Process Dynamics and Control

By: Gemechu B. (SCBE, AAiT, AAU)

Unit Three
Laplace Transforms and Transfer Function Models

October 25, 2023

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms andOctober
Transfer
25,Function
2023 Models
1 / 60
 3.1. Linearization of Non-linear Models

A linear differential equation is a differential equation that contains

variables only to the first power in any one term of the equation.

Example of non-linears are: squares, square roots, exponentials,

logarithmic, power, products of variables etc.

Many dynamic models contain one or more non linear terms,

therefore they are non-linear.

Most of significant developments in the control theories have occurred

for linear processes as easy analytical solutions are available for them.

Unfortunately, most of the chemical processes are non-linear in nature.

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms andOctober
Transfer
25,Function
2023 Models
2 / 60
 Linearization of Non-linear Systems

Linearization is the process by which we approximate non-linear

equation (terms) with linear ones.
The reasons for linearization are:
Non-linear systems do not have a closed form analytically solution
whereas linear systems have.
Control theory for linear systems is more developed than non-linear
systems.

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms andOctober
Transfer
25,Function
2023 Models
3 / 60
 The following figure pictorially describes the effect of linearization of a
nonlinear process

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms andOctober
Transfer
25,Function
2023 Models
4 / 60
 Linearization of non-linear systems with one variable

Taylor expansion of a single variable nonlinear function f(x):

x − xo ) (x − xo )2
f (x) = f (xo ) + f 0 (xo )[ ] + f 00 (xo )[ ] + ...
1! 2!

If we neglect all terms of order two and higher we get first order
Taylor approximation of the function as:

x − xo )
f (x) = f (xo ) + f 0 (xo )[ ]
1!

Therefore the linearization equation for a single variable becomes:

f (x) = f (xo ) + f 0 (xo )[x − xo ]

Linear approximation is satisfactory only when x is close to xo .

By: Gemechu B. (SCBE, AAiT, AAU) Process
Unit Dynamics
Three and ControlLaplace Transforms andOctober
Transfer
25,Function
2023 Models
5 / 60
 Linearization of non-linear systems with two variables

Consider a nonlinear function of two variables f (x1 , x2 ).

The linearization equation is obtained from the first order Taylor

expansion:

0 0
f (x1 , x2 ) = f (x1o , x2o ) + fx1 (x1o , x2o )[x1 − x1o ] + fx2 (x1o , x2o )[x2 − x2o ]

Linear approximation is satisfactory when (x1 , x2 ) is close to (x1o , x2o ).

First order Taylor approximation of nonlinear functions of more than

two variables can be developed similarly.

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms andOctober
Transfer
25,Function
2023 Models
6 / 60
 Deviation Variables: is the deviation (difference) of the actual value of
the variable from its steady state value.

In most control related problems, it is much easier to deal with

deviation variables than the actual variables.

A dynamic model is converted to its equivalent deviation variable

form by subtracting the steady state equation from linearized dynamic
model equation.
Deviation = Linearized dynamic model - Steady State

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms andOctober
Transfer
25,Function
2023 Models
7 / 60
 Steps for linearization and deviation variable form

1) Identify the nonlinear term in the equation

2) Find the first order Taylor series approximation of the term around
the steady state

3) Replace the nonlinear term by the approximate linear term

4) Subtract the steady state equation from the linearized dynamic

equation

5) Rearrange it and write it in a simplified form

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms andOctober
Transfer
25,Function
2023 Models
8 / 60
 Example 1: Consider the dynamical model equation of surge tank at
steady state height of hs
d(h) √
A = F i − Cv h
dt √
Here, h is a non-linear term in the model.
√
⇒ f (h) = h = f (hs ) + f 0 (hs )[h − hs ]
√ h − hs
⇒ f (h) ' hs + √
2 hs
d(h) √ h − hs
⇒A = Fi − Cv [ hs + √ ]- - - - - (1) Linearized dynamic equation
dt 2 hs
d(hs ) √ hs − hs
⇒A = Fis − Cv [ hs + √ ]- - - - (2) Steady state equation
dt 2 hs
d(h − hs ) √ √ h − hs
⇒A = [Fi − Fis ] − Cv [ hs − hs + √ ] - - - - [(1) - (2)]
dt 2 hs

