INTRODUCTION TO CONTROL

THEORY
INTRODUCTION
 Objectives

Signal analysis

Concept of signal

System analysis

Concept of ‘system’

Dynamical system

Abstract system definition

Continuous system control

Relation with other subjects

Mathematics

Physics

Applied science and technology

Control Theory
 Program

Transforms: Laplace Transform

Modeling: System modeling

Linearization.Transfer Function. Block Diagram

System Analysis: Temporal/Frequential Analysis

Feedback systems . Static Analysis of control : accuracy,

sentivity to perturbations. Root Locus. Frequential analysis of
feedback system.

Process Control: PID controllers. Phase lead and Phase lag

controllers.
 Basic references

J. Wilkie, M. Johnson and R. Katebi., “Control engineering: an introductory
course”. Ed. Palgrave Macmillan. 2002

OGATA, K., “Modern Control Engineering”, 5th edition. Ed. Prentice-Hall, 2010.

DORF, R. C., BISHOP, R. H. : “Modern Control Systems”. 12th edition. Ed.

Prentice Hall, 2010.

NISE, N. S., “Control Systems Engineering”, 6th edition. Ed. Wiley, 2011.

Khoo, M. C. K., “Physiological Control Systems: Analysis, Simulation, and

Estimation”. IEEE Press. 1999

Barrientos et al.: “Control de sistemas continuos. Problemas resueltos”. Ed.

McGrawHill, 1996.

DiStefano et al.: “Feedback and Control Systems ”. 2nd edition. Ed.

McGrawHill (Schaum’s Outlines), 1990.
 Examples

a thermostat, a device that regulates or controls the temperature in a

room or house using the central heating and/or cooling unit.
Thermostats are also found in most cars, ovens, freezers and
refrigerators, and of course hot water units.

a toilet flush, a control mechanism that ensures that the toilet gets
flushed, and that the toilet cistern is automatically refilled to a set level.
Similar mechanisms are used in irrigation canals, and wherever a fluid
level needs to be regulated.

a cruise controller, a device that more or less keeps the car traveling at
a constant speed as set by the driver, despite variations in the
inclination of the road. More advanced, but in a similar vein, is the auto
pilot used in aircraft to fly from way point to way point and the auto
rudder used in ships to maintain a given heading.
 Examples

In nature, feedback is at the core of

homeostasis, a term used to refer how in living organisms critically

important variables are kept within acceptable limits; such as our
body temperature or our blood glucose level for example.

the Earth’s hydrocycle; how water evaporates, then condenses

into clouds, moves around the globe, precipitates to the Earth, and
flows back to the oceans, its main reservoir.

prey-predator behavior, or how a predator species feeds on its

prey and therefore depends on that prey’s survival for its own
survival. Most ecologies depend on feedback to maintain the
balance of the species in the ecology.
 Studied Signals

Continuous time / discrete time

Periodic / not periodic

Determinist / Stocastic
 Studied systems

Linear / not linear

Concentrate parameters / distributed

Variable / invariant

Determinist / Stocastic

Monovariable / multivariable

Continuous time/ discrete
 Analysis Methods

Time domain / frequential

External Representation / internal
TRANSFORMS
 How to represent a signal

A signal can be represented in various ways

As a function of time
 Example: speed of a car at start

t
It seems the more representation

As a function of frequency
 Example : Sound

intensity

time/frequency are called ‘independent variable’

 How to represent a signal

It is possible to pass from a representation to the

other.

Which representation is ‘better’ depends on the

information you want to obtain from the signal.
 Transforms

They allow to pass, ‘transform’, from a kind of
representation to another.

Example: T(x1(time))=x2(frequency)

They aim to simplify the analysis of systems and

signals.

They don’t always have a physical meaning: they

can be ‘only’ mathematical representations.

Continuous Signals Transforms:

Fourier Transform

Laplace Transform
Table of Laplace transforms
System Modeling
Control Theory of Continuous
Linear Dynamic Systems
 Mathematical models

Concept of ‘model’

Mathematical model

Differential equations
Example.

d 2 x dx dy
3 2 + +2=0
dt dt dt

Variables are continuous time signals

Mathematical models
Mechanical Systems
Example
 Example

Y2 (s)
Findthetransferfunction U 2 (s)  ofsystemshowninthefigure,whereu1 y
u2areinputforcesandtheoutputsy1andy2arepositions.


 k1 u1

 m1 k2
y1


 b1
 u2

y2
m2
Example
 Example:
Mechanical accelerometer
Rotational Systems
Rotational Systems: Example
Rotational Systems: Example
Example
Rotational Systems: Pendulum
Rotational Systems:
Inverted Pendulum
Rotational Systems: Longitudinal
dynamics of an aircraft
 Rotational Systems: Satellite control.

