2008/09 MECH466 : Automatic Control 1
MECH466: Automatic Control
MECH466: Automatic Control
Dr. Ryozo Nagamune
Dr. Ryozo Nagamune
Department of Mechanical Engineering
Department of Mechanical Engineering
University of British Columbia
University of British Columbia Lecture 6
Lecture 6
Stability
Stability
Routh
Routh--Hurwitz stability criterionHurwitz stability criterion
2008/09 MECH466 : Automatic Control 2
Course roadmap
Course roadmap
Laplace transform Laplace transform Transfer function Transfer functionModels for systems
Models for systems
•
•electricalelectrical •
•mechanicalmechanical •
•electromechanicalelectromechanical Linearization
Linearization Modeling
Modeling AnalysisAnalysis DesignDesign
Time response
Time response
•
•TransientTransient •
•Steady stateSteady state Frequency response
Frequency response
•
•Bode plotBode plot Stability
Stability
•
•RouthRouth--HurwitzHurwitz
•
•NyquistNyquist
Design specs Design specs Root locus Root locus Frequency domain Frequency domain
PID & Lead
PID & Lead--laglag Design examples
Design examples
Matlab
Matlabsimulations & laboratoriessimulations & laboratories
Simple mechanical examples
Simple mechanical examples
We want mass to stay at x=0, but wind gave some We want mass to stay at x=0, but wind gave some initial speed (
initial speed (F(tF(t)=0). What will happen?)=0). What will happen?
How to characterize different behaviors with TF?How to characterize different behaviors with TF?
M
M
x(t
x(t)) f(t
f(t))
M
M
x(t
x(t))
f(t
f(t)) K
K
M
M
x(t
x(t)) f(t
f(t))
B
B
M
M
x(t
x(t))
f(t
f(t))
B B K K
Stability
Stability
Utmost important specification in control design!Utmost important specification in control design!
Unstable systems have to be stabilized by Unstable systems have to be stabilized by feedback.
feedback.
Unstable closedUnstable closed--loop systems are useless.loop systems are useless.
What happens if a system is unstable?What happens if a system is unstable?
•
• may hit mechanical/electrical may hit mechanical/electrical ““stopsstops””(saturation)(saturation)
•
2008/09 MECH466 : Automatic Control 5
What happens if a system is unstable?
What happens if a system is unstable?
Tacoma Narrows Bridge (July 1Tacoma Narrows Bridge (July 1--Nov.7, 1940)Nov.7, 1940)
2009
2009…… Wind
Wind--induced vibrationinduced vibration Collapsed!Collapsed!
2008/09 MECH466 : Automatic Control 6
Mathematical definitions of stability
Mathematical definitions of stability
BIBOBIBO(Bounded(Bounded--InputInput--BoundedBounded--Output) Output) stabilitystability: :
Any bounded input generates a bounded output.
Any bounded input generates a bounded output.
Asymptotic stability Asymptotic stability ::
Any ICs generates
Any ICs generates y(ty(t) converging to zero.) converging to zero. BIBO stable
BIBO stable
system
system
u(t
u(t)) ICs=0ICs=0 y(ty(t))
Asymp
Asymp. stable . stable system
system
u(t
u(t)=0)=0
y(t
y(t))
ICs
ICs
Some terminologies
Some terminologies
ZeroZero: roots of : roots of n(sn(s))
PolePole: roots of : roots of d(sd(s))
Characteristic polynomialCharacteristic polynomial: : d(sd(s))
Characteristic equationCharacteristic equation: : d(sd(s)=0)=0 Ex.
Ex.
Stability condition in s
Stability condition in s
-
-
domain
domain
(Proof omitted, and not required)
(Proof omitted, and not required)
For a system represented by a transfer For a system represented by a transfer function
function G(sG(s),),
system is BIBO stable system is BIBO stable
system is asymptotically stable system is asymptotically stable All the poles of
All the poles of G(sG(s) are in the open left ) are in the open left
half of the complex plane.
