0% found this document useful (0 votes)
19 views31 pages

Chapter 3

The document discusses using power series methods to solve differential equations. It provides examples of solving equations like y''+y=0 and y'+y=x^2 by assuming a power series solution and equating coefficients. It also discusses finding Taylor series representations of common functions.

Uploaded by

Dilasha Adhikari
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)
19 views31 pages

Chapter 3

The document discusses using power series methods to solve differential equations. It provides examples of solving equations like y''+y=0 and y'+y=x^2 by assuming a power series solution and equating coefficients. It also discusses finding Taylor series representations of common functions.

Uploaded by

Dilasha Adhikari
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/ 31

Differential Equations & Complex Variables

Saraswati Acharya, PhD


Department of Mathematics
School of Science, Kathmandu University
Kavre, Dhulikhel

Chapter 3:Series Solution of Differential Equations


Course Contents

Unit 3: Series Solutions of Differential Equations:


1 Introduction, Power series solution of D.E.,
2 Legendre’s equation and functions,
3 Bessels’ equations and functions, (Derivation not required),
Associated properties and its verifications.

Saraswati Acharya (Department MATH207


of Mathematics[2mm]School of Science, Kathmandu University[2mm]
2 / 31 K
Power Series

Definition: A power series in (x − a) is an infinite series of the form



X
cn (x − a)n = c0 + c1 (x − a) + c2 (x − a)2 + · · ·
n=0

where c0 , c1 , · · · are constants, called the coefficients of the series, a is a


constant called the center, and x is a variable.
In particular, a = 0, we obtain a power series in powers of x:

X
cn x n = c0 + c1 x + c2 x 2 + · · ·
n=0

Saraswati Acharya (Department MATH207


of Mathematics[2mm]School of Science, Kathmandu University[2mm]
3 / 31 K
Some Important Series

Taylor series: The Taylor series of f (x) about x = a is



X f n (a)
f (x) = (x − a)n
n!
n=0

f 0 (a) f 00 (a) f n (a)


= f (a) + (x − a) + (x − a)2 + · · · + (x − a)n + Rn
1! 2! n!
Maclaurin series: Taylor series about a = 0.

X f n (0)
f (x) = xn
n!
n=0

f 0 (0) f 00 (0) 2 f n (0) n


= f (0) + x+ x + ··· + x + Rn
1! 2! n!

Saraswati Acharya (Department MATH207


of Mathematics[2mm]School of Science, Kathmandu University[2mm]
4 / 31 K
Some familiar power series that are Maclaurin series:

1 X
1. = x n = 1 + x + x 2 + x 3 + · · · (|x| < 1)
1−x
n=0

X xn x x2 x3
2. e x = =1+ + + + ···
n! 1! 2! 3!
n=0

X (−1)n x 2n+1 x3 x5 x7
3. sin x = =x− + − + ···
(2n + 1)! 3! 5! 7!
n=0

X (−1)n x 2n x2 x4 x6
4. cos x = =1− + − + ···
(2n)! 2! 4! 6!
n=0

X x 2n+1 x3 x5 x7
5. sin hx = =x+ + + + ···
(2n + 1)! 3! 5! 7!
n=0

X x 2n x2 x4 x6
6. cos hx = =1+ + + + ···
(2n)! 2! 4! 6!
n=0

Saraswati Acharya (Department MATH207


of Mathematics[2mm]School of Science, Kathmandu University[2mm]
5 / 31 K
By using Power

Series Method, solve: y 0 − y = 0
X
Let y = cn x n = c0 + c1 x + c2 x 2 + c3 x 3 + c4 x 4 + · · · be solution
n=0
of y 0 − y = 0.
y 0 = c1 + 2c2 x + 3c3 x 2 + 4c4 x 3 + · · ·
Substituting the values of y and y 0 in y 0 − y = 0.
c1 + 2c2 x + 3c3 x 2 + 4c4 x 3 + · · · − (c0 + c1 x + c2 x 2 + c3 x 3 + c4 x 4 + · · · ) = 0
c1 − c0 + (2c2 − c1 )x + (3c3 − c2 )x 2 + (4c4 − c3 )x 3 + · · · = 0
Equating like terms, we get
c1 c0 c0
c1 = c0 , 2c2 = c1 =⇒ c2 = = =
2 2 2!
c0 c0
3c3 = c2 =⇒ c3 = =
2.3 3!
c0
c4 =
4!
c0 2 c0 3 1 1
∴ y = c0 + c0 x + x + x + · · · = c0 (1 + x + x 2 + x 3 + · · · )
2! 3! 2! 3!
y = c0 e x

