EACT633-N ONLINEAR C ONTROL S YSTEM

C HAPTER 2 :: D ESCRIBING F UNCTION A NALYSIS

Dr. Saravanakumar Gurusamy

Associate Professor,
Department of Electrical and Electronics Technology,
FDRE Technical and Vocational Training Institute,
Addis Ababa, Ethiopia

December 15, 2023

 C HAPTER 2 - D ESCRIBING F UNCTION

1 Nonlinearitiess . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

2 Describing Function Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

3 Describing Function Analysis for the Saturation Nonlinearity . . . . . . . . . . . . . . . . 11

4 Describing Function Analysis for the Dead zone Nonlinearity . . . . . . . . . . . . . . . . 18

5 Describing Function Analysis for the Other types Nonlinearity . . . . . . . . . . . . . . . . 23

D R .S ARAVANAKUMAR G URUSAMY EACT631 1 / 25

 T YPES OF N ONLINEARITIES

D R .S ARAVANAKUMAR G URUSAMY EACT631 2 / 25

 D EAD ZONE N ONLINEARITY

▶ The dead-zone is the region in which the output is zero for a given
input. When the input is increased beyond this dead- zone value, the
output will be linear. The region where the applied voltage has no
effect is called the deadband e.g. sensors, friction in all components
▶ Dead zone nonlinearity may occur in sensors, amplifiers and
actuators.
▶ In a de motor, we assume that any voltage applied to armature
winding will cause armature to rotate if the field current is maintained
constant.
▶ In reality, due to static friction at the motor shaft, rotation will occur
only if the torque provided by the motor is sufficiently large. This
corresponds to dead zone.
▶ This phenomenon occurs also in valve controlled pneumatic and
hydraulic actuators.

D R .S ARAVANAKUMAR G URUSAMY EACT631 3 / 25

 B ACKLASH N ONLINEARITY

▶ Backlash nonlinearity commonly occurs in mechanical components of control

systems. In gear trains, small gap exist between a pair of mating gears.
▶ As a result, when the driving gear a smaller angle than the gap H, the driven gear
does not move at all, which corresponds to the dead zone (OA).
▶ After the contact is established between the two gears, the driven gear follows the
rotation of the driving gear in a linear fashion (AB segment).
▶ When the driving gear rotates in the reverse direction, by a distance of 2H, the driven
gear again does not move, corresponding to the segment BC.
▶ After the contact between the two gears is reestablished, the driven gear linearly
follows the rotation of driving gear in the reverse direction (CD segment).
▶ Thus, if the driving gear is in periodic motion, the driven gear will move in the
fashion represented by closed path EBCD

D R .S ARAVANAKUMAR G URUSAMY EACT631 4 / 25

 F RICTION N ONLINEARITY

▶ Friction exists in a system, when there is relative motion between

contacting surfaces.
▶ The different types of friction are viscous friction, coulomb friction
and stiction. The viscous fiction is linear in nature and the frictional
force is directly proportional to relative velocity of sliding surfaces.
▶ The coulomb friction is a drag force which opposes motion (For eg:
drag due to rubbing contact between brushes and the commutator).
▶ Due to interlocking of surface irregularities, more force is required to
move an object from rest than to maintain it in motion. Hence, force of
stiction is always greater than that of Coulomb friction.

D R .S ARAVANAKUMAR G URUSAMY EACT631 5 / 25

 L INEARIZATION M ETHOD

▶ Analytical linearization
▶ Graphic linearization
▶ Harmonic linearization
• Describing function
• Statistical linearization
• Dual describing function

D R .S ARAVANAKUMAR G URUSAMY EACT631 6 / 25

 D ESCRIBING F UNCTION I

▶ Describing Function Method consists of replacing a nonlinear element within a system by an

"equivalent" linear time-invariant system which is in some sense the best possible linear
approximation of the given nonlinear system.
▶ Provides a linear approximation to the nonlinear element based on the assumption that the
input to nonlinear element is a sinusoidal input of known constant amplitude.
▶ This method is often used to predict the existence of periodic solutions in feedback systems.
▶ The fundamental harmonic element’s output is compared with input sinusoidal to determine
steady state amplitude and phase relation.
▶ This relation is the describing function for the nonlinear element.
▶ Thus, the method can also be viewed as ’harmonic linearization’ of a non- linear element.

r e y
+
−
G1 (s) N G2 (s)

▶ Let x = X sin ωt be a sinusoidal input given to the nonlinear system.

