Module 5: Linear Differential Equation of

Higher Order

Linear Differential Equation of Higher Order

with Constant Coefficients
General Form:
A linear differential equation with constant coefficients is given by:

dn y dn−1 y dy
an n
+ a n−1 n−1
+ · · · + a1 + a0 y = f (x)
dx dx dx
where a0 , a1 , . . . , an are constants.
Homogeneous Equation: When f (x) = 0:

an Dn y + an−1 Dn−1 y + · · · + a0 y = 0

Let F (D) = an Dn + an−1 Dn−1 + · · · + a0 .

The auxiliary (or characteristic) equation is:

F (m) = an mn + an−1 mn−1 + · · · + a0 = 0

Complementary Function (C.F.): If the roots of the auxiliary equa-

tion are m1 , m2 , . . . , mn :

• Distinct Real Roots:

yc = C1 em1 x + C2 em2 x + · · · + Cn emn x

• Repeated Real Roots: If a root m is repeated r times:

yc = (C1 + C2 x + C3 x2 + · · · + Cr xr−1 )emx

• Complex Roots: If the roots are m = α ± iβ:

yc = eαx (C1 cos βx + C2 sin βx)

Solution Method
1. Write the Auxiliary (Characteristic) Equation: Replace D by m:

F (m) = 0.

1
2. Find the roots m1 , m2 , . . . , mn .
3. Form the Complementary Function (C.F.) based on the nature of the
roots:



 y = C1 em1 x + C2 em2 x + · · · + Cn emn x , distinct real roots,



y = (C1 + C2 x + · · · + Cr xr−1 )emx , repeated roots,




y = eαx (C1 cos βx + C2 sin βx),

complex roots.

Examples Example 1: Distinct Real Roots

Given:
(D2 − 5D + 6)y = 0.
Solution:
Auxiliary equation:

m2 − 5m + 6 = 0 ⇒ (m − 2)(m − 3) = 0.

Thus, m1 = 2, m2 = 3.

y = C1 e2x + C2 e3x .
Example 2: Repeated Real Roots
Given:
(D − 2)2 y = 0.
Solution:
Auxiliary equation:

(m − 2)2 = 0 ⇒ m = 2 (repeated root).

For a repeated root m = 2 of multiplicity 2:

y = (C1 + C2 x)e2x .

Example 3: Complex Roots

Given:
(D2 + 4)y = 0.
Solution:
Auxiliary equation:

m2 + 4 = 0 ⇒ m = ±2i.

Complex roots m = 0 ± 2i ⇒ α = 0, β = 2.

2
 Hence,
y = C1 cos 2x + C2 sin 2x.
Example 4: Real and Complex Mixed Roots
Given:
(D − 1)(D2 + 9)y = 0.
Solution:
Auxiliary equation:
(m − 1)(m2 + 9) = 0 ⇒ m = 1, m = ±3i.
Hence, the solution is a combination of exponential and trigonometric
terms:
y = C1 ex + e0x (C2 cos 3x + C3 sin 3x) .
or simply,
y = C1 ex + C2 cos 3x + C3 sin 3x.
Example 5: Higher Order with Repeated and Complex Roots
Given:
(D2 + 2D + 2)2 y = 0.
Solution:
Auxiliary equation:
(m2 + 2m + 2)2 = 0.
Solving m2 + 2m + 2 = 0 ⇒ m = −1 ± i.
Each complex root is repeated twice.
So, the complementary function is:
y = e−x (C1 + C2 x) cos x + (C3 + C4 x) sin x .
 

y = e−x (C1 + C2 x) cos x + (C3 + C4 x) sin x .

 

Conclusion For a homogeneous differential equation:

F (D)y = 0,
the general solution is:

y = yc = (sum of terms corresponding to each root of F (m) = 0).

Particular Integral (P.I.)

