5.4 Regular Singular Points 5.

4 Regular Singular Points

5.5 Euler Equations 5.5 Euler Equations

Chapter 5
Series Solutions of Second Order Linear Equations

In this chapter we consider solving second order linear

equations with variable coefficients.
To find two fundamental solutions it is necessarily to search for
Chapter 5 solution in a larger set of functions than the set of elementary
functions.
Series Solutions of Second Power series representation of functions is our principal tool.
The basic idea is to look for a solution in the form of a power
Order Linear Equations series and try to find the coefficients in this series.

Second Semester 2020-2021

A® ‚Ë@ Õæ Aë áÜß @ . X . @
1 / 97 2 / 97

5.4 Regular Singular Points 5.4 Regular Singular Points

5.5 Euler Equations 5.5 Euler Equations

5.1 Review of Power Series

∞
X
In this section we summarize very briefly the main results about (2) A power series an (x − x0 )n certainly converges at x = x0 .
power series that we need. n=0
The set of all points at which a power series converges is called
Recall that a power series is an infinite sum of the form the interval of convergence.
∞ |an+1 | 1
X (3) Let L = lim . Then ρ = is called the radius of
an (x − x0 )n = 0. n→∞ |an | L
∞
n=0 X
convergence and the power series an (x − x0 )n converges in
The constant x0 is the center and the constants a0 , a1 , · · · are n=0
the coefficients. 1
the interval |x − x0 | < ρ = .
∞ L
X ∞ ∞
(1) A power series an (x − x0 )n converges at a point x if X
n
X
(4) If an (x − x0 ) and bn (x − x0 )n are two power series,
n=0
M n=0 n=0
X then
lim an (x − x0 )n exists. ∞ ∞ ∞
M→∞
n=0
X X X
an (x − x0 )n + bn (x − x0 )n = (an + bn )(x − x0 )n .
3 / 97
n=0 n=0 n=0 4 / 97
 5.4 Regular Singular Points 5.4 Regular Singular Points
5.5 Euler Equations 5.5 Euler Equations

∞
X
(5) If f (x) = an (x − x0 )n , |x − x0 | < ρ, then
(7) As long as we preserve the order of the terms of a series, we
n=0
can change the index of the series without altering its
∞
X convergence. Just as we make changes of the variable of
0
f (x) = nan (x − x0 )n−1 , |x − x0 | < ρ. integration in a definite integral, we can change the indices of a
n=1 series. For example
∞
X ∞
X ∞ ∞
n
(6) If an (x − x0 ) = bn (x − x0 )n , then an = bn for
X X
n
an (x − x0 ) = am−1 (x − x0 )m−1 .
n=0 n=0 n=0 m=1
n = 0, 1, 2, . . . . In particular, if
∞
X
an (x − x0 )n = 0, then an = 0 for n = 0, 1, 2, . . . . = a0 + a1 (x − x0 ) + a2 (x − x0 )2 + · · ·
n=0

5 / 97 6 / 97

5.4 Regular Singular Points 5.4 Regular Singular Points

5.5 Euler Equations 5.5 Euler Equations

5.2 Series Solutions near an Ordinary Point, Part I

In this section we consider how we can find a solution of the
second order linear equation
and the Legendre equation
P(x)y 00 + Q(x)y 0 + R(x)y = g (x). (1)
It is sufficient to consider the homogeneous equation (1 − x 2 )y 00 − 2xy 0 + α(α + 1)y = 0, (4)

P(x)y 00 + Q(x)y 0 + R(x)y = 0, (2) where α is a constant.

since the procedure for the corresponding nonhomogeneous For this reason, as well as to simplify the algebraic
equation is similar. computations, we primarily consider the case in which the
Many problems in mathematical physics leads to equations of functions P, Q, and R are polynomials.
the form (2) having polynomial coefficients; However, as we will see, the method of solution is also
for example, the Bessel equation applicable when P, Q, and R are general analytic functions.

x 2 y 00 + xy 0 + (x 2 − ν 2 )y = 0, (3)
where ν is a constant,
7 / 97 8 / 97
 5.4 Regular Singular Points 5.4 Regular Singular Points
5.5 Euler Equations 5.5 Euler Equations

Ordinary Points Examples

Example 1
Determine the singular points of the equation
Definition
Let P(x), Q(x), and R(x) be polynomials with no common factors. (x 2 − 2x − 3)y 00 + 2xy 0 + 3(x + 2)y = 0.
A point x0 is called an ordinary point of the equation
Solution
P(x)y 00 + Q(x)y 0 + R(x)y = 0 In this example
P(x) = x 2 − 2x − 3, Q(x) = 2x, R(x) = 3(x + 2)
if P(x0 ) 6= 0. On the other hand, if P(x0 ) = 0, then x0 is called a
are polynomials with no common factors.
singular point.
Thus the singular points are the solutions of P(x) = 0.
Since P(x) = x 2 − 2x − 3 = (x + 1)(x − 3), the singular
points of the equation are x = −1 and x = 3.

9 / 97 10 / 97

5.4 Regular Singular Points 5.4 Regular Singular Points

5.5 Euler Equations 5.5 Euler Equations

Series Solutions near an Ordinary Point

Let x0 be an ordinary point of P(x)y 00 + Q(x)y 0 + R(x)y = 0.
Example 2
Since P(x) is continuous, there is an interval I containing x0 in
Determine the singular points of the equation
which P(x) is never zero.
1 In that interval we can divide by P(x) and write equation (2) as
y 00 − y 0 + xy = 0.
x(x 2 + 1)
y 00 + p(x)y 0 + q(x)y = 0, (5)
Solution where p(x) = Q(x)/P(x) and q(x) = R(x)/P(x) are
In this example the equation can be written as continuous functions.
x(x 2 + 1)y 00 − y 0 + x 2 (x 2 + 1)y = 0. Hence, according to the existence and uniqueness Theorem,
Thus P(x) = x(x 2 + 1), Q(x) = −1, R(x) = x 2 (x 2 + 1) there exists in that interval a unique solution of equation (2)
are polynomials with no common factors. that also satisfies the initial conditions y (x0 ) = y0 , y 0 (x0 ) = y00
Thus the singular points are the solutions of P(x) = 0. It for arbitrary values of y0 and y00 .
follows that the singular point of the equation is x = 0. In this and the following section we discuss the solution of
equation (2) in the neighborhood of an ordinary point.
11 / 97 12 / 97
 5.4 Regular Singular Points 5.4 Regular Singular Points
5.5 Euler Equations 5.5 Euler Equations

Examples

We look for solutions of the form

∞
X Example 1
y= an (x − x0 )n = a0 + a1 (x − x0 ) + a2 (x − x0 )2 + . . . . (6)
n=0
Find a series solution of the equation

We assume that the series converges in the interval y 00 + xy 0 + y = 0

|x − x0 | < ρ for some ρ > 0.
about the point x0 = 0.
Our aim is to determine the coefficients an , n = 0, 1, 2, . . . .
This can be achieved by substituting the series and its Solution
derivatives for y and its derivatives.
In this example P(x) = 1, Q(x) = x, R(x) = 1.
We illustrate the method by the following examples.
Since P(0) = 1 6= 0, x0 = 0 is an ordinary point.
The operations, such as differentiation, that are involved in the
procedure are justified so long as we stay within the interval of
convergence.
13 / 97 14 / 97

