Second-Order Linear Ordinary

Differential Equations

Faculty : Dr Pui Tze Sian

School : School of Chemistry, Chemical
Engineering and Biotechnology
(CCEB)
Email : tspui@[Link]
Office : N1.3-B2-12
Outline

1. Basic Theory of 2nd Linear ODE

2. Reduction of Order

3. Homogeneous Linear DE with Constant Coefficients

4. Nonhomogeneous DE: Method of Undetermined Coefficients

Topic 1: Introduction 2
Definition: Second-Order Linear ODE

• Second-order linear ODE in the standard form:

y ''+ p ( x ) y '+ q ( x ) y = f ( x ) (1)

where p, q, and f are given continuous functions for x  I

• If f ( x )  0 , the linear ODE (1) is nonhomogeneous.
• If f ( x ) = 0 for all x  I

y ''+ p ( x ) y '+ q ( x ) y = 0 (2)

The linear ODE is homogeneous. We also say (2) is the associated

homogeneous ODE of (1)

Topic 1: Introduction 3
Initial value problem (IVP)

• As the general solution of a second-order ODE contains TWO arbitrary

constants, we need two conditions to determine a solution, i.e., the
particular solution.

• The initial value problem (IVP):

y ''+ p ( x ) y '+ q ( x ) y = f ( x ) , y ( xo ) = yo , y ' ( xo ) = y1 (3)

where xo, yo, and y1 are given constants.

Topic 1: Introduction 4
Basic Theory of Linear ODEs

Theorem
(Existence and Uniqueness) Let p(x), q(x) and f(x) be continuous
functions on an internal I. Then for each xo  I , the IVP

y ''+ p ( x ) y '+ q ( x ) y = f ( x ) , y ( xo ) = yo , y ' ( xo ) = y1 (4)

has a unique solution on I.

The proof of this theorem is fairly difficult. One may refer to, for example, E.A.
Coddington, An introduction to ODEs, Dover, 1961, for the proof.

Topic 1: Introduction 5
Superposition principle or linearity principle

Theorem
If y1(x) and y2(x) are two solutions to the homogeneous DE:

y ''+ p ( x ) y '+ q ( x ) y = 0, x I,
Then any linear combination of y1 and y2.
y ( x ) = c1 y1 ( x ) + c2 y2 ( x )

Is also a solution to the same equation, where c1 and c2 are arbitrary

constants.

Topic 1: Introduction 6
Superposition principle or linearity principle

Example
Verify that y ( x ) = c1 cos x + c2 sin x with c1 and c2 being arbitrary
constants is a solution to y ''+ y = 0.

Topic 1: Introduction 7
Linear dependence/Independence

Definition
Two functions y1 ( x ) , y2 ( x ) defined on an interval I are said to be linearly
independent (LI), if one of them is NOT a constant multiple of the other,
i.e., the ratio is not a constant, but a function of x:

y1 ( x ) y2 ( x )
 constant or  constant.
y2 ( x ) y1 ( x )

If the ratio is a constant, then y1 ( x ) , y2 ( x ) are said to be linearly

dependent (LD)

Topic 1: Introduction 8
Example

Are the following functions linearly dependent or linearly independent?

1 sin x and cos x Linearly independent

2 x and 3e x Linearly independent

3 e x and e − x Linearly independent

4 sin x and − sin x Linearly dependent

Topic 1: Introduction 9
Wronskian

Definition
Let y1 and y2 be differentiable functions on the interval I. The Wronskian
of y1 and y2, is defined by
y1 y2
W ( y1 , y2 )( x ) = ' '
= y y
1 2
'
− y '
1 y2 (5)
y1 y2

Note: W ( y1 , y2 )( x ) = −W ( y2 , y1 )( x )

Wronskian is a very important tool for the theory of linear DE.

