Material I - EEE 403 ACT Part 2 PDF

EEE 403 ADVANCED CONTROL THEORY
Design of Control Systems in State Space
Dr. P. Kavitha
Associate Professor
SELECT, VIT University
kavitha.p@vit.ac.in
POLE PLACEMENT
2 EEE 403 Design of Control Systems in State Space

Lecture Outline
 Introduction
 Pole Placement
Pole Placement Design Techniques

 Using Transformation Matrix P
 Direct Substitution Method
 Ackermann’s Formula
 Design of State Observer

Introduction
 One of the drawbacks of frequency domain methods of
design is that after designing the location of the dominant
second-order pair of poles, we keep our fingers crossed,
hoping that the higher-order poles do not affect the second-
order approximation.
 What we would like to be able to do is specify all closed-

loop poles of the higher-order system.

Introduction…
 Frequency domain methods of design do not allow us to

specify all poles in systems of order higher than 2 because
they do not allow for a sufficient number of unknown
parameters to place all of the closed-loop poles uniquely.
 One gain to adjust, or compensator pole and zero to select,

does not yield a sufficient number of parameters to place all
the closed-loop poles at desired locations.

Introduction…
 Remember, to place n unknown quantities, you need n

adjustable parameters.
 State-space methods solve this problem by introducing into the

system
 Other adjustable parameters and
 The technique for finding these parameter values
 On the other hand, state-space methods do not allow the

specification of closed-loop zero locations, which frequency
domain methods do allow through placement of the lead
compensator zero.
Introduction…
 Finally, there is a wide range of computational support for state-

space methods; many software packages support the matrix
algebra required by the design process.
 However, as mentioned before, the advantages of computer

support are balanced by the loss of graphic insight into a design
problem that the frequency domain methods yield.

Pole Placement
 In this lecture we will discuss a design method commonly

called the pole-placement or pole-assignment technique.
 We assume that all state variables are measurable and are

available for feedback.
 If the system considered is completely state controllable, then

poles of the closed-loop system may be placed at any desired
locations by means of state feedback through an appropriate
state feedback gain matrix.

Pole Placement …
 The present design technique begins with a determination of

the desired closed-loop poles based on the transient-response
and/or frequency-response requirements, such as speed,
damping ratio, or bandwidth, as well as steady-state
requirements.
 By choosing an appropriate gain matrix for state feedback, it is

possible to force the system to have closed-loop poles at the
desired locations, provided that the original system is
completely state controllable.

Pole Placement …
Consider a plant represented in state space by

𝒙 = 𝑨𝒙 + 𝑩𝑢
𝑦 = 𝑪𝒙

Pole Placement …
 In a typical feedback control system, the output, y, is fed back to

the summing junction.
 It is now that the topology of the design changes. Instead of

feeding back y, we feed back all of the state variables.
 If each state variable is fed back to the control, u, through a gain,

ki, there would be n gains, ki, that could be adjusted to yield the
required closed-loop pole values.

Pole Placement…

Pole Placement …
𝒙 = (𝑨 − 𝑩𝑲)𝒙
 The stability and transient response characteristics are

determined by the eigenvalues of matrix A-BK.
 If matrix K is chosen properly Eigenvalues of the system can be

placed at desired location.
 And the problem of placing the regulator poles (closed-loop

poles) at the desired location is called a pole-placement problem.

Pole Placement …
• There are three approaches that can be used to determine the

gain matrix K to place the poles at desired location.

 Ackermann’s formula
• All those method yields the same result.

Pole Placement
Method 1 Transformation Matrix

Pole Placement (Using Transformation Matrix P)
• Following are the steps to be followed in this particular

method.
1. Check the state controllability of the system
𝐶𝑀 = 𝐵 𝐴𝐵 𝐴2 𝐵 ⋯ 𝐴𝑛−1 𝐵

• Following are the steps to be followed in this particular method.
2. Transform the given system in CCF.
P  CM  W
 an 1 an  2  a1 1
a an  3  1 0
 n2
W       𝑠𝐼 − 𝐴 = 𝑠 𝑛 + 𝑎1 𝑠 𝑛−1 + 𝑎2 𝑠 𝑛−2 + ⋯ + 𝑎𝑛−1 s+𝑎𝑛
 
 a1 1  0 0
 1 0  0 0
A  P 1 AP B  P 1 B C  CP
3. Obtain the desired characteristic equation from desired

Eigenvalues.
 If the desired Eigenvalues are 𝜇1 , 𝜇2 , ⋯ , 𝜇𝑛
(𝑠 − 𝜇1 )( 𝑠 − 𝜇2 ) ⋯ 𝑠 − 𝜇𝑛 = 𝑠 𝑛 + 𝛼1 𝑠 𝑛−1 + 𝛼2 𝑠 𝑛−2 + ⋯ + 𝛼𝑛−1 𝑠+𝛼𝑛

