2020-2021

Control Engineering Design

Prof. Dr. Ammar A.

Aldair
Electrical Engineering –
Fourth Year
2020-2021
Contents
Chapter One: Design of the Control System in State Space.
1.1 Controllability and Observability Test: ........................................................................... 4

1.1.1 Controllability Test: ................................................................................................. 4

1.1.2 Observability Test: ................................................................................................... 5

1.2 State Feedback Controller (Poles Placement) ................................................................. 7

1.2.2 General Method to Determine the Matrix k: ................................................................. 9

1.2.3 Determination of Matrix k Using Ackermann’s Formula: .......................................... 10

1.2.4 Determination of Matrix k Using Transformation Matrix T: ...................................... 12

1.3 Design of Servo Systems: .............................................................................................. 14

1.4 State Observers: ............................................................................................................. 18

1.4.1 State Observer: ....................................................................................................... 18

1.4.2 Full-Order State Observer: ..................................................................................... 19

1.4.2.1 Effects of the Addition of the Observer on a closed-loop System: ...................... 22

1.4.2.2 Transfer Function of the observer based controller: ............................................. 24

1.4.3 Minimum-Order Observer: .................................................................................... 27

1.4.3.1 Observed-State Feedback Control System with Minimum-Order Observer: . 30

1.4.3.2 Transfer Function of Minimum-Order Observer Based Controller: ............... 32

1.4.3.3 Design of Regulator Systems with Observer: ................................................. 33

1.5 Design of Control System with Observers: ................................................................... 34

1.6 Quadratic Optimal Regulator System: ........................................................................... 38

2.1 Introduction: ....................................................................................................................... 42

2.2 Tuning Rules for PID Controllers: ................................................................................ 42

2.2.1 Ziegler-Nichols Rules for Tuning PID Controller: ................................................ 43

2.3 Modifications of PID Control Schemes: ....................................................................... 54

2.3.1 PI-D Controller:...................................................................................................... 54

2.3.2 I-PD Control: .......................................................................................................... 56

3.1 Introduction: ....................................................................................................................... 58

3.2 Root-Locus Approach to Control System Design: ............................................................. 59

2
 3.2.1 Effects of the Addition of Poles .................................................................................. 60

3.2.2 Effects of the Addition of Zeros .................................................................................. 60

3.3 Lead Compensation ............................................................................................................ 61

3.3.1 Lead Compensation Techniques Based on the Root-Locus Approach. ...................... 62

3.4 Lag Compensation .............................................................................................................. 68

3.4.1 Lag Compensation Techniques Based on the Root-Locus Approach: ........................ 69

3.4.2 Design Procedures for Lag Compensation by the Root-Locus Method: ..................... 70

3.5 Lag-Lead Compensation: ................................................................................................... 78

3.5.1Electronic Lag-Lead Compensator Using Operational Amplifiers: ............................. 79

