LINEAR CONTROL SYSTEMS

Unit -5
STATE SPACE ANALYSIS
Define state, state variable, state model and state diagram in state space
1 a
analysis.
1. State
The state of a system is defined as the smallest set of variables that completely
describes the system at any given time.

2. State Variable

State variables are the set of independent variables that represent the state of
the system. The number of state variables required to describe a system is equal
to the order of the system.

3. State Model
The state model is a mathematical representation of a system using two
equations namely, state equations and output equations.
o State Equation:
𝒙̇ (𝒕) = 𝑨𝒙(𝒕) + 𝑩 𝒖(𝒕)
o Output Equation:
𝒚(𝒕) = 𝑪𝒙(𝒕) + 𝑫 𝒖(𝒕)
where A = State matrix
u(t) = input vector B = Input matrix
x(t) = state vector C = output matrix
y(t) = output vector D= feedthrough matrix

4. State Diagram:
A state diagram is a graphical representation of the state-space model, showing
the relationship between state variables and system inputs/outputs.
 What is State transmission matrix? Discuss the properties of state transition
b
matrix.

2 a Discuss the concept of controllability and observability with an example.

Controllability and observability are fundamental concepts in control system theory. They
determine whether a system can be fully controlled and observed using input-output
relationships. These properties are crucial in state-space representation and for designing
controllers and observers in modern control systems.
1. Controllability
A system is controllable if it is possible to bring the system from one state to any other
desired state by varying inputs.
Mathematical Condition (Kalman’s Controllability Test)
A linear time-invariant (LTI) system represented in state-space form:
𝒙̇ (𝒕) = 𝑨𝒙(𝒕) + 𝑩 𝒖(𝒕)
where:
• x(t) = state vector,
• u(t) = input vector,
• A = state matrix,
• B = input matrix.
The system is controllable if the controllability matrix Qc has full rank (n).
(Where n is the number of states)
𝑸𝒄 = [𝑩 𝑨𝑩 𝑨𝟐 𝑩 ⋯ 𝑨𝒏−𝟏 𝑩]
2. Observability
A system is observable if the entire state vector x(t) can be determined from the system’s
output.
Mathematical Condition (Kalman’s Observability Test)
For an LTI system:
𝒚(𝒕) = 𝑪𝒙(𝒕) + 𝑫 𝒖(𝒕)
where:
• y(t) = output vector,
• C = output matrix,
• D = feedthrough matrix.
The system is observable if the observability matrix (Qo) has full rank (n).
𝑪
𝑪𝑨
𝑸𝒐 = 𝑪𝟐 𝑨
⋮
[𝑪 𝑨]
𝒏−𝟏

Find the controllability and observability of the system having transfer

b 𝒀(𝒔) 𝟐
function = (𝑺𝟑
𝑼(𝒔) +𝟔𝑺𝟐 +𝟏𝟏𝑺+𝟔)

3 a List out the advantages and disadvantages of state space techniques.

b Check whether given system is observable or controllable

 𝟎 𝟎 𝟏
𝑨 = [−𝟐 −𝟑 𝟎 ] and 𝑪 = [𝟏 𝟎 𝟎]
𝟎 𝟐 −𝟑

Obtain the state space model of the system for the given transfer function
4 𝟐
𝑮(𝒔) = (𝑺𝟐
+𝟓𝑺+𝟔)

Determine the state model of the system characterized by the differential

5
equation (s4 + 2s3 + 8s2 + 4s + 3) Y(s) = 10 U(s).

The state equations of a linear system are as follows

−𝟐 𝟎 𝟏 𝟏
𝒙̇ = [ 𝟏 −𝟑 𝟎] 𝒙 + [𝟎] 𝒖
6 𝟏 𝟏 𝟏 𝟏
𝒚 = [𝟐 𝟏 −𝟏]𝒙
𝒀(𝒔)
Determine the transfer function of the system
𝑼(𝒔)
The transfer function of the system from the state model is
𝒀(𝒔)
= 𝑪 (𝒔 𝑰 − 𝑨)−𝟏 𝑩 + 𝑫
𝑼(𝒔)
From the given state model
−2 0 1
𝐴 = [ 1 −3 0]
1 1 1
1
𝐵 = [0]
1

𝐶 = [2 1 −1]

𝐷= 0

The transfer function is

𝒀(𝒔)
= 𝑪 (𝒔 𝑰 − 𝑨)−𝟏 𝑩
𝑼(𝒔)