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms andOctober
Transfer
25,Function
2023 Models
9 / 60
 d(h0 ) Cv
⇒A = Fi0 − √ h0
dt 2 hs
d(h0 )
⇒A + αh0 = Fi0
dt
Cv
where α = √
2 hs
Thus, Deviation equation of surge tank at steady state height, hs is:

d(h0 )
A = Fi0 − αh0
dt

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
10 / 60
 3.2. The Laplace Transform of Representative Functions

Transforms: is a mathematical conversion from one way of thinking

to another to make a problem easier to solve.

Laplace transform is a transformation of a function from the time

domain to the s-domain where both t and s are independent variables.

Laplace transforms can be used in process control for solving linear

differential equations.

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
11 / 60
 The Laplace transform of a function f(t)is defined as
Z ∞
F (s) = L {f (t)} = f (t)e −st dt
0

F(s): symbol for the Laplace transform

s: complex independent variable
f(t): function of time to be transformed
L : operator, defined by the integral
f(t) must satisfy mild conditions that include mainly being piece wise
continuous for 0 < t < ∞.

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
12 / 60
 Table: Laplace Transforms of some commonly functions using
R∞
F (s) = 0 f (t)e −st dt

f(t) F(s)
1
1
s
1
e −at
s +a
a
sin(at)
s + a2
2
s
cos(at)
s 2 + a2
a
e −bt sin(at)
(s + b)2 + a2
s +b
e −bt cos(at)
(s + b)2 + a2
n!
tn n+1
s
n!
t n e −at
(s + a)n+1
By: Gemechu B. (SCBE, AAiT, AAU) Process
Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
13 / 60
 3.3. Solution of Differential Equations by Laplace Transform
Techniques

Linear differential equations can be solved easily using LT.

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
14 / 60
 The process of solving differential equation using Laplace transform
involves the following steps:
1) Taking Laplace transform of both sides of the equations substituting
values for the initial conditions in the derivative transforms.
N(s)
2) Rearranging the equations in the form: Y (s) =
D(s)
3) Use partial fraction expansion to break equation into simple fractions
that can be easily inverted using Laplace table
4) Find the inverse Laplace transform using Laplace tables

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
15 / 60
 3.3. Solving of Differential Equations by Laplace Transform
Laplace transform converts any function f(t) to F(s) and the inverse
Laplace transform converts F(s) back to f(t), the table provides an
organized way to carry out these transformations.
L[f 0 (t)] = sF (s) − f (0);
L[f 00 (t)] = s 2 F (s) − sf (0) − f 0 (0)
L[f n (t)] = s n F (s) − s n−1 f (0) − s n−2 f 0 (0) − ... − sf (n−2) (0) − f (n−1) (0)

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
16 / 60
 Example 1: Solve the differential equation using Laplace transforms

(a) y 00 + y = 1, y (0) = y 0 (0) = 0

(b) y 00 + 3y 0 + 2y = 24, y (0) = 10, y 0 (0) = 0

(c) y 00 + 2y 0 + 5y = 20, y (0) = 0, y 0 (0) = 10

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
17 / 60
 Solution (a): y 00 + y = 1, y (0) = y 0 (0) = 0
Taking the Laplace transform on both sides
⇒ L {y 00 } + L {y } = L {1}
1
⇒ [s 2 Y (s) − sy (0) − y 0 (0)] + Y (s) =
s
2 1
⇒ [s Y (s) − 0 − 0)] + Y (s) =
s
1
⇒ [s 2 + 1]Y (s) =
s
1 A Bs + c 1 s
⇒ Y (s) = 2
= + 2 = − 2
s(s + 1) s s +1 s s +1
Taking the inverse Laplace transform
1 s
⇒ L − {Y (s)} = L − { − 2 }
s s +1
1 s
⇒ y (t) = L − { } − L − { 2 }
s s +1
⇒ y (t) = 1 − cos(t)

