The University of Western Ontario

Faculty of Engineering
Department of Electrical and Computer Engineering
ECE3330A, Control Systems
Week 2 Material
Last updated in September 2022
Professor: Mehrdad R. Kermani

1 Introduction
The fist step in every control design case is developing a mathemati-
cal model for the system. To this effect, fundamental physical laws of
science and engineering are applied to the system to obtain a differ-
ential equation that can describe the relationship between the input
and output of the system. Ohm’s and Kirchhoff’s laws are examples
that are used to model electrical networks.
Example 1. Use your knowledge of physics and mechanics to obtain
a differential equation governing the rotation angle of the system.

fig_04_21

Figure 1: A rotational mechanical system

2 L-transformation
Although a differential equation represents the relationship between
the input and output of a system, it is not a satisfying representation
from a system perspective, since the input and output as well as the
coefficients of the equation (system parameters) appear throughout
the equation. We would prefer a mathematical representation such
as that shown in 2 where the input, output and system are separate
parts; in other words to represent the system as a block diagram. One
ECE3330A, Control Systems 2

main advantage of such a representation is simple representation of

the interconnection of several subsystems, being cascade or parallel.
Using Laplace transform we can achieve this objective.

Figure 2: The input r(t) stands for reference input and the output
c(t) for controlled output

The L-transformation is defined as,

Z ∞
L[f (t)] = F (s) = f (t)e−st dt
t=0−

where s = σ + jω is a complex variable representing the frequency

fig 02 01
fig_02_01

domain variable. The lower limit of the Laplace transform means

that even if f (t) is discontinuous at t = 0, we can still obtain trans-
formation so long as the integral converges.
When f (t) is also defined for t < 0, i.e., f (t) ̸= 0, the two-sided (as
apposed to one-sided) L-transform can be defined as follows,
Z ∞
L[f (t)] = f (t)e−st dt
t=−∞
In this course, unless stated otherwise, all transformations are one-
sided.
Example 2. Obtain the L-transform of impulse function defined as,
∞, 0− < t < 0+

δ(t) =
0, otherwise

Example 3. Obtain the L-transform of step function defined as,


1, t ⩾ 0
u(t) =
0, t < 0
ECE3330A, Control Systems 3

Example 4. Obtain the L-transform of ramp function defined as,


t, t ⩾ 0
r(t) =
0, t < 0

Example 5. Obtain the L-transform of exponential function defined

as,
Ae−at , t ⩾ 0

f (t) =
0, t<0

Example 6. Using the result of Example 5, obtain the L-transform

of sine function defined as,

A sin ωt, t ⩾ 0
f (t) =
0, t<0

Note the alternative definition of sinusoidal functions as sin ωt =

ejωt −e−jωt jωt −jωt
2j and cos ωt = e +e 2

2.1 Important Properties of L-Transformation

Some of important properties of the Laplace transformation are high-
lighted in the following,
ECE3330A, Control Systems 4

2.1.1 Linearity of L-transformation

Let F (s) = L[f (t)] and G(s) = L[g(t)], and α and β are two scalar
constants, then

L[αf (t) + βg(t)] = αF (s) + βG(s)

2.1.2 Differentiation theorem

n
dn f (t) n
X
L[ ] = s F (s) − sn−k f k−1 (0− )
dtn
k=1
3
e.g., L[ d dtf 3(t) ] = s3 F (s) − s2 f (0) − sf˙(0) − f¨(0)

2.1.3 Frequency-shift
L[e−at f (t)] = F (s + a)

2.1.4 Time-shift
L[f (t − T )] = e−sT F (s)

2.1.5 Initial value theorem

The initial value of a function can be obtained from the L-transformation
of that function as follows,
Z ∞
f (0+ ) = lim sF (s) = lim s f (t)s−st dt
s→∞ s→∞ t=0

2.1.6 Final value theorem

If F (s) has no poles with positive or zero real part and no double- or
higher-order poles at the origin, then the final value of f (t) is given
by,
f (∞) = lim f (t) = lim sF (s)
t→∞ s→0

2.1.7 Convolution
Z t Z t
L[ f1 (t − τ )f2 (τ )dτ ] = L[ f1 (τ )f2 (t − τ )dτ ] = F1 (s)F2 (s)
t=0 t=0

2.2 L-transform Inversion

The inverse Laplace transform for obtaining f (t) from its L-transformation
F (s) is defined as,
Z σ+j∞
−1 1
L [F (s)] = F (s)est ds = f (t)u(t)
2πj σ−j∞

where u(t) is the unit step function.