Topic 1: Introduction 10
Example

Compute the Wronskian of

1 sin x , cos x
2 x 3, − 2 x 3
Solution: Firstly, we have
sin x cos x sin x cos x
W ( sin x, cos x )( x ) = =
( sin x ) ' ( cos x ) ' cos x − sin x
= − sin 2 x − cos 2 x = −1
Similarly, we can calculate the Wronskian of the second pair:
x3 −2 x3 3
−2 x3
W ( x3 , −2 x 3 ) ( x ) = 3
x
= 2
( x ) ' ( −2 x ) ' 3x
3
−6 x 2
= −6 x5 + 6 x5 = 0
Topic 1: Introduction 11
Relationship between Wronskian and LD/LI

Theorem
Let y1 and y2 be the solution to the homogeneous equation:

y ''+ p ( x ) y '+ q ( x ) y = 0, x I, (6)

Then we have
W ( y1 , y2 )( x ) = 0  LD

W ( y1 , y2 )( x )  0  LI

Topic 1: Introduction 12
Example

Consider the second-order linear homogeneous ODE with constant

coefficients:
y ''− 3 y '+ 2 y = 0,
We can verify that y1 = e and y2 = e are two solutions. Are they linearly
x 2x

dependent or linearly independent?

Solution: We can calculate

W ( y1 , y2 ) = ... = e3 x  0

So these two functions as linearly independent.

Topic 1: Introduction 13
Theorem

Consider the general second-order linear homogeneous DE

y ''+ p ( x ) y '+ q ( x ) y = 0, (7)
where p, q are given continuous functions on an interval I. if
i. y1 and y2 are two solutions, and
ii. y1 and y2 are linearly independent (i.e., Wronskian is nonzero or the
ratio of y1, y2 is not a constant.
Then the general solution is
y ( x ) = c1 y1 ( x ) + c2 y2 ( x ) , (8)
Where c1 and c2 are two arbitrary constants. We call that y1, y2 are
fundamental solutions of (7).

Topic 1: Introduction 14
Example

Determine all the values of r such that xr is a solution to

x 2 y ''+ 3xy '− 8 y = 0, x0
Then find the general solution of this DE.

Solution: Insert y = xr into the given equation:

r ( r − 1) x r + 3rx r − 8 x r = 0  r 2 + 2r − 8 = 0

Therefore, the roots are r = -4, 2, which means y1 = x-4 and y2 = x2 are
two solutions of the equation. Moreover, they are linearly independent.
Thus the general solution is
y ( x ) = c1 x −4 + c2 x 2
Topic 1: Introduction 15
Exercise

Determine all the values of r such that erx is a solution to

y ''− 4 y '+ 3 y = 0
Then find the general solution of this DE.

Key: The GS is

y ( x ) = c1e x + c2e3 x

Topic 1: Introduction 16
Outline

1. Basic Theory of Linear ODE

2. Reduction of Order

3. Homogeneous Linear DE with Constant Coefficients

4. Nonhomogeneous DE: Method of Undetermined Coefficients

Topic 1: Introduction 17
Reduction of order

Given a nonzero solution y1(x) of the equation

y '' ( x ) + p ( x ) y ' ( x ) + q ( x ) y ( x ) = 0, (9)

The method of reduction of order can be used to find another linearly

independent solution y2(x). Then we can find its GS:

y ( x ) = c1 y1 ( x ) + c2 y2 ( x )

Topic 1: Introduction 18
Example

Find the general solution to

xy ''− 2 y '+ ( 2 − x ) y = 0, x0
Given that one solution is y1 ( x ) = e
x

Solution: we look for a second solution in the form

y2 ( x ) = u ( x ) y1 ( x ) = u ( x ) e x (Note: u  constant)
Insert it into the given DE and collecting the terms yield
x ( u ''+ 2u '+ u ) − 2 ( u '+ u ) + ( 2 − x ) u = 0
Which can be simplified to xu ''+ 2 ( x − 1) u ' = 0

Topic 1: Introduction 19
Example (Cont.’)

