0% found this document useful (0 votes)
405 views27 pages

Chapter 83 Power Series Methods of Solving Ordinary Differential Equations

Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
405 views27 pages

Chapter 83 Power Series Methods of Solving Ordinary Differential Equations

Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 27

CHAPTER 83 POWER SERIES METHODS OF SOLVING

ORDINARY DIFFERENTIAL EQUATIONS

EXERCISE 312 Page 864

t
1. Determine the following derivatives: (a) y (4) when y = e 2 x (b) y (5) when y = 8e 2

(a) If y = e ax , then y ( n ) = a n e ax . Hence, if y = e 2 x , then y (4) = ( 2 ) e 2 x = 16 e 2 x


4

5
1
 1  12 t 8 12 t 1 12 t
(b) If y = e ax , then y ( n ) = a n e ax . Hence, if y = 8e 2 , = =
t
then y (5) (8)   e e = e
2 32 4

1
2. Determine the following derivatives: (a) y (4) when y = sin 3t (b) y (7) when y = sin 5θ
50

 nπ 
=
(a) If y = sin ax, then y ( n ) a n sin  ax + 
 2 

 4π 
= 34 sin  3t + =
Hence, if y = sin 3t, then y (4)  81sin ( 3t + 2π ) = 81 sin 3t
 2 

1  1   7π   3π 
(b) If y = sin 5θ, then
= y (7)   57 sin  5θ=
+  1562.5sin  5θ +  = – 1562.5 cos 5θ
50  50   2   2 

2
3. Determine the following derivatives: (a) y (8) when y = cos 2x (b) y (9) when y = 3 cos t
3

 nπ 
(a) If y = cos ax, =
then y ( n ) a n cos  ax + 
 2 

 8π 
y (8) = ( 2 ) cos  2 x +
8
Hence, if y = cos 2x then  = 256 cos(2x + 4π) = 256 cos 2x
 2 

 2 9π  29 2 π 
9
2 2 29 2
(b) If y = 3cos t , then =
y (9) (3)   cos  t + =  cos  t +  = − sin t
3 3 3 2  38 3 2 38 3

1285 © 2014, John Bird


t7
4. Determine the following derivatives: (a) y (7) when y = 2x9 (b) y (6) when y =
8

a!
(a) If y = x a , then y ( n ) = x a −n
( a − n )!
9!
Hence, if y = 2x9 , then y (7) = (2) x 9−7 = ( 9!) x 2
( 9 − 7 )!

t7  1  7! 7 −6
(b) If y = , then y (6) =   t = 630 t
8  8  ( 7 − 6 )!

1
5. Determine the following derivatives: (a) y (7) when y = sinh 2 x (b) y (6) when y = 2 sinh 3x
4

(a) If y = sinh ax, then =


y(n)
an
2
{[1 + (−1)n ] sinh ax + [1 − (−1)n ] cosh ax}

 1  27
y (7)   {[1 + (−1)7 ] sinh 2 x + [1 − (−1)7 ] cosh 2 x}
1
Hence, if y = sinh 2 x =
then
4 4 2

= 16 (2 cosh 2x) = 32 cosh 2x

(b) If y = 2 sinh 3x, then=


y (6) (2)
36
2
{[1 + (−1)6 ] sinh 3x + [1 − (−1)6 ] cosh 3x}

= 36 {2sinh 3 x + 0} = 1458 sinh 3x

1
6. Determine the following derivatives: (a) y (7) when y = cosh 2x (b) y (8) when y = cosh 3 x
9

(a) If y = cosh ax, then =


y(n)
an
2
{[1 − (−1)n ] sinh ax + [1 + (−1)n ] cosh ax}

Hence, if y = cosh 2x, then y=


(7)
27
2
{[1 − (−1)7 ] sinh 2 x + [1 + (−1)7 ] cosh 2 x}

= 26 {2sinh 2 x + 0} =
27 sinh 2 x = 128 sinh 2x

 1  38
y (8)   {[1 − (−1)8 ] sinh 3 x + [1 + (−1)8 ] cosh 3 x}
1
(b) If =
cosh 3 x , then
9 9 2

1286 © 2014, John Bird


= 364.5 {0 + 2 cosh 3x} = 729 cosh 3x

1
7. Determine the following derivatives: (a) y (4) when y = 2 ln 3θ (b) y (7) when y = ln 2t
3

( n − 1)!
( −1)
n −1
(a) If y = ln ax, then y ( n ) =
xn
( 4 − 1)! = 3! 12
( 2 )( −1)
4 −1
If y = 2 ln 3θ, then y (4) = (−2) =−
θ4 θ4 θ4

1 1 7 −1 ( 7 − 1) ! 6! 240
  ( −1)
(b) If y = ln 2t , then y (7) = = =
3 3 t 7 3t 7 t7

1287 © 2014, John Bird


EXERCISE 313 Page 866

1. Obtain the nth derivative of: x 2 y

Since y = x 2 y then let v = x 2 and u = y

n(n − 1) ( n − 2) (2)
Thus, y ( n ) =
u ( n ) v + nu ( n −1) v (1) + u v + ...
2!

n(n − 1) ( n − 2)
= y ( n ) ( x 2 ) + n ( y ( n −1) 2 x ) + ( y 2)
2!

= x 2 y ( n ) + 2nxy ( n −1) + n(n − 1) y ( n − 2)

2. If y = x 3 e 2 x find y ( n ) and hence y (3)

Since y = x 3 e 2 x then let v = x3 and u = e 2 x and the nth derivative of e 2 x is 2n e 2 x

n(n − 1) ( n − 2) (2)
Thus, y ( n ) =
u ( n ) v + nu ( n −1) v (1) + u v + ...
2!

n(n − 1) n − 2 2 x n ( n − 1)( n − 2 ) ( n −3) 2 x


= 2n e 2 x ( x 3 ) + n ( 2n −1 e 2 x )( 3 x 2 ) + ( 2 e )( 6 x ) + 2 e (6)
2! 3!

= 2n x3 e 2 x + 3nx 2 2n −1 e 2 x + 3n(n − 1)2n − 2 e 2 x x + n ( n − 1)( n − 2 ) 2n −3 e 2 x

or=y ( n ) e 2 x 2n −3 {23 x 3 + 3nx 2 (2) 2 + 3n(n − 1) x(2) + n(n − 1)(n − 2)}

= e 2 x 2n −3 {8 x3 + 12nx 2 + n(n − 1)(6 x) + n(n − 1)(n − 2)}

y (3) e 2 x 20 {8 x3 + 36 x 2 + 3(2)6 x + 3(2)(1)}


Hence, =

= e 2 x {8 x3 + 36 x 2 + 36 x + 6}

3. Determine the 4th derivative of: y = 2 x3 e − x

Since y = 2 x3 e − x then let v = 2 x3 and u = e − x and the nth derivative of e − x is (−1) n e − x

n(n − 1) ( n − 2) (2)
Thus, y ( n ) =
u ( n ) v + nu ( n −1) v (1) + u v + ...
2!

