Control Engineering

Lecture 4

Muhammad Sajjad Sabir

1
 Approaches to solve control problems

Frequency Domain

rapidly provide immediately see

Advantages stability and the effects of
transient response varying system
information parameters until an
acceptable design
limited applied only to
Disadvantages applicability linear, time-
invariant systems

The state-space approach (also referred to as the modern, or time-domain, approach) is a

unified method for modeling, analyzing, and designing a wide range of systems

2
 Observations
1. We select a particular subset of all possible system variables and call the
variables in this subset state variables.

2. For an nth-order system, we write n simultaneous finite differential

equations in terms of the state variables. We call this system of simultaneous
differential equations state equations.

3. If we know the initial condition of all of the state variables at t0 as well as

the system input for t>t0, we can solve the simultaneous differential equations
for the state variables for t>t0.

4. We algebraically combine the state variables with the system's input and find
all of the other system variables for t>t0. We call this algebraic equation the
output equation.
5. We consider the state equations and the output equations a viable
representation of the system. We call this representation of the system a state-
space representation.
3
 state variables.
state equations.

FIGURE 3.7 T a la i al echa ical

initial conditione
FIGURE 3.7 T a la i al echa ical e
SOLUTION: output equation
SOLUTION:
First write the differentialstate-space
equations for therepresentation.
network in Figure 3
First write the differential
GURE 3.7 T a la i al echa icalmethods equations
e for the network in Figure 3.7, using the
of Chapter 2 to find the Laplace-transformed equation
methods of Chapter 2 to find the Laplace-transformed equations of motion. Next
UTION: State variables
take x1,take
the inverse: Laplacex the inverse Laplace transform of these equations, assuming
2 v1, v2 and
,
transform
conditions,
of these equations, assuming zero initial
obtain
conditions, and obtain
write the differential equations for the network in Figure 3.7, using the
ods of Chapter 2 to find the Laplace-transformed equations of motion. Next
state
the inverse equations.
Laplace transform of these equations, assuming zero initial (3.44)
itions, and obtain

(3.44)
(3.45)

(3.45)
initial
Nowcondition:
let Now let , and , and
, and then select 1, 1, ,2and
, then s
let and 2 ,as andNext
state variables.
and 2, as state
form
and variables.
thentwo of the
select Nextequations
state form twobyofsolving
the state
Eq.equations b
1, 1, 2,
(3.44)Next
2 as state variables. for form
d 1/d
twoand (3.44)
Eq.
of the for for
(3.45)
state d 1d/d
equations and
by2/d Eq.
Eq.(3.45)
. Finally,
solving addfor d 2/d . Finally,
and add
) for d output equation:
1/d and Eq. (3.45) forto
d complete
2/d . Finally,
theadd to complete
set of state and theHence,
equations. set of state equations. Hence,
to complete the set of state equations. Hence,
(3.46a)
4 (3.46a)
 State-space representation

x· = A x + B u state equations
y = C x + D u output equation

Equation
5 (3.18) is called the state equation, and the vector , the state vector,
 conditions, and obtain

(3.45)

, and , and then select 1, 1, 2,

s state variables. Next form two of the state equations by solving Eq.

x· = A x + B u
r d 1/d and Eq. (3.45) for d 2/d . Finally, addNow let and , and , and th
Now let , and , and
to complete the set of state equations. Hence,
and 2 as state variables. Next form two of the state equation
GURE 3.7 T a la i al echa ical e and 2 as state variables. Next form two of the state equat
(3.44) for d 1/d and Eq. (3.45) for d 2/d . Finally, add
UTION: (3.44) for d 1/dto and
to
Eq.
y
(3.45)
(3.46a)
complete
complete
write the differential equations for the network in Figure 3.7, using the the
=
the setfor
C x
d 2/d
of state
set of
+ D
equations.
state
u
. Finally, add
Hence,
equations. Hence
ods of Chapter 2 to find the Laplace-transformed equations of motion. Next
(3.46b)
the inverse Laplace transform of these equations, assuming zero initial
itions, and obtain

(3.46c)
(3.44)
(3.46d)
(3.46d)
In vector-matrix form, (3.45)

atrix form,
(3.47)
let , and , and then select 1, 1, 2,
2 as state variables. Next form two of the state equations by solving Eq. (3.47)
) for d 1/d and Eq. (3.45) for d 2/d . Finally, add and
to complete the set of state equations. Hence,

6 (3.46a)
where the dot indicates differentiation with respect to time. What is the output
 (3.6)1