For the non-homogeneous part f (x):
1
yp = f (x)
F (D)

3
Different Forms of Particular Integral (P.I.) for Linear
Differential Equations with Constant Coefficients
General Idea
For a linear differential equation with constant coefficients:

F (D)y = f (x)

where F (D) = an Dn +an−1 Dn−1 +· · ·+a0 , the Particular Integral (P.I.)

is defined as:
1
yp = f (x)
F (D)
The method of evaluating yp depends on the form of f (x).
Case 1: f (x) = eax
1 ax
yp = e
F (D)
Using the shift rule F (D)eax = eax F (D + a), we get:
eax
yp =
F (a)

Special Case: If F (a) = 0, that means a is a root of the auxiliary

equation. Then we multiply by x repeatedly until the denominator is non-
zero:
xr eax
yp =
r!F (r) (a)
where r is the number of times (D − a) is a factor in F (D).
Case 2: f (x) = sin bx or cos bx
Let’s consider:

F (D2 )y = sin bx or F (D2 )y = cos bx

If F (−b2 ) ̸= 0 Then,

1 sin bx 1 cos bx
yp = 2
sin bx = , yp = cos bx =
F (D ) F (−b2 ) 2
F (D ) F (−b2 )

Special Case: If F (−b2 ) = 0, then multiply the right-hand side by x

successively:
sin bx cos bx
yp = xr or yp = xr
r!F (r) (−b2 ) r!F (r) (−b2 )

4
If F (D) = D2 + b2 , then
1 −x 1 x
yp = sin bx = cos bx, yp = cos bx = sin bx
F (D) 2b F (D) 2b

Case 3: f (x) = Polynomial in x

If
f (x) = a0 + a1 x + a2 x2 + · · · + an xn
then assume
yp = A0 + A1 x + A2 x2 + · · · + An xn
Substitute into the given differential equation and equate coefficients of
like powers of x to solve for A0 , A1 , . . . , An .
Case 4: f (x) = eax P (x)
If the right-hand side is an exponential multiplied by a polynomial:

f (x) = eax P (x)

Then:
1 ax
yp = e P (x)
F (D)
Using the shift rule F (D)eax = eax F (D + a):
1
yp = eax P (x)
F (D + a)
1
Now apply the polynomial case to find F (D+a) P (x).
ax
Case 5: f (x) = e (sin bx or cos bx)
Let
f (x) = eax (sin bx or cos bx)
Then:
1 ax
yp = e (sin bx or cos bx)
F (D)
Using the shift rule again:
1
yp = eax (sin bx or cos bx)
F (D + a)

Case 6: f (x) = K (a constant)

When f (x) is a constant K:

K
yp = , provided F (0) ̸= 0
F (0)

5
 If F (0) = 0, multiply by x until it becomes nonzero.
Case 7: f (x) = eax Q(x) sin bx or eax Q(x) cos bx
For a mixed case of exponential, polynomial, and trigonometric terms:

f (x) = eax Q(x) sin bx or f (x) = eax Q(x) cos bx

then
1
yp = eax Q(x)(sin bx or cos bx)
F (D + a)
and evaluate as in previous cases.
General Solution Finally, the complete solution of

F (D)y = f (x)

is
y = y c + yp
where yc is the complementary function (solution of the homogeneous part),
and yp is the particular integral.

Examples of Particular Integral (P.I.) for Linear Dif-

ferential Equations with Constant Coefficients
General Reminder

For a linear differential equation:

F (D)y = f (x),

the Particular Integral (P.I.) is given by:

1
yp = f (x).
F (D)

Example 1: f (x) = eax

Given:
(D2 − 3D + 2)y = e2x .
Solution:
Here, F (D) = D2 − 3D + 2.

F (a) = a2 − 3a + 2 = (a − 1)(a − 2).

For a = 2:
F (2) = 4 − 6 + 2 = 0.

6
Since F (2) = 0, we multiply by x:

e2x
yp = x .
F ′ (2)

Now, F ′ (D) = 2D − 3 ⇒ F ′ (2) = 1.

yp = xe2x .
The complementary function (C.F.) is:

yc = C1 ex + C2 e2x .

Hence, the complete solution:

y = C1 ex + C2 e2x + xe2x .

Example 2: f (x) = sin bx or cos bx

Given:
(D2 + 9)y = sin 3x.
Solution:
We know:
1 sin 3x
yp = sin 3x = .
F (D) F (−b2 )
Here, F (D) = D2 + 9 ⇒ F (−b2 ) = (−9) + 9 = 0.
So, F (−b2 ) = 0, we multiply by x:
sin 3x
yp = x .
F ′ (−b2 )

Now, F ′ (D) = 2D ⇒ F ′ (−b2 ) = 2D, so we apply:

1 x sin 3x
yp = x · sin 3x = · .
2D 2 D
1
Since D
sin 3x = − 31 cos 3x,

x
yp = − cos 3x .
6

C.F. from F (D) = D2 + 9 = 0:

yc = C1 cos 3x + C2 sin 3x.

