + 12 y′ 1 + 14 xy 1 = 10
We havex
2
y′′ 1 + 12 xy′ 1 + 14 x
2
y 1 = 10. Therefore,b(x) 1 = 121 = 1 b
0
and c(x) 1 = 1 x
2
so that
c
0
1 = 10. The indicial equation isr
2
1 + 1 (b
0
1 − 1 1)r 1 + 1 c
0
1 = 1 r
2
1 + 1 r 1 = 10 , with rootsr
1
1 = 10
andr
2
1 = 1 −1.
0 Exercises 6–9
Once the values of the indicial parameter have been determined, the solution of
the differential equation can proceed as in the power-series method. The general
solution is
y(x) 1 = 1 c
1
y
1
(x) 1 + 1 c
2
y
2
(x) (13.9)
in which c
1
and c
2
are arbitrary constants, andy
1
(x)andy
2
(x)are independent
particular solutions. At least one of y
1
and y
2
has the form (13.4), and the other
depends on the solutions,r
1
andr
2
, of the indicial equation. We have the following
three cases.
1 Distinct roots not differing by an integer
Bothy
1
andy
2
have the form (13.4),
y
1
(x) 1 = 1 x
r
1
(a
0
1 + 1 a
1
x 1 + 1 a
2
x
2
1 +1-) (13.10a)
y
2
(x) 1 = 1 x
r
2
(A
0
1 + 1 A
1
x 1 + 1 A
2
x
2
1 +1-) (13.10b)
2 Double root
Ifr
1
1 = 1 r
2
1 = 1 r, one solution has the form (13.4),
y
1
(x) 1 = 1 x
r
(a
0
1 + 1 a
1
x 1 + 1 a
2
x
2
1 +1-) (13.11a)
and the second solution is
y
2
(x) 1 = 1 y
1
(x) 1 ln 1 x 1 + 1 x
r+ 1
(A
0
1 + 1 A
1
x 1 + 1 A
2
x
2
1 +1-) (13.11b)
xy xy y
2
1
2
1
2
′′− ′+= 0