MA3005: Control Theory

School of Mechanical and Aerospace

Engineering

Faculty: Asst Prof Tegoeh Tjahjowidodo

Email: [email protected]
Phone: (+65) 6790 4952
Office: N3-02c-68
Week 2 – Mathematical Modelling of Dynamic Systems
Lecture Map
2. Laplace
Transform
3. Modelling

1. Introduction to Systems

6. First Order 4. Response and

Stability

5. Routh Stability
8. Transient 7. Second Order
Response

9. Basic of PART II. CONTROLLER DESIGN

Control

Week 2 3
Learning Objectives
At the end of this week, you should be able to:
• Define different types of mathematical model for dynamic systems
• Translate different types of mathematical models
• Derive a transfer function from a block diagram model

𝑅(𝑠) SYSTEM 𝐶(𝑠)

𝐺(𝑠)

Week 2 4
Why Develop a Mathematical Model?
Reasons include:

• A controller design specifications are quantitatively described

and require a mathematical model (of the system).

• Systems that need to be studied, analysed or controlled can be

of different types (e.g. mechanical, electrical, hydraulic, etc.).

• One needs to understand how such varied systems respond to

different inputs, design system parameters based on system
response.

Week 2 5
Different Modelling Exercises in This Course
Mathematical models of physical dynamics of system may take the following forms:

𝑑𝑦 𝑄𝑜 𝐾 𝑅(𝑠) SYSTEM 𝐶(𝑠)

+ 𝑦 2 𝑥 = 2𝑥 = =
𝑑𝑥 𝑄𝑖 𝜏𝑠 + 1 𝐺(𝑠)

Differential Transfer Block

Equation Function Diagram

These models describe system response to an input.

Week 2 6
Transfer Function (TF)

Let’s consider a thermometer (essentially a system), where the input is temperature with an
output signal (it can be an electrical signal).

The dynamic characteristic of the system in differential equation is:

𝜏𝑞𝑜 + 𝑞𝑜 = 𝐾𝑞𝑖

In Laplace domain (assuming zero initial condition, q0(0) = 0):

𝜏[𝑠𝑄0 s − 𝑞𝑜 0 ] +𝑄𝑜 𝑠 = 𝐾𝑄𝑖 (𝑠)

𝜏𝑠 + 1 𝑄𝑜 = 𝐾𝑄𝑖

ℒ [𝑜𝑢𝑡𝑝𝑢𝑡] 𝑄𝑜 𝐾
= = = 𝐺(𝑠)
ℒ [𝑖𝑛𝑝𝑢𝑡] 𝑄𝑖 𝜏𝑠 + 1

Transfer function of the system

Week 2 7
Transfer Function (TF) Cont’d.
How a transfer function will help us in understanding the system?
• The ratio of Laplace Transform of output (response function) to Laplace transform of input
(driving function) for a Linear Time-Invariant (LTI) dynamic system, assuming zero initial
conditions, is expressed as: ℒ [𝑜𝑢𝑡𝑝𝑢𝑡] 𝐶(𝑠)
𝐺 𝑠 = =
ℒ [𝑖𝑛𝑝𝑢𝑡] 𝑅(𝑠)
• We can calculate the output of the system (with known transfer function) for any given
inputs (in Laplace): 𝐶 𝑠 = 𝐺(𝑠) ∙ 𝑅(𝑠)

Week 2 8
Block Diagram
System model can also be presented in pictorial form, referred as a block diagram:

𝑅(𝑠) SYSTEM 𝐶(𝑠)

𝐺(𝑠)

The block diagram represents the relationship between input/ output of the transfer function:

𝐶 𝑠 = 𝐺(𝑠) ∙ 𝑅(𝑠)

Week 2 9
Illustration of System Model (1)
Consider a dynamic system defined by the differential equation:
(𝑛) (𝑛 − 1) (𝑚) (𝑚 − 1)
𝑎0 𝑦 + 𝑎1 𝑦 + ⋯ + 𝑎𝑛−1 𝑦 + 𝑎𝑛 𝑦 = 𝑏0 𝑥 + 𝑏1 𝑥 + ⋯ + 𝑏𝑚−1 𝑥 + 𝑏𝑚 𝑥

where 𝑦 is the output and 𝑥 is the input.

