MECE 3350- Control Systems
Instructor: Dr. Shabnam Pejhan
Lecture 5- Effect of Pole Locations
Learning objectives
• By the end of this lecture, you
✓ will gain a good understanding of how the location of poles influence the temporal response of a given
system.
✓ will understand the theory of final value and concepts of steady state value and transient response.
3
Time Response
Reference
Altitude Actual Altitude
Actuator
Controller
(motors)
System
Sensor
Altitude Overshoot?
Steady state value or error?
Rise time?
Settling time?
Time
Transient Steady State
Transfer Function- Poles and Zeros
Input Output
𝐹(𝑆) 𝐶 𝑆 = 𝐹 𝑆 𝐺(𝑆)
G(S)
𝐶(𝑠) 𝑎𝑘 𝑠 𝑘 + 𝑎𝑘−1 𝑠 𝑘−1 + ⋯ + 𝑎1 𝑠 + 𝑎0
𝐺 𝑠 = = 𝑝>𝑘
𝐹(𝑠) 𝑏𝑝 𝑠 𝑝 + 𝑏𝑝−1 𝑠 𝑝−1 + ⋯ + 𝑏1 𝑠 + 𝑏0
The roots of 𝐹(𝑠) = 0 are the Poles (𝑝𝑖 ) of the transfer function. lim 𝐺 𝑠 = ∞
𝑠→𝑝𝑖
The roots of 𝐶(𝑠) = 0 are the Zeros (𝑧𝑖 ) of the transfer function. lim 𝐺 𝑠 = 0
𝑧→𝑝𝑖
5
Lecture 3 notes are revised.
6
Poles and Zeros on S Plane
𝑪(𝒔) 𝒔(𝒔 + 𝟑)
𝑮 𝒔 = = 𝟐
𝑭(𝒔) 𝒔 + 𝟐𝒔 + 𝟓
Zeros (𝑧𝑖 ) of the transfer function: 𝑠(𝑠 + 3)=0 ----- > 0, -3
Poles (𝑝𝑖 ) of the transfer function: 𝑠 2 + 2𝑠 +5=0 ----- > -1+2j , -1-2j
Imaginary axis
(ω)
Real axis
(σ)
7
Transient Response of a First Order System- Step Input
Input Output
𝐹(𝑠) 𝐼 𝑠 = 𝐹 𝑠 𝐺(𝑠)
𝐺(𝑠)
𝟏 ONLY
𝑮 𝒔 =𝒌 Pole: −𝟏/ 𝝉
𝝉 𝒔+𝟏 Real!
𝑉 𝑠 = 𝑅 + 𝐿𝑠 𝐼(𝑠) 𝜏 = 𝐿/𝑅
1 1 1 𝑘 = 1/𝑅
𝐺 𝑠 = =
𝐿𝑠 + 𝑅 𝑅 𝐿 𝑠 + 1
𝑅 Imaginary axis (ω) (sec-1)
Unit step input Left Half S Plane Right Half S Plane
𝐼 𝑠 = 𝑉 𝑠 𝐺(𝑠)
𝝉>𝟎 𝝉<𝟎
𝐼 𝑠 = 𝑈 𝑠 𝐺(𝑠)
1 1 1
𝐼 𝑠 = Converges Diverges
𝜏𝑅 𝑠 𝑠 + 1/𝜏
1
𝑉 𝑠 = Real axis
𝑠 (σ) (sec-1)
1 𝑡
𝑖 𝑡 = (1 − 𝑒 −𝜏 )
𝑅
8
Transient Response of a First Order System- Impulse Input
Input Output
𝐹(𝑠) 𝐼 𝑠 = 𝐹 𝑠 𝐺(𝑠)
𝐺(𝑠)
𝟏 ONLY
𝑮 𝒔 =𝒌 Pole: −𝟏/ 𝝉
𝝉 𝒔+𝟏 Real!
𝑉 𝑠 = 𝑅 + 𝐿𝑠 𝐼(𝑠) 𝜏 = 𝐿/𝑅
1 1 1 𝑘 = 1/𝑅
𝐺 𝑠 = =
𝐿𝑠 + 𝑅 𝑅 𝐿 𝑠 + 1
𝑅 Imaginary axis (ω) (sec-1)
Unit impulse input Left Half S Plane Right Half S Plane
𝐼 𝑠 = 𝐹 𝑠 𝐺(𝑠)
𝝉>𝟎 𝝉<𝟎
𝐼 𝑠 = 𝑈 𝑠 𝐺(𝑠)
1 1
𝐼 𝑠 = 1 Converges Diverges
𝑅 𝜏𝑠 + 1
𝑉 𝑠 =1 1 1 Real axis
= (σ) (sec-1)
𝜏𝑅 𝑠 + 1/𝜏
1 −𝑡 9
𝑖 𝑡 = (𝑒 𝜏 )
𝜏
First Order Transfer Functions Lecture 4
Standard Form k: Gain of system
𝝉: Time constant of the system: characterizes the
response to a step input of a first-order system.