Satellites require attitude control so that sensors,

antennas, etc., are properly pointed. Let’s consider one
axis of rotation

Note: Output of system θ(t) integrates torques

twice “double-integrator plant.”
Rotational Systems: Torsional pendulum
Rotational Systems: Satellite
EXAMPLE: NONLINEAR Rotational Pendulum
Electrical Models
Electrical Models
Electrical Models: Lead compensator
Electrical Models: Lag compensator
Electrical Models
EXAMPLE: DC Motor (servo-motor)
EXAMPLE: DC Motor (servo-motor)
EXAMPLE: DC Generator
EXAMPLE: DC Generator
 Analogous Systems

• The linearized differential equations of many very

different physical systems appear identical.

• One would suppose they behave in similar ways

(dynamic response) and can be controlled with similar
controllers.
 Linear model of respiratory mechanics.

The airways are divided into two categories:

the larger or central airways and the smaller or
peripheral airways, with fluid mechanical
resistances equal to Rc and Rp, respectively.
Air that enters the alveoli also produces an
expansion of the chest-wall cavity by the same
volume. This is represented by the connection
of the lung (CL ) and chest-wall (Cw)
compliances in series.
However, a small fraction of the volume of air
that enters the respiratory system is shunted
away from the alveoli as a result of the We account for this effect by placing a
compliance of the central airways and gas shunt compliance, Cs, in parallel with CL
compressibility. and Cw. The pressures developed at the
This shunted volume is very small under different points of this lung model are:
normal circumstances at regular breathing Pao at the airway opening, Paw in the
frequencies, but becomes progressively more central airways, PA in the alveoli and Ppi
substantial if disease leads to peripheral airway in the pleural space (between the lung
obstruction (i.e., increased Rp ) or a stiffening parenchyma and chest wall).
of the lungs or chest-wall (i.e., decreased CL or These pressures are referenced to Po, the
Cw). ambient pressure, which we can set to
zero.
Suppose the volume flow-rate of air entering the respiratory system is Q. Then, the
objective here is to derive a mathematical relationship between Pao and Q.
From Kirchhoff's Second Law (applied to the node Paw), if the flow delivered to the
alveoli is QA' then the flow shunted away from the alveoli must be Q – QA.
Applying, Kirchhoff's First Law to the closed circuit containing Cs, Rp, CL , and Cw, we
have

Applying Kirchhoff's First Law to the circuit containing Rc and Cs, we have

Differentiating these equations with respect to time, and subsequently reducing the two
equations to one by eliminating QA , we obtain the equation relating Pao to Q:
 Linear model of muscle mechanics.

The second example that we will consider is

the linearized physiological model of
skeletal muscle. Fo represents the force
developed by the active contractile element of
the muscle, while F is the actual force that
results after taking into account the mechanical
properties of muscle. R represents the viscous
damping inherent in the tissue, while Cp
(parallel elastic element) and Cs (series elastic
element) reflect the elastic storage properties
of the sarcolemma and the muscle tendons,
respectively.

First, consider the mechanical constraints placed on the model components as a result of the
parallel configuration.
If spring Cp is stretched by an incremental length x, the entire series combination of R
and Cs will also extend by the same length. Furthermore, the sum of the force transmitted
through the two branches of the parallel configuration must equal F. However, although the
sum of the extensions of Cs and R will have to equal x, the individual length contributions
from Cs and R need not be equal.
Thus, if we assume Cs is stretched a length x I' then the extension in the parallel
combination of R and Fo will be x –x1 . The velocity with which the dashpot represented by
R is extending is obtained by differentiating X- XI with respect to time, i.e., d(x –x1 )/dt.
Using the principle that the force transmitted through Cs must be equal to the force
transmitted through the parallel combination of Fo and R, we obtain the following equation:

Then, using the second principle, i.e., that the total force from both limbs of the parallel
combination must sum to F, we have

Eliminating x1 from these equations yields the following differential equation relating F to x
and Fo:
Linearization
 Linearization

Concept of linearization

Linearization

y = f (x1, x2 ..xn ) Non linear

∂y ∂y ∂y
y ≈ y0 + (x1 − x1 ) + (x2 − x2 ) + ...+ (xn − xn )
∂ x1 0 ∂ x2 0
0
∂ xn 0
0 0