The derivative of the current is

The General state-space representation (3.7)2

Linear
Thus, independence
knowing the state variable, i( ), and the input, ( ), we can find the value,
or a e, of any network variable at any time, . Hence, the algebraic
equations, Eqs. (3.5) through (3.7), are output equations.
State variables. The smallest set of linearly independent system va
System variable. Any variable that responds to an input or initial conditions
5. Since the variables of interest are completely describedthat
by Eqs. (3.1) and
the values (3.5)
of the members of the set at time 0 along with known
through (3.7), we say that the combined state equation (3.1) and the output
in a system. functions completely determine the value of all system variables for a
equations (3.5 through 3.7) form a viable representation of the network, which
we call a state-space representation. State vector. A vector whose elements are the state variables.
State variables.
Equation Thethesmallest
(3.1), which describes setnetwork,
dynamics of the of linearly
State space.independent system
The -dimensional space
is not unique. This variables
whose axes are the state varia
new term and is illustrated in Figure 3.3, where the state variables are
such that
equation the
could
substituting
be values of
written in termsthe
of anymembers
other network of
into Eq. (3.1) yields
a the
variable. set
For
resistor at time
example,
voltage, t capacitor
R, and a0
alongvoltage,
withCknown
. These variables form
forcing functions completely determinestate the value
space. of all
A trajectory system
can variables
be thought for
of as being mapped all
out by t
( ), for a range of . Also shown is the state vector at the particular ti
t>t0. (3.8)

State vector.
which can A vector
be solved knowing whose
that the elements are and knowing
initial condition
( ). In this case, the state variable is R( ). Similarly, all other network variables can
the state variables.
now be written in terms of the state variable, R( ), and the input, ( ). Let us now
extend our observations to a second-order system, State space.
such as that shown in Figure 3.2.

7
FIGURE 3.2 RLC e k
 Applying the state-space representation
A minimum number of state variables must be selected as components of the state vector. This
minimum number of state variables is sufficient to describe completely the state of the system

The components of the state vector (i.e., this minimum number of state variables) must be linearly
independent.

Minimum Number of State Variables

The minimum number required equals the order of the differential equation describing the system

The minimum number required equals the number of independent energy-storage elements in the system

8
Skill-A e men E e ci e 3.2
PROBLEM:
Represent the translational mechanical system shown in Figure 3.9 in state
space, where 3( ) is the output.

FIGURE 3.9 T a la i al echa ical e f Skill-A e e

E e ci e 3.2
ANSWER:

First step: write all motion equations

9
 FIGURE P3.2
3. Find a state-space representation for the system in Figure P3.3. Assume
the output is 1( ). [Section: 3.4]

FIGURE P3.3
4. Find a state-space representation for the system in Figure P3.4. Assume
the output is 2( ). [Section: 3.4]
10
 FIGURE P3.3
4. Find a state-space representation for the system in Figure P3.4. Assume
the output is 2( ). [Section: 3.4]

FIGURE
11 P3.4
10. Express each one of the systems in Figure P3.8 in state space phase-
variable form. [Section: 3.5]

Question is changed to simpler one, initially

C(s) 1
= 4
R(s) s + 8s 3 + s 2 + 7s + 10

s 4C(s) + 8s 3C(s) + s 2C(s) + 7sC(s) + 10C(s) = R(s)

FIGURE P3.8
Taking the inverse Laplace
11. Re ea P b e 10 MATLAB. [Sec : 3.5]

d 4c(t) d 3c(t) d 2c(t) dc(t) How many state variables?

Check + 8Answer! + +7 + 10c(t) = r(t)
dt 4 dt 3 dt 2 dt
Check book for further steps
12. Find a vector-matrix state-space representation for the transfer function.
[Section:
12 3.5]
10. Express each one of the systems in Figure P3.8 in state space phase-
variable form. [Section: 3.5]

Original question

C(s) 3s + 9 1
= 4 = * (3s + 9)
R(s) s + 8s 3 + s 2 + 7s + 10 s 4 + 8s 3 + s 2 + 7s + 10

R(s) C1(s) C(s)

FIGURE P3.8
11. Re ea P b e 10 MATLAB. [Sec : 3.5]

Check Answer!
Check book for further steps
12. Find a vector-matrix state-space representation for the transfer function.
[Section:
13 3.5]
Home Assignment (part of Mid Exam)

 Two forces in the system (state space representation)

 Two outputs in the system (state space representation)

14
 Two forces in the system (state space representation)

f1(t)

x· = A x + B u state equations
FIGURE P3.3
4. Find a state-space representation for the system in Figure P3.4. Assume
y=
the output is 2( ).C x+D
[Section: 3.4]u output equation
15
 Two outputs in the system (state space representation)

FIGURE P3.3
x· = A x + B u state equations
4. Find a state-space representation for the system in Figure P3.4. Assume
y = C x + D u output equation
the output is 2( ). [Section: 3.4]

16