1288 © 2014, John Bird


n(n − 1) n ( n − 1)( n − 2 )
= (−1) n e − x ( 2 x3 ) + n ( (−1) n −1 e − x )( 6 x 2 ) + ( (−1)n−2 e− x )(12 x ) + (−1)( n −3) e − x (12)
2! 3!

Hence, y (4) =(−1) 4 e − x ( 2 x 3 ) + 4 ( (−1)3 e − x )( 6 x 2 ) + 6 ( (−1) 2 e − x )(12 x ) + 4 ( (−1)1 e − x ) (12)

= e − x {2 x3 − 24 x 2 + 72 x − 48} or 2 e − x { x 3 − 12 x 2 + 36 x − 24}

4. If y = x 3 cos x , determine the 5th derivative.

 nπ 
Since y = x 3 cos x then let u = cos x and v = x3 and=
u ( n ) 1n cos  x + 
 2 
n(n − 1) ( n − 2) (2)
y(n) =
u n v + nu ( n −1) v (1) + u v + ...
2!
 nπ  3  (n − 1)π  n(n − 1)  (n − 2)π 
Hence, y ( n ) =cos  x +  ( x ) + n cos  x +  ( 3x ) + cos  x +  (6x)
2
 2   2  2!  2 
n(n − 1(n − 2)  (n − 3)π 
+ cos  x +  (6)
3!  2 

 5π   4π  5(4) 3π 2π
and y=
(5) x3 cos  x +  + 5 ( 3 x ) cos  x +
2
+ ( 6 x ) cos  x +  5(4)(3)
+ ( 6 ) cos  x + 
 2   2  2!  2  3!  2 

= − x3 sin x + 15 x 2 cos x + 60 x sin x + 60(− cos x)

= ( 60 x − x3 ) sin x + (15 x 2 − 60 ) cos x

5. Find an expression for y (4) if y = e − t sin t .

Since y = e − t sin t then let u = sin t and v = e − t and the nth derivative of e − t is (−1) n e − t
n(n − 1) ( n − 2) (2)
y(n) =
u n v + nu ( n −1) v (1) + u v + ...
2!
Hence,
 nπ  − t  (n − 1)π 
y ( n ) =sin  t +  ( e ) + n sin  t + (− e )
−t
 2   2 
n(n − 1)  (n − 2)π  − t n(n − 1)(n − 2)  (n − 3)π 
+ sin  t +  (e ) + sin  t + (− e )
−t
2!  2  3!  2 

n(n − 1(n − 2)(n − 3)  (n − 4)π  − t


+
sin  t +  (e )
4!  2 
 4π   3π  4(3) − t 2π π
( e ) sin  t +  − ( e ) sin  t + 
4(3)(2) − t
= e − t sin  t +
and y (4)  − 4 e sin  t +
−t
+
 2   2  2!  2  3!  2
1289 © 2014, John Bird
4(3)(2)(1)
+ sin ( t )( e − t )
4!

 3π   π  −t
= e − t sin t − 4 e − t sin  t +  + 6 ( e ) sin ( t + π ) − 4 ( e ) sin  t +  + e sin t
−t −t
 2   2

= e − t sin t + 4 e − t cos t − 6 e − t sin t − 4 e − t cos t + e − t sin t

= −4 e − t sin t

6. If y = x 5 ln 2 x find y (3)

a! 5!
Since y = x5 ln 2 x then let u = x5 and v = ln 2x =
and u n = x a−n x 5− n
( a − n )! (5 − n)!

n(n − 1) ( n − 2) (2)
y(n) =
u n v + nu ( n −1) v (1) + u v + ...
2!
 5!  5!  1  n(n − 1) 5!  1 
y(n) =  x 5− n  ln 2 x + n x 6− n   + x7−n  − 
 (5 − n)!  (6 − n)! x 2! (7 − n)!  x2 
n(n − 1)(n − 2) 5!  2
+ x8− n  
3! (8 − n)!  x3 
5! 2 5!  1  3(2) 5! 4  1  3(2)(1) 5! 5  2 
Hence,=
y3 x ln 2 x + (3) x 3   + x − + x  
2! 3!  x  2! (4)!  x 2  3! 5!  x 3 

= 60 x 2 ln 2 x + 60 x 2 − 15 x 2 + 2 x 2

= 60 x 2 ln 2 x + 47 x 2

i.e. y (3) x 2 ( 47 + 60 ln 2 x )
=

0 show that 2 x 2 y ( n + 2) + ( 4n + 1) x y ( n +1) + ( 2n 2 − n + 3) y ( n ) =


7. Given 2 x 2 y '' + xy '+ 3 y = 0

Differentiating each term of 2 x 2 y ''+ xy '+ 3 y =


0 n times, using Leibniz’s theorem of equation (13),

 n(n − 1) ( n ) 
gives: 2  y ( n + 2) x 2 + n y ( n +1) ( 2 x ) + y (2) + 0  + { y ( n +1) ( x) + ny ( n ) (1) + 0} + 3 { y ( n ) } = 0
 2! 
i.e. 2 x 2 y ( n + 2) + 4n x y ( n +1) + 2n(n − 1) y ( n ) + x y ( n +1) + n y ( n ) + 3 y ( n ) = 0

i.e. 2 x 2 y ( n + 2) + (4n + 1) x y ( n +1) + (2n 2 − 2n + n + 3) y ( n ) = 0

or 2 x 2 y ( n + 2) + (4n + 1) x y ( n +1) + (2n 2 − n + 3) y ( n ) = 0

1290 © 2014, John Bird


8. If y = ( x3 + 2 x 2 ) e 2 x determine an expansion for y (5)

Since y = ( x3 + 2 x 2 ) e 2 x then let u = e 2 x and v = ( x 3 + 2 x 2 ) and u n = 2n e 2 x

n(n − 1) ( n − 2) (2)
y(n) =
u n v + nu ( n −1) v (1) + u v + ...
2!
n(n − 1) n − 2 2 x
Hence,=
y(n) ( 2n e2 x )( x3 + 2 x 2 ) + n2n−1 e2 x ( 3x 2 + 4 x ) + 2 e ( 6x + 4)
2!
n(n − 1)(n − 2) n −3 2 x
+ 2 e (6)
3!
5(4) 3 2 x 5(4)(3) 2 2 x
and y (5) 25 e 2 x ( x3 + 2 x 2 ) + (5) 24 e 2 x ( 3 x 2 + 4 x ) +
= 2 e (6x + 4) + 2 e (6)
2 3!
= e 2 x {25 x 3 + 26 x 2 + (16)15 x 2 + (16)(20 x) + 60 x(8) + (8)(40) + 240}