1
𝑌(𝑠) 𝜏 𝑦 𝑡 = 1 − 𝑒 −𝑡/𝜏
𝐺 𝑠 = Y 𝑠 = 𝑈 𝑠 𝐺(𝑠) =
𝑈(𝑠)
1 1 𝑘 𝑘𝜏
𝑈 𝑠 = Y 𝑠 = 𝑘 𝑠(𝜏𝑠+1) = 𝑠 − 𝜏𝑠+1
𝑠
𝑘 𝑘𝜏
ℒ −1 {𝑌 𝑠 }= ℒ −1 −ℒ −1
𝑠 IF 𝝉 >0
𝜏𝑠 + 1
𝜏 2𝜏 3𝜏 4𝜏 5𝜏
Or poles are on LHS When k=1, y(t) is initially zero and gets to one
@t=time constant, y(t) reached 63.2% of its total change.
𝑒 −𝑡/𝜏 : Smaller Time Constant, the faster the system
response!
10
Transient Response of 2nd Order System-Poles
𝒌
𝑭(𝒕)
11 𝜔𝑛2 Natural frequency: 𝝎𝒏 =
𝒌
𝒎 𝑋 𝑠 = 𝒎
𝑠 𝑘 𝑠 2 + 2𝜁𝜔𝑛 𝑠 + 𝜔𝑛2 𝑏
Damping Ratio: 𝜻 =
𝒃 𝒙(𝒕) 2 𝑚𝑘
Poles of G(s): 𝒔𝟐 +𝟐𝜻𝝎𝒏 𝒔 + 𝝎𝟐𝒏 = 𝟎
−𝟐𝜻𝝎𝒏 ± 𝟐𝜻𝝎𝒏 𝟐 − 𝟒𝝎𝟐𝒏
𝒔=
𝟐
F 𝑠 = 1/𝑠
𝒔 = −𝜻𝝎𝒏± 𝝎𝒏 𝜻𝟐 − 𝟏
𝒔 = 𝝎𝒏 −𝜻 ± 𝜻𝟐 − 𝟏 Real and or
imaginary!
11
Transient Response of 2nd Order System-Poles
𝜔𝑛2 Natural frequency: 𝜔𝑛
𝐺 𝑠 = 2
𝑠 + 2𝜁𝜔𝑛 𝑠 + 𝜔𝑛2 Damping Ratio: 𝜁
𝒔 = 𝝎𝒏 −𝜻 ± 𝜻𝟐 − 𝟏
Case1- Over-damped 2nd order system (𝜻 > 𝟏) 𝜻𝟐 − 𝟏 > 𝟎
Case2- Critically damped 2nd order system (𝜻 = 𝟏) 𝜻𝟐 − 𝟏 = 𝟎
Case3- Under-damped 2nd order system (𝟎 < 𝜻 < 𝟏) 𝜻𝟐 − 𝟏 < 𝟎
Case4- Undamped 2nd order system (𝜻 = 𝟎) 𝜻𝟐 − 𝟏 < 𝟎 12
Transient Response of 2nd Order System-
Over-damped system (𝛇 > 𝟏) response to a unit step input
𝒔𝟏 = −𝝎𝒏 𝜻+ 𝝎𝒏 𝜻𝟐 − 𝟏 s1 𝑎𝑛𝑑 𝑠2
Poles of G(s): 𝒔 = 𝝎𝒏 −𝜻 ± 𝜻𝟐 −𝟏 ∈ ℝ 𝑎𝑛𝑑 𝑁𝑒𝑔𝑎𝑡𝑖𝑣𝑒:
𝒔𝟐 = −𝝎𝒏 𝜻- 𝝎𝒏 𝜻𝟐 − 𝟏 only exponential
temporal response
11 𝜔𝑛2
𝑋 𝑠 =𝑈 𝑠 𝐺 𝑠 =
𝑠 𝑘 𝑠 2 + 2𝜁𝜔𝑛 𝑠 + 𝜔𝑛2
11 𝜔𝑛2
=
𝑘 𝑠 (𝑠 − 𝑠1 )(𝑠 − 𝑠2 )
F 𝑠 = 𝑈 𝑠 = 1/𝑠
𝐾1 =
1
𝑥 𝑡 = (1 + 𝐾1 𝑒 −𝑠1𝑡 + 𝐾2 𝑒 −𝑠2𝑡 )
𝑘
𝐾2 = 13
Transient Response of 2nd Order System-
Over-damped system (𝛇 > 𝟏) response to a unit step input
𝒌
𝑭(𝒕) For example: m=1 Kg; k=1 N/m; b=3 Ns/m → ωn=1 and ζ=1.5
𝒎
s1=-0.38 and s2=-2.62
𝒃 𝒙(𝒕)
Left Half S Plane
𝑘
Natural frequency: 𝜔𝑛 =
𝑚
𝑏
Damping Ratio: 𝜁 =
2 𝑚𝑘
𝒔𝟏 = −𝝎𝒏 𝜻+ 𝝎𝒏 𝜻𝟐 − 𝟏
𝒔𝟐 = −𝝎𝒏 𝜻- 𝝎𝒏 𝜻𝟐 − 𝟏