∆y ≈ a1∆x1 + a2 ∆x2 + ...+ an ∆xn Linear

The xi represent variables as well as derivatives

The linearized equation is not unique, it depends on the

(tangent) point where the linearization is performed
 Linearization

Equilibrium point(s)

The variables of the linearized equation represent

increments w.r.t. the equilibrium point

In order to simplify the notation it is not used to

represent variables as increments

It is usually easier to linearize the equation term by term

In general the linearized and the non linear model have

different behaviors
 Transfer Function
u(t) y(t)
U(s) Y(s)

If the system is linear (or has been linearized) we have:

du (t ) d 2 u (t ) dy (t ) d 2 y (t )
a0u (t ) + a1 + a2 2
+ ... = b0 y (t ) + b1 + b2 2
+ ...
dt dt dt dt
 Transfer Function

u(t) y(t)
U(s) Y(s)

If we apply Laplace transform, assuming that the system is in

equilibrium for t<0, which implies that the value of the variable y and
its derivative are 0 for t=0- (as they represent increments w.r.t the
equilibrium point)
a 0 U ( s ) + a 1 sU ( s ) + a 2 s 2 U ( s )... = b 0 Y ( s ) + b1 sY ( s ) + b 2 s 2 Y ( s )...
( a 0 + a 1 s + a 2 s 2 ...) U ( s ) = ( b 0 + b 1 s + b 2 s 2 ...) Y ( s )
Y (s) a 0 + a 1 s + a 2 s 2 ...
= = G (s)
U (s) b 0 + b 1 s + b 2 s 2 ...

u(t) y(t)
G(s)
U(s) Y(s)

Y ( s ) = G ( s )U ( s ) ⇒ y ( t ) = g ( t )∗ u ( t )
 Transfer Function

u(t) y(t)
G(s)
U(s) Y(s)

Y(s) = G(s)U(s) ⇒ y(t) = g(t)∗ u(t)

a 0 U ( s ) + a 1 sU ( s ) + a 2 s 2 U ( s )... = b 0 Y ( s ) + b 1 sY ( s ) + b 2 s 2 Y ( s )...
( a 0 + a 1 s + a 2 s 2 ...) U ( s ) = ( b 0 + b1 s + b 2 s 2 ...) Y ( s )
Y (s) a 0 + a 1 s + a 2 s 2 ...
= = G (s)
U (s) b 0 + b 1 s + b 2 s ...
2
 Example: Impulse response

u(t) y(t)
G(s)
U(s) Y(s)

u( t ) = δ ( t ) ⇒ U ( s ) = 1
Y ( s ) = G ( s )U ( s ) = G ( s )
y(t ) = g(t )
 Transfer Function

For an impulsive input we have:

L [u (t )] → U ( s )
Y (s)
G (s) = ⇒ Y ( s ) = G ( s ) ⋅U ( s )
U (s)
y ( t ) = L − 1 [G ( s ) ⋅ U ( s ) ]

Utilizing an impulsive input we get:

u (t ) ≡ ∂ (t ) → U ( s ) = 1
Y ( s ) = G ( s ) ⋅U ( s ) = G ( s )
y (t ) = g (t ), g (t ) = L−1[G ( s )]

y (t ) = g (t ) ∗ u (t )
 Block Diagram

The block diagram de bloques is a way to represent
gráfically the relations in between the variables of a
system
 Block Diagrams

Example

Y(s)=A(s)X(s)+B(s)U(s)
X(s)=C(s)Z(s)-D(s)Y(s)
Z(s)=E(s)U(s)

+
B(s)
U(s) + Y(s)
A(s)
X(s)
Z(s)
E(s) C(s) D(s)
+ -
Level control of a tank
Example
Systems study
Operations with blocks
Operations with blocks
EXAMPLE: Simplify
EXAMPLE: Simplify
 How to obtain transfer functions

Working on equations

Reducing the block diagram

Mason method
 How to obtain transfer
functions: Mason’s Rule

If you don’t care to simplify a block diagram using block diagram
manipulation, you can use Mason’s Rule.

“Node” = Common input to several blocks, or output of summing
junction.

“Path” = Sequence of connected blocks, from one variable to another,

in the direction of signal flow, without including any variable more
than once.

“Forward Path” = Path from input to output.

“Loop” = Path from node back to itself.

“Path Gain” = Product of all transfer functions in a path.

