Chapter 6

Laplace Transform
Introduction
Laplace Transform is named after French mathematician Pierre-
Simon Laplace who applied the concept to his work on probability
theory. Though the work of Laplace led to what we now know as z-
transform, the current widespread use of Laplace Transform
developed during and after the Second World War. Presently,
Laplace Transform is an invaluable tool for solving initial-value
problems, as well as systems of ordinary differential equations. In
engineering, it provides a powerful tool for understanding the
behavior of systems to impulsive and discontinuous inputs.

Definition
If f(t) is a function defined for all t ≥ 0 its Laplace Transform is
defined as

(1)

We will say that the transform converges if the limit of integration

exists, otherwise it does not.

We also define an inverse Laplace Transform from the definition

of the Laplace Transform as
(2)

Example 1
Laplace transform of f(t) = 1.

Using the definition (1)

Since , we will say

Important note: As we have just computed that the Laplace

Transform of 1 is ; therefore, we will say that the inverse Laplace
Transform of is 1. The important point to note here is that the
function ‘1’ went through an integration operation before its
Laplace Transform could be computed; the inverse Laplace
Transform does not have any ‘definition’ and does not involve any
operation. This is exactly like differentiation and integration.
Because

which is computed from a definition of differentiation, we can

write

Differentiation has a definition; integration does not.

The other important idea here is to discuss ‘What is s? ‘s’ is an

abstract domain that has no physical significance. The way Laplace
transform will be put to use is that we will start with a function f(t)
in time-domain; we will take the Laplace Transform and move to
the s-domain. We will perform some algebraic operations (albeit
mathematical, not physical) in s-domain; finally perform the
inverse-transform, and come back to the time-domain. Therefore,
s-domain is just an intermediate domain.

Not for everyone: Do not confuse Laplace transform with Fourier

transform. In Fourier transform also, we start with a function f(t)
in time-domain, take the Fourier transform and go to
 (frequency)-domain. The -domain has a physical significance
where we perform operations like frequency filters, media
transmission and others. We may come back to the time-domain, or
end our exercise in -domain. In Laplace transform s-domain is
not like -domain in Fourier transform.

Example 2
Laplace transform of f(t) = eat, a is a constant, t ≥ 0.

( )

Properties of Laplace Transform – Linearity

Laplace transform is a linear operation. For two function f(t) and
g(t) whose transforms exist, and two constants a and b

The proof is direct from the definition of Laplace Transform.

Example 3
Laplace Transform of hyperbolic functions sinh(at) and cosh(at).

We are aware of the definition of hyperbolic functions

Therefore, using the results Example 2 and using the linearity

property of Laplace Transform,

Example 4
Laplace Transform of circular functions sin(at) and cos(at).

Using Euler’s formula

From Example 2, and using linearity property

Comparing the real part and imaginary parts

Example 5
Laplace transform of f(t) = tn.

Performing subsequent integrations continuously

Properties of Laplace Transform – s-domain shifting

Let us suppose that we have a function f(t) who Laplace transform
F(s) also exists. We want to ascertain what must happen to f(t) in t-
domain if we shift F(s) in s-domain by a to give us F(s – a).
We have

Therefore

( )
Therefore

where

From here we can write

Some examples
We have

Therefore

We have

Therefore

We have

Therefore
Inverse Laplace transform
An important exercise in application of Laplace transform to
engineering systems would be to compute the inverse Laplace
transform. The problem here would be that there would be an F(s);
we will have to manipulate the given F(s) into combinations of s-
functions whose inverse transform we know; then write their
corresponding f(t). Let us see this through several examples. In this
exercise, partial fractions would occasionally play an important
role.

Examples of Inverse Laplace transform

We have used partial fractions in the last problem.

Laplace transform of derivatives

Step-by-step we are moving towards finding the solution of initial-
value problems with discontinuous forcing function. To that end,
we evaluate the Laplace transform of derivatives of f(t). By
definition,
Using integration by parts

Therefore

Using this, we can further write

or
We can continue like this to obtain
( )

( )

Initial value problems

One of the biggest applications of Laplace transform in
engineering is to solve initial-value problems in order to
understand the behavior of systems to particular inputs. Coupled
with convolution, step function and delta function, Laplace
transform gives us an invaluable tool to understand the dynamic
behavior of systems. We begin with a simple examples, and we
will gradually move to a more complicated system.

