Sie sind auf Seite 1von 18

Teoría de Control

Root Locus design

Germán Andrés Ramos, PhD.


Design criteria

• Steady state performance – exactitud - tracking error –


step, ramp
- Steady state error:

r(t) − yss (t)


e = lim
t→∞ a

- Position error: step input r(s)=a


- Velocity error: ramp input r(s)=at

• Dada
α 0 + α1s ++ α n s n
Go (s) = n≥m
β0 + β1s ++ β m s m
Design criteria

• Position error: step input r(s)=a


α 0 − β0
e p = 1− Go (0) =
α0
• Velocity error: ramp input r(s)=at

d
ev = (1− Go (0)) t − Go (s)
ds s=0

" α 0 − β0 % α 0 β1 − β0α1
ev = $ 't − 2
# α 0 & α 0
Design criteria

• The plant output y(t) is said to track asymptotically the


reference input r(t) if

lim r(t) − y(t) = 0


t→∞

• step input r(s)=a • parable input r(s)=at2


α 0 = β0 α 0 = β0
• ramp input r(s)=at α1 = β1
α 0 = β0 α 2 = β2
α1 = β1
Design criteria

• System type: number of poles at the origen


- Loop transfer function Gl(s)

Tipo\ess ep ev ea
0 K ∞ ∞
1 0 K ∞
2 0 0 K
Design criteria

• Well-posedness:

- All transfer function in the system must be proper and

Δ(∞) = 1+ C(∞)G(∞) ≠ 0

• Total stability

-  Poles of Go(s) and its missing poles are all stable


-  Avoid unstable pole-zero cancellations
Design criteria

• Transient state – speed of response – step response


Root Locus Method
• Find the closed-loop pole location for any k in

Closed-loop transfer function Closed-loop poles:


kG(s)
Go (s) = 1+ kG(s) = 0
1+ kG(s)
Then: 1
G(s) = −
1 k
G(s) = −
k " 1%
Ĝ(s) = arg $ − ' = ±180
# k&
Root Locus Method

• In general
- The roots of the characteristic equation as a function of
kx
p(s) + kx q(s) = 0

Then:
1
G pq (s) = −
q(s) 1 kx
G pq (s) = =−
p(s) kx " 1%
Ĝ pq (s) = arg $ − ' = ±180
# kx &
Root Locus Method

• Procedure:
- Find the design region
- Plot the roots of the characteristic equation as a
function of k
p(s) + kq(s) = 0

• The region is defined by the time domain criteria (step


signal response) :
- ts settling time
- Overshoot
- tr rise time
Root Locus Method

• ts settling time

4.5 4.5
−a = − ts = < ts0
ts a
Root Locus Method

• Given
ω 2n
Go (s) = 2
s + 2ζω n s + ω 2n
Its poles are
−ζω n ± ω n ζ 2 −1

Also we have:
− πζ
1−ζ 2
Mp = e
M p ≤ 5% → ζ ≥ 0.7 → θ ≤ 45
ζ = cosθ
Root Locus Method

• Rise time tr
The farther away the closest pole from the origin, the smaller the rise
time ?

1.8
tr ≈
ωn
Root Locus Method

• Desired pole region

ζ < ζo
ts < tso
Root Locus Method

• Desired pole region

ζ < ζo
ts < tso
ωn < ωo

σ2 σ1
Root Locus method
1
• Ejemplo: G(s) =
s(s + 2)

• Position error ep=0

• Overshoot Mp<5%

• Settling time ts<9 s

• Rise time as small as possible


Root Locus method
1
• Ejemplo: G(s) =
(s +1)

• Position error ep=0

• Overshoot Mp<5%

• Settling time ts<2 s

• Rise time as small as possible


Root Locus method
2
• Ejemplo: G(s) =
s(s +1)(s + 5)

• Position error ep=0

• Overshoot Mp<5%

• Settling time ts<5 s

• Rise time as small as possible