Control Systems II Modeling in the

Time Domain
EEB 441/511

Dr Mandu A. Jeffrey
.....
Department of Electrical Engineering
Faculty of Engineering and Technology
University of Botswana
Modeling in the Time Domain Learning Outcomes
At the end of this section, you should be able to:
1) Find the state-space representation for a linear, time invariant
system
2) Model electrical and mechanical systems in state space
3) Convert a state-space representation to a transfer function
4) Convert a transfer function to state space
5) Linearize a nonlinear state-space representation

2
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain Frequency Domain Modeling
Two approaches are available for the analysis and design of
feedback control systems.

The first is known as the classical, or frequency-domain, technique.

 This approach is based on converting a system’s differential
equation to a transfer function - TF
 TF is a mathematical model of the system that algebraically
relates a representation of the output to a representation of the
input.
The second is known as the modern, or time-domain, technique.
 This is based on representing a system by a set of simultaneous,
first-order differential equations
3
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain Frequency Domain Modeling
Consider a general nth order, linear, time-invariant differential
equation:

where: c(t) is the output and r(t) is the input.

Taking the Laplace transform and assuming zero initial conditions

gives:

.. and the transfer

function:
4
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain Frequency Domain Modeling

 A major advantage of frequency-domain techniques is that they

readily provide stability and transient response information.
The effects of varying system parameters until an
acceptable design is met can be immediately seen.
 The primary disadvantage of this classical approach is it’s
limited applicability:
It can be applied only to linear, time-invariant systems or
systems that can be approximated as such.
5
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain Time Domain Modeling
 The state-space approach is also referred to as the modern or
time-domain approach.
 The state-space approach can be used to represent nonlinear
systems that have backlash, saturation, and dead zone.
 It can handle, conveniently, systems with nonzero initial
conditions.
 Time-varying systems, (for example, missiles with varying fuel
levels) can be represented in state space.
 Multiple-input, multiple-output (MIMO) systems can be
compactly represented in state space

6
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain Time Domain Modeling
TERMS
 System variable:
Any variable that responds to an input or initial conditions in a
system.
 State variables:
The smallest set of linearly independent system variables such
that the values of the members of the set at time t0 along with
known forcing functions completely determine the value of all
system variables for all t t0 .
 State vector:
A vector whose elements are the state variables.

7
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain Time Domain Modeling
TERMS
 State space:
The n-dimensional space whose
axes are the state variables.
 State equations:
A set of n simultaneous, first order
differential equations with n
variables, where the n variables to
be solved are the state variables.
 Output equation:
The algebraic equation that expresses the output variables of a
system as linear combinations of the state variables and the inputs.
8
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain General State Equations
The general mathematical description General form:
of the system is expressed as a set of
n coupled first-order ordinary
differential equations,
known as the state equations in which:
the time derivative of each state
variable is expressed in terms of the 
state variables
x1(t), . . . , xn(t)
and the system inputs functions f1(x,u,t), . . . ,
u1(t), . . . , ur(t). fn(x,u,t) can be
nonlinear or linear
9
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain Linear State Equations
The time derivative of each state variable is expressed as a linear
combination of the state variables x1(t), . . . , xn(t) and the system
inputs u1(t), . . . , ur(t).





10
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain Output Equations
 A system output is any system variable of interest.
 The output equation is expresses as a linear combination of the
state variables and the inputs
 For a system with m outputs:

11
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain State Space Representation
Multiple Input Multiple Output (MIMO)

State variables x1(t), . . . , xn(t)  nth order system

System inputs u1(t), . . . , ur(t)  r inputs
System outputs y1(t), . . . , ym(t)  m outputs

12
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain State Space Representation

Recap:
 State variables fully describe the system and its response to
any given set of inputs
 State variables are an internal description of the system which
completely characterize the system state at any time t, and
 Any output variables yi(t) may be computed from the state
variables.
13
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain State Space Representation
Linear Time Invariant (LTI) Linear Time Variant
systems: (or General) systems:
𝐱=𝐀 𝒕 𝐱+𝐁 𝒕 𝐮
𝐲=𝐂 𝒕 𝐱+𝐃 𝒕 𝐮
..
x = state vector
𝐱 = derivative of the state vector (w.r.t)
y = output vector
u = input or control vector
A = system matrix
.. B = input matrix
C = output matrix
D = feedforward matrix
14
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain State Space Representation

General Linear System

LTI system with D=0

15
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain State Space Representation
System with initial
conditions x (0)

Discuss:
 .
 .
 .
 .
 .
 .