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
18 / 60
 Solution (b): y 00 + 3y 0 + 2y = 24, y (0) = 10, y 0 (0) = 0
Taking the Laplace transform on both sides
L {y 00 + 3y 0 + 2y = 24}
⇒ L {y 00 } + L {3y 0 } + L {2y } = L {24}
24
⇒ [s 2 Y (s) − sy (0) − y 0 (0)] + 3[sY (s) − y (0)] + 2Y (s) =
s
2 24
⇒ [s Y (s) − 10s − 0] + 3[sY (s) − 10] + 2Y (s) =
s
2 24
⇒ s Y (s) − 10s + 3sY (s) − 30 + 2Y (s) =
s
2 24
⇒ [s + 3s + 2]Y (s) = + 10s + 30
s
10s 2 + 30s + 24
⇒ [s 2 + 3s + 2]Y (s) =
s
10s 2 + 30s + 24 10s 2 + 30s + 24
⇒ Y (s) = =
s(s 2 + 3s + 2) s(s + 1)(s + 2)

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
19 / 60
 A B C
⇒ Y (s) = + +
s s +1 s +2
12 4 2
⇒ Y (s) = − +
s s +1 s +2
Taking the inverse Laplace transform
12 4 2
⇒ L − {Y (s)} = L − { − + }
s s +1 s +2
12 4 2
⇒ y (t) = L − { } − L − { } + L −{ }
s s +1 s +2
1 1 1
⇒ y (t) = 12L − { } − 4L − { } + 2L − { }
s s +1 s +2
⇒ y (t) = 12 − 4e −t + 2e −2t

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
20 / 60
 Solution (c): y 00 + 2y 0 + 5y = 20, y (0) = 0, y 0 (0) = 10
Taking the inverse Laplace transform
L {y 00 + 2y 0 + 5y = 20}
⇒ L {y 00 } + L {2y 0 } + L {5y } = L {20}
20
⇒ [s 2 Y (s) − sy (0) − y 0 (0)] + 2[sY (s) − y (0)] + 5Y (s) =
s
2 20
⇒ [s Y (s) − 0 − 10] + 2[sY (s) − 0] + 5Y (s) =
s
2 20
⇒ [s + 2s + 5]Y (s) = + 10
s
10s + 20
⇒ [s 2 + 2s + 5]Y (s) =
s
10s + 20 A Bs + c 4 2 − 4s
⇒ Y (s) = = + 2 = + 2
s(s 2 + 2s + 5) s s + 2s + 5 s s + 2s + 5

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
21 / 60
 4 −4(s − 1/2)
⇒ Y (s) = +
s (s + 1)2 + 22
4 (s + 1 − 3/2)
⇒ Y (s) = − 4
s (s + 1)2 + 22
4 6 (s + 1)
⇒ Y (s) = + 2 2
−4
s (s + 1) + 2 (s + 1)2 + 22
4 2 (s + 1)
⇒ Y (s) = + 3 −4
s (s + 1)2 + 22 (s + 1)2 + 22
Taking the inverse Laplace transform
4 2 (s + 1)
⇒ L − {Y (s)} = L − { + 3 2 2
−4 }
s (s + 1) + 2 (s + 1)2 + 22
1 2 s +1
⇒ y (t) = 4L − { } + 3L − { 2 2
} − 4L − { }
s (s + 1) + 2 (s + 1)2 + 22
⇒ y (t) = 4 + 3e −t sin(2t) − 4e −t cos(2t)

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
22 / 60
 3.4. Other Laplace Transform Properties

Final and Initial Value Theorem

The final value theorem allows one to compute the value that a
function approaches as t → ∞.

The Initial value theorem allows one to compute the value that a
function approaches as t → 0.

If y(t) and Y(s) is Laplace transform pairs such that

Y (s) = L − [y (t)], then final value (FV) and initial value (IV) are
given by:
FV = lim y (t) = lim sY (s)
t→∞ s→0

IV = lim y (t) = lim sY (s)

t→0 s→∞

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
23 / 60
 Example 2: Find the final and initial value for the differential equations in
example 1.