Separating the variables yields

= 2 ( x −1 − 1)
u ''
u'
Integrating both sides leads to
ln u ' = 2 ( ln x − x ) + c,

Which can be written as u ' = c1 x 2e −2 x . Integrating again gives

u ( x ) = − c1e (1 + 2 x + 2 x 2 ) + c2
1 −2 x
4
As we need one nonzero u(x), we take c1 = −4, c2 = 0. Then
y2 ( x ) = u ( x ) e x = e − x (1 + 2 x + 2 x 2 )
Consequently, the general solution to the given DE is
y ( x ) = C1e x + C2e − x (1 + 2 x + 2 x 2 )
Topic 1: Introduction 20
A general formula for Reduction of Order

Theorem
Let y1 ( x )  0 be a solution to the second-order linear DE in the standard
form
y ''+ p ( x ) y '+ q ( x ) y = 0, (11)
Then a second linearly independent solution is
1 −  p ( x ) dx
y2 ( x ) = y1 ( x )  2 e dx (12)
y1 ( x )
Then the general solution is
y ( x ) = c1 y1 ( x ) + c2 y2 ( x )

Remarks: it only applied to (i) the standard form (i.e., the coefficient of y’’
is 1; and (ii) homogeneous DE.

Topic 1: Introduction 21
Exercise

Use the formula (12) in the Theorem to find y2 of the equation (10) in the
previous example.

Key: y2 ( x ) = e − x (1 + 2 x + 2 x 2 )

Given 1 ( ) = 2
y x x , find a second LI solution to
x 2 y ''− 3xy '+ 4 y = 0, x0

Key: y2 ( x ) = x 2 ln x

Topic 1: Introduction 22
Outline

1. Basic Theory of Linear ODE

2. Reduction of Order

3. Homogeneous Linear DE with Constant Coefficients

4. Nonhomogeneous DE: Method of Undetermined Coefficients

Topic 1: Introduction 23
Homogeneous DE with Constant Coefficients

We now discuss how to solve second-order linear DE, and start with a simple
but important case in the form

ay ''+ by '+ cy = 0 (14)

where a, b, c are given real constants.

Based on the theory, to find its general solution, it is suffices to find two
linearly independent solutions, say y1 ( x ) & y2 ( x )
Then the GS is
y ( x ) = c1 y1 ( x ) + c2 y2 ( x )

Topic 1: Introduction 24
Theorem

The function y ( x ) = e is a solution to

rx

ay ''+ by '+ cy = 0, (15)

If and only if r is a root of the Characteristic Equation (CE):

ar 2 + br + c = 0
Correspondence:
ay ''+ by '+ cy = 0,  ar 2 + br + c = 0

Topic 1: Introduction 25
Example

Solve the following equations:

(1) y ''+ 2 y '− 15 y = 0; ( 2) 9 y ''− 12 y '+ 4 y = 0,
( 3) y ''+ 4 y '+ 13 y = 0

Solution: (1) The CE is

r 2 + 2r − 15 = ( r + 5 )( r − 3) = 0,
Which has two distinct roots: r1 = -5 and r2 = 3. By the theorem, we know
that e −5x and e3x are two solutions, and they are linearly independent
(as the ratio of two function is not a constant or the Wronskin is
nonzero). Thus, the GS is
y ( x ) = c1e3 x + c2 −5 x

Topic 1: Introduction 26
 (2) The CE is
9r − 12r + 4 = ( 3r − 2 ) = 0,
2 2

2
2 x
Which has only one root: r = . By the linear theory, y1 = e 3 is a solution. We
3
lack another LI solution. Using the Reduction of order, we can find the other
one is 2
x
y2 = xy1 = xe 3
(16)

According to the theory, the general solution is

2
y ( x ) = ( c1 + c2 x ) e
x
3

Topic 1: Introduction 27
 (3) The CE is
r + 4r + 13 = ( r + 2 ) + 9 = 0
2 2

 ( r + 2 ) = −9
2

 r + 2 =  −9 = 3 −1 = 3i,

Where the complex unit is defined as i = −1

Thus the CE has the two conjugated complex roots: −2  3i. By the linear
theory, we know that
y1 ( x ) = e( −2+3i ) x , y2 ( x ) = e( −2−3i ) x (17)

