Design via frequency response Transient response via gain adjustment Consider a unity feedback system, where G(s) = - The closed loop transfer function is T (s) = ω s + 2ζωs + ω

Download as pdf or txt
Download as pdf or txt
You are on page 1of 12

Design via frequency response

Transient response via gain adjustment

Consider a unity feedback system, where G(s) =


2
ωn
s(s+2ζωn)
. The closed loop transfer function is
2
ωn
T (s) = 2 2
s + 2ζωs + ωn

Figure above; The time response of the second


order underdamped system
1
The percentage overshoot, %OS, is given by
cmax − cf inal
%OS =
cf inal
Note that %OS is a function only of the damp-
ing ratio, ζ.

−(ζπ/ 1−ζ 2)
%OS = e × 100
The inverse is given by
− ln(%OS/100)
ζ=q
π 2 + ln2(%OS/100)
There is also a relationship between damping
ratio and phase margin. The phase margin is
obtained by solving |G(jω)| = 1 to obtain the
frequency as
r q
ω1 = ωn −2ζ 2 + 1 + 4ζ 4
The phase margin is

ΦM = arctan r q
−2ζ 2 + 1 + 4ζ 4

2
Thus if we can vary the phase margin, we can
vary the percent overshoot, via a simple gain
adjustment.

Figure above; Bode plots showing gain adjust-


ment for a desired phase margin.

3
Problem: For the position control system shown
below, find the value of preamplifier gain, K,
to yield a 9.5% overshoot in the transient re-
sponse for a step input. Use only frequency
response methods.

Figure above; Bode plots for the example above.


4
100K
Solution: 1. G(s) = s(s+36)(s+100) . Choose
K = 3.6 to start the magnitude plot at 0dB.

2. Use
− ln(%OS/100)
ζ=q
π 2 + ln2(%OS/100)
to find ζ = 0.6 for %OS/100 = 0.095, and then
use

ΦM = arctan r q
−2ζ 2 + 1 + 4ζ 4
to find ΦM = 59.2◦ for ζ = 0.6.

3. Locate on the phase plot the frequency that


yields a 59.2◦ margin. This means −120.8◦
phase angle at frequency of 14.8 rad/s.

4. At 14.8 rad/s on the magnitude plot, the


gain is found to be -44.2dB. This magnitude
has to be raise to 0dB to yield the required
phase margin. 44.2dB increase is 162.2 in
gain, Thus K = 3.6 × 162.2 = 583.9.
5
Lag compensation

The steady error constants are:

position constant

Kp = lim G(s),
s→0
velocity constant

Kv = lim sG(s),
s→0
and acceleration constant

Ka = lim s2G(s).
s→0
The value of the steady-state error decreases
as the steady error constants increases.

The function of the lag compensator is to (i)


improve the steady error constant by increas-
ing only the low-frequency gain without any
resulting instability, and (2) increase the phase
margin of the system to yield the desired tran-
sient response.
6
Visualizing lag compensation

The transfer function of the lag compensator


is
s + T1
Gc(s) = 1
s + αT
where α > 1.

Figure above; Frequency response of a lag com-


s+0.1
pensator s+0.01

7
In the figure below, the uncompensated system
is unstable since the gain at 180◦ is greater
than 0dB. The lag compensator, while not
changing the low-frequency gain, does reduce
the high frequency gain. The magnitude curve
can be shaped to go through 0dB at the de-
sired phase margin to obtain the desired tran-
sient response.

Figure above; visualizing lag compensation

8
Design procedure

1. Set the gain,K, to the value that satis-


fies the steady-state error specification and
plot the Bode plots based on this gain.

2. Find the frequency where the phase margin


is 5◦ to 12◦ greater than the phase margin
that yields the desired transient response.

3. Select a lag compensator whose magnitude


response yields a composite Bode diagram
that goes 0dB at the frequency found in
step 2; (For detail see the next example)

4. Reset the system gain, K, to compensate


for any attenuation in the lag network in
order to keep the static error constant the
same as that found in step 1.

9
Problem: For the same previous position con-
trol system, use Bode Diagrams to design a lag
compensator to yield a a ten-fold improvement
in steady-state error over the gain-compensated
system while keeping the percent overshoot at
9.5%.

Solution: 1. The gain-compensated system is


58390
G(s) =
s(s + 36)(s + 100)
Thus Kv = 16.22, a ten-fold improvement changes
Kv = 162.2. Thus K needs to be set at K =
5839, and the open loop transfer function
583900
G(s) =
s(s + 36)(s + 100)

10
Figure above; Bode plots showing lag compen-
sator design.

2. The phase margin required for %OS/100 =


0.095 (ζ = 0.6) is ΦM = 59.2◦. We increase
this value by 10◦ to ΦM = 69.2◦ (i.e. phase
angle −110.8◦). The frequency is 9.8rad/s.
At this frequency the magitude plot must go
through 0dB. The magnitude at 9.8rad/s is
24dB. Thus the lag compensator must provide
-24dB attenuation at 9.8rad/s.
11
3.&4. We now design the compensator. First
draw the high frequency asymptote at -24dB.
Arbitrarily select the higher break frequency to
be about one decade below the phase margin
frequency, i.e. at 0.98rad/s, to start drawing a
-20dB/decade until 0dB is reached. The lower
break frequency is found to be 0.062rad/s.
Hence the lag compensator’s transfer function
is
0.063(s + 0.98)
Gc(s) =
(s + 0.062)
where the gain 0.063 of the compensator is set
to achieve a dc gain of unity.

The compensated system’s forward transfer func-


tion is thus
36786(s + 0.98)
Gc(s)G(s) =
s(s + 36)(s + 100)(s + 0.062)

12

You might also like