4. Compute the gain matrix K.
𝑲 = 𝛼𝑛 − 𝑎𝑛 𝛼𝑛−1 − 𝑎𝑛−1 ⋯ 𝛼2 − 𝑎2 𝛼1 − 𝑎1

• Example-1: Consider the regulator system shown in following
figure. The plant is given by
𝑥1 0 1 0 𝑥1 0
𝑥2 = 0 0 1 𝑥2 + 0 𝑢(𝑡)
𝑥3 −1 −5 −6 𝑥3 1
• The system uses the state feedback control u=-Kx. The desired
eigenvalues are 𝜇1 = −2 + 𝑗4, 𝜇2 = −2 − 𝑗4 ,𝜇3 = −10. Determine
the state feedback gain matrix K.
• Example-1: Step-1
𝑥1 0 1 0 𝑥1 0
𝑥2 = 0 0 1 𝑥2 + 0 𝑢(𝑡)
𝑥3 −1 −5 −6 𝑥3 1
 First, we need to check the controllability matrix of the system.
Since the controllability matrix CM is given by
0 0 1
𝐶𝑀 = 𝐵 𝐴𝐵 𝐴2 𝐵 = 0 1 −6
1 −6 31
 We find that rank(CM)=3. Thus, the system is completely state
controllable and arbitrary pole placement is possible.

• Example-1: Step-2 (Transformation to CCF)
𝑥1 0 1 0 𝑥1 0
𝑥2 = 0 0 1 𝑥2 + 0 𝑢(𝑡)
𝑥3 −1 −5 −6 𝑥3 1
 The given system is already in CCF

𝑥1 0 1 0 𝑥1 0
𝑥2 = 0 0 1 𝑥2 + 0 𝑢(𝑡)
𝑥3 −1 −5 −6 𝑥3 1
 Determine the characteristic equation
𝑠𝐼 − 𝐴 = 𝑠 3 + 6𝑠 2 + 5𝑠 + 1 = 0
𝑠𝐼 − 𝐴 = 𝑠 3 + 𝑎1 𝑠 2 + 𝑎2 𝑠 + 𝑎3
 Hence
𝑎1 = 6, 𝑎2 = 5, 𝑎3 = 1

Example-1: Step-4
 The desired characteristics polynomial can be computed using
desired eigenvalues
𝜇1 = −2 + 𝑗4 𝜇2 = −2 − 𝑗4 𝜇3 = −1
(𝑠 − 𝜇1 )( 𝑠 − 𝜇2 ) ⋯ 𝑠 − 𝜇𝑛 = (𝑠 + 2 − 4𝑗)( 𝑠 + 2 + 4𝑗) 𝑠 + 10
𝑠 + 2 − 4𝑗 𝑠 + 2 + 4𝑗 𝑠 + 10 = 𝑠 3 + 14𝑠 2 + 60𝑠 + 200

= 𝑠 3 + 𝛼1 𝑠 2 + 𝛼2 𝑠 + 𝛼3
 Hence
𝛼1 = 14, 𝛼2 = 60, 𝛼3 = 200

 State feedback gain matric K is then calculated as
𝑎1 = 6, 𝑎2 = 5, 𝑎3 = 1
𝛼1 = 14, 𝛼2 = 60, 𝛼3 = 200
𝑲 = 𝛼3 − 𝑎 3 𝛼2 − 𝑎2 𝛼1 − 𝑎1
𝑲 = 199 55 8

𝑥1 0 1 0 𝑥1 0
𝑥2 = 0 0 1 𝑥2 + 0 𝑢(𝑡)
𝑥3 −1 −5 −6 𝑥3 1
• State diagram of the given system
+ 𝑥3 𝑥3 𝑥2 𝑥2 𝑥1 𝑥1
𝑢(𝑡)
+
∫ ∫ ∫
+
-6
+
+
-5
+
+
-5
𝑢 = −𝑲𝑥 𝑲 = 199 55 8 𝑥1
𝑢 = − 199 55 8 𝑥2
𝑥3
199
+
+
55
+
+
8
+
𝑢(𝑡) + 𝑥3 𝑥3 𝑥2 𝑥2 𝑥1 𝑥1
-1
+
∫ ∫ ∫
+
-6
+
+
-5
+
27 EEE 403 Design of+
Control Systems in State Space
-5
Pole Placement
Method 2 Direct Substitution

Pole Placement (Direct Substitution Method)

Pole Placement (Direct Substitution Method)
2. Define the state feedback gain matrix as
𝑲 = 𝑘1 𝑘2 𝑘3 ⋯ 𝑘𝑛
 And equating 𝑠𝐼 − 𝐴 + 𝐵𝐾 with desired characteristic
equation.
(𝑠 − 𝜇1 )( 𝑠 − 𝜇2 ) ⋯ 𝑠 − 𝜇𝑛 = 𝑠 𝑛 + 𝛼1 𝑠 𝑛−1 + 𝛼2 𝑠 𝑛−2 + ⋯ + 𝛼𝑛−1 s+𝛼𝑛