= e 2 x {25 x 3 + 304 x 2 + 800 x + 560}

= e 2 x {25 x3 + 24 (19 x 2 ) + 24 (50)( x) + 24 (35)}

= e 2 x 24 {2 x3 + 19 x 2 + 50 x + 35}

1291 © 2014, John Bird


EXERCISE 314 Page 869

d2 y dy
1. Determine the power series solution of the differential equation: + 2x +y=0
dx 2 dx
dy
using the Leibniz–Maclaurin method, given that at x = 0, y = 1 and =2
dx

d2 y dy
+ 2x +y=0
d x2 dx

(i) The differential equation is rewritten as: y′′ + 2xy′ + y = 0 and from the Leibniz theorem of

equation (13), page 865 of textbook, each term is differentiated n times, which gives:

y ( n + 2) + 2 { y ( n +1) ( x) + n y ( n ) (1) + 0} + y ( n ) =
0

i.e. y ( n + 2) + 2 x y ( n +1) + (2n + 1) y ( n ) =


0 (1)

(ii) At x = 0, equation (1) becomes:

y ( n + 2) + (2n + 1) y ( n ) =
0

from which, y ( n + 2) =
−(2n + 1) y ( n )

This is the recurrence formula

(iii) For n = 0, ( y '')0 = − ( y )0


n = 1, ( y ''')0 = −3 ( y ')0

n = 2, ( y (4) )0 =
−5 ( y '')0 =
5( y )0

n = 3, ( y (5) )0 = { }
−7 ( y ''')0 = −7 −3 ( y ')0 =3 × 7 ( y ')0

n = 4, ( y (6) )0 = { }
−9 ( y (4) )0 = −9 5 ( y )0 =−5 × 9 ( y )0

n = 5, ( y (7) )0 = { }
−11( y (5) )0 = −11 3 × 7 ( y ')0 =−3 × 7 ×11( y ')0

n = 6, ( y (8) )0 = { }
−13 ( y (6) )0 = −13 −5 × 9 ( y )0 = 5 × 9 ×13 ( y )0

x2 x3 x4
(iv) Maclaurin’s theorem is: y = ( y )0 + x ( y ' )0 + ( y '')0 + ( y ''')0 + ( y (4) )0 + ...
2! 3! 4!

Thus, y = ( y )0 + x ( y ' )0 +
x2
2!
{
− ( y )0 + }
x3
3!
{
−3 ( y ')0 +
x4
4!
}
5 ( y )0 +
x5
5!
{ }
3 × 7 ( y ' )0 { }
+
x6
6!
{
−5 × 9 ( y )0 +}x7
7!
{
−3 × 7 ×11( y ')0 }
1292 © 2014, John Bird
(v) Collecting similar terms together gives:
 x 2 5 x 4 5 × 9 x 6 5 × 9 ×13 x8 
y = ( y )0 1 − + − + − ...
 2! 4! 6! 8! 
 3 x3 3 × 7 x 5 3 × 7 ×11 x 7 
+ ( y ' )0  x − + − + ...
 3! 5! 7! 
dy
At x = 0, y = 1 and = 2, hence, ( y )0 = 1 and ( y ' )0 = 2 .
dx
d2 y dy
Hence, the power series solution of the differential equation: + 2x +y=0 is:
d x2 dx
 x 2 5 x 4 5 × 9 x 6 5 × 9 ×13 x8   3 x3 3 × 7 x 5 3 × 7 ×11 x 7 
y = 1 − + − + − ... + 2  x − + − + ...
 2! 4! 6! 8!   3! 5! 7! 

d2 y dy
2. Show that the power series solution of the differential equation: ( x + 1) + ( x − 1) − 2y =
0,
d x2 dx
using the Leibniz–Maclaurin method, is given by: y =1 + x 2 + e − x given the boundary
dy
conditions that at x = 0, y = 2 and = –1
dx

d2 y dy
( x + 1) + ( x − 1) − 2y =
0
dx 2 dx

(i) The differential equation is rewritten as: (x + 1) y′′ + (x – 1)y′ – 2y = 0 and from the Leibniz

theorem of equation (13), page 865 of textbook, each term is differentiated n times, which

gives:

{ y ( n+ 2) ( x + 1) + ny ( n+1) (1) + 0} + { y ( n+1) ( x − 1) + n y ( n ) (1) + 0} − 2 y ( n ) =


0

i.e. (x + 1) y ( n + 2) + (n + x − 1) y ( n +1) + (n − 2) y ( n ) =
0 (1)

(ii) At x = 0, equation (1) becomes:

y ( n + 2) + (n − 1) y ( n +1) + (n − 2) y ( n ) =
0

from which, y ( n + 2) = (1 − n) y ( n +1) + (2 − n) y ( n )

This is the recurrence formula

(iii) For n = 0, ( y=
(2)
)0 ( y (1) )0 + 2( y )0
n = 1, ( y (3) )0 = ( y (1) )0

1293 © 2014, John Bird


n = 2, − ( y (1) )0
− ( y (3) )0 =
( y (4) )0 =
n = 3, ( y (5) )0 =
−2 ( y (4) )0 − ( y (3) )0 =
2 ( y (1) )0 − ( y (1) )0 =
( y (1) )0
n = 4, −3 ( y (5) )0 − 2 ( y (4) )0 =
( y (6) )0 = − ( y (1) )0
−3 ( y (1) )0 + 2 ( y (1) )0 =

n = 5, ( y (7) )0 =
−4 ( y (6) )0 − 3 ( y (5) )0 =
4 ( y (1) )0 − 3 ( y (1) )0 =
( y (1) )0
n = 6, ( y (8) )0 =
−5 ( y (7) )0 − 4 ( y (6) )0 =
−5 ( y (1) )0 + 4 ( y (1) )0 =
− ( y (1) )0

x 2 (2) x3 x4
(iv) Maclaurin’s theorem is: y = ( y )0 + x ( y (1) )0 + ( y )0 + ( y (3) )0 + ( y (4) )0 + ...
2! 3! 4!
x2
{( y ) } x3
{( y ) } + x4! {− ( y ) } + x5! {( y ) }
4 5
Thus, y = ( y )0 + x ( y (1) )0 + (1)
0
+ 2 ( y )0 + (1)
0
(1)
0
(1)
0
2! 3!
x6
{ x7
} {( y ) } + x8 {− ( y ) } + ...
8
+ − ( y (1) )0 + (1)
0
(1)
0
6! 7!
(v) Collecting similar terms together gives:
 x2   x 2 x3 x 4 x5 x6 x7 
y = ( y )0 1 + (2)  + ( y (1) )0  x + + − + − + − ...
 2!   2! 3! 4! 5! 6! 7! 
dy
At x = 0, y = 2 and = –1, hence, ( y )0 = 2 and ( y (1) )0 = −1
dx
d2 y dy
Hence, the power series solution of the differential equation: ( x + 1) + ( x − 1) − 2y =
0
dx 2 dx
is:
 x 2 x3 x 4 x5 x6 x7 
y = 2 {1 + x 2 } −  x + + − + − + − ...
 2! 3! 4! 5! 6! 7! 

x 2 x3 x 4 x5 x6 x7
= 2 + 2x 2 – x – − + − + − + ...
2! 3! 4! 5! 6! 7!

