University of the West of England

CONTROL SYSTEMS DESIGN- UFMFW7-15-3

Element Description : Coursework Report

Student Name : HPRS Siriwardana
Academic Supervisor : ms Anuththara Ariyarathne

1
AIM
Present consultant report for designing and simulating a control system.

OBJECTIVE

Assuming the role of control engineering consultants, we would create a report

including the key points outlined below
1. Define a practical engineering plant which would feature similar dynamical
behavior to
the theoretical dynamics given in the plant description below. Briefly describe the
operation of the plant.

2. Draw two equivalent control system block diagrams, which features the
output feedback
and the state feedback, respectively. Compare the similarity and difference.

3. Analyze the plant performance such stability, observability, controllability,

and time
response to a unit step reference input.

4. Design a state feedback controller. (Specify the reasonable design criteria).

5. Design a corresponding observer and explain when it will be used.
6. Provide relevant performance data and analysis from your simulated studies.
7. Discuss the importance of computational experiments (Simulation studies).

2
THEORY

1. Stability analysis
Routh stability criterion

The system stability analysis is the one of the most important studies in the control
system which is indicated what are the conditions will a system become as unstable
so the system is unstable how can be stabilized the system. If all poles are in the left
half of S plane the system is stable, it is a common criterion to evaluate a system
stability. Since most linear systems have the transfer function form.

Where α’s and β’s is constant and m≤n, is the first necessary step to factorize the
denominator polynomial A(s) to find the transfer function poles before evaluating
the system stability. When a polynomial of degree greater than second this process
is very time consuming therefore a simple criterion, known as Routh's stability
criterion, can be used to determine the number of poles that lie in right half s plane
without having to factories the polynomial A(s). The procedure to use Routh's
criterion is as follows

The denominator polynomial in form of

𝑎𝑂𝑠𝑛 + 𝑎1𝑠𝑛−1 + ⋯ + 𝑎𝑛−1𝑠 + 𝑎𝑛 = 0 (1.1)

3
Construct Routh’s array

𝑆𝑛 𝑎0 𝑎2 𝑎4 𝑎6 … … …
𝑆𝑛−1 𝑎1 𝑎3 𝑎5 𝑎7 … … …
𝑆𝑛−2 𝑏1 𝑏2 𝑏3 𝑏4 … … …
𝑆𝑛−3 𝑐1 𝑐2 𝑐3 𝑐4 … … …
𝑆𝑛−4 𝑑1 𝑑2 𝑑3 𝑑4 … … … (1.2)
...
...
...
𝑆2 𝑒1 𝑒2
𝑆1 𝑓1
𝑆2 𝑔1

Where coefficients 𝑏1 ,𝑏2 ,𝑏3 and so on, are evaluated as follows

The evaluation of the b will be continued until the remaining ones are all zero. The
same pattern of cross multiplying the coefficients of two previous rows is followed
in evaluating the c's, d's, e's, and so on.

4
Therefore,

And

This process is continued until the nth row shall be completed. The complete array
of the coefficients is triangular.
Exam the signs of the first column of the Routh’s array
Routh's stability criterion states that the number of roots of (1.1) with positive real
parts is equal to the number of changes in sign of the coefficients of the first column
of the array. It should be noticed that the exact values of the terms in the first column
need not be known, instead only the signs are needed. The necessary and sufficient
condition that all roots of (1.1) lie in the left half s plane is that all the coefficients
of (1.1) be positive and all terms in the first column of the array have positive signs.

5
2. Performance analysis
Controllability

Once an acceptable mathematical description of a system has been established, the

designer usually examines certain structural properties of the system (as
represented by its model) to ascertain the degree of control that can be exerted on
the system. This is accomplished through the controllability and observability
properties of the system, which form the basis of many design techniques. When
the state space equations are built up, some of the information can be evaluated
from the matrices associated with the equations.

Definition 1: state controllability

A dynamic system is to be completely state controllable if for any time t 0, it is

possible to construct an unconstrained control vector u(t) that will transfer any given
initial state x(t0) to any final state x(T) in a finite time interval t0 < t < T.

A criterion to check the state controllability

P = [ 𝐵 𝐴𝐵 … 𝐴𝑛−1𝐵]
The p is descripted as state controllability matrix when n is the dimension of matrix
A as well as system A is completely state controllable if the rank of matrix p equal
to n.

Moreover (n*m) matrix A have rank r if the maximum number of linearly

independent columns
(rows) are r, r<n, r<m.

Definition 2: output controllability

For any time t0, the system is to be completely state controllable which is possible
to build an unconstrained control vector u(t) which shall transfer for any given
initial output y(t0) to any final output y(T) for a finite time interval t0 < t < T.