Example 6
Solve
yʺ – y = t y(0) = 1 yʹ(0) = 1
Solution
The crux of the solution is to apply the expressions for Laplace
transform of derivatives those we obtained in this section. The
solution procedure will involve taking the Laplace transform of the
differential equation. This will convert the differential equation
into an algebraic equation of Y(s), where Y(s) is the Laplace
transform of y(t). Then we will obtain the inverse transform of Y(s)
that will hopefully lead us to the solution y(t).
Step 1. Take the Laplace transform of the given differential
equation

Step 2. Substitute the initial values and obtain an expression for

Y(s). In our problem y(0) = 1, y′(0) = 1. Therefore

This expression of Y(s) is referred to as the transfer function.

Step 3. Perform the inverse Laplace transform of the transfer

function to obtain solution of y(t). It is very common to apply
partial fractions at this step.

Taking the inverse Laplace transform

 y(t) = et + sinh t – t

Example 7
Find solution of
y − 3y′ − 10y = 2 y(0) = 1 y′(0) = 2

Step 1. Take Laplace transform

Therefore

Step 2. Substitute the initial conditions, obtain the transfer

function.

Step 3.
Take the inverse Laplace transform.

Example 8
Find solution of
y − y′ − 2y = e2t y(0) = 0 y′(0) = 1

Therefore

Using partial fractions

Taking the inverse Laplace transform.

Example 9
Find solution of
y − 2y′ + 2y = e−t y(0) = 0 y′(0) = 1

Therefore

Using partial fractions

Taking the inverse Laplace transform

Example 10
Find solution of
y + y = cos 2t y(0) = 2 y′(0) = 1
Therefore

Substituting initial conditions, finding the transfer function

Using partial fractions

Taking the inverse Laplace transform

In examples 6 through 10 we have obtained solutions of

nonhomogeneous second order differential equation with constant
coefficients. We have obtained solutions of these equations in
earlier chapter in this course. The solution procedure included
obtaining the homogeneous solution and the particular solution
separately. The homogeneous solutions involved finding the roots
of a characteristics equation; that was a quadratic equation.
Depending on the types of roots – real and distinct, repeated, or
complex conjugates – three different types of solution were
obtained.
Finding the particular solution involved method of
undetermined coefficients.
 Solution using Laplace transform does not involve any of
these. In fact, we have obtained solution for all three types of roots.
But the solution procedure was totally blind to all this. The biggest
challenge in obtaining solution of initial value problems using
Laplace transform is to obtain the inverse transform of the transfer
function that involves finding functions in t-domain whose Laplace
transform in s-domain.

Step function
In this and next sections, we are
going to discuss some tools that u(t)
1
will help us investigate behavior
of engineering systems with
discontinuous forcing function.
We will discuss Heaviside step 0
function in this section, and
Dirac delta function in the next uc(t) 1
section.

The Heaviside unit step function 0 c

is defined as
Figure 1. Unit step function u(t)
u(t) = 0 t≤0 and shifted unit step function uc(t)
=1 t>0
Here, we define a shifted unit step function, uc(t) as
uc(t) = 0 0≤t<c
=1 t≥0
The functions are shown graphically in Figure 1.

Before embarking upon applying Laplace transform to uc(t) and

using this function to obtain solution of engineering systems with
discontinuous forcing function, we see how we can apply uc(t) to
define discontinuous forcing functions.

Example 11
If a function f(t), t ≥ 0 is shifted by a delay c in t-domain, the new
function can be written as uc(t)f(t). The functions f(t) and uc(t)f(t)
are shown in Figure 2.

c
f(t) uc(t)f(t)
Figure 2. A function f(t); and the shifted function uc(t)f(t).

Example 12
We define a complimentary unit
step-function as 1 – uc(t). It is 1
defined as
1 − uc(t)
1 – uc(t) = 1 0≤t≤c
=0 t>c 0 c
This function is shown in Figure
3. Figure 3. The function 1 − uc(t)

Example 13
A rectangular pulse of width from t = a to t = b can be defined as
ua(t) – ub(t). The function is shown in Figure 4.
 a b
Figure 4. A rectangular pulse ua(t) – ub(t).
Example 14
If a function f(t) is defined between a ≤ t ≤ b as
g(t) = f(t) a≤t≤b
=0 otherwise
This function can also be written as
g(t) = (ua(t) – ub(t))f(t)

Example 15
The idea of a rectangular pulse from Example 11 can be used to
define more complicated discontinuous functions, such as
f(t) = 8sin t 0≤t≤1
= et 1≤t≤3
= t2 3≤t≤5
We can write this in terms of a rectangular pulse as
f(t) = (u0(t) – u1(t) 8sin t + (u1(t) – u3(t)) et + (u3(t) – u5(t)) t2
The function is shown in Figure 5.

