1.

Non-linear second order ordinary differential

equations
The most common homogeneous form of the non-linear second order
ordinary differential equation
2
2dy dy
x + ax + by = 0, where a, b are constant numbers, (1)
dx2 dx
is called the homogeneous Cauchy Euler equation. This equation can
be transformed into linear homogeneous differential equation with con-
stant coefficient by changing the independent variable: In fact, in order
to solve this equation, let us consider the change of variable
t = ln(x) ⇔ x = et.
Differentiating gives
dy dy dt 1 dy dy dy
= = ⇒ x = , (2)
dx dt dx x dt dx dt
( ) ( ) ( )
d2y d dy d 1 dy d −t dy dt
= = = e .
dx2 dx dx dx x dt dt dt dx
From which it follows
( ) ( 2 )
d2 y d 2
y dy dt d y dy
2
= e−t 2 − e−t = e−t 2
− e−t.
dx dt dt dx dt dt
Consequently, in view of x = et (hence x2 = e2t), we get
( 2 ) ( 2 )
d2 y d y dy 1 d 2
y d y dy
2
= 2
− 2t
⇒ x2 2 = − . (3)
dx dt dt e dx dt2 dt
Substituting from (2) and (3) into the differential equation (1), would
leads to ( 2 )
d y dy dy
− + a + by = 0.
dt2 dt dt
That is, we arrive at
d2 y dy
+ (a − 1) + by = 0. (4)
dt2 dt
This equation in y(t) is solved via its characteristic polynomial
λ2 + (a − 1)λ + b = 0.
1
 2

Now let λ1 and λ2 denote the two roots of this polynomial. We analyze
the three main cases: distinct real roots, complex roots and double roots:

If the roots are real and distinct, the general solution of (9) is
y = c1eλ1t + c2eλ2t.
If the roots are complex α ± iβ, the general solution of (9) is
y = eαt (c1 cos βt + c2 sin βt) .
If the roots are equal, the general solution of (9) is
y = c1eλ1t + c2teλ1t.
In all cases, the solution of the differential equation (1) may be found
by setting t = ln(x). Hence, in the first case, the general solution of (1)
is
y(x) = c1xλ1 + c2xλ2 ,
and in the second case, the general solution of (1) is
y = xα (c1 cos(β ln x) + c2 sin(β ln x)) .
Also in the third case, the general solution of (1) is
y(x) = c1xλ1 + c2 ln(x)xλ1 = xλ1 (c1 + c2 ln(x))) .
Example 1.1. Find the solution of
x2y ′′ − xy ′ − 3y = 0.
First we recognize that the equation is an Euler-Cauchy equation, with
a = −1 and b = −3.
1. Characteristic equation is λ2 − 2λ − 3 = 0 and the roots are −1, 3.
2. Since −1, 3 are distinct roots, the solution is
c1
y(x) = c1e−t + c2e3t = + c2x3.
x
Example 1.2. Find the solution of
x2y ′′ − 3xy ′ + 5y = 0.
First we recognize that the equation is an Euler- Cauchy equation, with
a = −3 and b = 5.
1. Characteristic equation is λ2 − 3λ + 5 = 0 and the roots are 2 ± i.
 3