“Loop Gain” = Product of all transfer functions in a loop.
Mason’s Rule
Mason’s Rule
 How to obtain transfer functions

Mason’s formula

Y(s) ∑M k ∆k
= k
U(s) ∆

∆ = 1 − ∑ Bi + ∑ Bi Bj + ...
Operations with flow diagrams
 Equivalence between block
diagrams and flow diagrams
 Key Concept: Analogous Quantities

Mechanical
Electrical Mechanical Analog I
Analog II
Quantity (Force-Current)
(Force Voltage)
Voltage, e Velocity, v Force, f
Current, i Force, f Velocity, v
Lubricity, 1/B
Resistance, R Friction, B
(Inverse friction)

Compliance, 1/K
Capacitance, C Mass, M
(Inverse spring constant)

Compliance, 1/K
Inductance, L Mass, M
(Inverse spring constant)

Transformer, N1:N2 Lever, L1:L2 Lever, L1:L2

Key Concept: Analogous Equations
Example: Conversion from Electrical to Mechanical 1 -- Mathematical
Method
 Insulin delivery control system

• Control systems have been utilized in the biomedical field to create implanted
automatic drug-delivery systems to patients.
• Automatic systems can be used to regulate blood pressure, blood sugar level, and heart
rate.
• A common application of control engineering is in the field of open-loop system drug
delivery, in which mathematical models of the dose-effect relationship of the drugs are
used. A drug-delivery system implanted in the body uses an open-loop system, since
miniaturized glucose sensors are not yet available.
• The best solutions rely on individually programmable, pocket-sized insulin pumps that
can deliver insulin according to a preset time history.
• More complicated systems will use closed-loop control for the measured blood glucose
levels.
• Control Goal
Design a system to regulate the blood sugar concentration of a diabetic by controlled
dispensing of insulin.
• Variable to Be Controlled
Blood glucose concentration
• Control Design Specifications
Provide a blood glucose level for the diabetic that closely approximates (tracks) the
glucose level of a healthy person.
Simple Feedback System Example
Simple Feedback System Example
Simple Feedback System Example
Simple Feedback System Example
Simple Feedback System
Simple Feedback System
Simple Feedback System
Simple Feedback System
Simple Feedback System
Simple Feedback System
Simple Feedback System
 Water clock
• Automatic control of water level
using a float level was used in
the Middle East for a water
clock.

• The water clock was used from

sometime before Christ until the
17th century.

• Discuss the operation of the

water clock, and establish how
the float provides a feedback
control that maintains the
accuracy of the clock.

• Sketch a block diagram of the

feedback system.
• The accuracy of the clock is dependent upon a constant flow from the orifice;
the flow is dependent upon the height of the water in the float tank.

• The height of the water is controlled by the float.

• The control system controls only the height of the water.

• The control system can be seen as:

SYSTEM ANALYSIS
 INTRODUCTION TO
TEMPORAL ANALYSIS
 Introduction
u(t) y(t)
G(s)
U(s) Y(s)

−1
Y (s) = G(s)U (s) ⇒ y(t) = g(t ) ∗ u(t) = L (G(s)U (s))

Temporal Analysis Concept y(t) = yrt (t) + yrp (t)
lim yrt (t) = 0
t →∞

yrp (t) = lim y(t)

t →∞
S te p R e s p o n s e
S te p R e s p o n s e 0 .7
2

1 .8 0 .6

1 .6
0 .5
1 .4

1 .2 0 .4

Amplitude
Amplitude

1
0 .3
0 .8

0 .6 0 .2

0 .4
0 .1
0 .2

0 0
0 0 .1 0 .2 0 .3 0 .4 0 .5 0 .6 0 .7 0 .8 0 .9 1 0 2 4 6 8 10 12

Continuous
T im e ( s e c ) T im e ( s e c )

System
u(t) y(t)
 Normalized input signals

Impulse u (t ) = δ (t ) ⇒ U (s) = 1

 0, t<0 1

Step u (t ) = u 0 (t ) =  ⇒ U (s) =
 1, t>0 s

Ramp 1
u (t ) = tu 0 (t ) ⇒ U (s) = 2
s
Step Response of continuous time
system

u(t) y(t)

G(s)
U(s) Y(s)

G (s)
Y (s ) = U (s )G (s ) =
s
 G (s) = t
∫
−1
y(t) = L   g (τ ) d τ
 s  0
 Step Response of continuous time
systems