In Laplace domain (with zero initial condition), it can be written as:

𝑎0 𝑠 𝑛 𝑌 + 𝑎1 𝑠 𝑛−1 𝑌 + ⋯ + 𝑎𝑛−1 𝑠𝑌 + 𝑎𝑛 𝑠𝑌 = 𝑏0 𝑠 𝑚 𝑋 + 𝑏1 𝑠 𝑚−1 𝑋 + ⋯ + 𝑏𝑚−1 𝑠𝑋 + 𝑏𝑚 𝑋

𝑌(𝑠) 𝑏0 𝑠 𝑚 + 𝑏1 𝑠 𝑚−1 + ⋯ + 𝑏𝑚−1 𝑠 + 𝑏𝑚

or = = 𝐺(𝑠)
𝑋(𝑠) 𝑛
𝑎0 𝑠 + 𝑎1 𝑠 𝑛−1 + ⋯ + 𝑎𝑛−1 𝑠 + 𝑎𝑛
where 𝐺(𝑠) is the transfer function of the system.

Week 2 10
Illustration of System Model (2)
In block diagram, the system is represented as:

𝑅(𝑠) 𝑏0 𝑠 𝑚 + 𝑏1 𝑠 𝑚−1 + ⋯ + 𝑏𝑚−1 𝑠 + 𝑏𝑚 𝐶(𝑠)

𝐺 𝑠 =
𝑎0 𝑠 𝑛 + 𝑎1 𝑠 𝑛−1 + ⋯ + 𝑎𝑛−1 𝑠 + 𝑎𝑛

Week 2 11
Example of Transfer Function
Equation of motion
𝑚𝑥 + 𝑏𝑥 + 𝑘𝑥 = 𝑓(𝑡)
Laplace Transform with zero initial conditions
𝑘 𝑏
𝑚𝑠 2 + 𝑏𝑠 + 𝑘 𝑋 𝑠 = 𝐹(𝑠)

𝑋(𝑠) 1 𝑚
=
𝐹(𝑠) 𝑚𝑠 2 + 𝑏𝑠 + 𝑘 𝑓(𝑡) 𝑥

• For a known force input (in Laplace domain), 𝐹(𝑠), the response, 𝑋(𝑠), can be calculated if
the TF is known.
• The response 𝑥(𝑡) can be calculated once 𝑋(𝑠) is obtained.

Week 2 12
Transfer Function (Cont’d.)
𝑏0 𝑠 𝑚 + 𝑏1 𝑠 𝑚−1 + ⋯ + 𝑏𝑚−1 𝑠 + 𝑏𝑚
A transfer function 𝐺 𝑠 = can be written in factorised
𝑎0 𝑠 𝑛 + 𝑎1 𝑠 𝑛−1 + ⋯ + 𝑎𝑛−1 𝑠 + 𝑎𝑛
pole-zero form:
𝐾(𝑠 + 𝑧1 )(𝑠 + 𝑧2 ) ⋯ (𝑠 + 𝑧𝑚 ) 𝑁(𝑠)
𝐺 𝑠 = = 𝑠 2 + 4𝑠 + 3 = (𝑠 + 3)(𝑠 + 1)
(𝑠 + 𝑝1 )(𝑠 + 𝑝2 ) ⋯ (𝑠 + 𝑝𝑛 ) 𝐷(𝑠)

• Poles: −𝑝1 , −𝑝2 , ⋯ are values of 𝑠 that make the TF 𝐺(𝑠) singular or 𝐺(𝑠) infinite
(= roots of denominator).
• Zeros: −𝑧1 , −𝑧2 , ⋯ are values of 𝑠 that make the TF 𝐺(𝑠) = 0 (= roots of numerator).
• When 𝑛 > 𝑚, then 𝐺(𝑠) has 𝑚 finite zeros and (𝑛 − 𝑚) zeros at 𝑠 = ∞ (infinite zeros).