2. Since 2 ± i are distinct roots, the solution is

y = e2t (c1 cos(t) + c2 sin(t)) = x2 (c1 cos(ln x) + c2 sin(ln x)) .
Example 1.3. Find the solution of
x2y ′′ − xy ′ + y = 0.
First we recognize that the equation is an Euler-Cauchy equation, with
a = −1and b = 1.
1. Characteristic equation is λ2 − 2λ + 1 = 0 and the roots are 1, 1.
2. Since 1 is a double root, the solution is
y(x) = c1x + c2 ln(x)x = x (c1 + c2 ln(x)) .
In what follows, we consider the non-homogeneous case of the Cauchy Euler e
namely we deal with
2
2d
y dy
x + ax + by = f (x), where a, b are constant numbers, (5)
dx2 dx
This equation can be transformed into linear non-homogeneous differ-
ential equation with constant coefficient by changing the independent
variable t = ln x(or x = et) as follows
d2 y dy
2
+ (a − 1) + by = f (et). (6)
dt dt
In this case we have to find the general solution yG.S. for (6) (hence, the
general solution for (5)) as we will show in the following
Example 1.4. Find the solution of
2
y2d dy
x − 2x + 2y = 4x 3
.
dx2 dx
First we recognize that the equation is a non-homogeneous Euler-Cauchy
equation, with a = −2 and b = 2. The change of variable
t = ln(x) ⇔ x = et.
convert this differential equation into
d2y dy
− 3 + 2y = 4(et 3
) = 4e 3t
. (7)
dt2 dt
1. Characteristic equation is λ2 − 3λ + 2 = 0 and the roots are 1, 2.
 4

2. Since Since 1, 2 are distinct roots, the complementary solution will

be
yc(x) = c1et + c2e2t = c1x + c2x2.
Now, to find the particular solution yp, we note that
1 { 3t} 1 { 3t} 4e3t
yp = 2 4e = 2 4e = .
D − 3D + 2 3 − 3(3) + 2 2
Hence
4x3
yp = = 2x3.
2
Then
yG.S. = yc + yp = c1x + c2x2 + 2x3.

Example 1.5. Find the solution of

x2y ′′ − xy ′ + y = 2 ln(x).
First we recognize that the equation is a non-homogeneous Euler-
Cauchy equation, with a = −1 and b = 1. The change of variable
t = ln(x) ⇔ x = et.
convert this differential equation into
d2 y dy
+ (−1 − 1) + y = 2t. (8)
dt2 dt
1. Characteristic equation is λ2 − 2λ + 1 = 0 and the roots are 1, 1.
2. Since 1 is a double root, the complementary solution will be
yc(x) = c1et + c2tet = x (c1 + c2 ln(x)) .
Now, to find the particular solution yp, we note that
1 1
yp = {2t} = {2t} = 2(D − 1)−1(D − 1)−1 {t} .
D − 2D + 1
2 (D − 1) 2

That is
( )( )
yp = 2 1 + D + D + · · · 1 + D + D + · · · {t}
2 2
( )
= 2 1 + D + D2 + · · · (t + 1) = 2 (t + 1 + 1)
Hence
yp = 2 (ln x + 2)
 5

Then
yG.S. = yc + yp = x (c1 + c2 ln x) + 2 (ln x + 2) .

Example 1.6. Find the solution of

x2y ′′ − xy ′ − 3y = x sin(ln x).
First we recognize that the equation is a non-homogeneous Euler-
Cauchy equation, with a = −1and b = 1. The change of variable
t = ln(x) ⇔ x = et.
covert the differential equation into
d2 y dy
+ (−1 − 1) − 3y = e t
sin t. (9)
dt2 dt
1. Characteristic equation is λ2 − 2λ − 3 = 0 and the roots are −1, 3.
2. Since −1, 3 are distinct roots, the complementary solution will be
c1
y(x) = c1e−t + c2e3t = + c2x3.
x
Now, to find the particular solution yp, we note that
1 { t } 1 1
yp = 2 e sin t = et {sin t} = et 2 {sin t} .
D − 2D − 3 (D + 2)(D − 2) D −4
Hence
1 −et −x sin(ln x)
yp = e t
{sin t} = sin t = .
−1 − 4 5 5
Then
c1 x sin(ln x)
yG.S. = yc + yp = + c2x3 − .
x 5

PROBLEMS
Solve the following Cauchy Euler equation differential equations
1. x2y ′′ + 3xy ′ + y = 24x.
2. x2y ′′ − 3xy ′ + 4y = (1 + x)2.
3. x2y ′′ − 2xy ′ + 2y = (ln x)2 + ln x2.
4. x2y ′′ + 8xy ′ + 12y = 3x5 − 2 cos(ln x).
5. x2y ′′ + 4xy ′ + 2y = x ln x.
6. x2y ′′ + 5xy ′ + 3y = x + sin(ln x).