Are two solutions (which are linearly independent). Note that y1 is the
conjugate of y2 and vice versa.
Topic 1: Introduction 28
 However, we would like to express the solution in terms of real functions for
the reason that the given DE has real coefficients. Recall the Euler’s formula
for complex exponential:
e a +bi = e a  ebi = e a  ( cos b + i sin b ) , (18)
For any real numbers a,b. Apply it to y1 :
y1 ( x ) = e( −2+3i ) x = e −2 x e3 xi = e −2 x ( cos ( 3x ) + i sin ( 3x ) )
= e −2 x cos ( 3 x ) + e −2 x sin ( 3 x ) i
Then we take the real and imaginary parts to form the GS:
y ( x ) = ( c1 cos 3x + c2 sin 3 x ) e −2 x
Note: We can apply (18) to y2(x), which leads to the same GS.
Topic 1: Introduction 29
Summary of the Rules

Consider the second-order ODE:

ay ''+ by '+ cy = 0, a  0. (19)

Its characteristic Equation (CE) is

ar 2 + br + c = 0 (20)

(i) If the CE (20) has two real, distinct roots: r1  r2 (i.e., when b 2 − 4ac  0 ),
then the GS of (19) is

y ( x ) = c1e r1x + c2e r2 x (21)

Topic 1: Introduction 30
Summary of the Rules

b
(ii) If the CE (20) has only on root: r1 (i.e., when b 2
− 4ac = 0 , so r1 = − ),
2a
then the GS of (19) is

y ( x ) = ( c1 + c2 x ) e r1x (22)

(iii) If the CE (20) has two complex conjugated roots:    i with real  , 
(i.e., when b 2 − 4ac  0 ), then the GS of (19) is

y ( x ) = ( c1 cos  x + c2 sin  x ) e x (23)

Note: here, we used the Euler’s formula (18) to extract the real and imaginary
parts to form the GS as with the previous Example.

Topic 1: Introduction 31
Reduction of Order for Rule (ii)

We next justify why the GS in Case (ii) takes the form (22). By Theorem on
Slide 29, (19) has one solution y1 ( x ) = e
r1 x

Following reduction of order, we look for a second LI solution:

y2 ( x ) = u ( x ) e r1x, and substitute it into the equation:
a (u ( x ) e ) ''+ b (u ( x ) e ) '+ c (u ( x ) e ) = 0
r1 x r1 x r1 x

a ( u ''+ 2r u '+ r u ) e + b ( u '+ r u ) e + cue = 0

1 1
2 r1 x
1
r1 x r1 x

au ''+ ( 2ar1 + b ) u ' = 0  au '' = 0  u '' = 0

Then u ( x ) = C1 x + C2. As we only need one u(x), we take it to be u(x) = x. Here,
b
we used the fact: r1 = − .Thus, we have y2 = xe r x 1

2a

Topic 1: Introduction 32
Application example: Simple Harmonic Motion

Applying Hooke’s law, the

restoring force is
𝐹𝑠 = −𝑘(𝑠 + 𝑥)

In motion, by Newton’s law,

𝑚𝑥 ′′ = −𝑘 𝑠 + 𝑥 + 𝑚𝑔
𝑚𝑥 ′′ = −𝑘𝑠 − 𝑘𝑥 + 𝑚𝑔
At equilibrium, 𝑚𝑔 = 𝑘𝑥 , then
DE becomes

𝑚𝑥 ′′ + 𝑘𝑥 = 0

Topic 1: Introduction 33
Application example: Simple Harmonic Motion

Rearrange the equation and introduce new variable, called the angular frequency, 𝜔

𝑚𝑥 ′′ + 𝑘𝑥 = 0

𝑥 ′′ + 𝜔2 𝑥 = 0 where 𝜔 = 𝑘/𝑚

The general solution is

𝑥 𝑡 = 𝐶1 𝑐𝑜𝑠𝜔𝑡 + 𝐶2 𝑠𝑖𝑛𝜔𝑡

𝑥 𝑡 is the position of the mass at any point in time.