16
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain State Space Representation
LTI 2nd Order SISO System

Notes:
 .
 .
 .
 .
 .
 .
17
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain Example
(Example 3.1 in Textbook)
Find a state-space representation for the electrical network if the
output is the current through the resistor.

Step 1 Label all of the branch currents in the network. These

include iL, iR and iC, as shown in the figure.
?? Minimum number of state variables??
18
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain RECAP – State Variables
STATE VARIABLES
1. … set of linearly independent system variables
2. … the values of the members of the set at time t0 along with
known forcing functions completely determine the value of all
system variables for all t t0 - ie. must be sufficient to
completely describe the system.

MINIMUM NUMBER OF STATE VARIABLES

1. Typically equals the order of the differential equation describing
the system
2. Can be the number of independent energy-storage elements in
the system.
19
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain RECAP – Electrical Components

20
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain Example
Step 2: Select the state variables by writing the derivative equation
for all energy-storage elements, that is, the inductor and the
capacitor.

Thus,

Choose the state variables as the quantities that are differentiated,

namely vC and iL.
The state-space representation would be complete if the right-hand
sides of the equations can be written as linear combinations of the
state variables and the input.
21
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain Example
Since iC and vL are not state variables, the next step is to express
iC and vL as linear combinations of the state variables, vC and iL,
and the input, v(t).

Step 3 Apply network theory, such as Kirchhoff’s voltage and

current laws, to obtain iC and vL in terms of the state variables, vC
and iL.

and

22
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain Example
Step 4: Substitute the results into equations in Step 2, and
simplify to obtain the following state equations:

Step 5: Find the output (iR ) equation as a linear combination

of the state variables vC and iL, and the input, v(t ):

23
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain Example
The final result for the state-space representation in vector-matrix
form is as follows:

The chosen state variables:

 Do they completely define the system?
 Are they unique?
24
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain Example
Past Exam question
Find a state-space representation for the electrical network if
the output is the voltage across the resistor R2

25
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain Example
Solution:
Find a state-space representation for the electrical network
if the output is the voltage across the resistor R2
3 state variables minimum - Choose iL , vC1 and vC2
𝑑𝑖𝐿 𝑑𝑣𝑐1 𝑑𝑣𝑐2
 i) 𝐿 = 𝑣𝐿 ii) 𝐶1 = 𝑖𝑐1 iii) 𝐶2 = 𝑖𝑐2
𝑑𝑡 𝑑𝑡 𝑑𝑡
Express 𝑣𝐿 𝑖𝑐1 𝑖𝑐2 as functions of the state variables and the input
 𝑣𝐿 = 𝑣𝑐1 − 𝑣𝑐2
𝑣𝑖 −𝑣𝑐1
 𝑖𝑐1 = 𝑖𝑅1 − 𝑖𝐿 = − 𝑖𝐿
𝑅1
𝑣𝑐2
 𝑖𝑐2 = 𝑖𝐿 −𝑖𝑅2 = 𝑖𝐿 −
𝑅2
Output: 𝑣𝑜 = 𝑣𝑅2 = 𝑣𝑐2
26
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain Example
𝑑𝑖𝐿
Solution:  𝐿 𝑑𝑡
= 𝑣𝑐1 − 𝑣𝑐2
𝑑𝑣𝑐1 𝑣𝑖 −𝑣𝑐1 𝑣𝑖 𝑣𝑐1
 𝐶1 = − 𝑖𝐿 = − − 𝑖𝐿
𝑑𝑡 𝑅1 𝑅1 𝑅1
𝑑𝑣𝑐2 𝑣𝑐2
 𝐶2 = 𝑖𝐿 −
𝑑𝑡 𝑅2
 𝑣𝑜 = 𝑣𝑐2


0 1 −1 𝐿
𝑖𝐿 𝐿 0
𝐱 = 𝑣𝑐1 𝐀 = −1 𝐶1 −1 𝑅1 𝐶1 0 𝐁= 1
𝑅1 𝐶1
𝑣𝑐2 1
𝐶2 0 − 1 𝑅2 𝐶2 0

𝐂= 0 0 1 𝐃= 0
27
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain Application – Drug Distribution
 A simple model divides the process into compartments: dosage,
absorption site, blood, peripheral compartment, and urine.
 The rate of change of the
amount of a drug in a
compartment is equal to
the input flow rate
diminished by the output
flow rate.
 Each xi represents the
amount of drug in that
particular compartment
(Lordi, 1972).
28
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain Application – Drug Distribution
Net flow rates into compartments

 X

 X

 X

 X

 X

 B = ??