7
Final solution:
x
y = C1 cos 3x + C2 sin 3x − cos 3x .
6

Example 3: f (x) = Polynomial in x

Given:
(D2 + 1)y = x2 .
Solution:
Assume yp = Ax2 + Bx + C.
Then:

yp′′ = 2A, (D2 + 1)yp = 2A + (Ax2 + Bx + C).

Equating with RHS x2 :

Ax2 + Bx + (C + 2A) = x2 .

Compare coefficients:

A = 1, B = 0, C + 2A = 0 ⇒ C = −2.

yp = x2 − 2.
C.F. from F (D) = D2 + 1 = 0 ⇒ D = ±i.

yc = C1 cos x + C2 sin x.
Hence:
y = C1 cos x + C2 sin x + x2 − 2.
Example 4: f (x) = eax P (x)
Given:
(D2 − 2D + 1)y = ex (x + 1).
Solution:
Using shift rule:
1
yp = ex (x + 1).
F (D + 1)
Compute:

F (D + 1) = (D + 1)2 − 2(D + 1) + 1 = D2 .

Hence:
1
yp = ex (x + 1).
D2
8
 We know:
x3 x2
Z Z
1
(x + 1) = (x + 1)dxdx = + .
D2 6 2
Thus: !
x3 x 2
yp = ex + .
6 2

C.F. from (D − 1)2 = 0 ⇒

yc = (C1 + C2 x)ex .

Final solution:
!
x3 x2
y = (C1 + C2 x)ex + ex + .
6 2

Example 5: f (x) = eax sin bx or eax cos bx

Given:
(D2 − 2D + 2)y = ex sin x.
Solution:
By shift rule:
1
yp = ex sin x.
F (D + 1)
Compute:

F (D + 1) = (D + 1)2 − 2(D + 1) + 2 = D2 + 1.

Thus:
sin x sin x sin x
yp = ex 2
= ex = ex .
F (−b ) (−1) + 1 0
Since denominator is zero, multiply by x:
sin x
yp = xex .
F ′ (−1)

Now F ′ (D) = 2D ⇒ F ′ (−1) = 2D.

Hence:
xex cos x
yp = − .
2
C.F. from (D2 − 2D + 2) = 0 ⇒ D = 1 ± i.

9
 yc = ex (C1 cos x + C2 cos x).
Final solution:
xex sin x
y = ex (C1 cos x + C2 sin x) − .
2

Example 6: f (x) = K (Constant)

Given:
(D2 + 2D + 1)y = 4.
Solution:
We know:
4 4
yp = = = 4.
F (0) 1
C.F. from (D + 1)2 = 0:

yc = (C1 + C2 x)e−x .

Hence:
y = (C1 + C2 x)e−x + 4.
Conclusion: For any linear differential equation with constant coeffi-
cients:
F (D)y = f (x),
the complete solution is:
y = y c + yp
where yc is found from the auxiliary equation F (m) = 0, and yp is found
using the appropriate form according to f (x).

Method of variation of Parameters:

Variation of Parameters for ay ′′ + by ′ + cy = f (x)
Assume a ̸= 0 and let the homogeneous equation

ay ′′ + by ′ + cy = 0

have two linearly independent solutions y1 (x), y2 (x) (found from the char-
acteristic equation ar2 + br + c = 0).
The general solution of the non-homogeneous equation is

y(x) = yc (x) + yp (x) = C1 y1 (x) + C2 y2 (x) + yp (x),

10
where a particular solution yp is sought in the form

yp (x) = u1 (x)y1 (x) + u2 (x)y2 (x),

with u1 , u2 functions to determine. Impose the standard two conditions to

make the system solvable:

u′1 (x)y1 (x) + u′2 (x)y2 (x) = 0,

f (x)
u′1 (x)y1′ (x) + u′2 (x)y2′ (x) =
 .
a
Write this linear system in matrix form:
 
y1 y2
! ′
u1
! 0
=  f (x)  .
 