D R .S ARAVANAKUMAR G URUSAMY EACT631 7 / 25

 D ESCRIBING F UNCTION II

▶ The output ’y’ of the nonlinear element will be a non-sinusoidal periodic function, which may
be expressed in terms of Fourier series.
y = A0 + A1 sin ωt + B1 cos ωt + A2 sin ωt + B2 cos ωt + A3 sin ωt + B3 cos ωt + . . .
▶ If the nonlinearity is symmetrical the average value of y is zero and hence the output y is given
by
y = A1 sin ωt + B1 cos ωt + A2 sin ωt + B2 cos ωt + A3 sin ωt + B3 cos ωt + . . .
▶ In the absence of external input (i.e., when r = 0) the output y of the nonlinearity N is feedback
to input through G1 (s) and G2 (s) in tandem.
▶ If G1 (s)G2 (s) has low pass characteristics, then all the harmonics of y are filtered, so that the
input x to the nonlinear element N is mainly contributed by fundamental component of y and
hence x remains sinusoidal.
▶ Under such conditions the harmonics of the output are neglected and the fundamental
component of y alone considered for the purpose of analysis.
y = y1 = A1 sin qωt + B1 cos ωt = Y1 < ϕ1 = Y1 sin(ωt + ϕ)
Where, Y1 = A21 + B21 and ϕ1 = tan−1 AB11
Y1 = Amplitude of the fundamental harmonics component of the output.
ϕ1 = Phase shift of the fundamental harmonic component of the output with respect to the
input.

D R .S ARAVANAKUMAR G URUSAMY EACT631 8 / 25

 D ESCRIBING F UNCTION III

▶ The coefficients A1 and B1 of the Fourier series are given by

Z 2π
2
A1 = y sin(ωt) d(ωt)
2π 0
Z 2π
2
B1 = y cos(ωt) d(ωt)
2π 0
▶ When the input x to the nonlinearity is sinusoidal, the describing function of the nonlinearity is
defined as,

Y1
KN (x, jω) =∠ϕ1
X
▶ The nonlinear element N in the system can be replaced by the following describing function.

r e y
+
−
G1 (jω) KN (x, jω) G2 (jω)

D R .S ARAVANAKUMAR G URUSAMY EACT631 9 / 25

 D ESCRIBING F UNCTION IV

▶ If the nonlinearity is replaced by the describing function then all linear theory frequency
domain techniques can be used for the analysis of the system.
▶ The describing functions are used only for stability analysis and it is not directly applied to the
optimization of the system design.

D R .S ARAVANAKUMAR G URUSAMY EACT631 10 / 25

 D ESCRIBING FUNCTION FOR THE SYSTEM WITH S ATURATION
N ONLINEARITY I

The input-output relation is linear for x = 0 to S. When the input x > S,

the output reaches a saturated value of KS. The input is a sinusoidal,

x = X sin ωt (1)

where X is the maximum value of input. When ωt = β, x = S Hence

equation (1) can be written as,

S = X sin β (2)
 
S −1 S
sin β = or β = sin (3)
X X

D R .S ARAVANAKUMAR G URUSAMY EACT631 11 / 25

 D ESCRIBING FUNCTION FOR THE SYSTEM WITH S ATURATION
N ONLINEARITY II

The output y of the nonlinearity can be

divided into three regions in a period of π.
The output equation for the three regions are
given by equi (4)

Kx 0 ≤ wt ≤ β

y = KS β < wt < (π − β) (4)

Kx (π − β) < wt < π


D R .S ARAVANAKUMAR G URUSAMY EACT631 12 / 25

 D ESCRIBING FUNCTION FOR THE SYSTEM WITH S ATURATION
N ONLINEARITY III

Let Y1 = Amplitude of the fundamental harmonic component of the output ϕ1 = Phase shift of the
fundamental harmonic component of the output with respect to the input
The describing function is given by,