2008/09 MECH466 : Automatic Control 9
Idea of stability condition
Idea of stability condition
Asym
Asym. Stability: . Stability: (
(U(sU(s)=0))=0)
BIBO Stability: BIBO Stability:
(y(0)=0) (y(0)=0)
Example
Example
Bounded if
Bounded if ReRe(α(α)>)>00
2008/09 MECH466 : Automatic Control 10
Remarks on stability
Remarks on stability
For a general system (nonlinear etc.), BIBO For a general system (nonlinear etc.), BIBO stability condition and asymptotic stability
stability condition and asymptotic stability
condition are different.
condition are different.
For For linear timelinear time--invariant (LTI) systemsinvariant (LTI) systems(to which (to which we can use Laplace transform and we can
we can use Laplace transform and we can
obtain a transfer function), the conditions
obtain a transfer function), the conditions
happen to be the same.
happen to be the same.
In this course, we are interested in only LTI In this course, we are interested in only LTI systems, we use simply
systems, we use simply ““stablestable””to mean both to mean both
BIBO and asymptotic stability.
BIBO and asymptotic stability.
Remarks on stability (cont
Remarks on stability (cont’
’d)
d)
Marginally stableMarginally stableifif
G(sG(s) has no pole in the open RHP (Right Half Plane), &) has no pole in the open RHP (Right Half Plane), &
G(sG(s) has at least one simple pole on ) has at least one simple pole on jwjw--axis, &axis, &
G(sG(s) has no multiple pole on ) has no multiple pole on jwjw--axis.axis.
UnstableUnstableif a system is neither stable nor if a system is neither stable nor marginally stable.
marginally stable. Marginally stable
Marginally stable NOT marginally stableNOT marginally stable
Stability summary
Stability summary
(BIBO, asymptotically) stable(BIBO, asymptotically) stableifif Re(s
Re(sii)<0 for all i.)<0 for all i.
marginally stablemarginally stableifif
Re(sRe(sii)<=0 for all i, and)<=0 for all i, and
simple root for simple root for Re(sRe(sii)=0)=0
unstableunstableifif
it is neither stable nor
it is neither stable nor
marginally stable.
marginally stable. Let
Let ssiibe be polespolesof G. of G. Then, G is
2008/09 MECH466 : Automatic Control 13
Mechanical examples: revisited
Mechanical examples: revisited
M
M
x(t
x(t)) f(t
f(t))
M
M
x(t
x(t))
f(t
f(t)) K
K
M
M
x(t
x(t)) f(t
f(t))
B
B
M
M
x(t
x(t))
f(t
f(t))
B B K K Poles= Poles= stable? stable? Poles= Poles= stable? stable? Poles= Poles= stable? stable? Poles= Poles= stable? stable?
2008/09 MECH466 : Automatic Control 14
Examples
Examples
Stable/marginally stable Stable/marginally stable /unstable /unstable ? ? ? ? ? ? ? ? ??? ???Course roadmap
Course roadmap
Laplace transform Laplace transform Transfer function Transfer functionModels for systems
Models for systems
•
•electricalelectrical
•
•mechanicalmechanical
•
•electromechanicalelectromechanical
Linearization
Linearization Modeling
Modeling AnalysisAnalysis DesignDesign
Time response
Time response
•
•TransientTransient
•
•Steady stateSteady state
Frequency response
Frequency response
•
•Bode plotBode plot
Stability
Stability
•
•RouthRouth--HurwitzHurwitz •
•NyquistNyquist
Design specs Design specs Root locus Root locus Frequency domain Frequency domain
PID & Lead
PID & Lead--laglag
Design examples
Design examples
Matlab
Matlabsimulations & laboratoriessimulations & laboratories
Routh
Routh-
-Hurwitz criterion
Hurwitz criterion
This is for LTI systems with a polynomialThis is for LTI systems with a polynomial
denominator (without sin,
denominator (without sin, coscos, exponential etc.), exponential etc.)
It determines if all the roots of a polynomial It determines if all the roots of a polynomial
lie in the open LHP (left halflie in the open LHP (left half--plane),plane),
or equivalently, have negative real parts.or equivalently, have negative real parts.
It also determines the number of roots of a It also determines the number of roots of a polynomial in the open RHP (right half
polynomial in the open RHP (right half--plane).plane).
It does It does NOTNOTexplicitly compute the roots.explicitly compute the roots.