′ ′ ′
y1 y2 u2
a

Let the Wronskian be W (x) = y1 y2′ − y1′ y2 (assumed nonzero). Solving

gives
f (x) y2 (x) f (x) y1 (x)
u′1 (x) = − , u′2 (x) = .
a W (x) a W (x)
Integrate:
Z Z
f (x) y2 (x) f (x) y1 (x)
u1 (x) = − dx, u2 (x) = dx.
a W (x) a W (x)

Finally, the particular solution is

Z Z
f (x) y2 (x) f (x) y1 (x)
yp (x) = −y1 (x) dx + y2 (x) dx,
a W (x) a W (x)

and the full solution is

Z
f (x) y2 (x)
y(x) =C1 y1 (x) + C2 y2 (x) − y1 (x) dx
a W (x)
Z
f (x) y1 (x)
+ y2 (x) dx.
a W (x)

Notes:

• This works for constant or variable coefficients as long as you have

two independent solutions y1 , y2 of the homogeneous equation and
W ̸= 0.

11
 • If the characteristic equation has repeated roots, pick y1 = erx and
y2 = xerx and apply the same formulas.

Example
y ′′ + y = sin x
Step 1: Solve the homogeneous equation

y ′′ + y = 0

The auxiliary equation is r2 + 1 = 0 ⇒ r = ±i. Hence,

y1 = cos x, y2 = sin x.

Step 2: Compute the Wronskian

y1 y2 cos x sin x
W = ′ ′ = = cos2 x + sin2 x = 1.
y1 y2 − sin x cos x

Step 3: Apply variation of parameters

y2
u′1 = −f (x) = − sin x · sin x = − sin2 x,
W
y1
u′2 = f (x) = sin x · cos x = sin x cos x.
W
Integrate:

1 − cos(2x)
Z Z
2 x sin(2x)
u1 = − sin x dx = − dx = − + ,
2 2 4
Z Z
1 1
u2 = sin x cos x dx = sin(2x) dx = − cos(2x).
2 4
Step 4: Particular solution
 
x sin(2x) 1
yp = u1 y1 + u2 y2 = − + cos x − cos(2x) sin x.
2 4 4

Step 5: General solution

x sin(2x) 1
y = C1 cos x + C2 sin x − cos x + cos x − cos(2x) sin x.
2 4 4

Example 2:
y ′′ − y = ex

12
 Homogeneous equation:

y ′′ − y = 0 =⇒ r2 − 1 = 0 =⇒ r = ±1.

Hence,
y1 = ex , y2 = e−x .
Wronskian:
ex e−x
W = = −2.
ex −e−x
Variation of parameters:

f (x)y2 ex e−x 1 f (x)y1 ex ex e2x

u′1 = − =− = , u′2 = = =− .
W −2 2 W −2 2
Integrate:
x 1
u1 = , u2 = − e2x .
2 4
Particular solution:
 
x 1 x 1 x 1
yp = u1 y1 + u2 y2 = ex − e2x e−x = ex − ex = ex − .
2 4 2 4 2 4

General solution:
 
x −x x x 1
y = C 1 e + C2 e +e − .
2 4

Solution of Second-Order Differential Equa-

tions with Variable Coefficients
General Form
A second-order linear differential equation with variable coefficients is given
by:
d2 y dy
a(x) 2 + b(x) + c(x)y = f (x),
dx dx
or equivalently,
d2 y dy
+ P (x) + Q(x)y = R(x),
dx2 dx
b(x) c(x) f (x)
where P (x) = , Q(x) = , and R(x) = .
a(x) a(x) a(x)
Types of Equations and Methods of Solution

13
 There is no single method that works for all variable-coefficient equa-
tions. We use one of the following approaches depending on the form of
the equation.
(a) Reduction of Order (When One Solution is Known)
If one solution y1 of the homogeneous equation

d2 y dy
2
+ P (x) + Q(x)y = 0
dx dx
is known, we can find the second solution y2 by assuming

y = vy1 , where v is a function of x.

Procedure:
y ′ = v ′ y1 + vy1′ ,
y ′′ = v ′′ y1 + 2v ′ y1′ + vy1′′ .
Substitute into the given equation and simplify using the fact that y1 sat-
isfies it. This leads to a first-order equation in v ′ , which can be solved by
standard methods.
Example 1: Solve:

x2 y ′′ − 3xy ′ + 4y = 0,

given that one solution is y1 = x2 .

Solution:
Let y = vy1 = vx2 .
Then:
y ′ = v ′ x2 + 2vx, y ′′ = v ′′ x2 + 4v ′ x + 2v.
Substitute into the equation:

x2 (v ′′ x2 + 4v ′ x + 2v) − 3x(v ′ x2 + 2vx) + 4(vx2 ) = 0.