5.4 Regular Singular Points 5.4 Regular Singular Points

5.5 Euler Equations 5.5 Euler Equations

Example 1 (continue) Example 1 (continue)

We look for a solution in the form of a power series about To combine the three series we need to rewrite them so that all
x0 = 0, series display the same generic term.
∞
X As a first step, we multiply the second series by x:
2 3
y = a0 + a1 x + a2 x + a3 x + · · · = an x n . X∞ ∞
X ∞
X
n=0 n(n − 1)an x n−2 + x nan x n−1 + an x n = 0.
Differentiating term by term yields n=2 n=1 n=0
∞ ∞
X X∞ ∞
X
n(n − 1)an x n−2 + nan x n + an x n = 0.
X
y 0 = a1 + 2a2 x + 3a3 x 2 + · · · = nan x n−1 ,
n=1 n=2 n=1 n=0
∞ (7)
X As a second step, we rewrite the summations so that x in each
y 00 = 2a2 + 6a3 x + · · · = n(n − 1)an x n−2 .
of these summations will have the exponent n.
n=2
Thus, in the first sum, we shift the index of summation by
Substituting in the differential equation gives replacing n by n + 2. (n = m + 2)
X∞ X∞ X∞
n−2 n−1
n(n − 1)an x +x nan x + an x n = 0. ∞
X ∞
X ∞
X
n=2 n=1 n=0 n(n − 1)an x n−2 = (m + 2)(m + 1)am+2 x m = (n + 2)(n + 1)an+2 x n
n=2 m=0 n=0
15 / 97 16 / 97
 5.4 Regular Singular Points 5.4 Regular Singular Points
5.5 Euler Equations 5.5 Euler Equations

Example 1 (continue) Example 1 (continue)

As a result we obtain This leads to the conditions
X∞ ∞
X ∞
X
(n + 2)(n + 1)an+2 x n + nan x n + an x n = 0. 2a2 +a0 = 0, (n+2)(n+1)an+2 +nan +an = 0, n = 1, 2, . . . .
n=0 n=1 n=0
As a next step, we adjust the summations so that the ranges of 1
It follows that a2 = − a0 and
the summations are all the same. 2
The common range is from 1 to ∞. −1
In this example we have an+2 = an , n = 1, 2, . . . .
n+2
∞ ∞ ∞
The last relation is called a recurrence relation and it can be
X X X
n n
2a2 + (n + 2)(n + 1)an+2 x + nan x + a0 + an x n = 0.
n=1 n=1 n=1 used to calculate the coefficients an , n = 3, 4, . . . , recursively.
We can now combine like powers of x and write this −1
n=1: a3 = a1
equation as 3
  
∞ −1 −1 −1 1
X n = 2 : a4 = a2 = a0 =⇒ a4 = a0
2a2 + a0 + [(n + 2)(n + 1)an+2 + nan + an ] x n = 0. 4 4 2 8
n=1
  
−1 −1 −1 1
The coefficient of each power of x in the left member of the n = 3 : a5 = a3 = a1 =⇒ a5 = a1
17 / 97 5 5 3 15 18 / 97
last equation must be equated to zero.

5.4 Regular Singular Points 5.4 Regular Singular Points

5.5 Euler Equations 5.5 Euler Equations

Example 1 (continue) Example 1 (continue)

Substituting the values of an , n ≥ 2 into the assumed solution

∞
X
2 3
y = a0 + a1 x + a2 x + a3 x + · · · = an x n ,
n=0

we have The two series in parentheses are the power series expansions of
two linearly independent solutions of the differential equation,
1 1 1 1
y = a0 + a1 x − a0 x 2 − a1 x 3 + a0 x 4 + a1 x 5 + . . . and the constants a0 and a1 are arbitrary constants.
2 3 8 15
Collecting terms in a0 and a1 , we have finally
   
1 2 1 4 1 3 1 5
y = a0 1 − x + x + . . . + a1 x − x + x + . . . ,
2 8 3 15

which gives the solution of the differential equation in powers

of x.
19 / 97 20 / 97
 5.4 Regular Singular Points 5.4 Regular Singular Points
5.5 Euler Equations 5.5 Euler Equations

5.2 Series Solutions near an Ordinary Point, Part I

Example 2 Differentiating term by term yields
∞
Find a series solution of the equation X
y0 = nan x n−1 ,
y 00 + xy 0 + (x 2 + 2)y = 0 n=1
∞ (8)
X
n−2
y 00 = n(n − 1)an x .
about the point x0 = 0.
n=2

Solution Substituting in the differential equation gives

∞ ∞ ∞
In this example P(x) = 1, Q(x) = x, R(x) = x 2 + 2.
X X X
n−2 n−1 2
n(n − 1)an x +x nan x + (x + 2) an x n = 0.
Since P(0) = 1 6= 0, x0 = 0 is an ordinary point. n=2 n=1 n=0
We look for a solution in the form of a power series about To combine the three series we need to rewrite them so that all
x0 = 0, series display the same generic term.
As a first step, we multiply the second series by x and the third
∞
2 3
X series by x 2 + 2 and we obtain
y = a0 + a1 x + a2 x + a3 x + · · · = an x n .
∞ ∞ ∞ ∞
n=0 X X X X
21 / 97 n(n − 1)an x n−2 + nan x n + an x n+2 + 2an x n = 0.22 / 97
n=2 n=1 n=0 n=0

5.4 Regular Singular Points 5.4 Regular Singular Points

5.5 Euler Equations 5.5 Euler Equations

As a second step, we rewrite the summations so that x in each

of these summations will have the exponent n. ∞
X X∞ X∞ X∞

Thus, in the first sum, we shift the index of summation by (n+2)(n+1)an+2 x n + nan x n + an−2 x n + 2an x n = 0.
replacing n by n + 2 and in third sum, we shift the index of n=0 n=1 n=2 n=0

summation by replacing n by n − 2. In this example we have

∞
X
∞ ∞ ∞ ∞
2a2 + (2)(3)a3 x + (n + 2)(n + 1)an+2 x n + a1 x +
n=2
X X X X
n−2 n n+2
n(n − 1)an x + nan x + an x + 2an x n = 0. ∞
X ∞
X ∞
X
n=2 n=1 n=0 n=0 nan x n + an−2 x n + 2a0 + 2a1 x + 2an x n = 0.
As a result we obtain n=2 n=2 n=2
X∞ ∞
X X∞ X∞ We can now combine like powers of x and write this equation as
n n n
(n+2)(n+1)an+2 x + nan x + an−2 x + 2an x n = 0. ∞
X
n=0 n=1 n=2 n=0 2(a2 +a0 )+3(2a3 +a1 )x+ [(n+1)(n+2)an+2 +nan +an−2 +2an ]x n = 0.
As a next step, we adjust the summations so that the ranges of n=2

the summations are all the same. The coefficient of each power of x in the left member of the
The common range is from 2 to ∞. last equation must be equated to zero.
We now write out individually the terms in each summation 23 / 97 24 / 97

that do not belong to this common range, and we continue to

 5.4 Regular Singular Points 5.4 Regular Singular Points
5.5 Euler Equations 5.5 Euler Equations

This leads to the conditions a2 + a0 = 0, 2a3 + a1 = 0, n=5:

(n + 1)(n + 2)an+2 + (n + 2)an + an−2 = 0, n = 2, 3, . . . . −1 −1 21 1 −a1
a7 = [7a5 + a3 ] = [ a1 − a1 ] =⇒ a7 =
1 6·7 42 40 2 (40)(42)
It follows that a2 = −a0 , a3 = − a1 and the recurrence
2 Substituting the values of an , n ≥ 2 into the assumed solution
relation is ∞
1
X
2 3
an+2 = − [(n + 2)an + an−2 ], n = 2, 3, . . . . y = a0 + a1 x + a2 x + a3 x + · · · = an x n ,
(n + 1)(n + 2) n=0
n=2: we have
−1 −1 1
a4 = [4a2 + a0 ] = [−4a0 + a0 ] =⇒ a4 = a0 1 1 3 1
3·4 12 4 y = a0 + a1 x − a0 x 2 − a1 x 3 + a0 x 4 + a1 x 5 − a0 x 6 + · · · .
n=3: 2 4 40 60
−1 −1 5 3 Collecting terms in a0 and a1 , we have finally
a5 = [5a3 + a1 ] = [− a1 + a1 ] =⇒ a5 = a1
4·5 20 2 40    
2 1 4 1 3 3 5
n=4: y = a0 1 − x + x + . . . + a1 x − x + x + . . . .
−1 −1 6 −1 4 2 40
a6 = [6a4 + a2 ] = [ a0 − a0 ] =⇒ a6 = a0
5·6 30 4 60
25 / 97 26 / 97

5.4 Regular Singular Points 5.4 Regular Singular Points

5.5 Euler Equations 5.5 Euler Equations

∞
Example 3 Differentiating y =
X
an x n gives
Solve the initial value problem n=0
∞
y 00 − xy 0 − y = 0, y (0) = 2, y 0 (0) = 1.
X
y0 = nan x n−1 ,
n=1 (9)
∞
Solution X
n−2
y 00 = n(n − 1)an x .
First of all, since the initial conditions are y (0) = 2, y 0 (0) = 1,
n=2
we have to find a series solution about x0 = 0.
Applying the initial conditions implies y (0) = a0 and
In this example P(x) = 1, Q(x) = −x, R(x) = −1. Since y 0 (0) = a1 . Thus we have a0 = 2 and a1 = 1 .
P(0) = 1 6= 0, x0 = 0 is an ordinary point. Substituting in the differential equation gives
Thus we look for a solution in the form of a power series ∞
X ∞
X X∞

∞
n(n − 1)an x n−2 − x nan x n−1 − an x n = 0.
X
2 3 n=2 n=1 n=0
y = a0 + a1 x + a2 x + a3 x + · · · = an x n .
n=0 We multiply the second series by x we get
X∞ X∞ ∞
X
n−2 n
27 / 97 n(n − 1)an x − nan x − an x n = 0. 28 / 97

n=2 n=1 n=0

 5.4 Regular Singular Points 5.4 Regular Singular Points
5.5 Euler Equations 5.5 Euler Equations

We multiply the second series by x we get This leads to the conditions

∞ ∞ ∞
X X X 2a2 −a0 = 0, (n+1)(n+2)an+2 −(n+1)an = 0, n = 1, 2, . . . .
n(n − 1)an x n−2 − nan x n − an x n = 0.
n=2 n=1 n=0 It follows that a2 = 12 a0 =⇒ a2 = 1 and the recurrence
As a second step, we rewrite the summations so that x in each relation is
an
of these summations will have the exponent n. an+2 = , n = 1, 2, . . . .
(n + 2)
∞ ∞ ∞
a1 1
X X X
(n + 2)(n + 1)an+2 x n − nan x n − an x n = 0. n=1: a3 = =⇒ a3 =
n=0 n=1 n=0 3 3
As a next step, we adjust the summations so that the ranges of a2 1
n=2: a4 = =⇒ a4 =
the summations are all the same. 4 4
The common range is from 1 to ∞. a3 1
∞ n=3: a5 = =⇒ a5 =
X 5 15
2a2 − a0 + [(n + 1)(n + 2)an+2 − (n + 1)an ]x n = 0. a4 1
n=1 n=4: a6 = =⇒ a6 =
6 24
The coefficient of each power of x in the left member of the
29 / 97 30 / 97
last equation must be equated to zero.

5.4 Regular Singular Points 5.4 Regular Singular Points

5.5 Euler Equations 5.5 Euler Equations

Example 4
Solve the equation y 00 − xy 0 − y = 0 by means of a power
Substituting the values of an , n ≥ 0 into the assumed solution
solution about x = 1.
∞
X
y = a0 + a1 x + a2 x 2 + a3 x 3 + · · · = an x n , Solution
n=0 In this example P(x) = 1, Q(x) = −x, R(x) = −1. Since
we have P(1) = 1 6= 0, x0 = 1 is an ordinary point.
We look for a solution in the form of a power series about
1 1 1 1
y = 2 + x + x2 + x3 + x4 + x5 + x6 + · · · . x0 = 1,
3 4 15 24
∞
X
y = a0 +a1 (x −1)+a2 (x −1)2 +a3 (x −1)3 +· · · = an (x −1)n .
n=0

31 / 97 32 / 97
 5.4 Regular Singular Points 5.4 Regular Singular Points
5.5 Euler Equations 5.5 Euler Equations

∞
X As a first step, we multiply the second series by x = (x − 1) + 1
Differentiating y = an (x − 1)n yields and we obtain
n=0 X∞ X∞
n−2
∞
n(n − 1)an (x − 1) − [(x − 1) + 1] nan (x − 1)n−1 −
n=2 n=1
X
y0 = nan (x − 1)n−1 , ∞
X
n=1
∞ (10) an (x − 1)n = 0
X
n=0
y 00 = n(n − 1)an (x − 1)n−2 . and hence
n=2 X∞ ∞
X
n(n − 1)an (x − 1)n−2 − nan (x − 1)n
Substituting in the differential equation gives n=2 n=1
∞ ∞
∞ ∞ ∞ X X
X X X − nan (x − 1)n−1 − an (x − 1)n = 0.
n(n−1)an (x−1)n−2 −x nan (x−1)n−1 − an (x−1)n = 0.
n=1 n=0
n=2 n=1 n=0 ∞
X ∞
X
n
(n + 2)(n + 1)an+2 (x − 1) − nan (x − 1)n
To combine the three series we need to rewrite them so that all
n=0 n=1
series display the same generic term. X∞ ∞
X
33 / 97 − (n + 1)an+1 (x − 1)n − an (x − 1)n = 0. 34 / 97
n=0 n=0