3.5.2 Lag-Lead Compensation Techniques Based on the Root-Locus Approach: ............... 80

3
 Chapter One
“Design of the Control System in State
Space”

1.1 Controllability and Observability Test:

1.1.1 Controllability Test:

A system is said to be controllable at 𝑡 = 𝑡0 if it is possible means of a control input 𝑢(𝑡) to
transfer the system from any initial state 𝑥(𝑡0 ) to any other state in a finite interval of time.
Consider the single input single output system:
𝒙̇ = 𝑨𝒙 + 𝒃𝒖
𝒚 = 𝒄𝒙
where 𝐴 𝑖𝑠 𝑛 × 𝑛 𝑚𝑎𝑡𝑟𝑖𝑥 , 𝑏 𝑖𝑠 𝑛 × 1 𝑐𝑜𝑙𝑢𝑚𝑛 𝑣𝑒𝑐𝑡𝑜𝑟 𝑎𝑛𝑑 𝑐 𝑖𝑠 1 × 𝑛 𝑟𝑜𝑤 𝑣𝑒𝑐𝑡𝑜𝑟
It can be showed that if the matrix 𝑆 forms as:
𝑆 = [𝑏 𝐴𝑏 𝐴2 𝑏 … 𝐴𝑛−1 𝑏](𝑛×𝑛)
is of rank n (when n is the order of the system) then the system is completely state
controllable. The rank of matrix 𝑆 is n, if the matrix 𝑆 is non-singular or det[𝑆] ≠ 0.
Example 1.1: Given the following system:
𝑥̇ 1 1 𝑥1 1
[ 1] = [ ][ ] + [ ]𝑢
𝑥̇ 2 0 −1 𝑥2 0
Is the system completely controllable?
Solution:
Since the order of the given system is n=2
∴ 𝑆 = [𝑏 𝐴𝑏]
1 1 1 1
𝐴𝑏 = [ ][ ] = [ ]
0 −1 0 0
1 1 1 1
Therefore, 𝑆 = [ ] → 𝑑𝑒𝑡 [ ]=0
0 0 0 0
Since the det[𝑆] = 0, so the system is not completely controllable.
Example 1.2: Given the following system:
𝑥̇ 1 1 𝑥1 0
[ 1] = [ ] [𝑥 ] + [ ] 𝑢
𝑥̇ 2 2 −1 2 1
Is the system completely controllable?
4
Solution:
Since the order of the given system is n=2
∴ 𝑆 = [𝑏 𝐴𝑏]
1 1 0 1 0 1 0 1
𝐴𝑏 = [ ][ ] = [ ] ⟹ 𝑆 = [ ] → 𝑑𝑒𝑡 [ ] = −1
2 −1 1 −1 1 −1 1 −1
Since the det[𝑆] ≠ 0, so that the given system is completely controllable.
If the given system is multi input multi output system:
𝒙̇ = 𝑨𝒙 + 𝑩𝒖
𝒚 = 𝑪𝒙
where 𝐴 𝑖𝑠 𝑛 × 𝑛 𝑚𝑎𝑡𝑟𝑖𝑥 , 𝑏 𝑖𝑠 𝑛 × 𝑚 𝑚𝑎𝑡𝑟𝑖𝑥 𝑎𝑛𝑑 𝑐 𝑖𝑠 𝑟 × 𝑛 𝑚𝑎𝑡𝑟𝑖𝑥
The system is said completely controllable if:
𝑑𝑒𝑡[𝑆𝑆 𝑇 ] ≠ 0
Example 1.3: Given the following system:

𝑥̇ 0 1 𝑥1 1 0 𝑢1
[ 1] = [ ] [𝑥 ] + [ ][ ]
𝑥̇ 2 −2 −3 2 0 1 𝑢2
Is the system completely controllable?
Solution:
Since the order of the given system is n=2
𝑆 = [𝐵 𝐴𝐵]
0 1 1 0 0 1
𝐴𝐵 = [ ][ ]=[ ]
−2 −3 0 1 −2 −3
1 0 0 1
∴𝑆=[ ]
0 1 −2 −3
Because 𝑆 is not square matrix we have to find the determinate of 𝑆𝑆 𝑇
1 0
1 0 0 1 0 1 2 −3
𝑆𝑆 𝑇 = [ ][ ]=[ ]
0 1 −2 −3 0 −2 −3 14
1 −3
2 −3
𝑑𝑒𝑡[𝑆𝑆 𝑇 ] = 𝑑𝑒𝑡 [ ] = 19
−3 14
Since the det[𝑆𝑆 𝑇 ] ≠ 0, so that the given system is completely controllable.

1.1.2 Observability Test:

A system is said to be Observable if the state 𝑥(𝑡) can be determined from knowledge of the
input 𝑢(𝑡) and the output 𝑦(𝑡) over a finite interval of time. Consider the single input single
output system:
𝒙̇ = 𝑨𝒙 + 𝒃𝒖