Saraswati Acharya (Department MATH207


of Mathematics[2mm]School of Science, Kathmandu University[2mm]
6 / 31 K
By using Power

Series Method, solve: y 00 + y = 0
X
Let y = cn x n = c0 + c1 x + c2 x 2 + c3 x 3 + c4 x 4 + · · · be the
n=0
solution of given ODE
y 0 = c1 + 2c2 x + 3c3 x 2 + 4c4 x 3 + 5c5 x 4 + · · ·
y 00 = 2c2 + 6c3 x + 12c4 x 2 + 20c5 x 3 + · · ·
Substituting the values of y and y 00 in y 00 + y = 0.
2c2 + 6c3 x + 12c4 x 2 + 20c5 x 3 + · · · + (c0 + c1 x + c2 x 2 + c3 x 3 + c4 x 4 + · · · ) = 0
Equating like terms, we get
c0 c1 c1
2c2 = −c0 , c2 = − 6c3 + c1 = 0 =⇒ c3 = − =−
2 6 3!
c2 c0
12c4 = −c2 =⇒ c4 = − =
12 4!
c3 c1
20c5 = −c3 =⇒ c5 = − =
20 5! 
x2 x4 x3 x5
 
∴ y = c0 1 − + + · · · + c1 x − + + ···
2! 4! 3! 5!
= c0 cos x + c1 sin x.

Saraswati Acharya (Department MATH207


of Mathematics[2mm]School of Science, Kathmandu University[2mm]
7 / 31 K
Solve initial value problem y 0 = y + x 2 , y (0) = 1 by using
power series method.


X
Assume that y (x) = cn x n = c0 + c1 x + c2 x 2 + · · · be the solution
n=0
of given ODE. Then,

c1 + 2c2 x + 3c3 x 2 + 4c4 x 3 + · · · = (c0 + c1 x + c2 x 2 + c3 x 3 + · · · ) + x 2


c1 + 2c2 x + 3c3 x 2 + 4c4 x 3 + · · · = c0 + c1 x + (c2 + 1)x 2 + c3 x 3 + · · ·

This yields,
(c1 − c0 ) + (2c2 − c1 )x + (3c3 − c2 − 1)x 2 + (4c4 − c3 )x 3 + · · · = 0
So, c1 − c0 = 0, 2c2 − c1 = 0, 3c3 − c2 − 1 = 0, 4c4 − c3 = 0 + · · ·

Saraswati Acharya (Department MATH207


of Mathematics[2mm]School of Science, Kathmandu University[2mm]
8 / 31 K
Solution (contd.)
We obtain
c1 c2 + 1 c0 + 2 c3 c0 + 2
c1 = c0 , c2 = , c3 = = , c4 = = ,···
2! 3 3! 4 4!
Substituting these values in (1)
c1 2 c0 + 2 3 c0 + 2 4
y (x) = c0 + c0 x + x + x + x + ···
2! 3! 4!
To solve the IVP, we set x = 0 and obtain;
1 = y (0) = c0
Thus, the solution of IVP is;
x2 3 4
+ x3 + x4 + · · ·
y (x) = 1 + x +
2! 3! 4!
x2
   
1 3 1
= 1+x + + x + x4 + · · ·
2! 2! 3!
Saraswati Acharya (Department MATH207
of Mathematics[2mm]School of Science, Kathmandu University[2mm]
9 / 31 K
Excercise

1. y 0 = x 2 y
3 /3
Ans: y = c0 e x
2. y 0 − 2xy = 0
2
Ans: y = a0 e x

Saraswati Acharya (Department MATH207


of Mathematics[2mm]School of Science, Kathmandu University[2mm]
10 / 31 K
Legendre’s Equation and Legendre Polynomial

Legendre’s Equation of order n

(1 − x 2 )y 00 − 2xy 0 + n(n + 1)y = 0 (n constant) (∗)

Legendre Polynomial
The solution of Legendre’s Equation (*) is known as Legendre Polynomial
or Legendre Function of degree n and is denoted by Pn (x) and defined by
M
X (2n − 2m)!
Pn (x) = (−1)m x n−2m
2n m!(n − m)!(n − 2m)!
m=0

n n−1
where, M= (if n is even) or (if n is odd) whichever is an integer.
2 2

Saraswati Acharya (Department MATH207


of Mathematics[2mm]School of Science, Kathmandu University[2mm]
11 / 31 K
Legendre Polynomials

The first few of Legendre Functions are

1. P0 (x) = 1 2. P1 (x) = x

1 1
3. P2 (x) = (3x 2 − 1) 4. P3 (x) = (5x 3 − 3x)
2 2

1 1
5. P4 (x) = (35x 4 − 30x 2 + 3) 6. P5 (x) = (63x 5 − 70x 3 + 15x)
8 8
and so on.
U
se Legendre’s Polynomial to show the above relation

Saraswati Acharya (Department MATH207


of Mathematics[2mm]School of Science, Kathmandu University[2mm]
12 / 31 K
Rodrigue’s Formula

The expression for Pn (x) given by

1 dn
Pn (x) = [(x 2 − 1)n ], n = 0, 1, 2, · · ·
2n n! dx n

is called Rodrigue’s Formula.