Simplify:

x4 v ′′ + 4x3 v ′ + 2x2 v − 3x3 v ′ − 6x2 v + 4x2 v = 0.

x4 v ′′ + x3 v ′ = 0.
Divide by x3 :
xv ′′ + v ′ = 0.
Let v ′ = p, so v ′′ = p′ :
dp dx
xp′ + p = 0 ⇒ =− .
p x

14
 Integrate:
C1
ln p = − ln x + C ⇒ p = .
x
Then:
C1
v′ = ⇒ v = C1 ln x + C2 .
x
Hence:
y = x2 (C1 ln x + C2 ) = C1 x2 ln x + C2 x2 .

y = C1 x2 ln x + C2 x2 .

Non-Homogeneous Euler–Cauchy Equations:

1. General Form
A second-order non-homogeneous Euler–Cauchy equation is given by

x2 y ′′ + axy ′ + by = f (x),

where a, b are constants and f (x) is a known function.

General Form

x2 y ′′ + axy ′ + by = f (x)
Let
x = et ⇒ t = ln x, y(x) = Y (t)
Step 2: Change of derivatives
dy dY dt 1
= · = Y′
dx dt dx x
1
y′ = Y ′
x
 ′
d Y 1 dY ′ Y ′
y ′′ = = − 2
dx x x dx x
Since
dY ′ dY ′ dt Y ′′
= · = ,
dx dt dx x
we get
Y ′′ Y ′ 1
y ′′ = 2
− 2 = 2 (Y ′′ − Y ′ )
x x x
15
 Step 3: Substitute into the equation

x2 y ′′ + axy ′ + by = f (x)
   ′
2 1 ′′ ′ Y
x 2
(Y − Y ) + ax + bY = f (et )
x x
(Y ′′ − Y ′ ) + aY ′ + bY = f (et )
Y ′′ + (a − 1)Y ′ + bY = f (et )
Step 4: Constant-Coefficient Equation
The homogeneous part is

Y ′′ + (a − 1)Y ′ + bY = 0

with characteristic equation

m2 + (a − 1)m + b = 0

3. Non-Homogeneous Part: Substitution Method To find a

particular integral, use the substitution

x = et ⇒ t = ln x.

Then !
d 1 d d2 1 d2 d
= , 2
= 2 2
− .
dx x dt dx x dt dt
Substitute into the equation:

x2 y ′′ + axy ′ + by = f (x)

to obtain !
2
d y dy dy
− +a + by = f (et ),
dt2 dt dt
which simplifies to
y ′′ + (a − 1)y ′ + by = f (et ).
This is a linear differential equation with constant coefficients.
4. Operator Form Let Dt = dtd . Then

(Dt2 + (a − 1)Dt + b)y = f (et ).

Hence the particular integral is

1
yp = f (et ),
Dt2 + (a − 1)Dt + b

16
which can be solved using standard methods for constant-coefficient ODEs.
Finally, replace t = ln x to express yp in terms of x.
Example:
x2 y ′′ − 3xy ′ + 4y = x2
Substitute x = et , y = Y (t):

a = −3 ⇒ a − 1 = −4
Y ′′ − 4Y ′ + 4Y = e2t
Step 1: Homogeneous Solution

Y ′′ − 4Y ′ + 4Y = 0

r2 − 4r + 4 = 0 ⇒ (r − 2)2 = 0
Yh = (C1 + C2 t)e2t
Step 2: Particular Solution
RHS = e2t already appears in Yh , so try

Yp = At2 e2t

Compute derivatives:

Yp′ = Ae2t (2t + 2t2 )

Yp′′ = Ae2t (2 + 8t + 4t2 )

Substitute into the left-hand side:

Y ′′ − 4Y ′ + 4Y = Ae2t [(2 + 8t + 4t2 ) − 4(2t + 2t2 ) + 4t2 ]

Simplify:
1
Ae2t (2) = e2t ⇒ A=
2
1
Yp = t2 e2t
2
Step 3: General Solution
1
Y = (C1 + C2 t)e2t + t2 e2t
2
Step 4: Back-Substitute t = ln x

y = x2 C1 + C2 ln x + 12 (ln x)2
 

17
 y = (C1 + C2 ln x)x2 + 21 x2 (ln x)2
Example Solve
x2 y ′′ + 3xy ′ + y = x2 .
Homogeneous part:

m(m − 1) + 3m + 1 = 0 ⇒ m2 + 2m + 1 = 0 ⇒ (m + 1)2 = 0.