(a) y 00 + y = 1, y (0) = y 0 (0) = 0

(b) y 00 + 3y 0 + 2y = 24, y (0) = 10, y 0 (0) = 0

(c) y 00 + 2y 0 + 5y = 20, y (0) = 0, y 0 (0) = 10

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
24 / 60
 Solution (a): y 00 + y = 1, y (0) = y 0 (0) = 0
1
Y (s) = 2
and y (t) = 1 − cos(t)
s(s + 1)
Final value = FV = limt→∞ y (t) = lims→0 sY (s)
1
⇒ FV = limt→∞ [1 − cos(t)] = lims→0 s[ 2 ]
s(s + 1)
⇒ FV = 1

Initial value = IV = limt→0 y (t) = lims→∞ sY (s)

1
⇒ IV = limt→0 [1 − cos(t)] = lims→∞ s[ 2 ]
s(s + 1)
⇒ IV = 0

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
25 / 60
 Solution (b): y 00 + 3y 0 + 2y = 24, y (0) = 10, y 0 (0) = 0
10s 2 + 30s + 24
Y (s) = and y (t) = 12 − 4e −t + 2e −2t
s(s 2 + 3s + 2)
Final value = FV = limt→∞ y (t) = lims→0 sY (s)
10s 2 + 30s + 24
⇒ FV = limt→∞ [12 − 4e −t + 2e −2t ] = lims→0 s[ ]
s(s 2 + 3s + 2)
⇒ FV = 12

Initial value = IV = limt→0 y (t) = lims→∞ sY (s)

10s 2 + 30s + 24
⇒ IV = limt→0 [12 − 4e −t + 2e −2t ] = lims→∞ s[ ]
s(s 2 + 3s + 2)
⇒ IV = 10

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
26 / 60
 Solution (c): y 00 + 2y 0 + 5y = 20, y (0) = 0, y 0 (0) = 10
10s + 20
Y (s) = and y (t) = 4 + 3e −t sin(2t) − 4e −t cos(2t)
s(s 2 + 2s + 5)
Final value = FV = limt→∞ y (t) = lims→0 sY (s)

10s + 20
⇒ FV = lim [4 + 3e −t sin(2t) − 4e −t cos(2t)] = lim s[ ]
t→∞ s→0 s(s 2 + 2s + 5)

⇒ FV = 4

Initial value = IV = limt→0 y (t) = lims→∞ sY (s)

10s + 20
⇒ IV = lim [4 + 3e −t sin(2t) − 4e −t cos(2t)] = lim s[ ]
t→0 s→∞ s(s 2 + 2s + 5)

⇒ IV = 0

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
27 / 60
 Transform of an Integral and Time Delay (Translation in Time)

Transform of an Integral

It is necessary to find the Laplace transform of a function that is

integrated with respect to time.

The transform of an integral is given by:

Z t
1
L{ f (t)dt} = F (s)
0 s

Note that Laplace transformation of an integral function of time leads

to division of the transformed function by s.

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
28 / 60
 Time Delay (Translation in Time)

Time delays play an important role in process dynamics and control.

They commonly occur as a result of the transport time required for a

fluid to flow through a pipe.

If L {f (t)} = F (s), then the Laplace transform of a time-delayed

function is:
Fd (s) = L {f (t − θ)} = e −sθ F (s)

Note that Laplace transformation of an time delay function leads to

multiply of the transformed function by e −sθ .

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
29 / 60
By: Gemechu B. (SCBE, AAiT, AAU) Process
Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
30 / 60
 3.5. Transfer Function Models
Transfer function (TF) is the ratio of the output (response function)
to the input (driving function) under the assumption that all initial
conditions are zero.
an algebraic expression for dynamic relation between a selected input
and output of the process model.
can only be derived for a linear model equations.
does not provide any information on the physical structure of the
system.