2𝜋
The period of motion is 𝑇 =
𝜔
1 𝜔
The frequency is 𝑓 = =
𝑇 2𝜋

Topic 1: Introduction 34
Application example 1: Simple Harmonic Motion

A 200 gram mass stretches a spring 5 cm. Find the equation of motion of the mass if it is
released from rest from a position 10 cm below the equilibrium position. What is the frequency
of the motion?
Solution

Topic 1: Introduction 35
Modeling a motorcycle suspension system
(damped system)

The suspension system is modeled as a damped spring-mass system

The ODE is
𝑚𝑥" + 𝑏𝑥′ + 𝑘𝑥 = 0

𝑚 = mass

𝑏 = Shock absorber
constant

𝑘 = Spring constant

Topic 1: Introduction 36
Application example 2: Damped system

Suppose that the spring in the previous example is immersed in a fluid with damping constant
𝑏 = 20. Find the position of the mass at any time 𝑡 if it starts from the equilibrium position and is
given a push to start it with an initial velocity of 0.6 m/s.
Solution

Topic 1: Introduction 37
Outline

1. Basic Theory of Linear ODE

2. Reduction of Order

3. Homogeneous Linear DE with Constant Coefficients

4 Nonhomogeneous DE: Method of Undetermined Coefficients

Topic 1: Introduction 38
Theorem

Consider the nonhomogeneous DE:

y ''+ p ( x ) y '+ q ( x ) y = f ( x ) , (25)
Where p, q, f are given continuous functions on an interval I.
Then its GS can be written in the form
y ( x ) = y p ( x ) + c1 y1 ( x ) + c2 y2 ( x ) (26)

yc ( x )

where yp(x) is a particular solution of (25), and yc(x) is the GS to the

associated homogeneous DE: y ''+ p ( x ) y '+ q ( x ) y = 0, or called the
complementary function for equation (25).

Topic 1: Introduction 39
 ▪ Consider the nonhomogeneous DE with constant coefficients:

ay ''+ by '+ cy = f ( x )  0, (27)

where a, b, c are given constants and f(x) is known.
▪ Base on the above theorem, its GS consists of two parts:
y ( x ) = y p ( x ) + yc ( x )

▪ We know how to find yc(x), i.e., GS of ay ''+ by '+ cy = 0.

▪ It suffices to find one solution yp(x) of (27).

Topic 1: Introduction 40
Method of undetermined coefficients

It can be applied, if f(x) is of one of the following types:

• Type I: f(x) = polynomial, e.g., f(x) = 1, 1 – x2, …
x −x
• Type II: f(x) = exponential: ao e , e.g., −e ,3e ,...
2x

• Type III: f(x) = trigonometric: ao cos ( x ) , bo sin (  x ) , e.g.,

f ( x ) = sin ( 3 x ) ,3cos ( x / 2 ) ,sin ( 5 x ) − cos x
• Type (IV): f(x) product of Type I, II or III, e.g.,
f ( x ) = 2 xe −10 x , x 2 sin x, xe x cos ( 2 x ) ,...
• Type V: f(x) = sum of Type I, II, III or IV , e.g.,
f ( x ) = 1 + e − x , 2e x + x cos x,...