To find (sI-A)-1
𝑠 0 0 −2 0 1 𝑠+2 0 −1
(𝑠 𝐼 − 𝐴) = [0 𝑠 0] − [ 1 −3 0] = [ −1 𝑠 + 3 0 ]
0 0 𝑠 1 1 1 −1 −1 𝑠 − 1
𝑎𝑑𝑗(𝑠𝐼 − 𝐴)
(𝑠 𝐼 − 𝐴)−1 =
|(𝑠𝐼 − 𝐴)|
 |(𝒔𝑰 − 𝑨)| = (𝒔 + 𝟐)[(𝑠 + 3)(𝑠 − 1) − 0] − 𝟎[ ] + (−𝟏)[1 + 𝑠 + 3]

|(𝒔𝑰 − 𝑨)| = (𝑠 + 2)(𝑠 + 3)(𝑠 − 1) − (𝑠 + 4)

|(𝒔𝑰 − 𝑨)| = (𝑠 2 + 3𝑠 + 2𝑠 + 6)(𝑠 − 1) − (𝑠 + 4)

|(𝒔𝑰 − 𝑨)| = (𝑠 2 + 5𝑠 + 6)(𝑠 − 1) − (𝑠 + 4)

|(𝒔𝑰 − 𝑨)| = (𝑠 3 + 5𝑠 2 + 6𝑠 − 𝑠 2 − 5𝑠 − 6) − (𝑠 + 4)

|(𝒔𝑰 − 𝑨)| = (𝑠 3 + 4𝑠 2 + 𝑠 − 6) − (𝑠 + 4)

|(𝒔𝑰 − 𝑨)| = 𝑠 3 + 4𝑠 2 + 𝑠 − 6 − 𝑠 − 4

|(𝒔𝑰 − 𝑨)| = 𝑠 3 + 4𝑠 2 − 10

𝑠+2 0 −1
(𝑠 𝐼 − 𝐴) = [ −1 𝑠+3 0 ]
−1 −1 𝑠−1
𝐶11 𝐶12 𝐶13 𝑇
𝑇
𝑎𝑑𝑗(𝑠𝐼 − 𝐴) = (𝑐𝑜𝑓𝑎𝑐𝑡𝑜𝑟 𝑜𝑓 (𝑠𝐼 − 𝐴)) = [𝐶21 𝐶22 𝐶23 ]
𝐶31 𝐶32 𝐶33

𝑠+3 0
𝐶11 = + | | = (𝑠 + 3)(𝑠 − 1) = (𝑠 2 − 𝑠 + 3𝑠 − 3) = (𝑠 2 + 2𝑠 − 3)
−1 𝑠 − 1
−1 0
𝐶12 = − | | = −(−𝑠 + 1) = 𝑠 − 1_
−1 𝑠 − 1
−1 𝑠 + 3
𝐶13 = + | |= 1+𝑠+3=𝑠+4
−1 −1
0 −1
𝐶21 = − | | = −(0 − 1) = 1
−1 𝑠 − 1
𝑠 + 2 −1
𝐶22 = + | | = (𝑠 + 2)(𝑠 − 1) − 1 = (𝑠 2 − 𝑠 + 2𝑠 − 2) − 1 = 𝑠 2 + 𝑠 − 3
−1 𝑠 − 1
𝑠+2 0
𝐶23 = − | | = −(−𝑠 − 2) = 𝑠 + 2
−1 −1
0 −1
𝐶31 = + | |=𝑠+3
𝑠+3 0
𝑠 + 2 −1
𝐶32 = − | | = −(−1) = 1
−1 0
𝑠+2 0
𝐶33 = + | | = (𝑠 + 2)(𝑠 + 3) = 𝑠 2 + 3𝑠 + 2𝑠 + 6 = 𝑠 2 + 5𝑠 + 6
−1 𝑠 + 3
 𝑇
(𝑠2 + 2𝑠 − 3) 𝑠−1 𝑠+4
𝑎𝑑𝑗(𝑠𝐼 − 𝐴) = [ 1 𝑠2 + 𝑠 − 3 𝑠+2 ]
2
(𝑠 + 3) 1 𝑠 + 5𝑠 + 6

(𝑠2 + 2𝑠 − 3) 1 𝑠+3
𝑎𝑑𝑗(𝑠𝐼 − 𝐴) = [ 𝑠−1 𝑠2 + 𝑠 − 3 1 ]
2
(𝑠 + 4) 𝑠+2 𝑠 + 5𝑠 + 6

𝑎𝑑𝑗(𝑠𝐼 − 𝐴)
(𝑠 𝐼 − 𝐴)−1 =
|(𝑠𝐼 − 𝐴)|
(𝑠2 + 2𝑠 − 3) 1 𝑠+3
1
(𝑠 𝐼 − 𝐴 )−1 = [ 𝑠−1 2
𝑠 +𝑠−3 1 ]
𝑠3 + 4𝑠2 − 10 2
(𝑠 + 4) 𝑠+2 𝑠 + 5𝑠 + 6

The transfer function is

𝒀(𝒔)
= 𝑪 (𝒔 𝑰 − 𝑨)−𝟏 𝑩
(
𝑼 𝒔)