Pole Placement (By Direct Substitution Method )
𝑥1 0 1 0 𝑥1 0
𝑥2 = 0 0 1 𝑥2 + 0 𝑢(𝑡)
𝑥3 −1 −5 −6 𝑥3 1
eigenvalues are 𝜇1 = −2 + 𝑗4 , 𝜇2 = −2 − 𝑗4 , 𝜇3 = −1 0.
Determine the state feedback gain matrix K.
Pole Placement (By Direct Substitution method)
𝑥1 0 1 0 𝑥1 0
𝑥2 = 0 0 1 𝑥2 + 0 𝑢(𝑡)
𝑥3 −1 −5 −6 𝑥3 1
0 0 1
𝐶𝑀 = 𝐵 𝐴𝐵 𝐴2 𝐵 = 0 1 −6
1 −6 31

Pole Placement (Using Direct Substitution method)
• Let K be 𝑲 = 𝑘1 𝑘2 𝑘3
𝑠 0 0 0 1 0 0
𝑠𝐼 − 𝐴 + 𝐵𝐾 = 0 𝑠 0 − 0 0 1 + 0 𝑘1 𝑘2 𝑘3
0 0 𝑠 −1 −5 −6 1
= 𝑠 3 + 6 + 𝑘3 𝑠 2 + 5 + 𝑘2 𝑠 + 1 + 𝑘1
 Desired characteristic polynomial is obtained as
𝑠 + 2 − 4𝑗 𝑠 + 2 + 4𝑗 𝑠 + 10 = 𝑠 3 + 14𝑠 2 + 60𝑠 + 200
 Comparing the coefficients of powers of s
14 = 6 + 𝑘3 𝑘3 = 8
60 = 5 + 𝑘2 𝑘2 = 55
200 = 1 + 𝑘1 𝑘1 = 199
Pole Placement
Method 3 Ackermann’s Formula

Pole Placement (Ackermann’s Formula)

2. Use Ackermann’s formula to calculate K
−1
𝐾 = 0 0 ⋯0 1 𝐵 𝐴𝐵 𝐴2 𝐵 ⋯ 𝐴𝑛−1 𝐵 ∅(𝐴)
∅ 𝐴 = 𝐴𝑛 + 𝛼1 𝐴𝑛−1 + ⋯ + 𝛼𝑛−1 𝐴 + 𝛼𝑛 𝐼

Pole Placement (By Ackermann’s Method )
𝑥1 0 1 0 𝑥1 0
𝑥2 = 0 0 1 𝑥2 + 0 𝑢(𝑡)
𝑥3 −1 −5 −6 𝑥3 1
eigenvalues are 𝜇1 = −2 + 𝑗4 , 𝜇2 = −2 − 𝑗4 , 𝜇3 = −1 0.
Determine the state feedback gain matrix K.
𝑥1 0 1 0 𝑥1 0
𝑥2 = 0 0 1 𝑥2 + 0 𝑢(𝑡)
𝑥3 −1 −5 −6 𝑥3 1
0 0 1
𝐶𝑀 = 𝐵 𝐴𝐵 𝐴2 𝐵 = 0 1 −6
1 −6 31

2. Use Ackermann’s formula to calculate K
−1
𝐾= 0 0 1 𝐵 𝐴𝐵 𝐴2 ∅(𝐴)
∅ 𝐴 = 𝐴3 + 𝛼1 𝐴2 + 𝛼2 𝐴 + 𝛼3 𝐼
• 𝛼𝑖 are the coefficients of the desired characteristic polynomial.
𝑠 + 2 − 4𝑗 𝑠 + 2 + 4𝑗 𝑠 + 10 = 𝑠 3 + 14𝑠 2 + 60𝑠 + 200
𝛼1 = 14, 𝛼2 = 60, 𝛼3 = 200

𝑥1 0 1 0 𝑥1 0
𝑥2 = 0 0 1 𝑥2 + 0 𝑢(𝑡)
𝑥3 −1 −5 −6 𝑥3 1
∅ 𝐴 = 𝐴3 + 14𝐴2 + 60𝐴 + 200𝐼

3 2
0 1 0 0 1 0 0 1 0 1 0 0
∅ 𝐴 = 0 0 1 + 14 0 0 1 + 60 0 0 1 + 200 0 1 0
−1 −5 −6 −1 −5 −6 −1 −5 −6 0 0 1
199 55 8
∅ 𝐴 = −8 159 7
−7 −34 117