Topic 1: Introduction 41
Example

Determine the general solution to each of the equations:

(i ) y ''− y = 16e3 x ; ( ii ) y ''− y = 16e − x

Solution: (i) we first obtain the complementary function by solving the

homogeneous DE. The CE is r 2 − 1 = 0  r = 1, so that
yc ( x ) = c1e x + c2e − x . (28)
Given f(x) = 16e3x, we search for yp(x), which after performing the
operation: y p ''− y p , it equals to f(x). We guess yp(x) should be a constant
multiple of the same exponential function:
y p ( x ) = Ae3x (29)
where the constant A is an undetermined coefficient.
Topic 1: Introduction 42
 We call (29) a trial solution. We now substitute it into the given DE to
determine the constant A. Differentiating yp yields

y ' p = 3 Ae3 x , y '' p = 9 Ae3 x

As yp is a solution of Equation (i), we have

y '' p − y p = 8 Ae = 16e
3x 3x
 A=2  y p = 2e 3x

Therefore, the GS of Equation (i) is

y ( x ) = y p ( x ) + yc ( x ) = 2e3 x + c1e x + c2e − x

Topic 1: Introduction 43
 −x
We now solve (ii) y ''− y = 16e . Similar to (i), we take the trial solution:
y p ( x ) = Be − x , and substitute it into the equation:

LHS = y '' p − y p = ( Be − x ) ''− ( Be − x ) = Be − x − Be − x = 0

But RHS = 16e − x . In other words, the trial is wrong.

y
It fails to work, because p ( x ) = Be −x
, happens to be a solution of
homogeneous DE: y ''− y = 0, obtain from (28) by taking
c1 = 0, c2 = B.

Topic 1: Introduction 44
 In this case, we have to modify the above attempt: y p ( x ) = Be .
−x

The rule is multiply it by x, we try y p ( x ) = Bxe , and find

−x

y ' p ( x ) = B (1 − x ) e − x , y '' p = ( B (1 − x ) e − x ) ' = B ( x − 2 ) e − x .

Substituting them into the equation (ii),

y '' p − y p = B ( x − 2 ) e − x − ( Bxe − x ) = −2 Be − x = 16e − x

 B = −8  y p ( x ) = −8 xe − x
The GS of (ii) is
y ( x ) = y p ( x ) + yc ( x ) = −8 xe − x + c1e x + c2e − x

Topic 1: Introduction 45
Trial solution yp(x) for different types of f(x)

Type I: f(x) = a polynomial of degree n, i.e.,

f ( x ) = an x n + an −1 x n −1 + ... + a1 x + ao , an  0.

where the coefficients are given (might be zero). Usual trial

y p ( x ) = An x n + An −1 x n −1 + ... + A1 x + A0 ,

e.g., f ( x ) = 2 x 2 − 1. Then the usual trial solution:

y p ( x ) = A2 x 2 + A1 x + Ao

It might fail, if there is any term being a solution of the homogeneous DE.
Modify it by multiplying usual yp by x!

Topic 1: Introduction 46
Trial solution yp(x) for different types of f(x)

Type II: f ( x ) =  e  x , where  ,  are known real constants

Usual trial solution

y p ( x ) = Ae  x

e.g., f ( x ) = 5e − x . Then the usual trial solution:

y p ( x ) = Ae − x

We have to modify it by multiplying x, if it is a solution to the

homogeneous DE.

Topic 1: Introduction 47
Trial solution yp(x) for different types of f(x)

Type III: f(x) = trigonometric polynomial, i.e.,

f ( x ) = C cos ( x ) + D sin ( x )

where  is known and C, D are known constants (can be zero).

Usual trial solution
y p ( x ) = A cos ( x ) + B sin ( x )

e.g., f ( x ) = 5cos ( 3x ) Then the usual trial solution:

y p ( x ) = A cos ( 3x ) + B sin ( 3x )

We have to modify it by multiplying x, if it contains a term which is a

solution to the homogeneous DE.

Topic 1: Introduction 48
Trial solution yp(x) for different types of f(x)

Type IV: f(x) = product of previous types. For example,

f ( x ) = cos ( 3 x ) e 2 x .Then, the usual trial solution:

y p ( x ) = ( A cos ( 3x ) + B sin ( 3x ) ) e 2 x

For example, if f ( x ) = x cos ( 3x ), then

y p ( x ) = ( A1 x + B1 ) cos ( 3x ) + ( A2 x + B2 ) sin ( 3x )
We have to modify it by multiplying x, if it contains a term which is a
solution to the homogeneous DE.