Saraswati Acharya (Department MATH207


of Mathematics[2mm]School of Science, Kathmandu University[2mm]
13 / 31 K
Problems

Use Rodrigue’s Formula to prove the following relations:

1. P0 (x) = 1 2. P1 (x) = x

1 1
3. P2 (x) = (3x 2 − 1) 4. P3 (x) = (5x 3 − 3x)
2 2

1 1
5. P4 (x) = (35x 4 − 30x 2 + 3) 6. P5 (x) = (63x 5 − 70x 3 + 15x)
8 8

Saraswati Acharya (Department MATH207


of Mathematics[2mm]School of Science, Kathmandu University[2mm]
14 / 31 K
Bessel’s differential equation

Any differential equation of the form

x 2 y 00 + xy 0 + (x 2 − ν 2 )y = 0 (1)

is known as Bessel’s differential equation.


1 The parameter ν is real and non-negative.
2 It appears in connection with electric fields vibrations, heat
conduction in cylindrical object.
3 As (1) is a second order differential equation, so there must be two
linear independent solutions.
4 Bessel’s functions of the first kind is denoted as Jν (x) are the
solutions of the Bessel’s differential equation that are
1 finite at the origin x = 0 for integer ν and
2 diverges as x → 0 for negative non-integer ν.

Saraswati Acharya (Department MATH207


of Mathematics[2mm]School of Science, Kathmandu University[2mm]
15 / 31 K
Bessel’s Equation and Bessel Functions Jν (x)

Bessel’s Equation:
A second order homogeneous differential equation of the form

x 2 y 00 + xy 0 + (x 2 − ν 2 )y = 0 (∗)

where the parameter ν ≥ 0 is a real number, is called Bessel’s Equation.


Bessel Function or Polynomial:
The solution of Bessel’s Equation (*) denoted by Jν (x) is called Bessel
Polynomial and is given by
∞ ∞
X (−1)m x 2m X (−1)m x 2m+ν
Jν (x) = x ν =
22m+ν m!Γ(ν + m + 1) 22m+ν m!Γ(ν + m + 1)
m=0 m=0

Saraswati Acharya (Department MATH207


of Mathematics[2mm]School of Science, Kathmandu University[2mm]
16 / 31 K
Bessel Polynomial

If we replace ν by −ν
∞ ∞
−ν
X (−1)m x 2m X (−1)m x 2m−ν
J−ν (x) = x =
22m−ν m!Γ(−ν + m + 1) 22m−ν m!Γ(m − ν + 1)
m=0 m=0

For ν = n (an integer),


∞ ∞
X (−1)m x 2m+n X (−1)m x 2m+n
Jn (x) = =
22m+n m!Γ(n + m + 1) 22m+n m!(n + m)!
m=0 m=0

Saraswati Acharya (Department MATH207


of Mathematics[2mm]School of Science, Kathmandu University[2mm]
17 / 31 K
1 Γ(ν) is the gamma function, a generalization of the factorial function
to non-integer values and is defined as
Z ∞
Γ(ν) = e −t t ν−1 dt
0
Γ(ν + 1) = νΓ(ν)
Γ(n + 1) = n!
2 It converges very rapidly due to the factorial in denominator.

G
General Solution of Bessel’s equation
For non-integer ν, the functions Jν (x) and J−ν (x) are linearly independent
and are therefore the two solutions of the differential equation. That is

y (x) = c1 Jν (x) + c2 J−ν (x). (2)

But if ν is an integer, then 2 is not a general solution because of linear


dependence.
Saraswati Acharya (Department MATH207
of Mathematics[2mm]School of Science, Kathmandu University[2mm]
18 / 31 K
Linear dependence of Bessel function Jn and J−n

Theorem
For integer ν = n, Jn and J−n are linearly dependent because

J−n (x) = (−1)n Jn (x).

Proof: We have,

−ν
X (−1)m x 2m
J−ν (x) = x (3)
22m−ν m!Γ(m − ν + 1)
m=0

If ν = n, then the gamma function in the coefficients of the first n


terms become infinite.
The coefficients becomes zero and the summation starts with m = n
and Γ(m − ν + 1) = (m − n)!