⇒ yh = (C1 + C2 ln x)x−1 .
Particular integral: Try yp = Axr . Substitute:

x2 (Ar(r − 1)xr−2 ) + 3x(Arxr−1 ) + Axr = x2 ,

which simplifies to

A[r(r − 1) + 3r + 1]xr = x2 .

Let r = 2:
1
A[2(1) + 6 + 1] = A(9) = 1 ⇒ A= .
9
Hence
1
y p = x2 .
9
Complete solution:

1
y = (C1 + C2 ln x)x−1 + x2 .
9

Solve the non-homogeneous Euler–Cauchy equation

x2 y ′′ + 2xy ′ + 2y = sin(ln x), x > 0.

Step 1: Solve the homogeneous equation

Consider the homogeneous equation

x2 y ′′ + 2xy ′ + 2y = 0.

Try y = xm . Then y ′ = mxm−1 , y ′′ = m(m − 1)xm−2 . Substituting yields

the indicial equation

m(m − 1) + 2m + 2 = 0 =⇒ m2 + m + 2 = 0.

18
Solve for m: √ √
−1 ± 1−8 −1 ± i 7
m= = .
2 2
√
Write m = α ± iβ with α = − 21 and β = 27 . Thus the homogeneous
solution is
 √ √

yh (x) = x−1/2 C1 cos 27 ln x + C2 sin 27 ln x .

Step 2: Transform to a constant-coefficient ODE Use the sub-

stitution x = et (so t = ln x). With respect to t,

d 1 d d2 1  d2 d
= , = − .
dx x dt dx2 x2 dt2 dt
Let Y (t) = y(et ). Substituting into the original equation gives
 d2 d d
2
− Y + 2 Y + 2Y = sin t.
dt dt dt
Simplify:

Y ′′ + (2 − 1)Y ′ + 2Y = sin t =⇒ Y ′′ + Y ′ + 2Y = sin t.

So we must solve the constant-coefficient ODE

Y ′′ + Y ′ + 2Y = sin t.

Step 3: Particular integral in the t–variable We find a particular

solution Yp (t) of
Dt2 + Dt + 2 Y = sin t,

d
where Dt = .
dt
Use the complex exponential method. Consider eit and evaluate the
operator polynomial at Dt = i:

P (i) = (i)2 + (i) + 2 = −1 + i + 2 = 1 + i ̸= 0.

eit
Thus a particular solution is the imaginary part of :
P (i)
!
eit
Yp (t) = Im .
1+i

1−i
Compute 1/(1 + i) = .
2
19
 Hence
eit 1−i
= (cos t + i sin t).
1+i 2
Multiply out:

(1 − i)(cos t + i sin t) = (cos t + sin t) + i (sin t − cos t) .

| {z } | {z }
real imag

Taking the imaginary part and dividing by 2 gives

1

Yp (t) = sin t − cos t .
2

Step 4: Back substitute t = ln x Recall Y (t) = y(x) with t = ln x.

Therefore
1

yp (x) = Yp (ln x) = sin(ln x) − cos(ln x) .
2

Step 5: General solution

Combine the homogeneous and particular parts:
 √ √

−1/2 7 7 1


y(x) = x C1 cos 2 ln x + C2 sin 2 ln x + 2
sin(ln x)


− cos(ln x) .

Euler-Cauchy Equation:
(ax + b)2 y ′′ + (ax + b)y ′ + y = f (ax + b) by putting et = ax + b.
Substitution: set

et = ax + b ⇐⇒ t = ln(ax + b), y(x) = Ye (t).

Compute derivatives using the chain rule. Since t = ln(ax + b) we have

dt a
= .
dx ax + b
Thus
dYe dt a
y′ = = Ye ′ (t) = Ye ′ ,
dx dx ax + b

20
and !
d

a

d e′
Y
y ′′ = Ye ′ =a
dx ax + b dx ax + b
1  dYe ′ e ′ 
=a (ax + b) −Y a .
(ax + b)2 dx
dYe ′ dYe ′ dt a
But = = Ye ′′ . Therefore
dx dt dx ax + b
1 
e ′′ a − aYe ′ =
 a2
y ′′ = a (ax + b) Y (Ye ′′ − Ye ′ ).
(ax + b)2 ax + b (ax + b)2

Now substitute into the ODE:

a2
(ax + b)2 y ′′ + (ax + b)y ′ + y = (ax + b)2 2
(Ye ′′ − Ye ′ )
(ax + b)
a
+ (ax + b)Ye ′ + Ye
ax + b
= a2 (Ye ′′ − Ye ′ ) + aYe ′ + Ye .