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
31 / 60
 The following are the steps to drive transfer functions:

1) Write the dynamic model describing the system

2) Linearize equations using Taylor series expansion

3) Write the linearilized equations in dynamic equation

4) Write its steady state equation

5) Find the deviation equation

6) Operate Laplace transform on the deviation equations

7) Obtain the ratio of Laplace transform output to input

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
32 / 60
 Transfer function of the differential equation:
an y n + ... + a1 y 0 + ao y = bm u m + ... + b1 u 0 + bo u
where y is the system output and u is the input of the system
⇒ [an s n + ... + a1 s + ao ]Y (s) = [bm s m + ... + b1 s + bo ]U(s)
bm s m + ... + b1 + bo
⇒ Y (s) = [ ]U(s)
an s n + ... + a1 s + ao
Y (s) bm s m + bm−1 s m−1 + ... + b1 + bo
⇒ G (s) = =
U(s) an s n + an−1 s n−1 + ... + a1 s + ao
Y (s) Laplaceofoutput
⇒ G (s) = = Transfer Function =
U(s) Laplaceofinput
G(s) is called transfer function of the process

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
33 / 60
By: Gemechu B. (SCBE, AAiT, AAU) Process
Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
34 / 60
 Transfer Function for first order control system

A first order control system is a type of control system whose

input-output relationship is a first-order differential equation.

Consider a process with an input u(t) and an output y(t):

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
35 / 60
 Let y(t) is in first order differential equation
a1 y 0 + ao y = bu(t)

Take the Laplace

⇒ a1 sY (s) + ao Y (s) = bU(s)
a1 b
⇒ sY (s) + Y (s) = U(s)
ao ao
b a1
⇒ τ sY (s) + Y (s) = KU(s); K = and τ =
ao ao
⇒ (τ s + 1)Y (s) = KU(s)
K
⇒ Y (s) = U(s)
τs + 1
Y (s) K
⇒ = = G (s)
U(s) τs + 1

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
36 / 60
 A general transfer function of first-order system is:

K
G (s) =
τs + 1

The two parameters of this transfer function are:

τ (time constant)
K (steady state gain)

Time constant of a process, τ : is a measure of how quickly a first

order system response to a unit step input.

Steady state gain, K: is the ration between the input signal and the
steady state value of output.

The unit of gain is the ratio of the units of output to that of input,
whereas the unit of time constant is same as that of time.
By: Gemechu B. (SCBE, AAiT, AAU) Process
Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
37 / 60
 Transfer Function for second order control system

A second-order system is one whose output is described by a

second-order differential equation.

The second order systems encountered in a chemical plant come from

processes that consist of two first-order systems in series.

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
38 / 60
 Let y(t) is in second order differential equation:
a2 y 00 + a1 y 0 + ao y = bu(t)
⇒ a2 s 2 Y (s) + a1 sY (s) + ao Y (s) = bU(s)
a2 a1 b
⇒ s 2 Y (s) + sY (s) + Y (s) = U(s)
ao ao ao
⇒ τ 2 s 2 Y (s) + 2ξτ sY (s) + Y (s) = KU(s)
⇒ [τ 2 s 2 + 2ξτ s + 1]Y (s) = KU(s)
K
⇒ Y (s) = 2 2 U(s)
τ s + 2ξτ s + 1
Y (s) K
⇒ = 2 2 = G (s)
U(s) τ s + 2τ ξs + 1

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
39 / 60
 A general transfer function of second-order system is:

K
G (s) =
τ 2s 2 + 2ξτ s + 1

There are three parameters of this transfer function are:

K = Process steady-state gain
τ = Time constant
ξ = Damping factor

K and τ have the same importance as for a first-order transfer

function.

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
40 / 60
 The steady-state gain (K):
the process gain affects the magnitude of the response.

Time constant (τ ):
determines the speed of response of the system.

The damping coefficient (ξ ) is dimensionless.

the value of ξ completely determines the degree of oscillation in a
process response after a perturbation.