5.4 Regular Singular Points 5.4 Regular Singular Points

5.5 Euler Equations 5.5 Euler Equations

 
The common range is from 1 to ∞. 1 1 1 1
n = 2 : a4 = [a2 + a3 ] = (a0 + a1 ) + (a0 + 3a1 )
∞
X 4 4 2 6
2a2 −a1 −a0 + [(n+1)(n+2)an+2 −(n+1)an+1 −(n+1)an ](x−1)n = 0. 1
=⇒ a4 = (2a0 + 3a1 )
n=1 12
The coefficient of each power of (x − 1) in the left member of Substituting the values of an , n ≥ 2 into the assumed solution
the last equation must be equated to zero. ∞
X
This leads to the conditions y = a0 +a1 (x −1)+a2 (x −1)2 +a3 (x −1)3 +· · · = an (x −1)n ,
n=0
2a2 −a1 −a0 = 0, (n+1)(n+2)an+2 −(n+1)an+1 −(n+1)an = 0, n = 1, 2, 3, . . . .
we have
a0 + a1 1 1
It follows that a2 = and the recurrence relation is y = a0 + a1 (x − 1) + (a0 + a1 )(x − 1)2 + (a0 + 3a1 )(x − 1)3
2 2 6
an + an+1 1
an+2 = , n = 1, 2, 3, . . . . + (2a0 + 3a1 )(x − 1)4 + · · · .
(n + 2) 12
Collecting terms in a0 and a1 , we have finally
n=1:
y = a0 1 + 12 (x − 1)2 + 16 (x − 1)3 + 16 (x − 1)4 + . . .
 

a1 + a2 1 1 1
a3 = = a1 + (a0 + a1 ) =⇒ a3 = (a0 + 3a1 )
+a1 (x − 1) + 12 (x − 1)2 + 12 (x − 1)3 + 14 (x − 1)4 + . . . . 36 / 97


3 3 2 6 35 / 97
 5.4 Regular Singular Points 5.4 Regular Singular Points
5.5 Euler Equations 5.5 Euler Equations

∞
Example 5 Differentiating y =
X
an x n gives
Solve the initial value problem n=0
∞
(x 2 + 4)y 00 + 6xy 0 + 4y = 0, y (0) = 1, y 0 (0) = 1.
X
y0 = nan x n−1 ,
n=1 (11)
∞
Solution X
n−2
y 00 = n(n − 1)an x .
First of all, since the initial conditions are y (0) = 1, y 0 (0) = 1,
n=2
we have to find a series solution about x0 = 0.
Applying the initial conditions implies y (0) = a0 and
In this example P(x) = x 2 + 4, Q(x) = 6x, R(x) = 4. Since y 0 (0) = a1 . Thus we have a0 = 1 and a1 = 1 .
P(0) = 4 6= 0, x0 = 0 is an ordinary point. Substituting in the differential equation gives
Thus we look for a solution in the form of a power series X∞ ∞
X X∞
2 n−2 n−1
∞
(x + 4) n(n − 1)an x + 6x nan x +4 an x n = 0.
X
n=2 n=1 n=0
y = a0 + a1 x + a2 x 2 + a3 x 3 + · · · = an x n .
n=0 We multiply the series by the coefficients we get
∞
X X∞ X∞ X∞
37 / 97 n(n−1)an x n + 4n(n−1)an x n−2 + 6nan x n + 4an x n =
38 /0.
97

n=2 n=2 n=1 n=0

5.4 Regular Singular Points 5.4 Regular Singular Points

5.5 Euler Equations 5.5 Euler Equations

As a second step, we rewrite the summations so that x in each The coefficient of each power of x in the left member of the
of these summations will have the exponent n. last equation must be equated to zero. This leads to the
∞
X conditions
4a0 + (6a1 + 4a1 )x + 4(n + 2)(n + 1)an+2 x n
8a2 +4a0 = 0, 10a1 +24a3 = 0, 4(n+1)(n+2)an+2 +(n2 +5n+4)an = 0, n =
n=0
∞ 1
It follows that a2 = − 21 a0 =⇒ a2 = − ,
X
+ [n(n − 1) + 6n + 4]an x n = 0. 2
n=2
5 5
As a next step, we adjust the two summations so that the a3 = − 12 a1 =⇒ a2 = − and the recurrence relation is
12
ranges of the summations are all the same.
(n + 4)an
4a0 + 10a1 x + 4(1)(2)a2 + 4(2)(3)a3 x an+2 = − , n = 2, 3, . . . .
4(n + 2)
∞
X
+ [4(n + 1)(n + 2)an+2 + (n2 + 5n + 4)an ]x n = 0. 6a2 3
n=2
n=2: a4 = − =⇒ a4 =
4·4 16
39 / 97 40 / 97
 5.4 Regular Singular Points 5.4 Regular Singular Points
5.5 Euler Equations 5.5 Euler Equations

5.3 Series Solutions near an Ordinary Point, Part II

In Section 5.2 we studied a method to find solutions of

7a3 7
n = 3 : a5 = − =⇒ a5 =
4·5 48 P(x)y 00 + Q(x)y 0 + R(x)y = 0, (12)
Substituting the values of an , n ≥ 0 into the assumed solution
where P, Q, and R are polynomials, in the neighborhood of an
∞
X ordinary point x0 .
y = a0 + a1 x + a2 x 2 + a3 x 3 + · · · = an x n ,
n=0
Assuming that the equation has a solution y = φ(x) that has a
Taylor series
∞
we obtain X
y = φ(x) = an (x − x0 )n ,
1 5 3 7
y = 1 + x − x2 − x3 + x4 + x5 + · · · . n=0
2 12 16 48 which converges for |x − x0 | < ρ, for some ρ > 0, we found
that the coefficients an can be determined by directly
substituting the series for y in the equation.

41 / 97 42 / 97

5.4 Regular Singular Points 5.4 Regular Singular Points

5.5 Euler Equations 5.5 Euler Equations

Starting from the assumption that

X∞
In this section we will consider how we might justify the y = φ(x) = an (x − x0 )n
statement that if x0 is an ordinary point of equation (12), then n=0
there exist solutions of the form is a solution of equation (12), then by differentiating
∞
X m times and sitting x = x0 we find
y = φ(x) = an (x − x0 )n . m!am = φ(m) (x0 ).
n=0
∞
X
In addition we consider the question of radius of convergence of y = φ(x) = an (x − x0 )n =
such series. n=0
a0 + a1 (x − x0 ) + a2 (x − x0 )2 + a3 (x − x0 )3 + · · ·
In doing this we are led to a generalization of the definition of
=⇒ φ(x0 ) = a0
an ordinary point. X∞
φ0 (x) = an (x − x0 )n = a1 + 2a2 (x − x0 ) + 3a3 (x − x0 )2 + · · ·
n=0
=⇒ φ0 (x0 ) = a1
43 / 97 44 / 97
 5.4 Regular Singular Points 5.4 Regular Singular Points
5.5 Euler Equations 5.5 Euler Equations

In order to determine φ(n) (x0 ) and the corresponding an for

∞
X n = 2, 3, . . . , we use the equation
φ00 (x) = an (x − x0 )n = 2a2 + 6a3 (x − x0 ) + · · ·
n=0
P(x)y 00 + Q(x)y 0 + R(x)y = 0.
=⇒ φ00 (x0 ) = 2a2 Since φ is a solution, we have
Thus, in order to compute an we must determine P(x)φ00 (x) + Q(x)φ0 (x) + R(x)φ(x) = 0.
φ(n) (x0 ) for n = 0, 1, 2, . . . from the differential equation.
For the interval about x0 for which P(x) 6= 0 we can
Suppose that y = φ(x) satisfies the initial conditions write this equation in the form
y (x0 ) = y0 , y 0 (x0 ) = y00 .
φ00 (x) = −p(x)φ0 (x) − q(x)φ(x), (13)
Then a0 = y0 and a1 = y00 .
Q(x) R(x)
If we are solely interested in finding a solution without where p(x) = , q(x) = .
P(x) P(x)
specifying and initial conditions, then a0 and a1 remain
Setting x = x0 in the above equation gives
arbitrary.
φ00 (x0 ) = −p(x0 )φ0 (x0 ) − q(x0 )φ(x0 ).
1 00
Hence we can find a2 = 2! φ (x0 ).
45 / 97 46 / 97