The verifying criteria for output controllability

Q = [ 𝐶𝐵 𝐶𝐴𝐵 … 𝐶𝐴𝑛−1𝐵𝐷]

For a system if the rank of matrix Q is equal to I which system output completely
controllable where I is the dimension output/number of output.
6
Observability
Definition 3: Observability

A system internal states measuring ability is known as the observability and a

system completely observable on t0 < t < T if, for every t0 and some T, every state
of vector (t0) can be determined from the result of the output vector y (t) on t0 < t
< T. In physical terms, a system is completely observable if every transition of the
system state eventually affects the output.

A criterion to check the observability

𝑅 = [𝐶 𝐶𝐴 … 𝐶𝐴𝑛−1]𝑇

If the rank of matrix R equals to n the system is completely observable.

State feedback controller design

The linear time invariant systems general state space description as mentioned
below.
𝑆(𝐴, 𝐵, 𝐶, 𝐷):

𝑥̇(𝑡) = 𝐴𝑥(𝑡) + 𝐵𝑢(𝑡) (𝑠𝑡𝑎𝑡𝑒 𝑒𝑞𝑢𝑎𝑡𝑖𝑜𝑛)

𝑦(𝑡) = 𝐶𝑥(𝑡) + 𝐷𝑢(𝑡) (𝑜𝑢𝑡𝑝𝑢𝑡 𝑒𝑞𝑢𝑎𝑡𝑖𝑜𝑛) (1.6)

The vector is defined respectively according to the input, output and state variable
𝑢(𝑡) = [𝑢1(𝑡) … 𝑢𝑚(𝑡)]𝑇
𝑦(𝑡) = [𝑦1(𝑡) … 𝑦𝑙 (𝑡)]𝑇 (1.7)
𝑥(𝑡) = [𝑥1(𝑡) … 𝑥𝑛(𝑡)]𝑇

Dynamic matrix 𝑨𝒏∗𝒏 : Describes the control trajectory of the state vector x(t) and
dynamics of the system
Dynamic matrix 𝑩𝒏∗𝒎 : Describes each control input effects the state variables of
the systems.
Output matrix 𝑪𝒍∗𝒏 : Transform of the state vector x(t) into the output vector y(t).
Transmission matrix 𝑫𝒍∗𝒎 : Describes the direct (feed forward) effect of control
inputs to output vector y(t).

7
The above-mentioned matrices usually known as system metrices and these
matrices are denoted a system by S (A, B, C, D).

𝑣(𝑡) = [𝑣1(𝑡) … 𝑣𝑚(𝑡)]𝑇 (1.8)

Is called reference vector and it should be noticed that there are two gain matrices
are introduced, Fs is the m*n state feedback gain matrix to specify the poles of
the closed loop system as well as H: is the m*m input feed forward gain matrix
to specify the zeros of the closed loop system.

The transfer function matrix between output Y(s) and reference V(s) is given by

Here
𝐹𝑠 = 𝐴 + 𝐵𝐹𝑠 (2.0)

Therefore, the controller system design can be divided into the assignment of poles
and zeros according to some specifications, which also is called pole and zero
assignment approach. The derivation of equation (1.9) is based on the state
feedback control law

𝑢(𝑡) = 𝐹𝑠𝑥(𝑡) + 𝐻𝑣(𝑡) (2.1)

Assign closed loop poles

Inspection of equation (1.9) the poles of the closed loop are determined by the
denominator (closed loop characteristic polynomial)

𝛥(𝑠) = 𝑑𝑒𝑡[𝑠𝐼 − 𝐹𝑠] (2.2)

8
The desired poles of the closed loop as mentioned below

The state feedback matrix Fs have obtain from Equation 2.2 and 2.3

Equal the coefficients associated with 𝑆𝑖on both side of (2.4) to give a set of n
linear equations
(Equation associated with 𝑆𝑛 can be removed because of 𝛼0= 1)

𝜇1(𝐹𝑠) = 𝛼1
.
.
.
𝜇𝑛(𝐹𝑠) = 𝛼𝑛 (2.5)

Here 𝜇𝑖 (. ) is the linear function of 𝐹𝑠

Theorem : A system S(A, B, C, D) as mentioned in (1.6) is completely controllable

therefore closed loop poles would be arbitrarily assigned through the state
feedback low as mentioned in (2.1)

Assign closed loop poles

In order to ensure the actual output y(t) the input feed usually gain H is carried out
as well as y(t) is exactly equal to the reference v(t) once steady state conditions are
reached, i.e. H is chosen to counteract the inherent steady state gain of the closed
loop system. Therefore it is straightforward, with reference to eqn (1.9), and
assuming a step reference v(t), to select

9
Plant description

The system plant was described as follows

A system plant is described as follows.