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
41 / 60
 a) Multiplicative rule of transfer function when connected in series:

Y (s) = G1 (s)G2 (s)U(s)

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
42 / 60
 b) Additive rule of transfer function when connected in parallel:

Y (s) = G1 (s)U1 (s) + G2 (s)U2 (s)

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
43 / 60
 Developing Transfer Function

Consider a process with one input f(t) and one output y(t)

Y (s) = G (s)F (s)

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
44 / 60
 Consider a process with two inputs f(t), h(t) and one output y(t):

Y (s) = G1 (s)F (s) + G2 (s)H(s)

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
45 / 60
 A process with two inputs f(t), h(t) and two outputs x(t), y(t):

X (s) = G11 (s)F (s) + G12 (s)H(s)

Y (s) = G21 (s)F (s) + G22 (s)H(s)

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
46 / 60
 Example 1: Let consider the mathematical model of surge tank flow
system:
dh
A + αh = Fi
dt
.

a) Drive the transfer function of the process

b) Determine the time constant and steady static gain

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
47 / 60
 a) Drive transfer function of:
dh
Dynamic model: A + αh = Fi
dt
Take the Laplace transform:
⇒ AsH(s) + αH(s) = Fi (s)
⇒ [As + α]H(s) = Fi (s)
H(s) 1
⇒ =
Fi (s) As + α
1
⇒ G (s) =
As + α

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
48 / 60
 1
b) Time constant and steady static gain: G (s) =
As + α
K
compare with the first order TF: G (s) =
τs + 1
1 K
⇒ =
As + α τs + 1
1/α K
⇒ A =
αs + 1
τs + 1
1 A
⇒ K = and τ =
α α

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
49 / 60
 Example 2: Let consider the mathematical model of a stirred-tank
heating process with constant volume:

dT F Q
= (Ti − T ) +
dt V ρCp V

Where Ti and Q are input variables,

T is output variable
F, V, CP and ρ are parameters (constants).
a) Drive the transfer functions of the process
b) Determine the time constant and static gain

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
50 / 60
 dT F Q
a) Transfer function of: = (Ti − T ) +
dt V ρCp V
dT
⇒ ρCp V = ρCp F (Ti − T ) + Q
dt
dT
⇒ ρCp V + ρCp FT = ρCp FTi + Q
dt
Take the Laplace transform:
⇒ (ρCp V )sT (s) + ρCp FT (s) = ρCp FTi (s) + Q(s)
⇒ [(ρCp V )s + ρCp F ]T (s) = ρCp FTi (s) + Q(s)
ρCp F 1
⇒ T (s) = Ti (s) + Q(s)
(ρCp V )s + ρCp F (ρCp V )s + ρCp F
⇒ T (s) = G1 (s)Ti (s) + G2 (s)Q(s)

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
51 / 60
 There are two transfer functions as of two inputs:

ρCp F 1
G1 (s) = = V
(ρCp V )s + ρCp F ( F )s + 1
1
1 ρC F
G2 (s) = = V p
(ρCp V )s + ρCp F ( F )s + 1
1
b) Time constant and steady static gain for: G1 (s) =
( VF )s + 1
V
⇒ K1 = 1 and τ1 =
F
1
ρCp F
Time constant and steady static gain for: G2 (s) =
( VF )s + 1
1 V
⇒ K2 = and τ2 =
ρCp F F

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
52 / 60
 Example 3: A second order system consists of two first-order processes
operating in series (τ1 = 4, τ2 = 1). Find the equivalent values of τ and ξ
for the corresponding second order system.