x 2 x3 x 4 x5 x6 x7
= 1 + x2 + 1 + x2 – x – − + − + − + ...
2! 3! 4! 5! 6! 7!

x 2 x3 x 4 x5 x6 x7
= 1 + x2 + 1 – x + − + − + − + ...
2! 3! 4! 5! 6! 7!

x 2 x3 x 4 x5 x6 x7
i.e. y = 1 + x 2 + e −x since e − x = 1 – x + − + − + − + ...
2! 3! 4! 5! 6! 7!

d2 y dy
3. Find the particular solution of the differential equation: ( x 2 + 1) +x − 4y =
0 using the
d x2 dx
dy
Leibniz–Maclaurin method, given the boundary conditions that at x = 0, y = 1 and =1
dx

1294 © 2014, John Bird


d2 y dy
( x 2 + 1) +x − 4y =
0
d x2 dx

i.e. ( x 2 + 1) y′′ + xy′ – 4y = 0

 2 n(n − 1) ( n ) 
i.e. ( x + 1) y ( n + 2) + ny ( n +1) (2 x) + y (2)  + { y ( n +1) x + ny n (1)} − 4 y ( n ) =
0
 2! 

i.e. ( x 2 + 1) y ( n+ 2) + ( 2nx + x ) y ( n+1) + (n(n − 1) + n − 4) y ( n ) =0

At x = 0, y ( n + 2) + ( n 2 − 4 ) y ( n ) =
0

from which, y ( n + 2)= ( 4 − n2 ) y (n) which is the recurrence formula

For n = 0, ( y '')0 = 4 ( y )0
n = 1, ( y ''')0 = 3 ( y ')0
n = 2, ( y (4) )0 = 0
n = 3, ( y (5) )0 = {
−5 ( y ''')0 = −5 −3 ( y ')0 = }
−5 ( 3)( y ')0

n = 4, ( y (6) )0 = −12 ( y (4) )0 = −12(0) =


0

n = 5, ( y (7) )0 = { }
−21( y (5) )0 = −21 −5 × 3 ( y ')0 =315 ( y ')0

x2 x3 x4
Maclaurin’s theorem is: y = ( y )0 + x ( y ' )0 + ( y '')0 + ( y ''')0 + ( y (4) )0 + ...
2! 3! 4!

Thus, y = ( y )0 + x ( y ')0 +
x2
2!
{ }
4 ( y )0 +
x3
3!
{ x4
3 ( y ')0 + {0} +
4!
}x5
5!
{
−3 × 5 ( y ')0 + 0 +
x7
7!
}
315 ( y ')0 { }
 x3 x5 x 7 
i.e. y = ( y )0 {1 + 2 x 2 } + ( y ')0  x + − + + ...
 2 8 16 
dy
At x = 0, y = 1 and = 1, hence, ( y )0 = 1 and ( y ' )0 = 1
dx
d2 y dy
Hence, the power series solution of the differential equation: ( x 2 + 1) +x − 4y =
0 is:
dx 2 dx
 x3 x5 x7 
y = {1 + 2x 2 } +  x + − + + ...
 2 8 16 

x3 x5 x7
i.e. y = 1 + x + 2x2 + − + + ...
3 8 16

4. Use the Leibniz–Maclaurin method to determine the power series solution for the differential

1295 © 2014, John Bird


d2 y d y dy
equation: x + + xy =
1 given that at x = 0, y = 1 and =2
d x2 d x dx

d2 y d y
x + + xy =
1
d x2 d x

i.e. x y′′ + y′ + xy = 0

i.e. { xy ( n+ 2) + ny ( n+1) (1)+)} + { y ( n+1) } + { xy ( n ) + ny ( n−1) (1)} =0

i.e. xy ( n + 2) + ( n + 1) y ( n +1) + xy ( n ) + ny ( n −1) =


0

At x = 0, ( n + 1) y ( n+1) + ny ( n−1) =
0

n ( n −1)
from which, y ( n +1) = − y which is the recurrence formula
n +1

1
For n = 1, ( y (2) )0 = − ( y )0
2
2 (1)
n = 2, ( y (3) )0 = − ( y )0
3
3 3 1  3
n = 3, − ( y (2) )0 =
( y (4) )0 = − − ( y )0  =( y )0
4 4 2  8
4 4  2 (1)  8 (1)
n = 4, ( y (5) )0 =
− ( y (3) )0 =
−  − ( y )0  = ( y )0
5 5 3  15
5  5 3 15
n = 5, − ( y (4) )0 =
( y (6) )0 = − ( y )0
−  ( y )0  =
6 6 8  48
6  6 8 16
n = 6, ( y (7) )0 =
− ( y (5) )0 = ( y (1) )0
−  ( y (1) )0  =
7 7 15  35
x 2 (2) x3 x4
Maclaurin’s theorem is: y = ( y )0 + x ( y (1) )0 + ( y )0 + ( y (3) )0 + ( y (4) )0 + ...
2! 3! 4!
Thus,
x2  1  x3  2  x4  3  x5  8 
y = ( y )0 + x ( y (1) )0 + − ( y )0  + − ( y (1) )0  +  ( y )0  +  ( y (1) )0 
2!  2  3!  3  4!  8  5! 15 
x 6  15  x 7  16 
− ( y )0  +  ( y (1) )0  + ...
6!  48  7!  35 
 1 1 4 1 6   x3 x5 x7 
i.e. y = ( y )0 1 − x 2 + x − x + ... + ( y (1) )0  x − + − + ...
 4 64 2304   9 225 11025 
dy
At x = 0, y = 1 and = 2, hence, ( y )0 = 1 and ( y (1) )0 = 2
dx
1296 © 2014, John Bird
d2 y d y
Hence, the power series solution of the differential equation: x + + xy =
1 is:
d x2 d x
 1 1 4 1   x3 x5 x7 
y = 1 − x 2 + x − x 6 + .. + 2  x − + − + ...
 4 64 2304   9 225 1025 

and it may be shown that this is equivalent to:

 1 1 1   x3 x5 x7 
y = 1 − x 2 + x4 − x 6 + ... + 2 x − + − ...
 2 2 2 ×4
2 2 2 ×4 ×6
2 2 2
  32 3 ×5 3 ×5 ×7
2 2 2 2 2

1297 © 2014, John Bird


EXERCISE 315 Page 875

1. Produce, using Frobenius’ method, a power series solution for the differential equation:
d2 y d y
2x + −y=0
d x2 d x

d2 y d y
2x + −y=0 may be rewritten as: 2xy′′ + y′ – y = 0
d x2 d x

(i) Let a trial solution be of the form y = xc{a 0 + a 1 x + a 2 x2 + a 3 x3 + … + a r xr+…}


where a 0 ≠ 0,

i.e. y = a 0 xc + a 1 xc+1 + a 2 xc+2 + a 3 xc+3 + … + a r xc+r +…

(ii) Differentiating gives:

y′ = a 0 cxc–1 + a 1 (c + 1)xc + a 2 (c + 2)xc+1 + … + a r (c + r)xc+r–1 + …

and y′′ = a 0 c(c – 1)xc–2 + a 1 c(c + 1)xc–1 + a 2 (c + 1)(c + 2)xc + …. + a r (c + r – 1)(c + r)xc+r–2 +