5.4 Regular Singular Points 5.4 Regular Singular Points

5.5 Euler Equations 5.5 Euler Equations

Example
Determine φ00 (0), φ000 (0), and φiv (0) if y = φ(x) is a solution of the
To determine a3 we differentiate equation (13) and then we initial value problem
set x = x0 obtaining
y 00 + xy 0 + ln(x + 1)y = 0, y (0) = 1, y 0 (0) = 2.
1 1 h 00
a3 = φ000 (x0 ) = − pφ + (p 0 + q)φ0 + q 0 φ .
3! 3! x=x0
Solution
Since P, Q, and R are polynomials and P(x0 ) 6= 0, all of the Since y = φ(x) is a solution of the initial value problem, we
derivatives of p and q exist at x0 . have
Hence we can continue to differentiate equation (13) φ(0) = 1, φ0 (0) = 2,
indefinitely, determining after each differentiation the successive and
coefficients a4 , a5 , . . . by setting x = x0 .
φ00 + xφ0 + ln(x + 1)φ = 0 or φ00 = −xφ0 − ln(x + 1)φ.

Thus
47 / 97 48 / 97
φ00 (0) = (0)φ0 (0) − ln(1)φ(0) =⇒ φ00 (0) = 0.
 5.4 Regular Singular Points 5.4 Regular Singular Points
5.5 Euler Equations 5.5 Euler Equations

φ(0) = 1, φ0 (0) = 2
In order to calculate φ000 (0) we differentiate
φ00 = −xφ0 − ln(x + 1)φ to obtain
Note that the important property that we used in determining
000 00 10 0
φ = −xφ − φ − ln(x + 1)φ − φ. the coefficients an was that we could compute infinitely many
x +1 derivatives of p(x) and q(x).
This gives
It might seem reasonable to relax the assumption that p and q
φ000 (0) = −(0)φ00 (0)−φ0 (0)−ln(1)φ0 (0)−φ(0) =⇒ φ000 (0) = −3. are rational functions and simply require that they be infinitely
differentiable in the neighborhood of x0 .
In order to calculate φiv (0) we differentiate
Unfortunately, this condition is too weak to ensure that we can
φ000 = −xφ00 − φ0 − ln(x + 1)φ0 − x+1 1
φ to obtain
prove the convergence of the resulting series expansion for
2 1 y = φ(x).
φiv = −xφ000 − 2φ00 − ln(x + 1)φ00 − φ0 + φ.
x +1 (x + 1)2
What is needed is to assume that the functions p and q are
This gives analytic at x0 .
φiv (0) = −(0)φ000 (0) − 2φ00 (0) − ln(1)φ00 (0) − 2φ0 (0) + φ(0)

=⇒ φiv (0) = −3.

49 / 97 50 / 97

5.4 Regular Singular Points 5.4 Regular Singular Points

5.5 Euler Equations 5.5 Euler Equations

Definition
(Ordinary Points)
Definition If p(x) and q(x) are analytic functions at point x0 , then x0 is called
an ordinary point of the equation
(Analytic Functions)
A function f (x) is said to be analytic at a point x = x0 if f (x) has a y 00 + p(x)y 0 + q(x)y = 0.
Taylor series expansion
∞ If x0 is not an ordinary point, then it is called a singular point.
X
n
f (x) = fn (x − x0 )
n=0
Example
Show that x = 0 is an ordinary point of the equation
that converge to it in some interval about x0 .
y 00 + e x y 0 + 3(x + 2)y = 0.

Solution
We have to show that p(x) = e x and
51 / 97 52 / 97
q(x) = 3(x + 2) are analytic functions at x = 0.
 5.4 Regular Singular Points 5.4 Regular Singular Points
5.5 Euler Equations 5.5 Euler Equations

Theorem
Let x0 be an ordinary point of the equation

y 00 + p(x)y 0 + q(x)y = 0.
∞
X xn
p(x) = e x = =⇒ p(x) is analytic at x = 0. Then
n!
n=0 (1) the general solution of the equation is given by
q(x) = 3(x + 1) = 3 + 3x =⇒ q(x) is analytic at x = 0.
∞
Therefore, x =0 is an ordinary point of the equation. X
y (x) = an (x − x0 )n = a0 y1 (x) + a1 y2 (x),
n=0

where a0 and a1 are arbitrary constants and y1 and

y2 are linearly independent series solutions that are analytic
at x0 .
(2) the radius of convergence of each series solution y1 and y2
53 / 97
is at least as large as the minimum of the radii of convergence 54 / 97
of the series of p(x) and q(x).

5.4 Regular Singular Points 5.4 Regular Singular Points

5.5 Euler Equations 5.5 Euler Equations

Lower bound for the radius of convergence

Let ρ1 be the radius of convergence of

∞
X Proposition
p(x) = pn (x − x0 )n and let ρ2 be the radius of
P(x)
n=0
∞
Let f (x) = , where P(x) and Q(x) are polynomials
X Q(x)
convergence of q(x) = qn (x − x0 )n . with no common factors. Then the radius of convergence of the
∞
n=0 X
Then the radius of convergence of the series solution power series f (x) = fn (x − x0 )n is the distance from x0 to
∞
X n=0
y (x) = an (x − x0 )n of the nearest zero of Q(x).
n=0
Remark
y 00 + p(x)y 0 + q(x)y = 0
In applying the above proposition we must consider real and
is ρ ≥ min {ρ1 , ρ2 }. complex roots of Q(x).
We can use the following proposition to find ρ1 and ρ2 .

55 / 97 56 / 97
 5.4 Regular Singular Points 5.4 Regular Singular Points
5.5 Euler Equations 5.5 Euler Equations

Examples
Example 1 Example 2
Determine a lower bound for the radius of convergence of the series Find a lower bound for the radius of convergence of the series
solution of solution of
(9 − x 2 )y 00 + 2y = 0 (x − 2)(x 2 + 1)y 00 − 4xy 0 + 6y = 0
about x = 1. about x = 0.

Solution Solution
2 −4x
In this example x0 = 1, p(x) = 0, and q(x) = . In this example x0 = 0, p(x) = , and
9 − x2 (x − 2)(x 2 + 1)
It follows that the radius of convergence for p(x) is 6
q(x) = .
ρ1 = ∞. (x − 2)(x 2 + 1)
The zeros of Q(x) = 9 − x 2 are x1 = 3 and x2 = −3. The zeros of Q(x) = (x − 2)(x 2 + 1) are x1 = 2, x2 = i,
Thus and x3 = −i.
ρ2 = min {d1 , d2 } where d1 = |3−1| = 2, d2 = |−3−1| = 4. Thus ρ1 = ρ2 = min {d1 , d2 , d3 }.
Hence ρ2 = 2. 57 / 97
First of all x0 = 0, x1 = 2 =⇒ d1 = |2 − 0| = 2. 58 / 97

Therefore, a lower bound is min {ρ1 , ρ2 } = min {∞, 2} = 2.