Saraswati Acharya (Department MATH207


of Mathematics[2mm]School of Science, Kathmandu University[2mm]
19 / 31 K

X (−1)m x 2m−n
J−n (x) = Let s = m − n, m = s + n
m=n
22m−n m!Γ(m − n + 1)

X (−1)n+s x 2s+n
=
22s+n (s + n)!s!
s=0

X (−1)s x 2s+n
= (−1)n
22s+n (s + n)!s!
s=0
∴ J−n (x) = (−1)n Jn (x).

Saraswati Acharya (Department MATH207


of Mathematics[2mm]School of Science, Kathmandu University[2mm]
20 / 31 K
Some Properties of Bessel Functions

d
P1. [x ν Jν (x)] = x ν Jν−1 (x)
dx
d
P2. [x −ν Jν (x)] = −x −ν Jν+1 (x)
dx

P3. Jν−1 (x) + Jν+1 (x) = Jν (x)
x
P4. Jν−1 (x) − Jν+1 (x) = 2Jν0 (x)

Saraswati Acharya (Department MATH207


of Mathematics[2mm]School of Science, Kathmandu University[2mm]
21 / 31 K
d
(x ν Jν (x)) = x ν Jν−1 (x)
dx

X (−1)m x 2m
Jν (x) = x ν (4)
22m+ν m! Γ(ν + m + 1)
m=0
Multiplying both sides of (4) by x ν

X (−1)m x 2m+2ν
x ν Jν (x) =
m=0
22m+ν m!Γ(ν + m + 1)

d (−1)m (2m + 2ν)x 2m+2ν−1
X
(x ν Jν (x)) =
dx m=0
22m+ν m!(ν + m)Γ(ν + m)

X (−1)m x 2m
= x 2ν−1
m=0
22m+ν−1 m!Γ(ν + m)

X (−1)m x 2m x ν−1
= xν
m=0
22m+ν−1 m!Γ(ν + m)
d
∴ (x ν Jν (x)) = x ν Jν−1 (x)·
Saraswati Acharya
dx (Department MATH207
of Mathematics[2mm]School of Science, Kathmandu University[2mm]
22 / 31 K
d
x −ν Jν (x) = −x −ν Jν+1 (x).

dx
Multiplying both sides of (4) by x −ν

X (−1)m x 2m
x −ν Jν (x) =
m=0
22m+ν m!Γ(ν + m + 1)

d X (−1)m (2m)x 2m−1
x −ν Jν (x) =

dx 22m+ν m(m − 1)!Γ(ν + m + 1)
m=0

X (−1)m x 2m−1
= Let, m − 1 = s, m = s + 1
m=1
22m+ν−1 (m − 1)! Γ(ν + m + 1)

X (−1)s+1 x 2s+1
=
s=0
22s+ν+1 s! Γ(ν + s + 2)

Equation (4) ν + 1 instead of ν and s instead of m


d
x −ν Jν (x) = −x −ν Jν+1 (x)

dx
Saraswati Acharya (Department MATH207
of Mathematics[2mm]School of Science, Kathmandu University[2mm]
23 / 31 K

Jν−1 + Jν+1 = Jν ·
x
We have
d
(x ν Jν (x)) = x ν Jν−1 (x)
dx
νx ν−1 Jν + x ν Jν0 = x ν Jν−1 (5)
d
x −ν Jν (x) = −x −ν Jν+1 (x)

Again,
dx
−ν−1
(−ν)x Jν + x −ν Jν0 = −x −ν Jν+1
Multiplying both sides by x 2ν
−νx ν−1 Jν + x ν Jν0 = −x ν Jν+1 (6)
Subtracting (5) from (6)
2νx ν−1 Jν = x ν (Jν−1 + Jν+1 )

Jν = Jν−1 + Jν+1
x

∴ Jν−1 + Jν+1 = Jν ·
x

Saraswati Acharya (Department MATH207


of Mathematics[2mm]School of Science, Kathmandu University[2mm]
24 / 31 K
Jν−1 + Jν+1 = 2Jν0
Adding (5) and (6), we get

2x ν Jν0 = x ν (Jν−1 + Jν+1 )


Jν−1 + Jν+1 = 2Jν0 .

exercise
Exercise: Prove the following:
r
2
1. J1/2 (x) = sin x
πx
r
2
2. J−1/2 (x) = cos x
πx

Saraswati Acharya (Department MATH207


of Mathematics[2mm]School of Science, Kathmandu University[2mm]
25 / 31 K
r
2
Proof of J−1/2 (x) = cos x
πx