14
Transient Response of 2nd Order System-
Critically-damped system (𝛇 = 𝟏) response to a unit step input
Poles of G(s): 𝒔 = 𝝎𝒏 −𝜻 ± 𝜻𝟐 − 𝟏
s1=s2 and REAL
and NEGATIVE:
𝒔𝟏 = 𝒔𝟐 = −𝝎𝒏 𝜻+ 𝝎𝒏 𝜻𝟐 − 𝟏 = −𝝎𝒏 Only exponential
temporal response
11 𝜔𝑛2 1 𝜔𝑛2
𝑋 𝑠 =𝑈 𝑠 𝐺 𝑠 = 2 2 = 2
𝑘 𝑠 𝑠 + 2𝜔𝑛 𝑠 + 𝜔𝑛 𝑠 𝑠 + 𝜔𝑛
1
𝑥 𝑡 = (1 − 𝑒 −𝜔𝑛𝑡 1 + 𝜔𝑛 𝑡 )
𝑘
15
Transient Response of 2nd Order System-
Critically-damped system (𝛇 = 𝟏) response to a unit step input
𝒌
𝑭(𝒕) For example: m=1 Kg; k=1 N/m; b=2 Ns/m → ωn=1 and ζ=1
𝒎
s1=-1
𝒃 𝒙(𝒕)
Left Half S Plane
𝑘
Natural frequency: 𝜔𝑛 =
𝑚
𝑏
Damping Ratio: 𝜁 =
2 𝑚𝑘
𝒔𝒑𝒐𝒍𝒆 = −𝝎𝒏
16
Over-damped (𝛇 > 𝟏) to Critically damped System (𝛇 = 𝟏)
response to a unit step input
For example: m=1 Kg; k=1 N/m 𝑠1 = −𝜔𝑛 𝜁+𝜔𝑛 𝜁2 − 1 𝑠2 = −𝜔𝑛 𝜁- 𝜔𝑛 𝜁2 − 1
17
Transient Response of 2nd Order System-
Under-damped system (𝟎 < 𝛇 < 𝟏) response to a unit step input
Poles of G(s): 𝒔 = 𝝎𝒏 −𝜻 ± 𝜻𝟐 − 𝟏 s1 and s2 have Real
component +
𝒔𝟏 = −𝝎𝒏 𝜻+ 𝒋𝝎𝒏 𝜻𝟐 − 𝟏 Imaginary component
(Real component<0):
𝒔𝟐 = −𝝎𝒏 𝜻- 𝒋𝝎𝒏 𝜻𝟐 − 𝟏 Exponential +Sinusoidal
temporal response
11 𝜔𝑛2
𝑋 𝑠 =𝑈 𝑠 𝐺 𝑠 =
𝑘 𝑠 𝑠 − 𝑠1 𝑠 − 𝑠2
1 1
𝑥 𝑡 = [1 − 𝑒 −𝜁𝜔𝑛 𝑡 sin 𝜔𝑛 1 − 𝜁 2 𝑡 + cos −1 𝜁 ]
𝑘 1 − 𝜁2
18
Transient Response of 2nd Order System-
under-damped system (0 < 𝛇 < 𝟏) response to a unit step input
𝒌
𝑭(𝒕) For example: m=1 Kg; k=1 N/m; b=0.5 Ns/m → ωn=1 and ζ=0.25
𝒎 s2=-0.25-j0.87
s1=-0.25+j0.87
𝒃 𝒙(𝒕) Left Half S Plane
𝑘
Natural frequency: 𝜔𝑛 =
𝑚
𝑏
Damping Ratio: 𝜁 =
2 𝑚𝑘
𝒔𝟏 = −𝝎𝒏 𝜻+j𝝎𝒏 𝜻𝟐 − 𝟏
𝒔𝟐 = −𝝎𝒏 𝜻-j 𝝎𝒏 𝜻𝟐 − 𝟏
19
Transient Response of 2nd Order System-
under-damped system (0 < 𝛇 < 𝟏) response to a unit step input
For example: m=1 Kg; k=1 N/m; 𝑠1 = −𝜔𝑛 𝜁+j𝜔𝑛 𝜁2 − 1 𝑠2 = −𝜔𝑛 𝜁-j 𝜔𝑛 𝜁2 − 1
20