29
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain Application – Aquifers
Underground Water
Modeling --
Water Conservation

Study and Model the

system in state space
(Kandel)

The aquifer is represented by three tanks

 Each hn is the water level (referred to as the head) in each tank
 Each qn is the natural water flow to the sea and is proportional to the
difference in head between two adjoining tanks:
qn = Gn (hn – hn-1 )
30
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain Application – Aquifers
 Each qon is the outflow
from the tanks for
irrigation, home, etc
 Each qin is the inflow
into the tanks from
rain, wells, etc
 Flow q21 is created for (Kandel)
conservation to
prevent loss to the sea
 The net flow into a tank is proportional to the rate of change of head
in each tank. Cndhn /dt = qin - qon + qn+1 - qn + q(n+1)n - qn(n-1)
State Space Model
The state variables and the outputs are the heads (hn) of each tank.
31
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain x
X
1) x
2) x
3) x
4) x
5) x

32
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain x
X
1) x
2) x
3) x
4) x
5) x

33
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain Application – Biogas Process
..
.

[?]-(?)
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015
by John Wiley & Sons. All rights reserved. 34
Modeling in the Time Domain Application – Biogas Process
Fermentation process stages

[10]-(Gerber & Span)

35
Modeling in the Time Domain Bio-digester : Design and Build
BEng final year student project

36
[16]-(Kwelagobe, et al)
Modeling in the Time Domain Biogas Model
5th order model

37
[30]-(Simeonov et al)
Modeling in the Time Domain Biogas Model
..
.

38
[30]-(Simeonov et al)
Modeling in the Time Domain Biogas Model – Introduction to LAB 1
MATLAB Simulation Simulate the BATCH process for the
nonlinear model
Simulink:  Study and characterize systems
response.
IF YOU WERE TO CONTROL THIS PROCESS,
WHAT WOULD BE YOUR OBJECTIVE(S)?
Simulate the CONTINUOUS process for the
nonlinear model: ie. vary D and Soi
(start with the initial So and varying D)
 Study and characterize systems
response.
Keep in mind….  Explain the effects of varying D and Soi
… individually and collectively

FORMULATE A CONTROL PROBLEM FOR THIS

PROCESS
39
--
Modeling in the Time Domain Canonical Forms - Converting
𝑠+3
.. 𝐺(𝑠) = 2
𝑠 + 10𝑠 + 24

40
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain Canonical Forms - Converting
-

Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015

by John Wiley & Sons. All rights reserved. 41
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain Converting – SS to TF
..
Given the state and output equations

take the Laplace transform assuming zero initial conditions:

Solving for X(s) in the state equation yields:

where I is the identity matrix.
or

42
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain Converting
..
– SS to TF
Substituting into the output equation
yields:

If U(s) and Y(s) are scalars, then the transfer function is,

⁃ ⁃ A solution y(t) can be obtained from the inverse Laplace

transform of the output Y(s)

43
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain Example – SS to TF
-
-

44
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain Example – SS to TF
Example
X

45
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain Example – SS to TF
Example

** Converting from State Space (ss) to Transfer Function (tf),

and vice-versa, is very simple in Matlab … very useful tool.

46
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain Example – TF to SS
Starting point … Transfer Function to State Space
Example 3.4

(a) Transfer function;

(b) Equivalent block diagram showing
phase-variables.
Note: y(t) = c(t)

47
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain Example – TF to SS
Next …
Example 3.5

(a) Transfer function;

(b) Decomposed transfer function;
(c) Equivalent block diagram
showing phase-variables.
Note: y(t) = c(t)

48
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain Converting – TF to SS
Self Study
Continue this section as a self study topic

Refer to Textbook – Nise.

 CH3.5: Converting a Transfer Function to State Space

… Pay attention to the canonical forms

49
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.
Modeling in the Time Domain Next Topic
Solution of State Equations

.. Nise: Ch4 – Time Response

 4.10 Laplace transform solution
 4.11 Time domain solution

50
Control Systems Engineering, 7th Edition by Norman S. Nise. Copyright © 2015. All rights reserved.