(iii) Substituting y, y′ and y′′ into each term of the given equation 2xy′′ + y′ – y = 0 gives:

2xy′′ = 2a 0 c(c – 1)xc–1 + 2a 1 c(c + 1)xc + 2a 2 (c + 1)(c + 2)xc+1 + …


+ 2a r (c + r –1)(c + r)xc+r–1 + … (a)

y′ = a 0 cxc–1 + a 1 (c + 1)xc + a 2 (c + 2)xc+1 + …. + a r (c + r)xc+r–1 + …

(b)

–y = –a 0 xc – a 1 xc+1 – a 2 xc+2 – a 3 xc+3 – … – a r xc+r –… (c)

(iv) The sum of these three terms forms the left-hand side of the equation. Since the right-hand side

is zero, the coefficients of each power of x can be equated to zero.

For example, the coefficient of xc–1 is equated to zero, giving: 2a 0 c(c – 1) + a 0 c = 0

or a 0 c [2c – 2 + 1] = a 0 c(2c – 1) = 0 (1)

1
Equation (1) is the indicial equation, from which, c = 0 or c =
2
The coefficient of xc is equated to zero, giving: 2a 1 c(c + 1) + a 1 (c + 1) – a 0 = 0

i.e. a 1 (2c2 + 2c + c + 1) – a 0 = a 1 (2c2 + 3c + 1) – a 0 = 0

1298 © 2014, John Bird


or a 1 (2c + 1)(c + 1) – a 0 = 0 (2)

Replacing r by (r + 1) will give:

in series (a), 2a r+1 (c + r + 1)(c + r)xc+r

in series (b), a r+1 (c + r + 1)xc+r

in series (c), –a r xc+r

Equating the total coefficients of xc+r to zero gives:

2a r+1 (c + r + 1)(c + r) + a r+1 (c + r + 1) – a r = 0

which simplifies to: a r+1 {(c + r + 1)(2c + 2r + 1)} – a r = 0 (3)

(a) When c = 0:

From equation (2), if c = 0, a 1 (1 × 1) – a 0 = 0, i.e. a 1 = a0


ar
From equation (3), if c = 0, a r+1 (r + 1)(2r + 1) – a r = 0, i.e. a r+1 = r≥0
(r + 1)(2r + 1)
a1 a0
Thus, when r = 1,=a2 = since a1 = a0
(2 × 3) (2 × 3)
a2 a0
when r = 2,=a3 =
(3 × 5) (2 × 3)(3 × 5)
a3 a0 a0
when r = 3,=a4 = = and so on
(4 × 7) (2 × 3)(3 × 5)(4 × 7) (2 × 3 × 4)(3 × 5 × 7)
The trial solution is: y = xc{a 0 + a 1 x + a 2 x2 + a 3 x3 + … + a r xr +…}

Substituting c = 0 and the above values of a 1 , a 2 , a 3 , … into the trial solution gives:
  a0  2  a0  3  a0  4 
y = x 0 a0 + a0 x +  x + x +  x + ...
  (2 × 3)   (2 × 3)(3 × 5)   (2 × 3 × 4)(3 × 5 × 7)  
 x2 x3 x4 
i.e. y = a0 1 + x + + + + ... (4)
 ( 2 × 3) ( 2 × 3)( 3 × 5) ( 2 × 3 × 4 )( 3 × 5 × 7 ) 
1
(b) When c = :
2

, a 1 ( 2 )  3  – a 0 = 0,
1 a0
From equation (2), if c = i.e. a 1 =
2 2 3

, a r+1  1 + r + 1 (1 + 2r + 1) – a r = 0,
1
From equation (3), if c =
2 2 

i.e. a r+1  r + 3  ( 2r + 2 ) – a r = a r+1 (2 r 2 + 5r +3) – a r = 0,


 2
1299 © 2014, John Bird
ar
i.e. a r+1 = r≥0
(2r + 3)(r + 1)
a1 a0 a0
Thus, when r = 1,=a2 = since a 1 =
(2 × 5) (2 × 3 × 5) 3
a2 a0
when r = 2,=a3 =
(3 × 7) (2 × 3 × 5)(3 × 7)
a3 a0
when r = 3,=a4 = and so on
(4 × 9) (2 × 3 × 4)(3 × 5 × 7 × 9)
The trial solution is: y = xc{a 0 + a 1 x + a 2 x2 + a 3 x3 + … + a r xr +…}

1
Substituting c = and the above values of a 1 , a 2 , a 3 , … into the trial solution gives:
2
1  
 a0   a0  2  a0  3  a0  4
y = x 2 a0 +   x +  x + x +  x + ...
  3  2 × 3× 5   (2 × 3 × 5)(3 × 7)   (2 × 3 × 4)(3 × 5 × 7 × 9)  
1  
x x2 x3 x4
i.e. y = a0 x 2 1 + + + + + ... (5)
 (1× 3) (1× 2)(3 × 5) (1× 2 × 3)(3 × 5 × 7) (1× 2 × 3 × 4)(3 × 5 × 7 × 9) 
Let a0 = A in equation (4), and a0 = B in equation (5)

 x2 x3 x4 
Hence, y = A 1 + x + + + + ...
 ( 2 × 3) ( 2 × 3)( 3 × 5) ( 2 × 3 × 4 )( 3 × 5 × 7 ) 
1  
x x2 x3 x4
+ B x 2 1 + + + + + ...
 (1× 3) (1× 2)(3 × 5) (1× 2 × 3)(3 × 5 × 7) (1× 2 × 3 × 4)(3 × 5 × 7 × 9) 

2. Use the Frobenius method to determine the general power series solution of the differential
d2 y
equation: +y=0
d x2

The differential equation may be rewritten as: y′′ + y = 0

(i) Let a trial solution be of the form y = xc{a 0 + a 1 x + a 2 x2 + a 3 x3 + … + a r xr +…} (1)


where a 0 ≠ 0,

i.e. y = a 0 xc + a 1 xc+1 + a 2 xc+2 + a 3 xc+3 + … + a r xc+r +… (2)

(ii) Differentiating equation (2) gives:

y′ = a 0 cxc–1 + a 1 (c + 1)xc + a 2 (c + 2)xc+1 + …. + a r (c + r)xc+r–1 + …

1300 © 2014, John Bird


and y′′ = a 0 c(c – 1)xc–2 + a 1 c(c + 1)xc–1 + a 2 (c + 1)(c + 2)xc + … + a r (c + r – 1)(c + r)xc+r–2 +

(iii) Replacing r by (r + 2) in a r (c + r – 1)(c + r)xc+r–2 gives: a r+2 (c + r + 1)(c + r + 2)xc+r

Substituting y and y′′ into each term of the given equation y′′ + y = 0 gives:

y′′ + y = a 0 c(c – 1)xc–2 + a 1 c(c + 1)xc–1 + [a 2 (c + 1)(c + 2) + a 0 ]xc + …

+ [a r+2 (c + r + 1)(c + r + 2) + a r ] xc+r + … = 0 (3)

(iv) The indicial equation is obtained by equating the coefficient of the lowest power of x to zero.