Week 2 13
Example 1: Poles of a System
Consider TF:
(𝑠 + 𝑧1 )(𝑠 + 𝑧2 )
𝐺 𝑠 = 2
𝑠 (𝑠 + 𝑝1 )(𝑠 + 𝑝2 )3

• As 𝑠 approaches 0, −𝑝1 𝑜𝑟 − 𝑝2 : 𝐺(𝑠) → ∞

• There are a total of 𝑛 = 6 poles:
 Single pole at 𝑠 = −𝑝1
 Twin poles at 𝑠 = 0
 Triplet poles at 𝑠 = −𝑝2

Week 2 14
Example 1: Zeros of a System
Consider TF:
(𝑠 + 𝑧1 )(𝑠 + 𝑧2 )
𝐺 𝑠 = 2
𝑠 (𝑠 + 𝑝1 )(𝑠 + 𝑝2 )3

• 𝐺 𝑠 = 0 at −𝑧1 and −𝑧2 (finite zeros).

• There are 𝑚 = 2 (finite) zeros.
• There are 𝑛 − 𝑚 = 4 zeros of 𝐺(𝑠) at 𝑠 = ∞, i.e., there are four infinite zeros.

Week 2 15
Example 2: Poles of a System
Consider TF:
𝑠 2 + 5𝑠 + 6
𝐺 𝑠 = 4
𝑠 + 6𝑠 3 + 8𝑠 2 + 4𝑠
• The TF can be rewritten as:
(𝑠 + 2)(𝑠 + 3)
𝐺 𝑠 =
𝑠(𝑠 + 4)(𝑠 + 1)2
• There are a total of 𝑛 = 4 poles:
 A single pole at 𝑠 = 0, −4
 Twin poles at 𝑠 = −1
 Two finite zeros at 𝑠 = −2, −3
 Two infinite zeros (= 4 poles – 2 finite zeros)

Week 2 16
Example 2: Poles of a System (Cont’d.)
Example
𝐾 𝑠 + 2 (𝑠 + 10)
𝐺 𝑠 =
𝑠 𝑠 + 1 (𝑠 + 5)(𝑠 + 15)2

• 𝐺(s) has (finite) zeros at 𝑠 =? −2; −10

• Single poles at 𝑠 =? 0; −1; −5
• Double/ twin poles at 𝑠 =? −15

Zero at Infinity
𝐾𝑠 2

𝐾
• When 𝑠 → ∞, lim 𝐺 𝑠 = = 0.
𝑠→∞ 𝑠5 𝑠3
• Therefore, there are three zeros at 𝑠 = ∞.

Week 2 17
Block Diagram
Block Diagram
Block diagram is a pictorial representation of the function performed by each system.
• It depicts the inter-relationships between systems and indicates the signal flows of the
actual system.
• Block diagram can be shown in two types – functional and cascaded.
Functional block, 𝐶 𝑠 = 𝐺 𝑠 𝑅(𝑠) Cascaded block

𝑅(𝑠) 𝐶(𝑠) 𝑅(𝑠) 𝑋(𝑠) 𝐶(𝑠)

𝐺(𝑠) 𝐺1 (𝑠) 𝐺2 (𝑠)

• As 𝑋 𝑠 = 𝐺1 𝑠 𝑅(𝑠) and 𝐶 𝑠 = 𝐺2 𝑠 𝑋(𝑠), therefore, 𝐶 𝑠 = 𝐺2 𝑠 𝐺1 𝑠 𝑅(𝑠):

𝑅(𝑠) 𝐶(𝑠)
𝐺2 (𝑠)𝐺1 (𝑠)

Week 2 19
Block Diagram (Cont’d.)
Summing Point:
𝑅2 (𝑠)
+
𝑅1 (𝑠) + 𝑅1 𝑠 + 𝑅2 𝑠 − 𝑅3 (𝑠)
-
𝑅3 (𝑠)
Branch Point:

𝑅(𝑠)
𝑅(𝑠) 𝑅(𝑠)
𝑅(𝑠)

Week 2 20
Block Diagram of Feedback Control

Week 2 21
Block Diagram for Feedback Control (Cont’d.)