5.4 Regular Singular Points 5.4 Regular Singular Points

5.5 Euler Equations 5.5 Euler Equations

x2 = i, x3 = −i
Example 3
Find a lower bound for the radius of convergence of the series
In the complex plane, x0 and x2 can be represented as solution of
the point x0 = (0, 0), x2 = (0, 1). (x 2 + 2x + 5)y 00 + 6xy 0 + 4x 2 y = 0
√
Thus d2 = 1 = 1. about x = 0.
In the same way, x0 and x3 can be represented as the
point x0 = (0, 0), x3 = (0, −1). Solution
√ In this example
Thus d3 = 1 = 1.
6x 4x 2
It follows that the radius of convergence for p(x) and q(x) x0 = 0, p(x) = , and q(x) = 2 .
+ 2x + 5x2 x + 2x + 5
are ρ1 = ρ2 = min {2, 1, 1} = 1.
The zeros of Q(x) = x 2 + 2x + 5 are
Thus ρ ≥ 1. x1 = −1 + 2i, x2 = −1 − 2i.
Thus ρ1 = ρ2 = min {d1 , d2 }.
Let us calculate the distance between
x0 = 0 and x1 = −1 + 2i.
59 / 97 60 / 97
 5.4 Regular Singular Points 5.4 Regular Singular Points
5.5 Euler Equations 5.5 Euler Equations

x1 = −1 + 2i, x2 = −1 − 2i

In the complex plane, x0 and x1 can be represented as

the point x0 = (0, 0), x1 = (−1, 2). In this section we consider the equation
√
Thus d1 = 5.
P(x)y 00 + Q(x)y 0 + R(x)y = 0, (14)
In the same way, x0 and x2 can be represented as the
point x0 = (0, 0), x2 = (−1, −2). in the neighborhood of a singular point x0 .
√
Thus d2 = 5. Recall that if P(x), Q(x), and R(x) are polynomials having no
It follows that the radius of convergence for common factors, the singular points of the above equation are
√
p(x) and q(x) are ρ1 = 5 = ρ2 . the points for which P(x) = 0.
√
Thus ρ ≥ 5.

61 / 97 62 / 97

5.4 Regular Singular Points 5.4 Regular Singular Points

5.5 Euler Equations 5.5 Euler Equations

Definition
(Regular Singular Points)
Remarks
Let P(x), Q(x), and R(x) be polynomials with no common factors.
(1) If x0 is a singular point of equation (14), then we can not use A singular point x0 of the equation
the method of Section 5.2 to find series solution about x = x0 .
P(x)y 00 + Q(x)y 0 + R(x)y = 0
Q(x)
(2) Without additional information about the behavior of
P(x) is called a regular singular point of the equation if
R(x)
and near x0 , it is impossible to describe the behavior of Q(x) R(x)
P(x) lim (x − x0 ) and lim (x − x0 )2
the solutions of equation (14) near x0 . x→x0 P(x) x→x0 P(x)

are finite. If a singular point is not regular, it is called irregular

singular point.

63 / 97 64 / 97
 5.4 Regular Singular Points 5.4 Regular Singular Points
5.5 Euler Equations 5.5 Euler Equations

Examples

Example 1 x = 0:
Find the singular points of the equation Q(x) (x − 1) −1
lim (x − 0) = lim x =
x→0 P(x) x→0 x(x + 2) 2
x(x + 2)y 00 + (x − 1)y 0 + 5y = 0
R(x) 5
and lim (x − 0)2 = lim x 2 = 0.
and determine whether they are regular or irregular. x→0 P(x) x→0 x(x + 2)
Thus x = 0 is a regular singular point.
Solution x=-2:
In this example P(x) = x(x + 2), Q(x) = x − 1, R(x) = 5 are Q(x) (x − 1) 3
lim (x + 2) = lim (x + 2) =
polynomials with no common factors. x→−2 P(x) x→−2 x(x + 2) 2
R(x) 5
Thus the singular points are the solutions of P(x) = 0. and lim (x + 2)2 = lim (x + 2)2 = 0.
x→−2 P(x) x→−2 x(x + 2)
It follows that the singular points of the equation are x = 0 and
Thus x = −2 is a regular singular point.
x = −2.

65 / 97 66 / 97

5.4 Regular Singular Points 5.4 Regular Singular Points

5.5 Euler Equations 5.5 Euler Equations

Example 2
x = 0:
Find all singular points of the equation Q(x) (x − 1) 1
lim (x − 0) = lim x = lim does not
x 2 (x − 1)2 y 00 + (x − 1)y 0 − 3xy = 0
x→0 P(x) x→0 x 2 (x − 1)2 x→0 x(x − 1)
exist.
and classify them. Thus x = 0 is an irregular singular point.
x = 1:
Solution Q(x) (x − 1)
lim (x − 1) = lim (x − 1) 2 =1
In this example P(x) = x 2 (x − 1)2 , Q(x) = x − 1, R(x) = −3x x→1 P(x) x→1 x (x − 1)2
are polynomials with no common factors. and
R(x) −3x −3
Thus the singular points are the solutions of P(x) = 0. lim (x − 1)2 = lim (x − 1)2 2 2
= lim = −3.
x→1 P(x) x→1 x (x − 1) x→1 x
It follows that the singular points of the equation are x = 0 and Thus x = 1 is a regular singular point.
x = 1.

67 / 97 68 / 97
 5.4 Regular Singular Points 5.4 Regular Singular Points
5.5 Euler Equations 5.5 Euler Equations

Example 3 As a second step, we rewrite the summations so that x in each

Show that the equation xy 00 + y0 + y = 0 has only one nonzero of these summations will have the exponent n.
∞
X ∞ ∞ ∞
an x n .
X X X
n−1 n−1
solution of the form y = n(n − 1)an x + nan x + an−1 x n−1 = 0.
n=0 n=2 n=1 n=1
As a next step, we adjust the summations so that the ranges of
Solution
the summations are all the same.
The point x = 0 is a regular singular point of the equation. In this example we have
Substituting in the differential equation gives X∞ ∞
X ∞
X
∞ ∞ ∞ n(n − 1)an x n−1 + a1 + nan x n−1 + a0 + an−1 x n−1 = 0.
X X X
x n(n − 1)an x n−2 + nan x n−1 + an x n = 0. n=2 n=2 n=2

n=2 n=1 n=0 We can now combine like powers of x and write this equation as
∞
As a first step, we multiply the first series by x and we obtain
X
a1 + a0 + [n(n − 1)an + nan + an−1 ]x n−1 = 0.
∞ ∞ ∞ n=2
X X X
n−1 n−1 n
n(n − 1)an x + nan x + an x = 0. 69 / 97 70 / 97
n=2 n=1 n=0

5.4 Regular Singular Points 5.4 Regular Singular Points

5.5 Euler Equations 5.5 Euler Equations

The coefficient of each power of x in the left member of the

last equation must be equated to zero. This leads to the Substituting the values of an , n ≥ 2 into the assumed solution
conditions ∞
X
2 3
a0 + a1 = 0, n2 an + an−1 = 0, n ≥ 2. y = a0 + a1 x + a2 x + a3 x + · · · = an x n ,
n=0
It follows that a1 = −a0 and
we have
−1
an = 2 an−1 , n = 2, 3, . . . . a0 2 a0 3 a0
n y = a0 − a0 x + x − x + x4 + . . .
4 36 16(36)
−1 a0
n=2: a2 = a1 = As a result we obtain the solution
4 4
  