0 0 1 199 55 8
𝐵 𝐴𝐵 𝐴2 𝐵 = 0 1 −6 ∅ 𝐴 = −8 159 7
1 −6 31 −7 −34 117
−1
𝐾= 0 0 1 𝐵 𝐴𝐵 𝐴2 ∅(𝐴)
−1
0 0 1 199 55 8
𝐾= 0 0 1 0 1 −6 −8 159 7
1 −6 31 −7 −34 117
𝐾 = 199 55 8

Pole Placement

𝑥1 1 2 1 𝑥1 1
𝑥2 = 0 1 3 𝑥2 + 0 𝑢(𝑡)
𝑥3 1 1 1 𝑥3 1
• Determine the state feedback gain for each state variable to place the
poles at -1+j, -1-j,-3. (Apply all methods)
Design of Control Systems in Sate Space
Observer Based Approach

Lecture Outline
 Introduction
 State Observer
 Topology of Pole Placement (Observer based)
 Full Order State Observer
 Reduced Order state Observer

Introduction
 In the pole-placement approach to the design of control systems,
we assumed that all state variables are available for feedback.
 In practice, however, not all state variables are available for

feedback.
 Then we need to estimate unavailable state variables.

Introduction…
 Estimation of unmeasurable state variables is commonly called observation.
 A device (or a computer program) that estimates or observes the state variables is
called a state estimator, state observer, or simply an observer.
 There are two types of state observers

 Full Order State Observer
 If the state observer observes all state variables of the system, regardless of
whether some state variables are available for direct measurement, it is
called a full-order state observer.
 Reduced Order State Observer

 If the state observer observes only those state variables which are not
available for direct measurement, it is called a reduced-order state observer.

State Feedback Control with Observer Based Approach
State feedback with state observer

State Feedback Control with Observer Based Approach
 State feedback Control

 State Feedback with observer

State Observer
 A state observer estimates the state variables based on the

measurements of the output and control variables.
 Here the concept of observability plays an important role.
 State observers can be designed if and only if the observability

condition is satisfied.

 Consider the plant defined by
𝑦 = 𝑪𝒙
 The mathematical model of the observer is basically the same as
that of the plant, except that we include an additional term that
includes the estimation error to compensate for inaccuracies in
matrices A and B and the lack of the initial error.
 The estimation error or observation error is the difference between

the measured output and the estimated output.
 The initial error is the difference between the initial state and the
initial estimated state.
 Thus we define the mathematical model of observer to be
𝒙 = 𝑨𝒙 + 𝑩𝑢 + 𝑲𝒆 (𝑦 − 𝑪𝒙)
 Where 𝒙 is estimated state vector, 𝑪𝒙 is estimated output and
𝑲𝒆 is observer gain matrix.

Full Order State Observer
 The order of the state observer that will be discussed here is the same as that of the
plant.
 Consider the plant define by following equations
𝒙 = 𝑨𝒙 + 𝑩𝑢 (1)
𝑦 = 𝑪𝒙
 Equation of state observer is given as
𝒙 = 𝑨𝒙 + 𝑩𝑢 + 𝑲𝑒 (𝑦 − 𝑪𝒙) (2)
 To obtain the observer error equation, let us subtract Equation (2) from Equation
(1):
𝒙 − 𝒙 = 𝑨𝒙 + 𝑩𝑢 − [𝑨𝒙 + 𝑩𝑢 + 𝑲𝑒 𝑦 − 𝑪𝒙 ]
𝒙 − 𝒙 = 𝑨𝒙 + 𝑩𝑢 − 𝑨𝒙 − 𝑩𝑢 − 𝑲𝑒 𝑪𝒙 − 𝑪𝒙
𝒙 − 𝒙 = 𝑨𝒙 + 𝑩𝑢 − 𝑨𝒙 − 𝑩𝑢 − 𝑲𝑒 𝑪𝒙 − 𝑪𝒙
 Simplifications in above equation yields
𝒙 − 𝒙 = 𝑨(𝒙 − 𝒙) − 𝑲𝑒 𝑪 𝒙 − 𝒙 (3)
 Define the difference between 𝒙 and 𝒙 as the error vector

e.
𝒆=𝒙−𝒙
 Equation (3) can now be written as
𝒆 = 𝑨𝒆 − 𝑲𝑒 𝑪𝒆
𝒆 = (𝑨 − 𝑲𝑒 𝑪)𝒆

Full Order State Observer…
𝒆 = (𝑨 − 𝑲𝒆 𝑪)𝒆
 From above we see that the dynamic behavior of the error vector is
determined by the eigenvalues of matrix A-KeC.
 If matrix A-KeC is a stable matrix, the error vector will converge to

zero for any initial error vector e(0).
 That is, 𝒙(𝒕) will converge to 𝒙(𝒕) regardless of the values of x(0).
 And if the eigenvalues of matrix A-KeC are chosen in such a way
that the dynamic behavior of the error vector is asymptotically stable
and is adequately fast, then any error vector will tend to zero (the
origin) with an adequate speed.