𝑟(𝑡) + 𝑒(𝑡) 𝑐(𝑡)

controller plant
- Transfer Function

sensor

A simplified version – sufficient for most cases of study:

𝑟(𝑡) + 𝑒(𝑡) 𝑐(𝑡)

controller plant
-

Unity Feedback

Week 2 22
Block Diagram: Analysis
Imagine 𝐺 𝑠 as a cascaded transfer function of the controller and the plant, and 𝐻(𝑠) is the transfer function of
the sensor.
𝑅(𝑠) + E(s)
G(s)
C(s)
-
B(s) 𝐻(𝑠)

The following relations may be derived:

𝐶 𝑠 = 𝐺 𝑠 𝐸(𝑠)
𝐶 𝑠 = 𝐺 𝑠 [𝑅 𝑠 − 𝐻 𝑠 𝐶 𝑠 ]
𝐸 𝑠 = 𝑅 𝑠 − 𝐵 𝑠 = 𝑅 𝑠 − 𝐻 𝑠 𝐶(𝑠)
𝐶 𝑠 + 𝐺 𝑠 𝐻 𝑠 𝐶 𝑠 = 𝐺 𝑠 𝑅(𝑠)
1 + 𝐺 𝑠 𝐻 𝑠 𝐶 𝑠 = 𝐺 𝑠 𝑅(𝑠)

Eliminating 𝐸 𝑠 to get relationship between input and output:

𝐶(𝑠) 𝐺(𝑠)
=
𝑅(𝑠) 1 + 𝐺 𝑠 𝐻(𝑠) Closed-Loop Transfer Function (CLTF)

Week 2 23
Block Diagram: Equivalent

𝑅(𝑠) + 𝐸(𝑠) 𝐶(𝑠)

𝐺(𝑠)
-
𝐵(𝑠) 𝐻(𝑠)

is equivalent to:

𝑅(𝑠) 𝐺(𝑠) 𝐶(𝑠)

1 + 𝐺 𝑠 𝐻(𝑠)

Return to slide on “Eliminating a feedback loop”

Week 2 24
Analysis – Alternative
Common examination/ real-life problem:

• Find the closed-loop transfer function for given block diagram.

Steps to solve such a problem:

1) Start from the output signal.
2) Follow the arrows while noting the state value.
3) Continue until you arrive again on the output signal.

Week 2 25
Analysis – Alternative (Cont’d.)

R(s) + E(s) C(s)

G(s)
-
B(s) H(s)

G(s)[ R(s) - H(s) C(s) ] = C(s)

G(s)R(s) - G(s)H(s)C(s) = C(s)
G(s)R(s) = [1 + G(s)H(s)]C(s)
𝐶(𝑠) 𝐺(𝑠)
=
𝑅(𝑠) 1 + 𝐺 𝑠 𝐻(𝑠)

Week 2 26
Terminologies
Open-Loop Transfer Function (OLTF):
R(s) + E(s) C(s)
B( s) G(s)
 G ( s) H ( s)
E ( s) -
B(s) H(s)
Forward Loop Transfer Function (FLTF):

C (s)
 G (s)
E ( s)

Note: If H(s) = 1, OLTF equals FLTF.

Week 2 27
Closed-Loop Transfer Function (CLTF)

𝑅(𝑠) + 𝐸(𝑠) 𝐶(𝑠)

𝐺(𝑠)
-
𝐵(𝑠) 𝐻(𝑠)

For negative (positive) feedback system,

𝐹𝑜𝑟𝑤𝑎𝑟𝑑 𝐿𝑜𝑜𝑝 𝑇𝐹 (+) for negative feedback
𝐶𝐿𝑇𝐹 = (–) for positive feedback
1 ± 𝑂𝑝𝑒𝑛 𝐿𝑜𝑜𝑝 𝑇𝐹

For unity feedback system,

𝐹𝑜𝑟𝑤𝑎𝑟𝑑 𝐿𝑜𝑜𝑝 𝑇𝐹
𝐶𝐿𝑇𝐹 =
1 ± 𝐹𝑜𝑟𝑤𝑎𝑟𝑑 𝐿𝑜𝑜𝑝 𝑇𝐹

Week 2 28
Example: Feedback System

𝑅(𝑠) + 5 𝐶(𝑠)
- 2𝑠 + 1

0.1𝑠
5
2𝑠 + 1 5
𝐶𝐿𝑇𝐹 = =
5
1+ ∙ 0.1𝑠 2.5𝑠 + 1
2𝑠 + 1
Equivalent system:
𝑅(𝑠) 5 𝐶(𝑠)
2.5𝑠 + 1