Type V: f(x) = sum of previous types. We treat it separately (see the last
example).
Topic 1: Introduction 49
Trial Functions

Right hand-side function Trial function Unknown parameters

Polynomial in t of degree p, Polynomial in t of degree p, Coefficients of the polynomial,

for example for example for example

6t + 4t − 2t + 5
3 2 Pt 3 + Qt 2 + Rt + S P, Q, R, and S

Exponential function of t, for Exponential function of t, for Coefficients of the

example example exponential, for example
e −3t Pe −3t P

Sine or cosine of a multiple of Linear combination of sine Coefficients of sine and

t, for example and cosine of the same cosine terms, for example
multiple of t, for example
P and Q
sin 5t P sin 5t + Q cos 5t
Topic 1: Introduction 50
Example

In each case, write an appropriate trial solution. Do not solve for the
coefficients in your propose trial solution:
( i ) y ''+ 5 y ' = 4 x 2
( ii ) y ''− y = 3xe x
( )
iii y ''− 4 y '+ 5 y = 3e 2x
sin x

Solution: (i) The CE: r 2 + 5r = 0 has the two roots r = 0, −5, so the
complementary function
yc ( x ) = c1 + c2e −5 x . (30)
As f ( x ) = 4 x , we propose the usual trial solution
2

y p ( x ) = Ao + A1 x + A2 x 2
Topic 1: Introduction 51
 Example (Cont.’)

Observe that the first term Ao is a solution of the homogeneous DE as it

can be obtained from (30) by taking c1 = Ao , c2 = 0.

Thus we have to modify it as

y p ( x ) = x ( Ao + A1 x + A2 x 2 ) = Ao x + A1 x 2 + A2 x 3

No any term solves the homogeneous DE any more! It gives the trial
solution of (1).

Topic 1: Introduction 52
 Example (Cont.’)

(ii) The CE: r 2 − 1 = 0 has the roots r = 1, so the GS to the homogeneous
equation is
yc ( x ) = c1e x + c2e x

The usual trial solution is

y p ( x ) = ( A + Bx ) e x = Ae x + Bxe x

However, the first term solves the associated homogeneous DE, so we

use the modified form
y p ( x ) = x ( A + Bx ) e x = Axe x + Bx 2e x

Now no term is a solution to the associated homogeneous DE.

Topic 1: Introduction 53
 Example (Cont.’)

We now consider (iii): y ''− 4 y '+ 5 y = 3e 2 x sin x

The CE is r 2 − 4r + 5 = 0, with roots r = 2  i. Hence the complementary

functions is
yc ( x ) = e 2 x ( c1 cos x + c2 sin x )

The usual trial solution corresponding to f ( x ) = 3e 2 x sin x is

y p ( x ) = e 2 x ( A cos x + B sin x )
However, this form solves the associated homogeneous DE, so we must
modify it as
y p ( x ) = xe
2x
( A cos x + B sin x )
No term solves the homogeneous DE.

Topic 1: Introduction 54
 Type V: Sum of different types

Consider ay ''+ by '+ cy = f ( x )  0, (31)

where f = f1 + f 2, and f1 , f 2 are of two different types of I-IV.

We consider them separately and find the corresponding trial solutions

y p1 ( x ) , y p 2 ( x ) , respectively. That is

ay ''p1 + by 'p1 + cy p1 = f1 ( x ) , ay ''p 2 + by 'p 2 + cy p 2 = f 2 ( x )

Then the trial solution of (31) is

y p ( x ) = y p1 ( x ) + y p 2 ( x )

Topic 1: Introduction 55
Example

Determine an appropriate trial solution of

y ''+ y = 4e3 x + 5 x 2

Solution:
y p ( x ) = y p1 ( x ) + y p 2 ( x ) = Ao e3 x + ( B0 + B1 x + B2 x 2 )

End of Chapter Two: Part 1

Topic 1: Introduction 56
 End of Chapter Two: Part 1
Second order linear DE

Topic 1: Introduction 57