Now we proceed to compute the Laplace transform of uc(t).

To make use of this, we define a delayed function g(t), and compute
the Laplace transform of g(t). Let us define g(t) as
g(t) = 0 0≤t≤c
= f(t – c) t≥c
As we saw earlier, g(t) can also be written as uc(t) f(t – c). Therefore,

Let us define y = t – c, so t = y + c, and dt = dy. Therefore

( )

So, we conclude

which is equivalent to
We use the last two expressions for solving problems. First we
embark on computing some Laplace transform.

Example 16
Compute Laplace transform

The function is shown in Figure 6.

We write each interval using unit step-function

Taking the Laplace transform

Example 17
Compute Laplace transform

This can be written in terms of the unit step-function

f(t) = t(1 – u1(t)) + 1·u1(t) = t – (t – 1)u1(t)
Therefore, the Laplace transform is

Example 18
Compute Laplace transform

This can be written in terms of the unit step-function

We will write this in terms of the unit step-function as
f(t) = (t + 3)u2(t) = (t – 2)u2(t) + 5u0(t)
The Laplace transform is

Example 19
Compute Laplace transform

We write this as
f(t) = 1·(1 – u2(t)) + t2u2(t) = 1 + (t2 – 1)u2(t)
Now
t2 – 1 = (t – 2 + 2)2 – 1 = (t – 2)2 +4(t – 2) + 4 – 1
= (t – 2)2 + 4(t – 2) + 3
Therefore
f(t) = 1 + [(t – 2)2 + 4(t – 2) + 3]u2(t)
The Laplace transform is

Example 20
Compute the inverse Laplace transform

We write this as

Taking the inverse Laplace transform

Example 21
Compute the inverse Laplace transform
We write

Taking inverse Laplace transform

( ) ( )

( ) ( )

Example 22
Compute the inverse Laplace transform

We will write this as

Taking the inverse Laplace transform

( )

( )

Solution of initial value problems with discontinuous

input
Now that we have learnt about handling discontinuous functions,
let us look into solving initial value problems with discontinuous
input.

Example 23
Find solution of the initial-value problem

We initiate the solution taking Laplace transform of the differential

equation, and invoking the initial conditions.

This give the transfer equation

We use partial fractions

To bring this to invertible form, we write

Now taking the inverse Laplace transform

Further

( )

Combining the two transforms will give us the complete solution.

( ) √3 1 √3
𝑦(𝑡) = 𝑢 (𝑡) 1 − 𝑒 cos (𝑡 − 1) + sin (𝑡 − 1)
2 √3 2

√ √
+𝑒 cos 𝑡 + sin 𝑡
√

We can also write the solution as

⎧𝑒 √3 1 √3
cos 𝑡 + sin 𝑡 𝑡>0
⎪ 2 √3 2
𝑦(𝑡) =
⎨ ( ) √3 1 √3
⎪1 − 𝑒 cos ( )
𝑡−1 + sin (𝑡 − 1) 𝑡>1
2 √3 2
⎩
The solution is shown in Figure 7.
We use variation of Example 23 to further understand behavior of
systems. As a first variation, we change the initial condition.

Example 24

We will proceed as before – we will take the Laplace transform of

the differential equation, then invoke the initial condition.

We obtain the transfer equation.

Using partial fractions, we obtain

We have already converted this to an invertible form in
Example 23

Taking the inverse Laplace transform, we obtain our solution.

( ) √3 1 √3
𝑦(𝑡) = 𝑢 (𝑡) 1 − 𝑒 cos (𝑡 − 1) + sin (𝑡 − 1)
2 √3 2

The behavior is shown in Figure 8.

We have seen earlier in this chapter that a second order differential

equation can be a governing equation for a spring-mass-damper
system, an RLC circuit in series or parallel. The governing equation
for a spring-mass-damper system is

The governing equation of an RLC circuit in series is

The governing equation for an RLC circuit in parallel is

where R is the resistance, C is the capacitance, L is the inductance,

those are properties of the system. The output i is the current in the
loop, and v is the voltage across the circuit. The input vs(t) is the
source voltage, and is(t) is the source current. With these equations
we want to investigate the response i for the input vs(t); and the
response v for the input is(t). To understand the general behavior of
the system, we are going to change the properties of the system,
and observe how the behavior of the system responds.

Example 25
We increase the mass m in a spring-mass-damper system, L in a
series RLC circuit, or C is a parallel RLC circuit.