Week 2 29
Example: Unity Feedback System

𝑟(𝑡) + 3 𝑐(𝑡)
𝑠+2 𝑠(2𝑠 + 1)
-

3(𝑠 + 2)
𝐶(𝑠) 𝑠(2𝑠 + 1) 1.5(𝑠 + 2)
= = 2
𝑅(𝑠) 3(𝑠 + 2) 𝑠 + 2𝑠 + 3
1+
𝑠(2𝑠 + 1)

Equivalent system:
𝑅(𝑠) 1.5(𝑠 + 2) 𝐶(𝑠)
𝑠 2 + 2𝑠 + 3

Week 2 30
Block Diagram Reduction
Block Diagram Reduction
Complicated block diagram can be simplified by step-by-step rearrangement using rules of
block diagram algebra (Block Diagram Reduction section in Modern Control Engineering,
Katsuhiko Ogata, 4th ed., Prentice Hall).

• General rule is to move branch points and summing points, interchange summing
points, and eliminate internal feedback loops.

Week 2 32
Block Diagram Simplification
Sometimes we deal with a complex block diagram.
Example:
𝐺1
𝑅(𝑠) + + + 𝐺2
𝐶(𝑠)
-
𝐻1
+
-
𝐻2

Week 2 33
Block Diagram Simplification (Cont’d.)
Moving a summing point ahead of a block:

A + AG – B A + AG – B
G
-
 -
G

B 1 1
G G( A  B )
G
B = AG – B

Inputs and outputs are the same for both systems.

Week 2 34
Block Diagram Simplification (Cont’d.)
Consecutive summation blocks are swappable.

𝐴 + + 𝐴+𝐵−𝐶 𝐴 + + 𝐴+𝐵−𝐶

+ - - +
𝐵 𝐶 𝐶 𝐵

Week 2 35
Block Diagram Simplification (Cont’d.)
Moving a branch point ahead of a block:

A AG A AG
G  G

AG AG
G

Moving a branch point behind a block:

A
G
AG
 A
G
AG

A A

Week 2 36
Block Diagram Simplification (Cont’d.)
Eliminating a feedback loop:

A + A 𝐺1
𝐺1
B
 B
- 1 + 𝐺1 𝐺2
𝐺2

1 +
 A
𝐺2 𝐺1
B
𝐺2 -

Non-unity feedback system converted to unity feedback system

Return to slide on “Block Diagram: Equivalent”

Week 2 37
Example: Adapted from B-2-2 (pp. 70)
Obtain the CLTF 𝑪(𝒔)/𝑹(𝒔):

G1
R(s) + + + C(s)
G2
-
H1
+
-
H2

Week 2 38
Example: Adapted from B-2-2 (pp. 70) Cont’d.
Swap𝑯the
Sum 𝟏 and 𝑯𝟐 blocks:
summation blocks:

G1
R(s) + + + C(s)
G2
-
H1
+
- H1-H2
H2

Week 2 39
Example: Adapted from B-2-2 (pp. 70) Cont’d.
Summing blocks and sub-feedback system:

G2
1+G11 1  G2 ( H 1  H 2 )

Week 2 40
Example: Adapted from B-2-2 (pp. 70) Cont’d.
Combining two serial
Summing blocks blocks:
and sub-feedback system:

G2 (1  G1 ) G2
1+G1 1  G ( H  H1 ) G ( H  H )
2 1 2 2 1 2

Week 2 41
Summary
At the end of this week, you should be able to:
• Define different types of mathematical model for dynamic systems
• Translate different types of mathematical models
• Derive a transfer function from a block diagram model

𝑅(𝑠) SYSTEM 𝐶(𝑠)

𝐺(𝑠)

Week 2 42
Reference
No. Slide No. Reference

Figure B-2-2. An example of block diagram. Adapted from Modern Control Engineering (p. 70), by Katsuhiko Ogata, 2009, 5th Edition, New Jersey, NJ:
1. 38 – 41
Prentice Hall.

Week 2 43