X (−1)m x 2m+ν
We know that, Jν (x) = . So, Putting ν = −1/2;
m=0
22m+ν m!Γ(ν + m + 1)
∞ ∞
X (−1)m x 2m−1/2 x −1/2 X (−1)m x 2m
J−1/2 (x) = =
m=0
22m−1/2 m!Γ(m − 1/2 + 1) 2−1/2 m=0 22m m!Γ(m + 1/2)
r 
x2 x4 x6

2 1
= − + − +· · ·
x Γ(1/2) 22 1!Γ(1 + 1/2) 24 2!Γ(2 + 1/2) 26 3!Γ(3 + 1/2)
=r "
2 1 x2 x4 x6
− 2 + 4 − 6 +· ·
1 3
x Γ(1/2) 2 · 1! · 2 · Γ(1/2) 2 · 2! · 2 · 2 · Γ(1/2) 2 · 3! · 2 · 32 · 12 · Γ(1/2)
1 5

r " #
1 2 x2 x4 x6 √
= 1− 2 + − + · · · [Γ(1/2)= π]
Γ(1/2) x 2 · 1! · 12 24 · 2! · 32 · 21 26 · 3! · 52 · 32 · 12
r  r
x2 x4 x6

2 2
= 1− + − + ··· = cos x Proved
πx 2! 4! 6! πx

Saraswati Acharya (Department MATH207


of Mathematics[2mm]School of Science, Kathmandu University[2mm]
26 / 31 K
r
2
Proof of J1/2 (x) = sin x
πx

We know that,

X (−1)m x 2m
Jν (x) = x ν
m=0
22m+ν m!Γ(ν + m + 1)

1
If ν = 2


√ X (−1)m x 2m
J1/2 (x) = x
22m+1/2 m! Γ 21 + m + 1

m=0

Saraswati Acharya (Department MATH207


of Mathematics[2mm]School of Science, Kathmandu University[2mm]
27 / 31 K
r ∞
2X (−1)m x 2m+1
J1/2 (x) = 2m+1 m! Γ m + 3

x 2 2
m=0

Using Γ( 12 ) = π
      
3 1 1 3 1 1
Γ m+ = m+ m− ··· Γ
2 2 2 2 2 2
−(m+1) √
=2 (2m + 1)(2m − 1) · · · 3.1. π
Again 22m+1 m! = 2m+1 2m m!
= 2m+1 2m m(m − 1)(m − 2) · · · 3.2.1
= 2m+1 2m(2m − 2)(2m − 4) · · · 6.4.2

 
3
22m+1 m! Γ m + = (2m + 1)! π
2
r ∞ r
2 X (−1)m x 2m+1 2
J 1 (x) = ∴ J 1 (x) = sin x.
2 πx (2m + 1)! 2 πx
m=0

Saraswati Acharya (Department MATH207


of Mathematics[2mm]School of Science, Kathmandu University[2mm]
28 / 31 K
r
2
J− 1 (x) = cos x
2 πx r
2 √ q
The equation J 1 (x) = sin x can be written as xJ 1 (x) = π2 sin x.
2 πx 2
Then

d √  r2
xJ 1 (x) = cos x
dx 2 π
r
1 2
x 2 J 1 −1 = cos x
2 π
r
2
J− 1 = cos x
2 πx
r  
2 sin x
J3 = − cos x
2 πx x

Saraswati Acharya (Department MATH207


of Mathematics[2mm]School of Science, Kathmandu University[2mm]
29 / 31 K
Using

Jν−1 + Jν+1 = Jν (x) (7)
x
1
If ν = in (7),
2
1
J− 1 + J 3 = J1
2 2 x 2
1
J 3 = J 1 − J− 1
2 x r2 2
r
1 2 2
= sin x − cos x
x πx πx
r  
2 sin x
∴ J3 = − cos x .
2 πx x

Saraswati Acharya (Department MATH207


of Mathematics[2mm]School of Science, Kathmandu University[2mm]
30 / 31 K
r
2  cos x 
J− 3 = − + sin x
2 πx x
1
If ν = − in (7), we get
2
1
J− 1 −1 + J− 1 +1 = − J− 1
2 2 x 2
1
J− 3 = − J− 1 − J 1
2 xr2 2
r
1 2 2
=− cos x − sin x
x πx πx
r
2  cos x 
∴ J− 3 = − + sin x .
2 πx x

Saraswati Acharya (Department MATH207


of Mathematics[2mm]School of Science, Kathmandu University[2mm]
31 / 31 K

You might also like