𝒆 = (𝑨 − 𝑲𝒆 𝑪)𝒆
 If the plant is completely observable, then it can be proved that

it is possible to choose matrix Ke such that A-KeC has
arbitrarily desired eigenvalues.
 That is, the observer gain matrix Ke can be determined to yield

the desired matrix A-KeC.

Duality Property
 The design of the full-order observer becomes that of

determining an appropriate Ke such that A-KeC has desired
eigenvalues.
 Thus, the problem here becomes the same as the pole-

placement problem.
 In fact, the two problems are mathematically the same.
 This property is called duality.

Duality Property
 Consider the system defined by
𝑦 = 𝑪𝒙
 In designing the full-order state observer, we may solve the dual
problem, that is, solve the pole-placement problem for the dual
system.
𝒛 = 𝑨∗ 𝒛 + 𝑪∗ 𝑣
𝑛 = 𝑩∗ 𝒛
 Assuming the control signal 𝑣 to be
𝑣 = −𝑲𝒛
Duality Property
 If the dual system is completely state controllable, then the state
feedback gain matrix K can be determined such that matrix A*-
C*K will yield a set of the desired eigenvalues.
 If 𝜇1 , 𝜇2 ,…, 𝜇𝑛 , are the desired eigenvalues of the state observer

matrix, then by taking the same 𝜇𝑖′ 𝑠 as the desired eigenvalues of the
state-feedback gain matrix of the dual system, we obtain
 Noting that the eigenvalues of A*-C*K and those of A-K*C are the
same, we have
𝑠𝐼 − (𝑨∗ − 𝑪∗ 𝑲 ) = (𝑠 − 𝜇1 )( 𝑠 − 𝜇2 ) ⋯ 𝑠 − 𝜇𝑛
𝑠𝐼 − (𝑨∗ − 𝑪∗ 𝑲 ) = 𝑠𝐼 − (𝑨 − 𝑲∗ 𝑪 )

Duality Property
𝑠𝐼 − (𝑨∗ − 𝑪∗ 𝑲 ) = 𝑠𝐼 − (𝑨 − 𝑲∗ 𝑪 )
 Comparing the characteristic polynomial 𝑠𝐼 − (𝑨 − 𝑲∗ 𝑪 ) and the

characteristic polynomial for the observer system
𝑠𝐼 − (𝑨 − 𝑲𝑒 𝑪 ) , we find that Ke and K* are related by
𝑠𝐼 − (𝑨 − 𝑲∗ 𝑪 ) = 𝑠𝐼 − (𝑨 − 𝑲𝑒 𝑪 )
𝑲∗ = 𝑲𝑒
 Thus, using the matrix K determined by the pole-placement

approach in the dual system, the observer gain matrix Ke for the
original system can be determined by using the relationship Ke=K*.

Observer Gain Matrix
 Using Transformation Matrix Q

 Using Transformation Matrix Q
𝑲 = 𝛼𝑛 − 𝑎𝑛 𝛼𝑛−1 − 𝑎𝑛−1 ⋯ 𝛼2 − 𝑎 2 𝛼1 − 𝑎1
 Since
𝑲𝑒 = 𝑲∗
𝛼𝑛 − 𝑎𝑛
𝛼𝑛−1 − 𝑎𝑛−1
𝑲𝑒 = 𝑲∗ = ⋮
𝛼2 − 𝑎2
𝛼1 − 𝑎1
𝑘1
𝑘2
𝑲𝒆 = ⋮
𝑘3
𝑘𝑛

−1 ∅(𝐴)
𝐾 = 0 0 ⋯0 1 𝐵 𝐴𝐵 𝐴2 𝐵 ⋯ 𝐴𝑛−1 𝐵
 For the dual system

𝒛 = 𝑨∗ 𝒛 + 𝑪∗ 𝑣
𝑛 = 𝑩∗ 𝒛
𝑲 = 0 0 ⋯ 0 1 𝐶∗ −1 ∅(𝐴∗ )
𝐴∗ 𝐶 ∗ (𝐴∗ )2 𝐶 ∗ ⋯ (𝐴∗ )𝑛−1 𝐶 ∗
 Since 𝑲𝑒 = 𝑲∗
𝑲𝑒 = 𝑲∗ = 0 ∗
0 ⋯0 1 𝐶∗ 𝐴∗ 𝐶 ∗ (𝐴∗ )2 𝐶 ∗ ⋯ (𝐴∗ )𝑛−1 𝐶 ∗ −1
∅(𝐴∗ )

𝑲𝑒 = 𝑲∗ = 0 ∗
0 ⋯0 1 𝐶∗ 𝐴∗ 𝐶 ∗ (𝐴∗ )2 𝐶 ∗ ⋯ (𝐴∗ )𝑛−1 𝐶 ∗ −1
∅(𝐴∗ )
 Simplifying it further
𝐾𝑒 = ∅(𝐴∗ )∗ 𝐶 ∗ 𝐴∗ 𝐶 ∗ (𝐴∗ )2 𝐶 ∗ ⋯ (𝐴∗ )𝑛−1 𝐶 ∗ −1 ∗