5
 𝒚 = 𝒄𝒙
where 𝐴 𝑖𝑠 𝑛 × 𝑛 𝑚𝑎𝑡𝑟𝑖𝑥 , 𝑏 𝑖𝑠 𝑛 × 1 𝑐𝑜𝑙𝑢𝑚𝑛 𝑣𝑒𝑐𝑡𝑜𝑟 𝑎𝑛𝑑 𝑐 𝑖𝑠 1 × 𝑛 𝑟𝑜𝑤 𝑣𝑒𝑐𝑡𝑜𝑟 .
to test the Observability, from the matrix 𝑄 that give as:
𝑐
𝑐𝐴
𝑄 = 𝑐𝐴2
:
[𝑐𝐴𝑛−1 ]
If the rank of 𝑄 is n then the system is completely observable. In other words, if det[𝑄] ≠ 0
then the given system is completely observable.
Example 1.4: Consider the system:
𝑥̇ 1 1 𝑥1 0
[ 1] = [ ][ ] + [ ]𝑢
𝑥̇ 2 2 −1 𝑥2 1
𝑥1
𝑦 = [1 0 ] [ 𝑥 ]
2

Is the system completely observable?

Solution:
Since the order of the given system is n=2, then:
𝑐
𝑄=[ ]
𝑐𝐴
1 1
𝑐𝐴 = [1 0 ] [ ] = [1 1]
2 −1
1 0 1 0
∴𝑄=[ ] → 𝑑𝑒𝑡 [ ]
1 1 1 1
Since det[𝑄] ≠ 0, so that the given system is completely observable.
If the given system is multi input multi output system:
𝒙̇ = 𝑨𝒙 + 𝑩𝒖
𝒚 = 𝑪𝒙
where 𝐴 𝑖𝑠 𝑛 × 𝑛 𝑚𝑎𝑡𝑟𝑖𝑥 , 𝑏 𝑖𝑠 𝑛 × 𝑚 𝑚𝑎𝑡𝑟𝑖𝑥 𝑎𝑛𝑑 𝑐 𝑖𝑠 𝑟 × 𝑛 𝑚𝑎𝑡𝑟𝑖𝑥
The system is said completely observable if:
𝑑𝑒𝑡[𝑄 𝑇 𝑄] ≠ 0
Example 1.5: Given the following system:

𝑥̇ 0 1 𝑥1 1 0 𝑢1
[ 1] = [ ] [𝑥 ] + [ ][ ]
𝑥̇ 2 −2 −3 2 0 1 𝑢2
𝑦1 1 0 𝑥1
[𝑦 ] = [ ][ ]
2 0 1 𝑥2
Is the system completely controllable?
Solution:

6
Since the order of the given system is n=2
1 0
𝑐
𝑄=[ ]=[ 0 1]
𝑐𝐴 0 1
−2 −3
Because 𝑆 is not square matrix we have to find the determinate of 𝑄 𝑇 𝑄

1 0
1 0 0 −2 0 1 ] = [5 6
𝑄𝑇 𝑄 = [ ][ ]
0 1 1 −3 0 1 6 11
−2 −3
5 6
𝑑𝑒𝑡[𝑄 𝑇 𝑄] = 𝑑𝑒𝑡 [ ] = 19
6 11
Since the det[𝑄 𝑇 𝑄] ≠ 0, so that the given system is completely observable.

Home Work:
Test the Controllability and the Observability of the following systems:
0 1 0 1
1) 𝐴 = [0 0 1 ] ; 𝑏 = [3] ; 𝑐 = [1 −1 2]
0 −3 −2 1
2 0 0 0 1
1 0 0
2) 𝐴 = [0 2 0 ] ; 𝐵 = [ 1 0] ; 𝐶 = [ ]
0 1 0
0 3 1 0 1
3) 𝑦⃛ + 16𝑦̈ + 192𝑦̇ + 640𝑦 = 160𝑢̇ + 640𝑢.

1.2 State Feedback Controller (Poles Placement)

Let the single input single output system be given by:
𝒙̇ (𝑡) = 𝐴𝒙(𝑡) + 𝑏𝑢(𝑡) (1.1)
𝑦(𝑡) = 𝑐𝒙(𝑡) (1.2)
if state feedback control 𝑢∗ (𝑡) = −𝒌𝒙 , then the control input to given system is given by
𝑢(𝑡) = 𝑟(𝑡) + 𝑢∗ (𝑡), the system with state feedback system is given in Figure 1:

r(t) + y(t)

𝑢∗ (𝑡)

Figure 1 the system with state feedback controller

7