𝑌(𝑠) 1
= 𝑔(𝑠) = 2
𝑅(𝑠) 𝑠 + 𝑚𝑠 + 𝑛

6𝑡ℎ 𝑑𝑖𝑔𝑖𝑡 𝑜𝑓 𝑦𝑜𝑢𝑟 𝑟𝑒𝑔𝑖𝑠𝑡𝑟𝑎𝑡𝑖𝑜𝑛 𝑛𝑢𝑚𝑏𝑒𝑟

Where; 𝑚= 10
𝑛 = 𝑇ℎ𝑒 𝑙𝑎𝑠𝑡 𝑑𝑖𝑔𝑖𝑡 𝑜𝑓 𝑦𝑜𝑢𝑟 𝑟𝑒𝑔𝑖𝑠𝑡𝑟𝑎𝑡𝑖𝑜𝑛 𝑛𝑢𝑚𝑏𝑒𝑟

Task A
DC motor speed control plant.
The above-mentioned plant generally descripted as a typical DC (Direct Current)
motor voltage speed model.

The Transfer function =𝜔(𝑠)/𝐸(𝑠)

Where

ω – Speed of motor
E – Supply voltage

The system is closed loop control system. The motor speed was given to the tacho
generator as a feedback. According to that this system is observable. From the
potentiometer the Supply voltage was controlled According to that this system is
controllable. The system was used to control most heavy load (DC Motor) in the
industries.

10
Task B
Output Feedback Control System Block Diagram

State Feedback Control System Block Diagram

Fs – State Feedback gain matrix

H – Input Feed forward gain matrix

Fs Specify the poles of the closed loop system and H specify the zeros of the closed
loop system
The state feedback controller refers to the selection of individual feedback gains for
the complete set of state variables therefore it is considered that all the state variables
are available for observation.

11
State feedback control system shall be stabilized the closed loop system and can able
to achieve further design specification moreover the parameterization of state
feedback controller is an important aspect of control theory.

Task C
Stability

𝑌(𝑠) 1
= 𝑔(𝑠) = 2
𝑅(𝑠) 𝑠 + 𝑚𝑠 + 𝑛

6𝑡ℎ 𝑑𝑖𝑔𝑖𝑡 𝑜𝑓 𝑦𝑜𝑢𝑟 𝑟𝑒𝑔𝑖𝑠𝑡𝑟𝑎𝑡𝑖𝑜𝑛 𝑛𝑢𝑚𝑏𝑒𝑟

Where; 𝑚= 10
𝑛 = 𝑇ℎ𝑒 𝑙𝑎𝑠𝑡 𝑑𝑖𝑔𝑖𝑡 𝑜𝑓 𝑦𝑜𝑢𝑟 𝑟𝑒𝑔𝑖𝑠𝑡𝑟𝑎𝑡𝑖𝑜𝑛 𝑛𝑢𝑚𝑏𝑒𝑟

My register no 24000818
𝑌(𝑠) 1
= 𝑔(𝑠) = 2
𝑅(𝑠) 𝑠 + 0.8𝑠 + 8

𝑠𝑛 𝑎0 = 1 𝑎2 = 8 𝑎4 = 0
𝑛−1 𝑎1 = 0.8 𝑎3 = 0 𝑎5 = 0
𝑠
𝑠 𝑛−2 𝑏1 = 8 𝑏2 = 0 𝑏3 = 0
𝑠 𝑛−3 𝑐1 = 0 𝑐2 = 0 𝑐3 = 0
𝑠 𝑛−4 𝑑1 = 0 𝑑2 = 0 𝑑3 = 0

Calculation

𝑎1 𝑎2 − 𝑎0 𝑎3 𝑎1 𝑎4 − 𝑎0 𝑎5 𝑎1 𝑎6 − 𝑎0 𝑎7
𝑏1= 𝑏2= 𝑏3=
𝑎1 𝑎1 𝑎1

0.8 ∗ 8 − 1 ∗ 0 1∗ 0−1∗0 1∗ 0−1∗ 0

𝑏1= 𝑏2= 𝑏3=
0.8 0.8 0.8

12
 𝑏1 𝑎3 − 𝑎1 𝑏2 𝑏1 𝑎5 − 𝑎1 𝑏3 𝑏1 𝑎7 − 𝑎1 𝑏4
𝑐1= 𝑐2= 𝑐3=
𝑏1 𝑏1 𝑏1

1∗ 0−1∗0 1∗ 0−1∗ 0
8∗0−1∗0 𝑐2= 𝑐3=
𝑐1= 8 8
8

C1=0 C2=0 C3=0

𝑐1 𝑏2 − 𝑏1 𝑏𝑐2 𝑐1 𝑏3 − 𝑏1 𝑐3
𝑑1= 𝑑2=
𝑐1 𝑐1

0∗0−8∗0 0∗ 0−8∗0
𝑑1= 𝑑2=
0 0
d1=0 d2=0
According to above calculation

1. The polynomial coefficients A(s) are positive

2. The first column of Routh’s array values are positive

Therefore, the system is stable.

13
Time response to a unit step reference input

Via the MATLAB the unit step response was generated.

14