0 0 ⋯0 1 ∗
−1
𝐶 0
𝐶𝐴 0
𝐾𝑒 = ∅(𝐴) ⋮ ⋮
𝐶𝐴𝑛−2 0
𝐶𝐴𝑛−1 1

 The feedback signal through the observer gain matrix Ke serves as a

correction signal to the plant model to account for the unknowns in
the plant.
 If significant unknowns are involved, the feedback signal through the

matrix Ke should be relatively large.
 However, if the output signal is contaminated significantly by

disturbances and measurement noises, then the output y is not reliable
and the feedback signal through the matrix Ke should be relatively
small.

 The observer gain matrix Ke depends on the desired characteristic

equation
(𝑠 − 𝛽1 )( 𝑠 − 𝛽2 ) ⋯ 𝑠 − 𝛽𝑛 = 0
 The observer poles must be two to five times faster than the controller
poles to make sure the observation error (estimation error) converges
to zero quickly.
 This means that the observer estimation error decays two to five times
faster than does the state vector x.
 Such faster decay of the observer error compared with the desired
dynamics makes the controller poles dominate the system response.

 It is important to note that if sensor noise is considerable, we may

choose the observer poles to be slower than two times the controller
poles, so that the bandwidth of the system will become lower and
smooth the noise.
 In this case the system response will be strongly influenced by the

observer poles.
 If the observer poles are located to the right of the controller poles in
the left-half s plane, the system response will be dominated by the
observer poles rather than by the control poles.

 In the design of the state observer, it is desirable to determine several

observer gain matrices Ke based on several different desired
characteristic equations.
 For each of the several different matrices Ke , simulation tests must be

run to evaluate the resulting system performance.
 Then we select the best Ke from the viewpoint of overall system

performance.
 In many practical cases, the selection of the best matrix Ke boils down
to a compromise between speedy response and sensitivity to
disturbances and noises.
Example-1
• Consider the system
𝑥1 0 20.6 𝑥1 0
= 𝑥 + 𝑢(𝑡)
𝑥2 1 0 2 1
𝑥1
𝑦= 0 1 𝑥
2
• We use observer based approach to design state feedback control
such that
𝑢 = −𝑲𝒙
• Design a full-order state observer assume that the desired eigenvalues

of the observer matrix are 𝛽1 = −10, 𝛽2 = −10.

Example-1
𝑥1 0 20.6 𝑥1 0
= 𝑥 + 𝑢(𝑡)
𝑥2 1 0 2 1
𝑥1
𝑦= 0 1 𝑥
2
• Let us examine the observability matrix first
𝐶 0 1
𝑂𝑀 = =
𝐶𝐴 1 0
• Since rank(OM)=2 the given system is completely state observable

and the determination of the desired observer gain matrix is possible.

Example-1 (Method-1)
𝑥1 0 20.6 𝑥1 0
= 𝑥 + 𝑢(𝑡)
𝑥2 1 0 2 1
𝑥1
𝑦= 0 1 𝑥
2
 The given system is already in the observable canonical form. Hence,
the transformation matrix Q is I.

𝑥1 0 20.6 𝑥1 0
= 𝑥 + 𝑢(𝑡)
𝑥2 1 0 2 1
𝑥1
𝑦= 0 1 𝑥
2
 The characteristic equation of the given system is
𝑠𝐼 − 𝐴 = 𝑠 2 − 20.6 = 0
 We have
𝑎1 = 0, 𝑎2 = −20.6

𝑥1 0 20.6 𝑥1 0
= 𝑥 + 𝑢(𝑡)
𝑥2 1 0 2 1
𝑥1
𝑦= 0 1 𝑥
2
 The desired characteristic equation of the system is
(𝑠 − 𝛽1 )( 𝑠 − 𝛽2 ) = (𝑠 + 10) 𝑠 + 10
(𝑠 − 𝛽1 )( 𝑠 − 𝛽2 ) = 𝑠 2 + 20𝑠 + 100
 We have
𝛼1 = 20, 𝛼2 = 100

 Observer gain matrix Ke can be calculated using following formula
𝛼2 − 𝑎2
𝑲𝑒 = 𝛼 − 𝑎
1 1
 Where
𝑎1 = 0, 𝑎2 = −20.6 𝛼1 = 20, 𝛼2 = 100
100 − (−20.6)
𝑲𝑒 =
20 − 0
120.6
𝑲𝑒 =
20
𝑥1 0 20.6 𝑥1 0
= 𝑥 + 𝑢(𝑡)
𝑥2 1 0 2 1
𝑥1
𝑦= 0 1 𝑥
2
 The characteristic equation of observer error matric is
𝑠𝑰 − 𝑨 + 𝑲𝑒 𝑪 = 0
 Assuming
𝑘𝑒1
𝑲𝑒 =
𝑘𝑒2
𝑠 0 0 20.6 𝑘𝑒1
𝑠𝑰 − 𝑨 + 𝑲𝑒 𝑪 = − + 0 1
0 𝑠 1 0 𝑘𝑒2
= 𝑠 2 +𝑘𝑒2 𝑠 − 20.6 + 𝑘𝑒1