If σ i are the real poles of G(s) and αi±jβi the complex

ones with imaginary part (not multiples), we have
 Step Response of continuous time systems

y (t ) = G (0) + ∑ Bi eσ i t + ∑ Ei eαi t sen( β i t + ϕ i ) t≥0

Terms of the form B i e σ it
14 1

0 .9
12

10
σ > 0 0 .8

0 .7
e x p (-a * t )

8
0 .6
σ < 0
0 .5

6
0 .4

0 .3
4
0 .2
2
0 .1

0 0
0 0 .5 1 1.5 2 2.5 3 3 .5 4 4 .5 5 0 0.5 1 1.5 2 2.5 3 3 .5 4 4 .5 5

Terms of the form E i e α it s e n ( β i t + ϕ i )

1
10
0.8

5 0.6

0.4

0 0.2

0
-5
-0 . 2

-0 . 4
-1 0
-0 . 6

-0 . 8
-1 5 0 0.5 1 1.5 2 2.5 3 3 .5 4 4 .5 5
0 0 .5 1 1.5 2 2.5 3 3 .5 4 4 .5 5
 Complex Exponential

t=0:.1:100;
plot3(t,100*real(exp((-.02+.2*j)*t)),100*imag(exp((-.02+.2*j)*t)))
 Relation between temporal response and poles and
zeroes in continuous time systems

Stability

Poles in negative semiplane

Static Gain. Steady-state regime

G(0)

y rp ( t ) = lim y ( t ) = A
t→ ∞

G (s)
lim y ( t ) = lim s = G (0)
t→ ∞ s→ 0 s
 Relation between temporal response and poles and
zeroes in continuous time systems

Speed

Poles Distances from imaginary axes

e σ it , e α it
e −200 t (t = 1) = e −200
e −2 t (t = 1 0 0 ) = e −200

Oscillations

Imaginary Poles
E i s e n ( β it + ϕ i )
Example: Spring Mass System
Example: Spring Mass System
Example: Spring Mass System
Example: Spring Mass System
Example: Spring Mass System
Longitudinal Motion Equations of an Aircraft
Longitudinal Modes of an Aircraft
Time Response
Time Response
 Evaluating Response Using Poles

PROBLEM: Given the system of the figure, write the output, c(t), in
general terms. Specify the forced and natural parts of the solution.
Evaluating Response Using Poles
 First order systems

K
G(s) =
1 + Ts
 First order systems:
Impulse response

(t) y(t)
G(s)
1 Y(s)

−1 K  k − t
y(t ) = g(t) = L   = e T
t ≥ 0
1 + Ts T
 First order systems:
Step response
u0(t) y(t)
G(s)
1/s Y(s)

 G (s)
−1 −1  k  −1  1 T 
y(t ) = L   = L   = kL  −  =
 s   s (1 + T s )   s 1 + Ts 
− tT
= k (1 − e ) t ≥ 0
Time Response
 Time Response

0.632*0.72

τ
 First order systems:
Step response

Tangent at the origin

k
y ′(0 ) =
T

Stabilization time

π
ts ≈
σ
 First order systems:
Ramp response
tu0(t) y(t)
G(s)
1/s2 Y(s)

−1  G ( s )  −1  k  −1
 T 1 T 2

y(t ) = L  2  = L  2  = kL  − + 2 + =
 s   s (1 + Ts )   s s 1 + Ts 
− tT − tT
= k ( − T + t + Te ) = k ( t − T ) + kTe t≥0
 Second order systems

Caracteristic parameters of a second order

system

kω n 2
G(s) =
s 2 + 2ζω n s + ω n 2
k = Static gain.
ωn = undamped natural frequency
ζ = damping ratio
σ =ζωn = attenuation.
ωd = ωn 1 − ζ 2 = damped frequency.
ϑ = cos −1 (ζ )
 Time Response

k = Static gain.
ω n = undamped natural frequency
ζ = damping ratio
σ = ζω n = attenuation.
ω d = ω n 1 − ζ 2 = damped frequency.
ϑ = cos −1 (ζ )
 Second order systems

Poles

s 2 + 2ζ ω n s + ω n 2 = 0
s = −ζ ω n ± ζ 2ω n 2 − ω n2 = − ζ ω n ± ω n ζ 2 − 1 =
− ζ ω n ± jω n 1 − ζ 2
= − σ ± jω d
Time Response
 Second order systems classification

Overdampened (ζ > 1)

s
s = −ζ ω n ± ω n ζ 2
−1
⌠

Critically dampened (ζ = 1)

