CS-4 +WE Additional Exmple Included On RL
CS-4 +WE Additional Exmple Included On RL
Chalachew W.
CONTENTS
Frequency Response of a System
Bode Plot
Nyquist (1932) and Bode (1945) are based on frequency response, while Root-locus (1948) is
based on time response of a system.
Compared to time domain analysis frequency domain analysis helps to determine transfer
function of a physical system easily and accurately. It is also easy to visualize the effects of noise
and disturbance and parameter variations. The main drawback of frequency domain is it is not
as intuitive as root locus.
FREQUENCY RESPONSE OF A SYSTEM CONT.…
ω ω
r(t) G(s) c(t) R(s)=A 2 +ω2 +=A
( + ω)( − ω)
r(t)=Asin(ωt) ( )
G(s)=
( )
r(t)input to the system:- sinusoidal
signal amplitude ‘A’ and frequency ‘ω’ D(s)= + + + …
( ) N(s)
=G(s) ⇒C(s)=
( ) + + + …( )( )
1 2 3 1 2
⇒C(s)=G(s) R(s) C(s)= + + + +
+ + + +ω ω
FREQUENCY RESPONSE OF A SYSTEM CONT.…
Taking inverse LaPlace transform Transient response decays to
zero as time goes to infinity
This shows as long as the system is stable (i.e. a, b, c…positive) the steady state response becomes
c(t)= = +
C(s)= = + =G(s)R(s)=G(s)A
( )( )
( ) ( )
Using partial fractions = and =
FREQUENCY RESPONSE OF A SYSTEM CONT.…
( ) ( )
C(s)= + c(t)=A|G(j )| +
( ) ( )
( ) ( )
C(s)= + c(t)=A|G(j )|
G(-j )=|G(j )|
FREQUENCY RESPONSE OF A SYSTEM conclusion
If a linear, time-invariant and stable system
c(t)=A|G(jω)| sin( + )
subjected to a sinusoidal input the output will
be a sinusoidal output of same frequency
r(t)=Asin( )
with different amplitude(A|G(jω)|) and phase
shifted by ϴ at steady state.
CORRELATION BETWEEN TIME AND FREQUENCY RESPONSE
Frequency domain specifications
Resonant peak, : The maximum value of the magnitude |G(jω)| of closed loop transfer function is
called the resonant peak.
peak overshoot in time domain
Resonant frequency, : The frequency at which the resonant peak occurs is called resonant frequency.
frequency of oscillation in time domain
Bandwidth, : frequency at which the magnitude of |G(jω)|drops by 3dB below 0dB level. [ : cut-off
frequency, frequency at which the gain is -3dB],
large BW-small rise time which is fast response
Cut-off rate: the slope of the log-magnitude curve near the cut-off frequency. [ability of a system to
distinguish the signal from noise]
CORRELATION BETWEEN TIME AND FREQUENCY RESPONSE
Frequency domain specifications cont.…
General second order system 1
=
S-domain 2 1−
G(s)=
= 1−2
Frequency domain
G(jω)=
( ) = 1−2 + 2−4 +4
where u=
|G(jω)|= (1 − ) +4
GAIN AND PHASE MARGIN
Gain margin, GM: value of gain(k) added to the system in order to bring the system to the verge of
instability. Given by reciprocal of the magnitude of open loop transfer function at phase cross over
frequency ( ). [ frequency at which the phase of open loop transfer function is 180⁰.]
GM=| |
Phase margin, : value of phase(ϴ) added to the system in order to bring the system to the verge of
instability. Obtained by adding 180⁰ to the phase angle ϴ of the open loop transfer function at the gain
cross over frequency( ). [ frequency at which the magnitude of open loop transfer function is unity.]
=180⁰+<G(j )
BODE(LOGARITHMIC) PLOT
Bode plot is
a. A plot of the logarithmic (base 10) of the magnitude in decibels of sinusoidal transfer function versus
frequency in logarithmic scale (that is why we call it logarithmic plot) and
b. A plot of its’ phase angle versus frequency in logarithmic scale.
Given G(s)
Magnitude plot
Step 1: convert into frequency domain (i.e. substituting ‘jω’ in place of ‘s’) G(jω)
Step 2: derive magnitude of G(jω) and convert into dB:dB=20log |G jω |
Step 3: mark the points dB for values of ω=[0, 0.1, 1, 10, 100, ∞] and for the corner points then connect the
points.
BODE(LOGARITHMIC) PLOT
Phase plot
Step 5: mark the points angle for values of ω=[0, 0.1, 1, 10, 100, ∞] and for the corner points then connect
the points.
NYQUIST (POLAR) PLOT
The polar plot of a sinusoidal transfer function G(jω) is a plot of the magnitude of G(jω) versus
the phase angle of G(jω) on polar coordinates as ω is varied from zero to infinity.
For the given G(s)
Step 1: convert into frequency domain (i.e. substituting ‘jω’ in place of ‘s’) G(jω)
Step 2: derive magnitude of G(jω) |G(jω)|
Step 3: derive angle of G(jω) <G(jω)
Step 4: mark the points |G(jω)|<G(jω) for values of ω=[0, ∞]