 The desired characteristic polynomial is
(𝑠 − 𝛽1 )( 𝑠 − 𝛽2 ) = 𝑠 2 + 20𝑠 + 100
 Comparing coefficients of different powers of s
𝑠 2 + 20𝑠 + 100 = 𝑠 2 +𝑘𝑒2 𝑠 − 20.6 + 𝑘𝑒1
120.6
𝑲𝑒 =
20

𝑥1 0 20.6 𝑥1 0
= 𝑥 + 𝑢(𝑡)
𝑥2 1 0 2 1
𝑥1
𝑦= 0 1 𝑥
2
 Using Ackermann’s formula
−1
𝑪 0
𝑲𝑒 = ∅(𝑨)
𝑪𝑨 1
 Where
∅ 𝑨 = 𝐴2 + 𝛼1 𝐴 + 𝛼2 𝐼
𝛼1 = 20, 𝛼2 = 100
∅ 𝑨 = 𝐴2 + 20𝐴 + 100𝐼
∅ 𝑨 = 𝑨2 + 20𝑨 + 100𝑰
2
0 20.6 0 20.6 1 0
∅ 𝑨 = + 20 + 100
1 0 1 0 0 1
120.6 412
∅ 𝑨 =
20 120.6

 Using Ackermann’s formula
−1
𝑪 0
𝑲𝑒 = ∅(𝑨)
𝑪𝑨 1
−1
120.6 412 0 1 0
𝑲𝑒 =
20 120.6 1 0 1
120.6
𝑲𝑒 =
20

Example-1
 We get the same Ke regardless of the method employed.
 The equation for the full-order state observer is given by
𝒙 = 𝑨𝒙 + 𝑩𝑢 + 𝑲𝒆 (𝑦 − 𝑪𝒙)
𝒙𝟏 0 20.6 𝒙1 0 120.6 𝒙1
= + 𝑢(𝑡) + (𝑦 − 0 1 )
𝒙𝟐 1 0 𝒙2 1 20 𝒙2
𝒙𝟏 0 20.6 𝒙1 0 120.6 120.6 𝒙1

= + 𝑢(𝑡) + 𝑦− 0 1
𝒙𝟐 1 0 𝒙2 1 20 20 𝒙2
𝒙𝟏 0 −100 𝒙1 0 120.6
= + 𝑢(𝑡) + 𝑦
𝒙𝟐 1 −20 𝒙2 1 20
Example-2
• Design a regulator system for the following plant:
𝑥1 0 20.6 𝑥1 0
= 𝑥 + 𝑢(𝑡)
𝑥2 1 0 2 1
𝑥1
𝑦= 1 0 𝑥
2
 The desired closed-loop poles for this system are at 𝜇1 = −1.8 + 𝑗2.4 ,
𝜇2 = −1.8 − 𝑗2.4. Compute the state feedback gain matrix K to place the poles
of the system at desired location.
 Suppose that we use the observed-state feedback control instead of the actual-
state feedback. The desired eigenvalues of the observer matrix are 𝛽1 = −8,
𝛽2 = −8.
 Obtain the observer gain matrix Ke and draw a block diagram for the observed-
state feedback control system.

Reduced Order State Observer

Introduction
 The state observers discussed in previous lecture was designed to
estimate all the state variables.
 In practice however, some of the state variables may be accurately

measured.
 Therefore, such accurately measurable state variables need not be

estimated.
 In that case a reduced order state observer may be designed to

estimate only those state variables which are not directly measurable.

Introduction
 Suppose that the state vector x is an n-vector and the output vector
y is an m-vector that can be measured.
 Since m output variables are linear combinations of the state

variables, m state variables need not be estimated.
 We need to estimate only n-m state variables.
 Then the reduced-order observer becomes an (n-m)th order

observer.
 Such an (n-m)th order observer is the minimum-order observer.

Introduction
 Following figure shows the block diagram of a system with a
minimum-order observer.