Hence the equation becomes

a2 Ye ′′ + (−a2 + a)Ye ′ + Ye = f (et ).

Divide through by a2 :
1  1 1
Ye ′′ + − 1 Ye ′ + 2 Ye = 2 f (et ) .
a a a

This is a linear ODE with constant coefficients in t. The homogeneous part

is 1  1
Ye ′′ + − 1 Ye ′ + 2 Ye = 0,
a a
with characteristic polynomial
1  1
r2 + − 1 r + 2 = 0.
a a
Let the roots be r1 , r2 . Then

 C er1 t + C2 er2 t , r1 ̸= r2 ,
 1


Yeh (t) = (C1 + C2 t)ert , r1 = r2 = r,



eαt C1 cos(βt) + C2 sin(βt) , r = α ± iβ.


21
 A particular solution Yep (t) can be found by variation of parameters in
the general case. If Ye1 , Ye2 are two independent solutions of the homogeneous
equation, variation of parameters gives

Yep = u1 (t)Ye1 (t) + u2 (t)Ye2 (t),

with
Ye2 (t) a12 f (et ) Ye1 (t) a12 f (et )



u′1 (t) =− , u′2 (t) = ,
W (t) W (t)
where W (t) = Ye1 Ye2′ − Ye2 Ye1′ .
Finally substitute back t = ln(ax + b). The general solution is




y(x) = Ye ln(ax + b) = Yeh ln(ax + b) + Yep ln(ax + b) .

Example
(2x + 1)2 y ′′ + (2x + 1)y ′ + y = 2x + 1
Put et = 2x + 1, y(x) = Ye (t). Then

Ye ′′ − 21 Ye ′ + 41 Ye = 14 et .
√
Characteristic: r2 − 12 r + 14 = 0 ⇒ r = 14 ± i 43 . So
 √
√

t/4
Yh = e
e C1 cos 4 t + C2 sin 43 t .
3

Try Yep = Aet . Substitution gives 34 Aet = 14 et ⇒ A = 31 . OR

F (D) = D2 − 12 D + 14 , G(t) = 14 et .
1 t
e
F (1) = 3
4
⇒ Yp = 43 = 13 et . ⇒ yp (x) = 13 (2x + 1).
e
4
Thus  √ √

Ye = et/4 C1 cos 3 3
+ 13 et .


4
t + C2 sin 4
t

Back-substitute t = ln(2x
h + 1): √ √
i
y(x) = (2x + 1)1/4 C1 cos 43 ln(2x + 1) + C2 sin 3


4
ln(2x + 1) +
1
3
(2x + 1).

22
Simultaneous Differential Equations
Simultaneous differential equations involve two or more dependent vari-
ables satisfying two or more differential equations at the same time. Typi-
cal methods include elimination, matrix (eigenvalue) method, and Laplace
transforms.
Elimination Method Consider the system
dx dy
= x + y, = x − y.
dt dt
Differentiate the first equation:

d2 x d dx dy
2
= (x + y) = + .
dt dt dt dt
Substitute the given equations:

d2 x
= (x + y) + (x − y) = 2x.
dt2
Thus we obtain the second–order ODE:
d2 x
− 2x = 0.
dt2
Solve it: √
m2 − 2 = 0 ⇒ m = ± 2.
Hence √ √
x(t) = C1 e 2t
+ C 2 e− 2t
.
dx
Now substitute into dt
= x + y to find y:

dx √ √ √ √
y= − x = C1 ( 2 − 1)e 2t − C2 ( 2 + 1)e− 2t .
dt
dx ′′ d2 x
This gives the general solution. Notation: x′ = , x = 2 , etc.
dt dt
Example 1 (Linear, homogeneous)
Solve
x′ = x + 2y, y ′ = 3x + 4y.
Step 1: Express y from the first equation:

x′ − x
y= .
2

23
 Step 2: Differentiate the first equation to eliminate y ′ :

x′′ = x′ + 2y ′ .

Substitute y ′ = 3x + 4y:

x′′ = x′ + 2(3x + 4y) = x′ + 6x + 8y.

x′ − x
Replace y by (from Step 1):
2
x′ − x
8y = 8 · = 4x′ − 4x,
2
so
x′′ = x′ + 6x + (4x′ − 4x) = 5x′ + 2x.
Step 3: Rearranged second-order ODE for x:

x′′ − 5x′ − 2x = 0.