2008/09 MECH466 : Automatic Control 17
Polynomial and an assumption
Polynomial and an assumption
Consider a polynomialConsider a polynomial
AssumeAssume
If this assumption does not hold, Q can be factored asIf this assumption does not hold, Q can be factored as
where
where
The following method applies to the polynomialThe following method applies to the polynomial
2008/09 MECH466 : Automatic Control 18
Routh
Routh
array
array
From the given
From the given
polynomial
polynomial
Routh
Routh
array
array
(How to compute the third row)
(How to compute the third row)
Routh
Routh
array
array
(How to compute the fourth row)
2008/09 MECH466 : Automatic Control 21
Routh
Routh-
-Hurwitz criterion
Hurwitz criterion
The number of roots The number of roots in the open right half in the open right half--plane plane
is equal to is equal to
the number of sign changes the number of sign changes in the
in the first columnfirst columnof Routhof Routharray.array.
2008/09 MECH466 : Automatic Control 22
Example 1
Example 1
Routh
Routharrayarray
Two sign changes
Two sign changes
in the first column
in the first column
Two roots in RHP
Two roots in RHP
Example 2
Example 2
Routh
Routharrayarray
No sign changes
No sign changes
in the first column
in the first column
No roots in RHP
No roots in RHP
Always same!
Always same!
Example 3 (from slide 14)
Example 3 (from slide 14)
Routh
Routharrayarray
No sign changes
No sign changes
in the first column
in the first column No roots in RHPNo roots in RHP
Always same!
2008/09 MECH466 : Automatic Control 25
Simple & important criteria for stability
Simple & important criteria for stability
11ststorder polynomialorder polynomial
22ndndorder polynomialorder polynomial
Higher order polynomialHigher order polynomial
2008/09 MECH466 : Automatic Control 26
Examples
Examples
All roots in open LHP? All roots in open LHP?
Yes / No Yes / No
Yes / No Yes / No
Yes / No Yes / No
Yes / No Yes / No
Yes / No Yes / No
Why no proof in textbooks?
Why no proof in textbooks?
“
“most undergraduate students are exposed to the most undergraduate students are exposed to the
Routh
Routh––Hurwitz criterion in their first introductory Hurwitz criterion in their first introductory controls course. This exposure, however, is at the
controls course. This exposure, however, is at the
purely algorithmic level in the sense that no attempt
purely algorithmic level in the sense that no attempt
is made whatsoever to explain why or how such an
is made whatsoever to explain why or how such an
algorithm works.
algorithm works.””
An Elementary Derivation of the Routh–Hurwitz Criterion
Ming-Tzu Ho, Aniruddha Datta, and S. P. Bhattacharyya IEEE Transactions on Automatic Control
vol. 43, no. 3, 1998, pp. 405-409.
Why no proof in textbooks? (cont
Why no proof in textbooks? (cont’
’d)
d)
“
“The principal reason for this is that the classical The principal reason for this is that the classical proof of the
proof of the RouthRouth--Hurwitz criterion relies on the Hurwitz criterion relies on the
notion of Cauchy indexes and Sturm
notion of Cauchy indexes and Sturm’’s theorem, s theorem, both of which are beyond the scope of
both of which are beyond the scope of
undergraduate students.
undergraduate students.””
“
“RouthRouth--Hurwitz criterion has become one of the few Hurwitz criterion has become one of the few results in control theory that most control engineers
results in control theory that most control engineers
are compelled to accept on faith.
2008/09 MECH466 : Automatic Control 29
Summary and Exercises
Summary and Exercises
Stability for LTI systemsStability for LTI systems
(BIBO, asymptotically) stable, marginally stable, unstable(BIBO, asymptotically) stable, marginally stable, unstable
Stability for Stability for G(sG(s) is determined by poles of G.) is determined by poles of G.
RouthRouth--Hurwitz stability criterionHurwitz stability criterionto determine stability to determine stability without explicitly computing the poles of a system
without explicitly computing the poles of a system
Next, examples of Next, examples of RouthRouth--Hurwitz criterionHurwitz criterion
ExercisesExercises
Read Sections 6.1 & 6.2.Read Sections 6.1 & 6.2.
Solve Problems 6.1, 6.9 & 6.14.Solve Problems 6.1, 6.9 & 6.14.