Transient Response of 2nd Order System-
Undamped system (𝛇 = 𝟎) response to a unit step input
Poles of G(s): 𝒔 = 𝝎𝒏 −𝜻 ± 𝜻𝟐 − 𝟏
Two Imaginary poles:
ONLY Sinusoidal
𝒔𝟏 = +𝒋𝝎𝒏
temporal response
𝒔𝟐 =- 𝒋𝝎𝒏
11 𝜔𝑛2
𝑋 𝑠 =𝑈 𝑠 𝐺 𝑠 =
𝑘 𝑠 𝑠 2 + 𝜔𝑛2
1
𝑥 𝑡 = [1 − cos 𝜔𝑛 𝑡 ]
𝑘
21
Transient Response of 2nd Order System-
undamped system (𝛇 = 𝟎) response to a unit step input
𝒌
𝑭(𝒕) For example: m=1 Kg; k=1 N/m; b=0 Ns/m → ωn=1 and ζ=0
𝒎
s1=0+1j s2=0-1j
𝒃 𝒙(𝒕)
𝒌
Natural frequency: 𝝎𝒏 =
𝒎
𝑏
Damping Ratio: 𝜁 =
2 𝑚𝑘
𝒔𝟏 = +𝒋𝝎𝒏
𝒔𝟐 = −𝒋𝝎𝒏
The frequency of oscillation of undamped system is called Natural Frequency
22
Natural Frequency and Pole Locations
Natural Frequency: the frequency of oscillation of the system when ζ =0 (NO DAMPING!)
Let’s assume k=1 N/m
𝒔𝟏 = +𝒋𝝎𝒏
𝒔𝟐 =- 𝒋𝝎𝒏
23
CASE 5! Hypothetical Negative Damping 𝜻 < 𝟎
−1 < 𝜻 < 𝟎 𝜻 ≤ −𝟏
Poles of G(s): 𝒔 = 𝝎𝒏 −𝜻 ± 𝜻𝟐 − 𝟏 −𝝎𝒏 𝜻 ± j𝝎𝒏 𝜻𝟐 − 𝟏 −𝝎𝒏 𝜻 ± 𝝎𝒏 𝜻𝟐 − 𝟏
𝒔𝟏 = −𝝎𝒏 𝜻+ 𝝎𝒏 𝜻𝟐 − 𝟏
𝒔𝟐 = −𝝎𝒏 𝜻- 𝝎𝒏 𝜻𝟐 − 𝟏
11 𝜔𝑛2
𝑋 𝑠 =𝑈 𝑠 𝐺 𝑠 =
𝑘 𝑠 𝑠 2 + 𝜔𝑛2
1 1
𝑥 𝑡 = [1 − 𝑒 𝜁 𝜔𝑛 𝑡 sin 𝜔𝑛 1 − 𝜁 2 𝑡 + cos −1 𝜁 ]
𝑘 1 − 𝜁2
24
Let’s Summarize!
Damping Poles Location System is called:
x
x
25
Location of Poles on S Plane (Magnitude and Angle)
𝜔𝑑 = 𝜔𝑛 𝟏 − 𝜻𝟐 𝜔𝑑 is called damped frequency
For a second order system, the poles are
Magnitude of s:
The angle to the imaginary axis (θ)
26
Steady State Value
Lecture 3
Final value theorem gives the steady-state value without computing the inverse transform.
If the function converges, i.e., the poles have negative real parts, then
Steady State Value :
27
Steady State Value
𝒌
𝑭 𝒕 = 𝑼𝒏𝒊𝒕 𝑺𝒕𝒆𝒑 𝑰𝒏𝒑𝒖𝒕
𝒎
𝒃 𝒙(𝒕)
𝒗 𝒕 = 𝑼𝒏𝒊𝒕 𝑰𝒎𝒑𝒖𝒍𝒔𝒆 𝑰𝒏𝒑𝒖𝒕
28
MATLAB
The code shared with you on Canvas.
29
MATLAB
R= 10
L=0.2
30
Next Lecture: Block Diagrams
31