Minimum Order State Observer
 To present the basic idea of the minimum-order observer we will
consider the case where the output is a scalar (that is, m=1).
 Consider the system

𝑦 = 𝑪𝒙
 where the state vector x can be partitioned into two parts xa (a scalar)
and xb [an (n-1)-vector].
 Here the state variable xa is equal to the output y and thus can be
directly measured, and xb is the unmeasurable portion of the state
vector.
 Then the partitioned state and output equations become

 The equation of measured portion of the state is given as
 Or
 The terms on the left hand side of above equation can be measure,
therefore this equation serves as an output equation.
 The equation of unmeasurable portion of the state is given as
 Noting that terms Abaxa and Bbu are known quantities.
 Above equation describes the dynamics of the unmeasured portion of

the state.
 The design procedure can be simplified if we utilize the design
technique developed for the full-order state observer.
 Let us compare the state equation for the full-order observer with that
for the minimum-order observer.
 The state equation for the full-order state observer is
 The state equation for the minimum order state observer is
 The output equations for the full order and minimum order observers
are

 List of Necessary Substitutions for Writing the Observer Equation for
the Minimum-Order State Observer

 The observer equation for the full-order observer is given by :
 Then, making the substitutions of Table–1 into above equation, we

obtain
 Error dynamics are given as

 The characteristic equation for minimum order state observer is

 Design methods
1. Using Transformation Matrix

 Design methods
 1. Using Ackerman’s Formula

Example-1
 Consider a system
 Assume that the output y can be measured accurately so that

state variable x1 (which is equal to y) need not be estimated.
Let us design a minimum-order observer. Assume that we
choose the desired observer poles to be at

Example-1

End of Slides
Best Wishes

Material I - EEE 403 ACT Part 2 PDF

Transféré par

Informations du document

Description originale:

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Material I - EEE 403 ACT Part 2 PDF

Transféré par

Droits d'auteur :

Formats disponibles

EEE 403 ADVANCED CONTROL THEORY

Design of Control Systems in State Space

2 EEE 403 Design of Control Systems in State Space

Pole Placement Design Techniques

 Direct Substitution Method

 Design of State Observer

3 EEE 403 Design of Control Systems in State Space

 What we would like to be able to do is specify all closed-

4 EEE 403 Design of Control Systems in State Space

 Frequency domain methods of design do not allow us to

 One gain to adjust, or compensator pole and zero to select,

5 EEE 403 Design of Control Systems in State Space

 Remember, to place n unknown quantities, you need n

 State-space methods solve this problem by introducing into the

 On the other hand, state-space methods do not allow the

 Finally, there is a wide range of computational support for state-

 However, as mentioned before, the advantages of computer

7 EEE 403 Design of Control Systems in State Space

 In this lecture we will discuss a design method commonly

 We assume that all state variables are measurable and are

 If the system considered is completely state controllable, then

8 EEE 403 Design of Control Systems in State Space

 The present design technique begins with a determination of

 By choosing an appropriate gain matrix for state feedback, it is

9 EEE 403 Design of Control Systems in State Space

Consider a plant represented in state space by

10 EEE 403 Design of Control Systems in State Space

 In a typical feedback control system, the output, y, is fed back to

 It is now that the topology of the design changes. Instead of

 If each state variable is fed back to the control, u, through a gain,

11 EEE 403 Design of Control Systems in State Space

12 EEE 403 Design of Control Systems in State Space

 The stability and transient response characteristics are

 If matrix K is chosen properly Eigenvalues of the system can be

 And the problem of placing the regulator poles (closed-loop

13 EEE 403 Design of Control Systems in State Space

• There are three approaches that can be used to determine the

 Using Transformation Matrix P

• All those method yields the same result.

14 EEE 403 Design of Control Systems in State Space

Method 1 Transformation Matrix

• Following are the steps to be followed in this particular

1. Check the state controllability of the system

16 EEE 403 Design of Control Systems in State Space

• Following are the steps to be followed in this particular method.

2. Transform the given system in CCF.

• Following are the steps to be followed in this particular method.

3. Obtain the desired characteristic equation from desired

(𝑠 − 𝜇1 )( 𝑠 − 𝜇2 ) ⋯ 𝑠 − 𝜇𝑛 = 𝑠 𝑛 + 𝛼1 𝑠 𝑛−1 + 𝛼2 𝑠 𝑛−2 + ⋯ + 𝛼𝑛−1 𝑠+𝛼𝑛

18 EEE 403 Design of Control Systems in State Space

• Following are the steps to be followed in this particular method.

4. Compute the gain matrix K.

19 EEE 403 Design of Control Systems in State Space

21 EEE 403 Design of Control Systems in State Space

• Example-1: Step-2 (Transformation to CCF)

22 EEE 403 Design of Control Systems in State Space

23 EEE 403 Design of Control Systems in State Space

𝑠 + 2 − 4𝑗 𝑠 + 2 + 4𝑗 𝑠 + 10 = 𝑠 3 + 14𝑠 2 + 60𝑠 + 200

𝛼1 = 14, 𝛼2 = 60, 𝛼3 = 200

 State feedback gain matric K is then calculated as

25 EEE 403 Design of Control Systems in State Space

Method 2 Direct Substitution