(𝑠2 + 2𝑠 − 3) 1 𝑠+3 1
−𝟏
1 2
(𝒔 𝑰 − 𝑨) 𝑩 = [ 𝑠−1 𝑠 +𝑠−3 1 ] [0]
𝑠3 + 4𝑠2 − 10 2
(𝑠 + 4) 𝑠+2 𝑠 + 5𝑠 + 6 1

1 (𝑠2 + 2𝑠 − 3) + 0 + 𝑠 + 3
(𝒔 𝑰 − 𝑨)−𝟏 𝑩 = [ 𝑠−1+0+1 ]
𝑠3 + 4𝑠2 − 10
𝑠 + 4 + 0 + 𝑠2 + 5𝑠 + 6

1 𝑠2 + 3𝑠
−𝟏
(𝒔 𝑰 − 𝑨) 𝑩 = [ 𝑠 ]
𝑠3 + 4𝑠2 − 10 2
𝑠 + 6𝑠 + 10

1 𝑠2 + 3𝑠
𝑪 (𝒔 𝑰 − 𝑨)−𝟏 𝑩 = [2 1 −1] [ 𝑠 ]
𝑠3 + 4𝑠2 − 10 2
𝑠 + 6𝑠 + 10

1 𝑠2 + 3𝑠
−𝟏 [2
𝑪 (𝒔 𝑰 − 𝑨) 𝑩 = 1 −1] [ 𝑠 ]
𝑠3 + 4𝑠2 − 10 2
𝑠 + 6𝑠 + 10

1
𝑪 (𝒔 𝑰 − 𝑨)−𝟏 𝑩 = [2𝑠2 + 6𝑠 + 𝑠−𝑠2 − 6𝑠 − 10]
𝑠3 + 4𝑠2 − 10

1
𝑪 (𝒔 𝑰 − 𝑨)−𝟏 𝑩 = 2
[𝑠2 + 𝑠 − 10]
𝑠3 + 4𝑠 − 10
 𝑠2 + 𝑠 − 10
𝑪 (𝒔 𝑰 − 𝑨)−𝟏 𝑩 =
𝑠3 + 4𝑠2 − 10

The transfer function is

𝒀(𝒔) 𝑠2 + 𝑠 − 10
𝑮(𝒔) = = 𝑪 (𝒔 𝑰 − 𝑨)−𝟏 𝑩 = 3
𝑼(𝒔) 𝑠 + 4𝑠2 − 10

MATLAB Code (not required for exam):

% Define the State-Space Matrices

A = [-2 0 1;1 -3 0;1 1 1];

B = [1; 0; 1];

C = [2 1 -1];

D = 0;

% Convert State-Space to Transfer Function

sys = ss(A, B, C, D); % Create state-space system

tf_sys = tf(sys); % Convert to transfer function

% Display the Transfer Function

disp('Transfer Function:');

tf_sys

The state equation of a system is given by

𝒙̇ −𝟑 𝟏 𝒙𝟏 𝟎
[ 𝟏] = [ ] [𝒙 ] + [ ] 𝒖(𝒕), 𝒕>𝟎
7 𝒙𝟐̇ −𝟐 −𝟏 𝟐 𝟏
i. Is the system controllable
ii. Compute the state transition matrix

8 Explain the design of Lead-Lag compensator

9 Derive the transfer function for the Lead compensator

10 Explain PID Controllers with necessary expressions

11
The state equation of the system is
𝒙̇ (𝒕) = 𝑨𝒙(𝒕) + 𝑩 𝒖(𝒕)
Taking Laplace transform
𝒔 𝑿(𝒔) = 𝑨𝑿(𝒔) + 𝑩 𝑼(𝒔)
𝒔 𝑿(𝒔) − 𝑨𝑿(𝒔) = 𝑩 𝑼(𝒔)
(𝒔 𝑰 − 𝑨)𝑿(𝒔) = 𝑩 𝑼(𝒔)
𝑿(𝒔) = (𝒔 𝑰 − 𝑨)−𝟏 𝑩 𝑼(𝒔) ---(1)

The output equation of the system is

𝒚(𝒕) = 𝑪𝒙(𝒕) + 𝑫 𝒖(𝒕)
Taking Laplace transform
𝒀(𝒔) = 𝑪 𝑿(𝒔) + 𝑫 𝑼(𝒔) -----(2)
Substitute (2) in (1)
𝒀(𝒔) = 𝑪 (𝒔 𝑰 − 𝑨)−𝟏 𝑩 𝑼(𝒔) + 𝑫 𝑼(𝒔)
𝒀(𝒔) = 𝑪 (𝒔 𝑰 − 𝑨)−𝟏 𝑩 + 𝑫𝑼(𝒔)
𝒀(𝒔)
= 𝑪 (𝒔 𝑰 − 𝑨)−𝟏 𝑩 + 𝑫
𝑼(𝒔)
The transfer function of the system is,

𝒀(𝒔)
𝑮(𝒔) = = 𝑪 (𝒔 𝑰 − 𝑨)−𝟏 𝑩 + 𝑫
𝑼(𝒔)