KN (X, ω) = (Y1 /X)∠ϕ1

where p
Y1 = A2 + B2 and ϕ1 = − tan−1 (B/A).
The output y has half wave and quarter wave symmetries.
Therefore, B = 0 and
Z π/2
2
A1 = y sin(ωt)d(ωt) (5)
π/2 0
The output, y is given by two different expressions in the period 0 to π/2. Hence equation (5) can be
written as shown in (6)
Z β Z π/2
4 4
A1 = y sin(ωt)d(ωt) + y sin(ωt)d(ωt) (6)
π 0 π β

D R .S ARAVANAKUMAR G URUSAMY EACT631 13 / 25

 D ESCRIBING FUNCTION FOR THE SYSTEM WITH S ATURATION
N ONLINEARITY IV

On substituting the values of y from equ (5) in equ (6), we get,

π
Z β Z
4 4 2
A1 = Kx sin(ωt)d(ωt) + KS sin(ωt)d(ωt)
π 0 π β

On substituting x = X sin(ωt), we get,

π
Z β Z
4K 4KS 2
A1 = X sin(ωt)x sin(ωt)d(ωt) + sin(ωt)d(ωt)
π 0 π β

Z π
4KX β 2
Z
4KS 2
= sin (ωt)d(ωt) + sin(ωt)d(ωt)
π 0 π β

4KX β 1 − cos 2ωt 4KS β

Z   Z
d(ωt) + sin ωtd(ωt)
π 0 2 π 0

sin 2ωt β
 
2KX
= ωt − + 4KS − 4KS = 0
π 2 0

2KX sin 2β
= [β − ] + 4KS − cos π/2 + cos β
π 2
D R .S ARAVANAKUMAR G URUSAMY EACT631 14 / 25
 D ESCRIBING FUNCTION FOR THE SYSTEM WITH S ATURATION
N ONLINEARITY V

2KX
= [β − sin 2β] + 4KS cos β (7)
π
On substituting for S, (i.e, S = X sin β from equ. (2) in equ. (7) we get,
 
2KX sin 2β 4K
A1 = β− + X sin β cos β
π 2 π
 
2KX 2 sin βcosβ 4KX
A1 = β− + sin β cos β
π 2 π
2KX
A1 = [β − sinβcosβ + 2sinβcosβ]
π
2KX
A1 = [β + sinβcosβ] (8)
π
2KX
q q
Y1 = A21 + B21 = A21 + 0 = A1 = [β + sinβcosβ] (9)
π
B1
ϕ1 = arctan( ) = arctan 0 = 0 (10)
A1
D R .S ARAVANAKUMAR G URUSAMY EACT631 15 / 25
 D ESCRIBING FUNCTION FOR THE SYSTEM WITH S ATURATION
N ONLINEARITY VI

using eq (10 and 11), the describing function can be written as

Y1 2KX
KN (x, jω) = ∠ϕ1 = [β + sinβcosβ] (12)
X π
Depending on the maximum value of input X, the describing function can be written as,
π
If X < S, then β = , KN (X, ∞) = K (13)
2

2K
If X > S, KN (X, ∞) = [β + sin β cos β] (14)
π
The equation (14) can be expressed in another form as shown below. From equi (2) we get,

S
S = X sin β, then sin β = (15)
X

D R .S ARAVANAKUMAR G URUSAMY EACT631 16 / 25

 D ESCRIBING FUNCTION FOR THE SYSTEM WITH S ATURATION
N ONLINEARITY VII

On constructing right angle triangle with unity hypotenuse as shown

cosβ can be evaluated.
in fig.,q

2
adj = 1 − XS ,
q
2
adj
cosβ = hyp , adj = 1 − XS
In the describing function of equ(14), substitute for β, sin β and cos β
from equations(3),(15) and (16)
 s 
     2
2K  −1 S S S 
KN (X, ω) = sin + 1− forX > S (17)
π X X X