ECE3330A, Control Systems 5

2.3 Partial Fraction Expansion (PFE)

To find the inverse Laplace transform of a complicated function, the
function can be first converted to sum of simpler terms whose inverse
L-transformation is available or can be easily obtained. This method
is known as Partial Fraction Expansion.
Example 7. Obtain the inverse L-transform of,
2 2
F (s) = =
s2 + 3s + 2 (s + 1)(s + 2)

In general, given an F (s) whose denominator has real and distinct

roots, a PFE can be obtained as

N (s) N (s) k1 k2 kn
F (s) = = = + + ... +
D(s) (s + p1 )(s + p2 )...(s + pn ) s + p1 s + p2 s + pn

Subsequently, f (t) can be obtained as a sum of several terms each

related to one of the denominator’s root, i.e.,

f (t) = k1 e−p1 t + k2 e−p2 t + ... + kn e−pn t

Example 8. Obtain the inverse L-transform of,

2 2
F (s) = =
s3 + 5s2+ 8s + 4 (s + 1)(s + 2)2
ECE3330A, Control Systems 6

In general, given an F (s) whose denominator has real and repeated

roots, a PFE can be obtained as

N (s) N (s) k1 k2 kr kr+1 kn

F (s) = = r
= r
+ r−1
+ ... + + + ... +
D(s) (s + p1 ) (s + p2 )...(s + pn ) (s + p1 ) (s + p1 ) s + p1 s + p2 s + pn

Subsequently, f (t) can be obtained as a sum of several terms each

related to one of the denominator’s root, i.e.,

tr−1 −p1 t tr−2 −p1 t

f (t) = k1 e +k2 e +...+kr e−p1 t +kr+1 e−p2 t +...+kn+r e−pn t
(r − 1) (r − 2)

Example 9. Obtain the inverse L-transform of,

3
F (s) =
s(s2 + 2s + 5)
ECE3330A, Control Systems 7

In general, given an F (s) whose denominator has complex or purely

imaginary roots, a PFE can be obtained as
N (s) N (s) k1 k1∗ kn
F (s) = = ∗ = + ∗ + ... +
D(s) (s + p1 )(s + p1 )...(s + pn ) s + p1 s + p1 s + pn

where, we assume that p1 = ℜ(p) + jℑ(p) = |p|ej∡p and p2 = p∗1 =

ℜ(p) − jℑ(p) = |p|e−j∡p . Subsequently, f (t) can be obtained as a
sum of several terms each related to one of the denominator’s root,
i.e.,

f (t) = k1 e−(ℜ(p)+jℑ(p))t + k1∗ e−(ℜ(p)−jℑ(p))t + ... + kn e−pn t

The first 2 terms can then be further simplified to exponentially

decaying or pure sinusoidal terms. More on this can be found in
future examples as well as assignments.

3 Transfer Function
We are now ready to formulate a system representation, known as
Transfer Function, that algebraically relates a system’s input to its
output while keeping input, output and system representations sep-
arate. Let us begin by a general nth-order Linear Time-Invariant
(LTI) differential equations of the form,

dn y(t) dn−1 y(t) dm u(t) dm−1 u(t)

an +a n−1 +...+a0 y k = bm +bm−1 +...+b0 uk
dtn dtn−1 dtm dtm−1
(1)
where y(t) is the output, u(t) is the input, and ai and bi are constant
coefficients. In control theories a linear system is called causal if
n⩾m

Remark in some textbooks a different notation may be adopted.

You can use any notation so long as you keep the consistencies of
your preferred notation throughout the calculations. For example a
differential equation can also be defined as,

dn y(t) dn−1 y(t) dm u(t) dm−1 u(t)

+a 1 +...+an yk = b0 +b1 +...+bm uk
dtn dtn−1 dtm dtm−1
Taking L-transform of the differential equation (1), results in,

an sn Y (s)+an−1 sn−1 Y (s)+...+a0 Y (s)+ICy(0) = bm sm U (s)+bm−1 sm−1 U (s)+...+b0 U (s)+ICu(0)

(2)
where ICy(0) and ICu(0) are many terms due to the initial conditions
of y(t) and u(t), respectively. To obtain the transfer function of
the system, we assume all initial conditions are zero, i.e.,
ECE3330A, Control Systems 8

(an sn + an−1 sn−1 + ... + a0 )Y (s) = (bm sm + bm−1 sm−1 + ... + b0 )U (s)
(3)
and from which the ratio of the Laplace transform of the output to
the Laplace transform of input can be obtained as,

Yss (s) bm sm + bm−1 sm−1 + ... + b0 N (s)

G(s) ≜ = n n−1
= (4)
U (s) an s + an−1 s + ... + a0 D(s)

This ratio is known as the transfer function of the system. The

roots of N (s) = 0 (m roots) are called the zeros of the system and
the roots of the D(s) = 0 (n roots) are called the poles of the system.
ECE3330A, Control Systems 9

Example 10. Use the L-transformation to obtain the transfer func-

tion for the system in example 1.

Example 11. Use the L-transformation to obtain the transfer func-

tion for the following differential equation,

y (3) (t) + 2ÿ(t) + ẏ(t) = 2r̈(t) − r(t)

Example 12. Use the L-transformation to obtain Y (s) given the in-
put r(t) = 3δ̇(t), and the specified initial conditions. The differential
equation describing the system is given by,

y (3) (t) + 2ÿ(t) + ẏ(t) = 2r̈(t) − r(t)

y(0) = 1 ẏ(0) = −2 ÿ(0) = 0

ECE3330A, Control Systems 10

Example 13. Obtain the response of the previous system to a unit

impulse function δ(t) assuming all initial conditions are zero. Identify
the components of the response and observe their relation to the
system poles.

Example 14. Repeat the previous example for a unit step function
u(t). Notice the drastic changes in the behavior of the system re-
sponse. What can be the cause of this behavior? We will talk about
this later in this course.
ECE3330A, Control Systems 11

table_02_01

table_02_02