Nonlinear Systems and Control

Lecture # 2
Examples of Nonlinear Systems

p. 1/1

Pendulum Equation

mg

ml = mg sin kl
x1 = ,

x2 =

p. 2/1

x 1 = x2
x 2 =

g
l

sin x1

k
m

x2

Equilibrium Points:
0 = x2
0 =

g
l

sin x1

k
m

x2

(n, 0) for n = 0, 1, 2, . . .

Nontrivial equilibrium points at (0, 0) and (, 0)

p. 3/1

Pendulum without friction:

x 1 = x2
x 2 =

g
l

sin x1

Pendulum with torque input:

x 1 = x2
x 2 =

g
l

sin x1

k
m

x2 +

1
ml2

p. 4/1

Tunnel-Diode Circuit

i,mA
1

i=h(v)
0.5

0.5

iC = C

0.5

1 v,V

(b)

dvC
dt

vL = L

x1 = vC , x2 = iL ,

diL
dt

u=E

p. 5/1

iC + iR iL = 0 iC = h(x1 ) + x2
vC E + RiL + vL = 0 vL = x1 Rx2 + u
x 1 =
x 2 =

1
C
1
L

[h(x1 ) + x2 ]
[x1 Rx2 + u]

Equilibrium Points:
0 = h(x1 ) + x2
0 = x1 Rx2 + u

p. 6/1

h(x1 ) =
i

E
R

1
R

x1

R
1.2
1
0.8
0.6

Q
1

Q
2

0.4
Q
3

0.2
0
0

0.5

p. 7/1

MassSpring System

Ff

Fsp

m
y + Ff + Fsp = F

Sources of nonlinearity:
Nonlinear spring restoring force Fsp = g(y)
Static or Coulomb friction

p. 8/1

Fsp = g(y)
g(y) = k(1 a2 y 2 )y,

|ay| < 1 (softening spring)

g(y) = k(1 + a2 y 2 )y (hardening spring)

Ff may have components due to static, Coulomb, and
viscous friction

When the mass is at rest, there is a static friction force Fs

that acts parallel to the surface and is limited to s mg
(0 < s < 1). Fs takes whatever value, between its limits,
to keep the mass at rest
Once motion has started, the resistive force Ff is modeled
as a function of the sliding velocity v = y

p. 9/1

Ff

Ff

(b)

(a)

Ff

(c)

(d)

(a) Coulomb friction; (b) Coulomb plus linear viscous friction; (c) static, Coulomb, and linear
viscous friction; (d) static, Coulomb, and linear viscous frictionStribeck effect

p. 10/1

Negative-Resistance Oscillator

i = h(v)

(b)

h(0) = 0,

h (0) < 0

h(v) as v , and h(v) as v

p. 11/1

iC + iL + i = 0
C

dv
dt

v(s) ds + h(v) = 0

Differentiating with respect to t and multiplying by L:

CL

d2 v
dt2

+ v + Lh (v)

dv
dt

=0

= t/ CL
dv
d

CL

dv
dt

d2 v
d 2

= CL

d2 v
dt2

p. 12/1

Denote the derivative of v with respect to by v

v
+ h (v)v + v = 0,

L/C

Special case: Van der Pol equation

h(v) = v + 13 v 3
v
(1 v 2 )v + v = 0

State model: x1 = v,

x2 = v

x 1 = x2
x 2 = x1 h (x1 )x2

p. 13/1

Another State Model:

z1 = iL ,

z 1 =
z 2

Change of variables:

z2

= [z1 + h(z2 )]
z = T (x)

x1 = v = z2

dv
dv
x2 =
CL
=
=
d
dt
= [z1 h(z2 )]
T (x) =

z2 = vC

h(x1 ) 1 x2
x1

, T

L
C

[iL h(vC )]

(z) =

z2
z1 h(z2 )

p. 14/1

Adaptive Control
P lant :

y p = ap yp + kp u

Ref erenceM odel :

y m = am ym + km r

u(t) = 1 r(t) + 2 yp (t)

1

km
kp

and

am ap
kp

When ap and kp are unknown, we may use

u(t) = 1 (t)r(t) + 2 (t)yp (t)

where 1 (t) and 2 (t) are adjusted on-line

p. 15/1

Adaptive Law (gradient algorithm):

1 = (yp ym )r
2 = (yp ym )yp ,

>0

State Variables: eo = yp ym , 1 = 1 1 , 2 = 2 2
y m = ap ym + kp (1 r + 2 ym )
y p = ap yp + kp (1 r + 2 yp )
e o = ap eo + kp (1 1 )r + kp (2 yp 2 ym )
= + kp [2 yp 2 yp ]

= (ap + kp 2 )eo + kp (1 1 )r + kp (2 2 )yp

p. 16/1

Closed-Loop System:
e o = am eo + kp 1 r(t) + kp 2 [eo + ym (t)]
1 = eo r(t)
2 = eo [eo + ym (t)]

p. 17/1