D R .S ARAVANAKUMAR G URUSAMY EACT631 17 / 25

 D EADZONE N ONLINEARITY

D R .S ARAVANAKUMAR G URUSAMY EACT631 18 / 25

 D EADZONE N ONLINEARITY I

D
▶ Output = 0, when the input is less than 2 or greater than − D2 .
▶ Input-Output relationship is linear when the Input is greater than D2 .
▶ Input-Output characteristic of Response to sinusoidal input (X Sin ωt) is shown in figure
dead-zone nonlinearity.
Equation (1): x = X sin ωt where X is the maximum value of input. When ωt = α, x = D/2 ⇒
D
sin α = 2X and α = sin−1 2XD
. the output equation

0
 0 ≤ wt ≤ α
y = K(x − D
 2) α < wt < (π − α)
0 (π − α) < wt < π


Let Y1 = Amplitude of the fundamental harmonic component of the output ϕ1 = Phase shift of the
fundamental harmonic component of the output with respect to the input
The describing function is given by,

KN (X, ω) = (Y1 /X)∠ϕ1

where p
Y1 = A2 + B2 and ϕ1 = − tan−1 (B/A).

D R .S ARAVANAKUMAR G URUSAMY EACT631 19 / 25

 D EADZONE N ONLINEARITY II

The output y has half wave and quarter wave symmetries.

Z π
2 2
B1 = 0 and A1 = y sin ωtd(ωt)
π/2 0
Z π  
4 2 D
A1 = K x− sin ωd(ωt)
π α 2
Since the output, y, is zero in the range, 0 ≤ ωt ≤ α, the limits of integration in equation (6) can be
changed to α to π/2 instead of 0 to π/2. Put x = X sin ωt in equation (7):
"Z #
π/2  
4K D
A1 = X sin ωt − sin ωtd(ωt)
π α 2
"Z #
π/2 Z π/2
4K D
= X sin2 ωtd(ωt) − sin ωtd(ωt)
π α 2 α
" Z #
4K X π/2 D π/2
Z
= (1 − cos 2ωt)d(ωt) − sin ωtd(ωt)
π 2 α 2 α

D R .S ARAVANAKUMAR G URUSAMY EACT631 20 / 25

 D EADZONE N ONLINEARITY III

"  π #
4K X sin 2ωt 2 D π
= ωt − − [− cos ωt]α2
π 2 2 α 2
   
4K X π sin π sin 2α D π
= − −α+ − − cos + cos α
π 2 2 2 2 2 2
   
4K X π sin 2α D
= −α+ − (cos α)
π 2 2 2 2

   
4K X π sin 2α D
A1 = −α+ − (cos α)
π 2 2 2 2
 
2
Y1 = KX 1 − (α + sin α cos α)
π
ϕ1 = 0

D
From equation (3) we get, sin α = 2X ⇒ D = 2X sin α. On substituting for D from equation (9) in
equation (8):

D R .S ARAVANAKUMAR G URUSAMY EACT631 21 / 25

 D EADZONE N ONLINEARITY IV

   
4KX π α 2 sin α cos α 4KX π α sin α cos α
A1 = − + − sin α cos α − − −
π 4 2 4 π 4 2 2
 
2
= KX 1 − (α + sin α cos α)
π
 
2
q q
2 2 2
Y1 = A1 + B1 = A1 + 0 = A1 = KX 1 − (α + sin α cos α)
π
B1
ϕ1 = tan−1 = tan−1 0 = 0
A1
D
sinα = X
2
s
D 2
 
adj
cosα = = 1−
Hyp DX
substitute α,sin α, and cos α then,
  s 
     2
2  −1 D D D
KN (X, ω) = K 1 − sin + 1−  for X > S (17)
π 2X 2X 2X

D R .S ARAVANAKUMAR G URUSAMY EACT631 22 / 25

 D EADZONE AND S ATURATION N ONLINEARITY

D R .S ARAVANAKUMAR G URUSAMY EACT631 23 / 25

 B ACKLASH N ONLINEARITY

D R .S ARAVANAKUMAR G URUSAMY EACT631 24 / 25

 R ELAY WITH D EAD ZONE AND H YSTERESIS N ONLINEARITY

D R .S ARAVANAKUMAR G URUSAMY EACT631 25 / 25

 End of Chapter 2

D R .S ARAVANAKUMAR G URUSAMY EACT631 25 / 25