Hence, a 0 c(c – 1) = 0 from which, c = 0 or c = 1 since a 0 ≠ 0

For the term in xc–1, i.e. a 1 c(c + 1) = 0

With c = 1, a 1 = 0; however, when c = 0, a 1 is indeterminate, since any value of a 1 combined

with the zero value of c would make the product zero

− a0
For the term in xc, a 2 (c + 1)(c + 2) + a 0 = 0 from which, a2 = (4)
(c + 1)(c + 2)
For the term in xc+r, a r+2 (c + r + 1)(c + r + 2) + a r = 0
− ar
from which, ar + 2 = (5)
(c + r + 1)(c + r + 2)
(a) When c = 0: a 1 is indeterminate, and from equation (4)
− a0 − a0
=a2 =
(1× 2) 2!
− ar − a1 − a1 − a1
In general, ar + 2 = =
and when r = 1, a3 = =
(r + 1)(r + 2) (2 × 3) (1× 2 × 3) 3!
− a2 a0
when r = 2,=
a4 =
3× 4 4!
−a1

− a3 3! a1
when r = 3,=
a5 = =
4×5 4 × 5 5!
 a0 a1 a0 a1 
Hence, y = x 0 a0 + a1 x − x 2 − x 3 + x 4 + x 5 ... from equation (1)
 2! 3! 4! 5! 
 x2 x4   x3 x5 
= a0 1 − + − ... + a1  x − + − ...
 2! 4!   3! 5! 
Since a0 and a1 are arbitrary constants depending on boundary conditions, let a0 = A and

1301 © 2014, John Bird


 x2 x4   x3 x5 
a1 = B, then: y = A 1 − + − ... + B  x − + − ... (6)
 2! 4!   3! 5! 
− a0 − a0
(b) When c = 1: a 1 = 0, and from equation (4),=a2 =
(2 × 3) 3!
− ar − ar
=
Since c = 1, ar + 2 = from equation (5)
(c + r + 1)(c + r + 2) (r + 2)(r + 3)
− a1
and when r = 1, a3 = = 0 since a 1 = 0
(3 × 4)

 a0 
−− 
− a2 =3!  a0
when r = 2,=a4 =
(4 × 5) 4×5 5!
− a3
when r = 3,=a5 = 0
(5 × 6)

 a0 a0 
Hence, when c = 1, y = x1 a0 − x 2 + x 4 + ... from equation (1)
 3! 5! 
 x3 x5 
i.e. y = a0  x − + + ...
 3! 5! 
Again, a0 is an arbitrary constant; let a0 = K,

 x3 x5 
then y = K  x − + − ...
 3! 5! 
However, this latter solution is not a separate solution, for it is the same form as the second series in

equation (6) above. Hence, equation (6) with its two arbitrary constants A and B gives the general

solution.

d2 y
Hence the general power series solution of the differential equation: +y=0 is given by:
d x2
 x2 x4   x3 x5 
y = A 1 − + − ... + B  x − + − ...
 2! 4!   3! 5! 

or y = P cos x + Q sin x from the series expansions of cos x and sin x

d2 y dy
3. Determine the power series solution of the differential equation: 3 x +4 −y=0
dx 2 dx
using the Frobenius method.

d2 y dy
3x +4 −y=0 may be rewritten as: 3xy′′ + 4y′ – y = 0
dx 2 dx
1302 © 2014, John Bird
(i) Let a trial solution be of the form y = xc{a 0 + a 1 x + a 2 x2 + a 3 x3 + … + a r xr +…}

i.e. y = a 0 xc + a1 xc+1 + a 2 xc+2 + a 3 xc+3 + … + a r xc+r +…

(ii) Differentiating gives:

y′ = a 0 cxc–1 + a 1 (c + 1)xc + a 2 (c + 2)xc+1 + … + a r (c + r)xc+r–1 + …

and y′′ = a 0 c(c – 1)xc–2 + a 1 c(c + 1)xc–1 + a 2 (c + 1)(c + 2)xc + … + a r (c + r – 1)(c + r)xc+r–2 +

(iii) Substituting y, y′ and y′′ into each term of the given equation 3xy′′ + 4y′ – y = 0 gives:

3xy′′ = 3a 0 c(c – 1)xc–1 + 3a 1 c(c + 1)xc + 3a 2 (c + 1)(c + 2)xc+1 + …


+ 3a r (c + r –1)(c + r)xc+r–1 + … (a)

4y′ = 4a 0 cxc–1 + 4a 1 (c + 1)xc + 4a 2 (c + 2)xc+1 + … + 4a r (c + r)xc+r–1 + … (b)

–y = –a 0 xc – a 1 xc+1 – a 2 xc+2 – a 3 xc+3 – … – a r xc+r – …

(c)

(iv) The coefficient of xc–1 is equated to zero giving: 3a 0 c(c – 1) + 4a 0 c = 0

or a 0 c [3c – 3 + 4] = a 0 c(3c + 1) = 0

1
This is the indicial equation, from which, c = 0 or c = −
3
The coefficient of xc is equated to zero giving: 3a 1 c(c + 1) + 4a 1 (c + 1) – a 0 = 0

i.e. a 1 (3c(c + 1) +4(c+1)) – a 0 = a 1 (c + 1)(3c + 4) – a 0 = 0

or a 1 (c + 1)(3c + 4) – a 0 = 0 (1)

Equating the total coefficients of xc+r to zero gives:

3a r+1 (c + r)(c + r + 1) + 4a r+1 (c + r + 1) – a r = 0

i.e. a r+1 (c + r + 1)(3c + 3r + 4) – a r = 0

ar
which simplifies to: ar +1 = (2)
(c + r + 1)(3c + 3r + 4)

(a) When c = 0:

a0
From equation (1), if c = 0, a 1 (4) – a 0 = 0, i.e. a 1 =
4

1303 © 2014, John Bird


ar
From equation (2), if c = 0, ar +1 = r≥0
(r + 1)(3r + 4)
a1 a0 a0
=
Thus, when r = 1, a2 = since a1 =
(2 × 7) (2 × 4 × 7) 4

a2 a0 a0
when r = =
2, a3 = =
(3 ×10) (3 ×10)(2 × 4 × 7) (1× 2 × 3)(4 × 7 ×10)
a3 a0 a0
when r = =
3, a4 = = and so on
(4 ×13) (4 ×13)(3 ×10)(2 × 4 × 7) (2 × 3 × 4)(4 × 7 ×10 ×13)
The trial solution is: y = xc{a 0 + a 1 x + a 2 x2 + a 3 x3 + … + a r xr +…}