⌠
s s = −ζ ω n
 Second order systems classification

Underdampened (0 < ζ < 1)

ωn s
d
 s = − ζ ω n ± jω n 1−ζ 2
⌠

Oscillator (ζ = 0)
ζ = cosθ s = ± jω n
ωn s

Unstable (ζ < 0)

s = − ζ ω n ± jω n 1−ζ 2

α i = − ζ ω n > 0 (p o s itiv e re a l p a rt)

 Over dampened second order systems
Step Response
u0(t) y(t) p1 p2 s
G(s)
1/s Y(s)
⌠
 ω  e p1 t e p2t 
y ( t ) = k  1 + n
 −   t ≥ 0
 2 ζ −1
2
 p 1 p2 

ζ = 2.5

ζ = 1.5

2
G (s) =

Pendiente en el origen: s2 + 3s + 2

Tangent at the origin: y ′(0 ) = 0

 Critically dampened second order systems
Step Response

u0(t) y(t) s
G(s)
⌠
1/s Y(s)

y ( t ) = k (1 − e − σ t (1 + σ t ) ) t≥0

ζ =1

ζ = 1.5

2
G (s) =
s2 + 2s + 1

Tangent at the origin: y ′(0 ) = 0

 Under dampened second order systems
Step Response

u0(t) y(t) ωn d s
G(s) 
1/s Y(s)
⌠

 e −σt 
y ( t ) = k  1 − s e n ( ω d t + ϑ )  t ≥ 0
 1−ζ 2


ζ = 0.2
ζ = 0.5
 Oscillating second order systems
Step Response

u0(t) y(t)
G(s)
1/s Y(s) n

y ( t ) = k (1 − co s(ω n t )) t≥0

2
G (s) =
s2 + 1

Tangent at the origin: y ′(0 ) = 0

Time Response
Natural Responses
Time Response
Time Response
 Under dampened second order systems
Step Response
 Tangent in the origin:
y ′(0 ) = 0

 Stabilization time:
π
ts ≈ , (with σ =ζω n )
σ

 Rise time:
π −ϑ
tr ≈ , (with ϑ = cos −1 (ζ ))
ωd

 Peak time: tp =
π
ωd

ζπ π
− −
 Overshoot: M p = e 1−ζ 2
= e tg θ
Simplified Longitudinal Modes of
Motion of an Aircraft
Simplified Longitudinal Modes of
Motion of an Aircraft
Longitudinal Motion Equations of an Aircraft
Simplified Lateral Modes of Motion of an
Aircraft
Simplified Lateral Modes of Motion of an
Aircraft
Time Response
Design Synthesis
EXAMPLE: Converting specs. to s-plane
 Equivalent reduced order system

Concept of dominant pole(s)

Dominant

Deleting non dominant roots

A
s
s
B A<<B
Equivalent reduced order system
 Equivalent reduced order system

Deleting pairs polo/zero

s s
A

B A<<B

The equivalent system must keep the static gain

Equivalent reduced order system
Comparing Responses of Three-Pole Systems
Equivalent reduced order system
Effect of adding a zero to a two-pole system

1
T(s) =
(s+1+ 2.8 j )(s+1− 2.8 j )
Step response of a nonminimum-phase system
 Algebraic methods for the
stability analysis
• Continuous time
– Routh-Hurwitz method

Caracteristic polinomial (denominator of the transfer function)

 Routh-Hurwitz Method
Necessary but not sufficient conditions:

•All coefficients have the same sign

•None of them is null

Routh table
Routh-Hurwitz Method
Routh-Hurwitz Method
Routh-Hurwitz Method
Routh-Hurwitz Method
 Routh-Hurwitz Method

The system will be stable if all the coefficient of the first column
have the same sign.
 Routh-Hurwitz Method
Example

In the first column we have two change of sign. For that we have
two roots in the rightside semiplane : the system is unstable
Special Cases
 Special Cases

A zero in the first column:

•Substitute the zero with a small positive number . In order to

count the sign change s we make it to tend to 0.

•Make the variable substitution: s=1/x. We will apply Routh

criteria on the new polynomial p(x)
 Special Cases

Unstable system
Special Cases
 Special Cases

• A line of zeroes in the table:

• Take the coefficient of the upper row

• Write the auxiliary row
• Calculate the derivative of the auxiliary equation
• Substitute the line of zeroes with the coefficients of the derivative
equation
• Continue with the method
Special Cases
Stability as a function of parameters