Characteristic equation: r2 − 5r − 2 = 0, hence

√ √
5 ± 25 + 8 5 ± 33
r= = .
2 2
Thus the general solution for x is
√
5± 33
x(t) = C1 er1 t + C2 er2 t , r1,2 = .
2
Step 4: Recover y using y = (x′ − x)/2:

x′ (t) − x(t)
y(t) = ,
2
which gives y(t) explicitly once x(t) is substituted. Solve

x′ = 2x + y + et , y ′ = x + 3y + t.

Step 1: Solve the first equation for y:

y = x′ − 2x − et .

Step 2: Differentiate this expression for y:

y ′ = x′′ − 2x′ − et .

24
Step 3: Substitute into the second equation y ′ = x + 3y + t:
x′′ − 2x′ − et = x + 3(x′ − 2x − et ) + t.
Step 4: Expand and collect terms:
x′′ − 2x′ − et = x + 3x′ − 6x − 3et + t,
so
x′′ − 2x′ − et − 3x′ + 6x + 3et − t − x = 0.
Combine like terms:
x′′ − 5x′ + 5x + 2et − t = 0,
or equivalently
x′′ − 5x′ + 5x = t − 2et .
This is a linear nonhomogeneous second-order ODE for x(t). Solve by:
• finding complementary solution xc from r2 − 5r + 5 = 0,
• then a particular solution xp for the right-hand side t − 2et (use un-
determined coefficients or variation of parameters).
Sketch of solution: Characteristic roots for homogeneous part
√ √
5 ± 25 − 20 5± 5
r= = ,
2 2
so
xc (t) = Aer1 t + Ber2 t .
For xp , try xp (t) = at + b + cet (polynomial degree 1 for t, plus an et
term). Substitute into the ODE and solve for a, b, c. Once x(t) = xc + xp
is found, compute y(t) = x′ − 2x − et .
Superposition Principle and Linear Combination A differential
equation is said to be linear if it can be written in the form
L[y] = f (x),
where L is a linear differential operator,
L[y] = an (x)y (n) + an−1 (x)y (n−1) + · · · + a1 (x)y ′ + a0 (x)y.
If f (x) = 0, the equation is called homogeneous; if f (x) ̸= 0, the equation
is nonhomogeneous. Linear Combination
Given functions y1 (x), y2 (x), . . . , yn (x), any function of the form
C1 y1 (x) + C2 y2 (x) + · · · + Cn yn (x),
where C1 , . . . , Cn are constants, is called a linear combination of the
functions.

25
Superposition Principle (Homogeneous Case)
Theorem. Let L be a linear differential operator. If y1 (x) and y2 (x) are
solutions of the homogeneous equation

L[y] = 0,

then any linear combination

y(t) = C1 y1 (x) + C2 y2 (x)

is also a solution.
Proof. Since y1 and y2 satisfy L[y1 ] = 0 and L[y2 ] = 0, and L is linear, we
have

L[C1 y1 + C2 y2 ] = C1 L[y1 ] + C2 L[y2 ] = C1 (0) + C2 (0) = 0.

Thus y is a solution. □

Superposition for Nonhomogeneous Equations

For the nonhomogeneous equation

L[y] = f (x),

every solution can be written as

y(t) = yh (x) + yp (x),

where yh is the general solution of the homogeneous equation L[y] =

0, and yp is any particular solution of L[y] = f (x).

Superposition in Systems of Differential Equations

For the linear system
x′ = Ax,
if x1 (t) and x2 (t) are solutions, then any linear combination

x(t) = C1 x1 (t) + C2 x2 (t)

d
is also a solution. This follows from the linearity of the operator − A.
dt
Example 1: Second–order Linear Homogeneous DE

26
 Solve the differential equation

y ′′ − 3y ′ + 2y = 0.

The characteristic equation is

r2 − 3r + 2 = 0 ⇒ r = 1, 2.

So two independent solutions are

y1 = et , y2 = e2t .

By the superposition principle, the general solution is

y(t) = C1 et + C2 e2t .

Summary

• Linear differential equations admit linear combinations of solutions.

• The general solution of a homogeneous linear DE is a linear combi-

nation of linearly independent solutions.

• For nonhomogeneous equations, every solution is the sum of the ho-

mogeneous solution and a particular solution.

• Linear systems follow the same structure due to operator linearity.

27