After proceeding as before, we obtain the transfer equation

Bringing it to an invertible form

Taking the inverse Laplace transform

( ) √7 1 √7
𝑦(𝑡) = 𝑢 (𝑡) 1 − 𝑒 cos (𝑡 − 1) + sin (𝑡 − 1)
4 √7 4

The new behavior and its comparison with the original behavior are
shown in Figure 9.

Example 26
We decrease the mass m in a spring-mass-damper system, L in a
series RLC circuit, or C is a parallel RLC circuit.

We skip the details here. The final solution is

( )

The new behavior and its comparison with the original behavior are
shown in Figure 10.

Example 27
We increase the damping coefficient Cd in the spring-mass-damper
system, increase R in a series RLC circuit, or decrease R in a parallel
RLC circuit.

The transfer function is

Using partial fractions will lead to

Inverting this, we obtain the solution

( )

The new behavior and its comparison with the original behavior are
shown in Figure 11.

Example 28
We decrease the damping coefficient Cd in the spring-mass-damper
system, decrease R in a series RLC circuit, or increase R in a parallel
RLC circuit.

The transfer function is

Taking the inverse, we obtain

( ) √15 1 √15
𝑦(𝑡) = 𝑢 (𝑡) 1 − 𝑒 cos (𝑡 − 1) + sin (𝑡 − 1)
4 √15 4

The new behavior and its comparison with the original behavior are
shown in Figure 12.
Example 29
As our last set of examples, we increase the spring constant k in the
spring-mass-damper system; decrease C in a series RLC circuit; or
decrease L in a parallel RLC circuit.

Find the solution of the initial value problem

The transfer function is

Using partial fractions would lead to

 1 ( ) √7 1 √7
𝑦(𝑡) = 𝑢 (𝑡) 1−𝑒 cos ( 𝑡 − 1) + sin ( 𝑡 − 1)
2 2 √7 2

The new behavior and its comparison with the original behavior are
shown in Figure 13.

New behavior
Original behavior

Figure 13. Example 29

Example 30
As our last example in this series, we decrease the spring constant
k in the spring-mass-damper system; increase C in a series RLC
circuit; or increase L in a parallel RLC circuit.

Find the solution of the initial value problem

The transfer function is

Since we have gone through this exercise a number of times, we just
state the final solution here
1 ( ) 1 1
𝑦(𝑡) = 𝑢 (𝑡) 1 − 𝑒 cos (𝑡 − 1) + sin (𝑡 − 1)
2 2 2
The new behavior and its comparison with the original behavior are
shown in Figure 14.

Example 31
Find the solution of the initial value problem

This problem governs the charge on the capacitor in a simple

electric circuit with a unit voltage pulse for 5 ≤ t < 20.
Alternatively, y may represent the response of a damped oscillator
subject to the applied force g(t). Alternatively, y may represent the
response of a spring-mass-damper system (a car or a motorbike)
going over a bump.
To find a solution, we take the Laplace transform of the differential
equation.

Substituting the initial conditions, and solving for Y(s)

Computing the inverse Laplace transform will give us the solution

y(t). To do that, we write

where

The solution, therefore, would be

y(t) = u5(t)h(t – 5) – u20(t)h(t – 20)
where

To compute h(t), we write H(s) in terms of p[atrial fractions. The

details are as before, therefore are skipped here. Finally,

Taking the inverse Laplace transform

Applying this, the solution, once again, is
y(t) = u5(t)h(t – 5) – u20(t)h(t – 20)
The solution is shown in Figure 15.

Impulse Function
Besides the unit step function, another very common discontinuous
that appears in modeling of engineering systems is the impulse
function. The impulse function is modeled by the Dirac delta
function (t). During our discussion on Fourier transform, we
encountered the delta function, as defined by Paul Dirac
(1)

(2)

(3)
There are ambiguities in the definition of the function the way it
has been defined. Nevertheless, the delta-function is extremely
instrumental in understanding the behavior of system with impulse
input. To that end, we embark upon computing the Laplace
transform of (t) directly. Invoking the third property of the delta
function,

Example 32
Obtain solution of
2yʺ + yʹ + 2y = (t – 5), y(0) = 0, yʹ(0) = 0
This equation appears very commonly in study of electrical
circuits, or mechanical oscillators. To find a solution, we take
Laplace transform on both sides, and invoke the initial conditions.
(2s2 + s + 2)Y(s) = e−5s
Therefore

To write this in a manner convenient for the inverse transform

Taking the inverse transform

The behavior is shown in Figure 16.

Exercise
Boyce, DiPrima, Meade, p 253, problems 8-15, p 261, problems 1-
20, p 268, problems 1-16, p 274, problems 1-8, p 279, problems 1-
8.