Substituting c = 0 and the above values of a 1 , a 2 , a 3 , … into the trial solution gives:
 a0  a0  2  a0  3  a0  4 
y = x 0 a0 + x +  x + x +  x + ...
 4  (1× 2)(4 × 7)   (1× 2 × 3)(4 × 7 × 10)   (2 × 3 × 4)(4 × 7 × 10 × 13)  
 x x2 x3 x4 
i.e. y = a0 1 + + + + + ... (3)
 (1× 4) (1× 2)(4 × 7 ) (1× 2 × 3)( 4 × 7 ×10 ) (2 × 3 × 4)(4 × 7 ×10 ×13) 
1
(b) When c = − :
3

From equation (1), if c = − , a 1  2  ( 3) – a 0 = 0,


1 a0
i.e. a 1 =
3 3 2
1 ar ar ar
c = − , ar +1 =
From equation (2), if= = r≥0
 2 1 + +
 r +  ( 3r + 3) 3 (3r + 2)3(r + 1)
3 (3r 2)( r 1)
 3
a1 a0 a0 a0
Thus, when r = 1,=a2 = = since a 1 =
(5 × 2) (22 × 5) (1× 2)(2 × 5) 2
a2 a0
when r = 2,=a3 =
(8 × 3) (1× 2 × 3)(2 × 5 × 8)
a3 a0
=
when r = 3, a4 = and so on
(11× 4) (1× 2 × 3 × 4)(2 × 5 × 8 ×11)
The trial solution is: y = xc{a 0 + a 1 x + a 2 x2 + a 3 x3 + … + a r xr +…}

1
Substituting c = − and the above values of a 1 , a 2 , a 3 , … into the trial solution gives:
3
−  
1
a0 a0 a0 a0
y = x 3 a0 + x + x2 + x3 + + ...
 2 (1× 2)(2 × 5) (1× 2 × 3)(2 × 5 × 8) (1× 2 × 3 × 4)(2 × 5 × 8 ×11) 

−  
1
x x2 x3 x4
i.e. y = a0 x 3 1 + + + + + ... (4)
 (1× 2) (1× 2)(2 × 5) (1× 2 × 3)(2 × 5 × 8) (1× 2 × 3 × 4)(2 × 5 × 8 ×11) 
Let a0 = A in equation (3), and a0 = B in equation (4)

1304 © 2014, John Bird


 x x2 x3 
Hence, y = A 1 + + + + ...
 (1× 4 ) (1× 2 )( 4 × 7 ) (1× 2 × 3)( 4 × 7 ×10 ) 

−  
1
x x2 x3
+ B x 3 1 + + + + ...
 (1× 2) (1× 2)(2 × 5) (1× 2 × 3)(2 × 5 × 8) 

4. Show, using the Frobenius method, that the power series solution of the differential equation:
d2 y
−y=0 may be expressed as y = P cosh x + Q sinh x, where P and Q are constants. [Hint:
d x2
check the series expansions for cosh x and sinh x on page 221.]

The differential equation may be rewritten as: y′′ – y = 0

(i) Let a trial solution be of the form y = xc{a 0 + a 1 x + a 2 x2 + a 3 x3 + … + a r xr +…} (1)


where a 0 ≠ 0,

i.e. y = a 0 xc + a 1 xc+1 + a 2 xc+2 + a 3 xc+3 + … + a r xc+r +… (2)

(ii) Differentiating equation (2) gives:

y′ = a 0 cxc–1 + a 1 (c + 1)xc + a 2 (c + 2)xc+1 + … + a r (c + r)xc+r–1 + …

and y′′ = a 0 c(c – 1)xc–2 + a 1 c(c + 1)xc–1 + a 2 (c + 1)(c + 2)xc + … + a r (c + r – 1)(c + r)xc+r–2 +

(iii) Replacing r by (r + 2) in a r (c + r – 1)(c + r)xc+r–2 gives: a r+2 (c + r + 1)(c + r + 2)xc+r

Substituting y and y′′ into each term of the given equation y′′ – y = 0 gives:

y′′ – y = a 0 c(c – 1)xc–2 + a 1 c(c + 1)xc–1 + [a 2 (c + 1)(c + 2) – a 0 ]xc + …

+ [a r+2 (c + r + 1)(c + r + 2) – a r ] xc+r + … = 0 (3)

(iv) The indicial equation is obtained by equating the coefficient of the lowest power of x to zero

Hence, a 0 c(c – 1) = 0 from which, c = 0 or c = 1 since a 0 ≠ 0

For the term in xc–1, i.e. a 1 c(c + 1) = 0

With c = 1, a 1 = 0; however, when c = 0, a 1 is indeterminate, since any value of a 1 combined

with the zero value of c would make the product zero


1305 © 2014, John Bird
a0
For the term in xc, a 2 (c + 1)(c + 2) – a 0 = 0 from which, a2 = (4)
(c + 1)(c + 2)
For the term in xc+r, a r+2 (c + r + 1)(c + r + 2) – a r = 0
ar
from which, ar + 2 = (5)
(c + r + 1)(c + r + 2)
(a) When c = 0: a 1 is indeterminate, and from equation (4)
a0 a0
=a2 =
(1× 2) 2!
ar a1 a1 a1
In general, ar + 2 = =
and when r = 1, a3 = =
(r + 1)(r + 2) (2 × 3) (1× 2 × 3) 3!
a2 a0
when r = 2,=
a4 =
3× 4 4!
a1
a3 3! a1
when r = 3,=
a5 = =
4 × 5 4 × 5 5!
 a0 a1 a0 a1 
Hence, y = x 0 a0 + a1 x + x 2 + x3 + x 4 + x 5 ... from equation (1)
 2! 3! 4! 5! 
 x2 x4   x3 x5 
= a0 1 + + + ... + a1  x + + + ...
 2! 4!   3! 5! 
Since a0 and a1 are arbitrary constants depending on boundary conditions, let a0 = A and

 x2 x4   x3 x5 
a1 = B, then: y = A 1 + + + ... + B  x + + + ... (6)
 2! 4!   3! 5! 
−a0 a0
(b) When c = 1: a 1 = 0, and from equation (4),=a2 =
(2 × 3) 3!
ar ar
=
Since c = 1, ar + 2 = from equation (5)
(c + r + 1)(c + r + 2) (r + 2)(r + 3)
a1
and when r = 1, a3 = = 0 since a 1 = 0
(3 × 4)

 a0 
  a
= =
a2 3! 
when r = 2, =
0
a4
(4 × 5) 4 × 5 5!
a3
when r = 3,=a5 = 0
(5 × 6)

 a0 a0 
Hence, when c = 1, y = x1 a0 + x 2 + x 4 + ... from equation (1)
 3! 5! 
 x3 x5 
i.e. y = a0  x + + + ...
 3! 5! 
1306 © 2014, John Bird
Again, a0 is an arbitrary constant; let a0 = K,

 x3 x5 
then y = K  x + + − ...
 3! 5! 
However, this latter solution is not a separate solution, for it is the same form as the second series in

equation (6) above. Hence, equation (6) with its two arbitrary constants A and B gives the general

solution

d2 y
Hence the general power series solution of the differential equation: +y=0 is given by:
d x2
 x2 x4   x3 x5 
y = A 1 + + + ... + B  x + + + ...
 2! 4!   3! 5! 