Solution: Assume the two first-order transfer functions are:

K1 K2
G1 (s) = and G2 (s) =
τ1 s + 1 τ2 s + 1
If two first-order processes connected in series, then
G (s) = G1 (s)G2 (s)
K1 K2
⇒ G (s) = [ ][ ]
τ1 s + 1 τ2 s + 1
K1 K2
⇒ G (s) = 2
(τ1 )(τ2 )s + (τ1 + τ2 )s + 1

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
53 / 60
 Compare with the second order transfer function:
K
G (s) = 2 2
τ s + 2ξτ s + 1
⇒ τ 2 = (τ1 )(τ2 ) and 2ξτ = τ1 + τ2
⇒ τ 2 = (4)(1) and 2ξτ = 4 + 1
⇒ τ 2 = 4 and 2ξτ = 5
⇒ τ = 2 and 2ξτ = 5
⇒ τ = 2 and 2(2)ξ = 5
⇒ τ = 2 and ξ = 1.25

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
54 / 60
 Example 4: A second order system consists of two first-order processes
operating in series have the same time constants. Determine the damping
factor (ξ) for the corresponding second order system.

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
55 / 60
 Unit Three Problems
1) Solve the differential equation using Laplace transforms, find the final
and initial value.
a) 5y 0 + 4y = 2, y (0) = 1
b) y 00 + 3y 0 + 2y = 6e −t , y (0) = 1, y 0 (0) = 2
c) y 00 + 4y = cos(2t), y (0) = y 0 (0) = 1
d) y 00 − 2y 0 + y = e 2t , y (0) = y 0 (0) = 0
e) y 000 + 6y 00 + 11y 0 + 6y = 1, y (0) = y 0 (0) = y 00 (0) = 0
f) y iv + 16y 000 + 86y 00 + 176y 0 + 105y = 1680, y (0) = y 0 (0) = 0

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
56 / 60
 2) Consider the initial value problem

y 00 + by 0 + cy = 1, y (0) = m, y 0 (0) = n

s 2 + 2s + 1
Suppose that L [y (t)] = Y (s) = .
s 3 + 3s 2 + 2s
Determine the values of the constants b, c, m, n.

3) Find the the final value of y(t) for a unit step change [u(t) = 1] in
the input u(t) using the final value theorem for:
y 000 + 6y 00 + 11y 0 + 6y = 4u 0 + 2u,
where y (0) = y 0 (0) = y 00 (0) = 0 and u(0) = 0.

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
57 / 60
 4) For the integro-differential equation,
Rt
y 00 + 4y 0 + 3y = 2 0 e −t dt with y (0) = y 0 (0) = 0
(a) Find y(t).
(b) What is the limiting value as t → ∞ ?

5) A stirred-tank blending system initially is full of water and is being fed

pure water at a constant flow rate, q. At a particular time, an
operator adds caustic solution at the same volumetric flow rate q but
concentration Ct . If the liquid volume V is constant, the dynamic
VdC
model for this process is: + qC = qCi with c(0)=0.
dt
3 3
Data: V = 2m , q = 0.4m /min, Ci = 50kg /m.
(a) Find c(t).
(b) What is the final value as t → ∞ ?

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
58 / 60
 6) Develop the transfer function relating Qi (s) to Q2 (s) of the system
below. The system consists of two liquid surge tanks in series so that
the outflow from the first tank is the inflow to the second tank.
h1 h2
Given: q1 = and q2 =
R1 R2

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
59 / 60
 7) For the process modeled by:

2x 0 + 2x + 3y = 2u

y 0 + 6y − 4x = 2u + 4f

Find the four transfer functions relating two outputs (x, y ) to two
inputs (u, f ).

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
60 / 60
 8) The mathematical model for a stirred-tank heating process with
constant volume and flow rate is given by:

dT F Q
= (Ti − T ) +
dt V ρCp V

where Ti and Q are input variables, T is output variable and

F = 10m3 /hr , V = 5m3 , Cp = 2 × 10−3 kJ/(kg .o C )
ρ = 1000kg /m3 are parameters.

a) Drive the transfer function for the process model that relates input
temperature (Ti ) and heat supplied (Q) to output temperature (T ).
b) Determine T(t) of the process for step change magnitude of
Ti = 26o C and Q = 40kJ/hr . Determine the final output temperature.
c) How long will take to come to the final temperature of T = 21o C ?

By: Gemechu B. (SCBE, AAiT, AAU) Process

Unit Dynamics
Three and ControlLaplace Transforms and
October
Transfer
25, Function
2023 Models
61 / 60