−1 −1 a0 a0 
1 1 1

n=3: a3 = a2 = =⇒ a3 = − y = a0 1 − x + x 2 − x 3 + x4 + . . . .
9 9 4 36 4 36 16(36)
  
−1 −1 −a0 a0
n=4: a4 = a3 = =⇒ a4 =
16 16 36 16(36)
71 / 97 72 / 97
 5.4 Regular Singular Points 5.4 Regular Singular Points
5.5 Euler Equations 5.5 Euler Equations

5.5 Euler Equations

The simplest equation with a regular singular point at x = 0 is
the Cauchy-Euler equation, or the equidimensional equation
L[y ] = x 2 y 00 + αxy 0 + βy = 0, x 6= 0,
where α and β are constants. In any interval not including the origin, the equation L[y ] = 0
In this equation P(x) = x 2 , Q(x) = αx, and R(x) = β. has a general solution of the form
Thus, y = c1 y1 (x) + c2 y2 (x),
Q(x) R(x)
lim x =α and lim x 2 =β where y1 and y2 are linearly independent solutions.
x→0 P(x) x→0 P(x)
and hence x = 0 is a regular singular point.
The solution of the Cauchy-Euler equation is typical of solutions
of all differential equations with a regular singular point.
Therefore, it is useful to to study the solutions of this equation
in detail.
73 / 97 74 / 97

5.4 Regular Singular Points 5.4 Regular Singular Points

5.5 Euler Equations 5.5 Euler Equations

Transformation to linear equation with constant Real Distinct Roots

coefficients
If the roots r1 and r2 are real and distinct, then the solution is
Let y (x) = y (t), t = ln |x|. y (t) = c1 e r1 t + c2 e r2 t .
Then
dt 1 Thus
= , y (x) = c1 |x|r1 + c2 |x|r2 .
dx x
0 dy dt 1 dy
y = = ,
dt dx x dt 2 Proposition
1 dy 1 d y
y 00 = − 2 + 2 2.
x dt x dt If the characteristic equation r 2 + (α − 1)r + β = 0 has two real
Substituting into the equation distinct roots r1 and r2 , then the general solution of the differential
L[y ] = x 2 y 00 + αxy 0 + βy = 0 equation
x 2 y 00 + αxy 0 + βy = 0
yields a second order linear equation with constant coefficients
d 2y dy is given by
+ (α − 1) + βy = 0. y (x) = c1 |x|r1 + c2 |x|r2 ,
dt 2 dt
The characteristic equation reads r 2 + (α − 1)r + β = 0. where c1 and c2 are arbitrary constants.
75 / 97 76 / 97
 5.4 Regular Singular Points 5.4 Regular Singular Points
5.5 Euler Equations 5.5 Euler Equations

Examples

Example 1 Example 2
Solve x 2 y 00 − 4xy 0 + 4y = 0. Solve x 2 y 00 + 7xy 0 + 5y = x, x > 0.

Solution
Solution
First of all we find the general solution yc of the homogeneous
The given equation is a Cauchy-Euler equation with α = −4
equation x 2 y 00 + 7xy 0 + 5y = 0.
and β = 4.
This is a Cauchy-Euler equation with α = 7 and β = 5.
Its characteristic equation is
r 2 + (α − 1)r + β = r 2 − 5r + 4 = 0. Thus characteristic equation is
r 2 + (α − 1)r + β = r 2 + 6r + 5 = 0.
The roots of the characteristic equation are r1 = 1 and r2 = 4.
The roots of the characteristic equation are r1 = −1 and
Therefore, the general solution is
r2 = −5.
y = c1 |x| + c2 x 4 . Therefore, the general solution of the homogeneous equation is

yc = c1 x −1 + c2 x −5 .
77 / 97 78 / 97

5.4 Regular Singular Points 5.4 Regular Singular Points

5.5 Euler Equations 5.5 Euler Equations

Next we look for a particular solution yp of the It follows that

nonhomogeneous equation x 2 y 00 + 7xy 0 + 5y = x. −x −5 x
We have to use the method of variation of parameters. u10 = −7
= ,
x(−4x ) 4
1 x −1 1
Note that g (x) = . 0
u2 = = − x5
x x(−4x )−7 4
So we assume
and hence
yp = u1 y1 + u2 y2 = u1 x −1 + u2 x −5 , x
Z
1
u1 = dx = x 2 ,
where u1 and u2 satisfy 4Z 8
1 1
u10 = −y2 g (x) u2 = − x dx = − x 6 .
5
W (y1 ,y2 ) , 4 24
y1 g (x)
u20 = W (y1 ,y2 ) . Thus
x2 x6 x x 1
1 yp = x −1 − x −5 = − = x.
Here g (x) = , y1 (x) = x −1 , y2 (x) = x −5 . 8 24 8 24 12
x and the general solution is
Let us calculate W (y1 , y2 ):
1
x −1 x −5 y = c1 x −1 + c2 x −5 + x.
W (y1 , y2 ) = −2 = −4x −7 . 12
−x −5x −6 79 / 97 80 / 97
 5.4 Regular Singular Points 5.4 Regular Singular Points
5.5 Euler Equations 5.5 Euler Equations

Equal Roots
If the roots of r 2 + (α − 1)r + β = 0 are equal, we have Example 1
y (t) = c1 e r1 t + c2 te r1 t . Solve x 2 y 00 − 3xy 0 + 4y = 0.
It follows that
Solution
y (x) = c1 |x|r1 + c2 |x|r1 ln |x|.
The given equation is a Cauchy-Euler equation with α = −3
and β = 4.
Proposition Its characteristic equation is
r 2 + (α − 1)r + β = r 2 − 4r + 4 = 0.
If the characteristic equation r 2 + (α − 1)r + β = 0 has two equal
real r1 = r2 , then the general solution of the differential equation The roots of the characteristic equation are r1 = r2 = 2.
Therefore, the general solution is
x 2 y 00 + αxy 0 + βy = 0
y = c1 x 2 + c2 x 2 ln |x|.
is given by
y (t) = c1 |x|r1 + c2 |x|r1 ln |x|,
where c1 and c2 are arbitrary constants. 81 / 97 82 / 97

5.4 Regular Singular Points 5.4 Regular Singular Points

5.5 Euler Equations 5.5 Euler Equations

Example 2
Solve the initial value problem
x 2 y 00 + 3xy 0 + y = 0, y (−1) = 1, y 0 (−1) = 2. Thus
y = c1 (−x)−1 + c2 (−x)−1 ln(−x) = −c1 x −1 − c2 x −1 ln(−x).
Solution It follows that y (−1) = 1 =⇒ c1 = 1.
The given equation is a Cauchy-Euler equation with α = 3 and Next we find y 0 = c1 x −2 + c2 x −2 ln(−x) − c2 x −2 .
β = 1.
It follows that y 0 (−1) = 2 =⇒ c1 − c2 = 2 =⇒ c2 = −1.
Its characteristic equation is
Therefore, the solution of the initial value problem is
r 2 + (α − 1)r + β = r 2 + 2r + 1 = 0.
The roots of the characteristic equation are r1 = r2 = −1. y = |x|−1 − |x|−1 ln |x|.
Therefore, the general solution is

y = c1 |x|−1 + c2 |x|−1 ln |x|.