or y = P cosh x + Q sinh x from the series expansions of cosh x and sinh x

1307 © 2014, John Bird


EXERCISE 316 Page 879

d2 y dy
1. Determine the power series solution of Bessel’s equation: x 2 +x + ( x2 − v2 ) y =
0
dx 2 dx
when v = 2, up to and including the term in x 4

d2 y dy
The complete solution of Bessel’s equation: x 2 +x + ( x2 − v2 ) y =
0 is:
d x2 dx
 x2 x4 x6 
y = A x v 1 − + − + ...
 22 (v + 1) 24 × 2!(v + 1)(v + 2) 26 × 3!(v + 1)(v + 2)(v + 3) 
 x2 x4 x6 
+ B x − v 1 + + + + ...
 2 (v − 1) 2 × 2!(v − 1)(v − 2) 2 × 3!(v − 1)(v − 2)(v − 3)
2 4 6

 x2 x4 x6 
and y = A x v 1 − + − + ... when v is a
 2 (v + 1) 2 × 2!(v + 1)(v + 2) 2 × 3!(v + 1)(v + 2)(v + 3)
2 4 6

positive integer
 x2 x4 
Hence, when v = 2, y = A x 2 1 − + + ...
 22 (2 + 1) 24 × 2!(2 + 1)(2 + 2) 
 x2 x4   x4 x6 
i.e. y = A x 2 1 − + − ... or A  x2 − + − ...
 12 384   12 384 

2. Find the power series solution of the Bessel function: x 2 y ''+ xy '+ ( x 2 − v 2 ) y =
0 in terms of the

Bessel function J 3 ( x) when v = 3. Give the answer up to and including the term in x 4

x  1 
v
x2 x4
J v ( x) =    − + − ... provided v is not a negative integer
 2   Γ(v + 1) 22 (1!)Γ(v + 2) 24 (2!)Γ(v + 3) 

x  1 
3
x2 x4
Hence, when v = 3, J 3 ( x) =    − + − ...
 2   Γ(3 + 1) 22 (1!)Γ(3 + 2) 24 (2!)Γ(3 + 3) 

3
x  1 x2 x4  x3 x5 x7
i.e. J 3 ( x) =    − + − ... or − + − ...
 2   Γ 4 2 Γ5 2 Γ 6
2 5
 8 Γ 4 25 Γ5 28 Γ6

3. Evaluate the Bessel functions J 0 ( x) and J1 ( x) when x = 1, correct to 3 decimal places.

1308 © 2014, John Bird


x2 x4 x6
J 0 ( x) = 1 − + − + ...
22 (1!) 2 24 ( 2!)2 26 (3!) 2

12 14 16
and when x = 1, J 0 ( x) = 1 − + − + ...
22 (1!) 2 24 ( 2!)2 26 (3!) 2

= 1 – 0.25 + 0.015625 – 0.000434 + …

= 0.765 correct to 3 decimal places

x x3 x5 x7
J1 ( x ) = − + − + ...
2 23 (1!)(2!) 25 (2!)(3!) 27 (3!)(4!)

1 13 15 17
and when x = 1, J1 ( x) = − + − + ...
2 23 (1!)(2!) 25 (2!)(3!) 27 (3!)(4!)

= 0.5 – 0.0625 + 0.002604 – 0.000054

= 0.440 correct to 3 decimal places

1309 © 2014, John Bird


EXERCISE 317 Page 883

1. Determine the power series solution of the Legendre equation: (1 − x 2 ) y ''− 2 xy '+ k (k + 1) y =
0

when (a) k = 0 (b) k = 2, up to and including the term in x5

The power series solution of the Legendre equation is:

 k (k + 1) 2 k (k + 1)(k − 2)(k + 3) 4 
y = a0 1 − x + x − ...
 2! 4! 
 (k − 1)(k + 2) 3 (k − 1)(k − 3)(k + 2)(k + 4) 5 
+ a1  x − x + x − ...
 3! 5! 
 ( −1)( +2) ( −1)( −3)( +2)( +4) 
(a) When k = 0, y = a0 {1 − 0 + 0 − ...} + a1  x − x3 + x 5 − ...
 3! 5! 
 x3 x5 
i.e. y = a0 + a1  x + + + ... 
 3 5 
 2(3) 2 2(3)(0)(5) 4   (1)(4) 3 (1)(−1)(4)(6) 5 
(b) When k = 2, y = a0 1 − x + x − ... + a1  x − x + x − ...
 2! 4!   3! 5! 
 2 1 
i.e. y = a0 (1 − 3 x 2 ) + a1  x − x 3 − x5 − ... 
 3 5 

2. Find the following Legendre polynomials: (a) P1 ( x) (b) P4 ( x) (c) P5 ( x)

(a) Since in P1 ( x) , n = k = 1, then from the second part of equation (47), page 881 of textbook, i.e.
the odd powers of x:
y = a1 { x − 0} = a1 x

a1 is chosen to make y = 1 when x = 1


i.e. 1 = a1
Hence, P1 ( x) = x
(b) Since in P4 ( x) , n = k = 4, then from the first part of equation (47), page 881 of textbook, i.e. the
even powers of x:
 4(5) 2 4(5)(2)(7) 4   35 
y = a0 1 − x + x + 0  = a0 1 − 10 x 2 + x 4 
 2! 4!   3 
a0 is chosen to make y = 1 when x = 1
 35   2 8 3
i.e. 1 = a0 1 − 10 + = a0 1 − 10 + 11 = a0 , from which, a0 =
 3   3 3 8
1310 © 2014, John Bird
3 35 
Hence, P4 ( x) = 1 − 10 x 2 + x 4 
8 3 
1
or P4 ( x) = ( 35 x 4 − 30 x 2 + 3)
8

(c) Since in P5 ( x) , n = k = 5, then from the second part of equation (47), i.e. the odd powers of x:

 (k − 1)(k + 2) 3 (k − 1)(k − 3)(k + 2)(k + 4) 5 


y = a1  x − x + x − ...
 3! 5! 
 (4)(7) 3 (4)(2)(7)(9) 5   14 21 
i.e. y = a1  x − x + x − 0 = a1  x − x 3 + x 5 
 3! 5!   3 5 

a1 is chosen to make y = 1 when x = 1

 14 21   15 − 70 + 63  8 15
i.e. 1 = a1 =
1 − +  a1  =  a1 from which, a1 =
 3 5  15  15 8
15  14 3 21 5 
Hence, P5 ( x) = x− x + x 
8 3 5 
1
or P5 ( x) = ( 63x5 − 70 x3 + 15 x )
8

1311 © 2014, John Bird

You might also like