Now we apply the initial conditions.

First of all the initial conditions implies that we solve for x < 0.83 / 97 84 / 97
 5.4 Regular Singular Points 5.4 Regular Singular Points
5.5 Euler Equations 5.5 Euler Equations

Complex Roots
If the roots of r 2 + (α − 1)r + β = 0 are complex numbers Example 1
r1 = λ + iµ, r2 = λ − iµ, Solve x 2 y 00 + 3xy 0 + 5y = 0.
then we have
Solution
y (t) = c1 e λt cos(µt) + c2 e λt sin(µt).
The given equation is a Cauchy-Euler equation with α = 3 and
It follows that β = 5.
y (x) = c1 |x|λ cos(µ ln |x|) + c2 |x|λ sin(µ ln |x|). Its characteristic equation is
r 2 + (α − 1)r + β = r 2 + 2r + 5 = 0.

Proposition The roots of the characteristic equation are r1,2 = −1 ± 2i.

Therefore, the general solution is
If the characteristic equation r 2 + (α − 1)r + β = 0 has complex h i
roots r1 = λ + iµ and r2 = λ − iµ, then the general solution of the y = |x|−1 c1 cos(2 ln |x|) + c2 sin(2 ln |x|) .
differential equation x 2 y 00 + αxy 0 + βy = 0 is given by

y (x) = c1 |x|λ cos(µ ln |x|) + c2 |x|λ sin(µ ln |x|), 85 / 97 86 / 97

where c1 and c2 are arbitrary constants.

5.4 Regular Singular Points 5.4 Regular Singular Points
5.5 Euler Equations 5.5 Euler Equations

Example 2 Next we look for a particular solution yp of the

Solve 2x 2 y 00 + 2xy 0 + 8y = 1. nonhomogeneous equation x 2 y 00 + xy 0 + 4y = 12 .
We have to use the method of variation of parameters.
Solution 1
Note that g (x) = 2 .
First of all we find the general solution yc of the homogeneous 2x
So we assume
equation 2x 2 y 00 + 2xy 0 + 8y = 0.
This is a Cauchy-Euler equation with α = 1 and β = 4. yp = u1 y1 + u2 y2 = u1 cos(2 ln |x|) + u2 sin(2 ln |x|),
Thus characteristic equation is where u1 and u2 satisfy
r 2 + (α − 1)r + β = r 2 + 4 = 0. −y2 g (x)
The roots of the characteristic equation are r1 = 2i and u10 = ,
W (y1 , y2 )
r2 = −2i. y1 g (x)
u20 = .
Therefore, the general solution of the homogeneous equation is W (y1 , y2 )
1
yc = c1 cos(2 ln |x|) + c2 sin(2 ln |x|). Here g (x) = , y1 (x) = cos(2 ln |x|), y2 (x) = sin(2 ln |x|).
2x 2
87 / 97 88 / 97
 5.4 Regular Singular Points 5.4 Regular Singular Points
5.5 Euler Equations 5.5 Euler Equations

It follows that
− sin(2 ln |x|) sin(2 ln |x|)
u10 = 2 −1
=− ,
2x (2x ) 4x
Let us calculate W (y1 , y2 ): cos(2 ln |x|) cos(2 ln |x|)
u20 = =
2x 2 (2x −1 ) 4x
cos(2 ln |x|) sin(2 ln |x|) and hence
W (y1 , y2 ) =
− x2 sin(2 ln |x|) 2
x cos(2 ln |x|) sin(2 ln |x|)
Z
1
u1 = − dx = cos(2 ln |x|),
4x 8
2 2 2 cos(2 ln |x|)
Z
1
= cos2 (2 ln |x|) + sin2 (2 ln |x|) = . u2 = dx = sin(2 ln |x|).
x x x 4x 8
Thus
1 1 1
yp = cos(2 ln |x|) cos(2 ln |x|) + sin(2 ln |x|) sin(2 ln |x|) = .
8 8 8
and the general solution is
y = c1 cos(2 ln |x|) + c2 sin(2 ln |x|) + 18
89 / 97 90 / 97

5.4 Regular Singular Points 5.4 Regular Singular Points

5.5 Euler Equations 5.5 Euler Equations

Example 3
We have yc = c1 cos(2t) + c2 sin(2t).
Solve x 2 y 00 + xy 0 + 4y = sin(ln x), x > 0.
Next we look for a particular solution yp of the
nonhomogeneous equation.
Solution
In this example g (t) = sin t.
Let y (x) = y (t), t = ln x.
It follows that
Then
dt 1
dx
= .
x yp = t s (A0 cos t + B0 sin t).

y 0 = dy dt 1 dy
dt dx = x dt , Since r1,2 = ±2i, we have s = 0.
1 d 2y
y 00 = − x12 dy
dt + x 2 dt 2 . Substituting yp into the equation gives
Substituting into the equation
−A0 cos t − B0 sin t + 4(A0 cos t + B0 sin t) = sin t.
2 00 0
x y + xy + 4y = sin(ln x)
It follows that A0 = 0 and B0 = 13 .
d 2y
yields + 4y = sin(t). 91 / 97 92 / 97
dt 2
 5.4 Regular Singular Points 5.4 Regular Singular Points
5.5 Euler Equations 5.5 Euler Equations

General Form of Euler Equation

Hence The solution of an Euler equation of the form

1
yp = sin t
3
(x − x0 )2 y 00 + α(x − x0 )y 0 + βy = 0, x 6= x0
and
1
y = c1 cos(2t) + c2 sin(2t) + sin t. can be obtained in similar way.
3
We take t = ln |x − x0 | and the characteristic equation is the
Lastly we substitute t = ln x to obtain
same r 2 + (α − 1)r + β = 0.
y = c1 cos(2 ln x) + c2 sin(2 ln x) + 13 sin(ln x).

93 / 97 94 / 97

5.4 Regular Singular Points 5.4 Regular Singular Points

5.5 Euler Equations 5.5 Euler Equations

Example 1
Example 2
Solve (x + 1)2 y 00 + 3(x + 1)y 0 + 3
4y = 0. Solve (3x − 1)2 y 00 + (9x − 3)y 0 − 9y = 0.

Solution Solution
The given equation is a Cauchy-Euler equation with x0 = −1 Firstly we write the equation as
and α = 3 and β = 34 .
2
x − 13 y 00 + x − 13 y 0 − y = 0.


Its characteristic equation is Thus the given equation is a Cauchy-Euler equation with
3
r 2 + (α − 1)r + β = r 2 + 2r + 4 = 0. x0 = 13 and α = 1 and β = −1.
The roots √
of the characteristic equation are given by Its characteristic equation is r 2 + (α − 1)r + β = r 2 − 1 = 0.
r1,2 = −2±2 4−3 = −2±1
2 . The roots of the characteristic equation are r1 = 1 and
So r1 = − 23 and r2 = − 12 . r2 = −1.
Therefore, the general solution is 1 1 −1
Therefore, the general solution is y = c1 x − 3 + c2 x − 3
y = c1 |x + 1|−3/2 + c2 |x + 1|−1/2 , x 6= −1
95 / 97 96 / 97
5.4 Regular Singular Points
5.5 Euler Equations

Õç'

97 / 97