Automatic Control

Lecture 6:
Block diagrams & Time response

Saeed Behbahani
behbahani@cc.iut.ac.ir

Department of Mechanical Engineering

Isfahan University of Technology
Transfer Function (review)

Y(s)=R(s).G(s)
 Block Diagrams
 A control system consists of several
components, with different functions
 Block diagram is a pictorial representation of
the functions of the components, and flow of
signals
 The main source of energy is not explicitly
shown
 The block diagram of a system is not unique
 Basics of Block Diagrams
• Summing point
• Branch point

• Closed loop system

Closed loop transfer function

Example
Example
 Example
Active Suspension System
𝑀.𝑥¨1=−𝐾1(𝑥1−𝑥2)−D(𝑥˙1−𝑥2)+𝑢(𝑡)

𝑚.𝑥¨2=𝐾1(𝑥− 2)+D(𝑥˙1− 2) 𝐾2(𝑥−𝑦𝑅) 𝑢(𝑡)

Draw a block diagram for the system and find the

transfer function between X1(s) and U(s).

𝑀𝑠.𝑋1(𝑠)=−𝐾1[𝑋(𝑠)−𝑋2(𝑠)]−Ds[𝑋1(𝑠)−𝑋2(𝑠)]+𝑈(𝑠)
2

𝑚.𝑠𝑋2()=𝐾1[𝑋(𝑠)−2 ]+Ds[𝑋1(𝑠)−2 ]−𝐾2[𝑋(𝑠)−𝑌𝑅 ]−𝑈(𝑠)

2

(𝑀𝑠2+𝐷𝑠 𝐾1).𝑋1(𝑠)=𝐷𝑠+𝐾1)𝑋2(𝑠)❑+𝑈(𝑠)
(𝑚𝑠2+𝐷 𝐾1+2).𝑋2(𝑠=𝐷𝑠+𝐾1)𝑋(𝑠+𝐾2𝑌𝑅(𝑠)−𝑈(𝑠)
 (𝑀𝑠2+𝐷𝑠 𝐾1).𝑋1(𝑠)=𝐷𝑠+𝐾1)𝑋2(𝑠)❑+𝑈(𝑠)

(𝑚𝑠2+𝐷 𝐾1+2).𝑋2(𝑠)=𝐷𝑠+𝐾1𝑋(𝑠)+𝐾2𝑌𝑅(𝑠)−𝑈(𝑠)
YR(s)
𝐾2

- + X2(s)
+ 1 1
(𝐷𝑠+𝐾1) 2
𝑚 𝑠 + 𝐷 𝐾 1+ 2
𝑀𝑠2+𝐷 𝐾1 + +
+ X1(s)
U(s)

(𝐷𝑠+𝐾1)

1 𝐷𝑠+𝐾1
222
−.
𝑋1 ( 𝑠) 𝑀 + 𝐷 𝐾 1 𝑠 + 1 𝑚 𝐷 𝐾 1 2
=
𝑈 ( 𝑠) 𝐷 + 𝐾1 𝑠 1
1− 2 . 2
𝑀 𝑠 + 𝐷 𝐾 1 𝑚 1+ 2
2
𝑋1 ( 𝑠) 𝑚 𝑠 + 𝐾 2
=
𝑈 ( 𝑠) ( 𝑀 2 + 𝐷 𝐾1 ) 𝑚 2 𝑠 1 + 2 − ( 𝐷 𝐾1 𝑠 )
Typical Step Response
 Influence for Pole Location
• If all poles are in LHP, the system is stable.
• For Stable First Order System:
 ‫‪Example‬‬
‫معادله دیفرانسیل حاکم بر یک سیستم بصورت مقابل است‪:‬‬ ‫•‬
‫آیا سیستم پایدار است؟‬ ‫•‬
‫به منظور پایدار سازی و کنترل سیستم از خروجی سیستم فیدبک گرفته و‬ ‫•‬
‫از دستور منها میکنیم‪ .‬سپس خطا را در یک بهره ‪ K‬ضرب کرده و به عنوان‬
‫ورودی به سیستم ارسال میکنیم‪ .‬بلوک دیاگرام سیستم حلقه بسته را‬
‫ترسیم کنید‪.‬‬
‫مقدار بهره ‪ K‬چقدر باشد که سیستم پایدار شود و زمان نشست پاسخ به‬ ‫•‬
‫دستور پله واحد ‪ 0.4‬ثانیه شود‪.‬‬
‫‪𝑌 ( 𝑠 ) 10‬‬
‫=)(𝐺→𝑠𝑈‪𝑠𝑌()=2𝑆+10‬‬
‫‪𝑈 ( 𝑠) − 2‬‬

‫)‪R(s‬‬ ‫‪10‬‬ ‫)‪Y(s‬‬ ‫‪10‬‬

‫‪𝐾.‬‬
‫‪K‬‬ ‫𝐾‪𝑌(𝑠) −2 10.‬‬
‫‪𝑠−2‬‬ ‫= =)𝑠(𝑙𝑐𝐺‬
‫𝐾‪𝑅(𝑠) 10𝑠−2+‬‬
‫‪-‬‬ ‫‪1+ 𝐾 .‬‬
‫‪𝑠 −2‬‬
‫‪14‬‬
‫‪𝜏 = → 𝑇 𝑠 =0 . 4 → 𝐾 1 2‬‬
‫‪10 𝐾 − 2‬‬
 ‫‪Pole Placement Idea‬‬
‫‪𝑌 ( 𝑠 ) 12‬‬
‫= )𝑠(𝑙𝑐𝐺‪𝑤𝑖𝑡h𝐾=1.2,‬‬
‫‪𝑅 ( 𝑠 ) + 10‬‬

‫‪For unit step command‬‬ ‫‪10-‬‬ ‫‪2+‬‬

‫‪1 12‬‬
‫‪𝑅 ( 𝑠) = →𝑌 .‬‬
‫‪𝑠 𝑠+10‬‬

‫‪lim𝑦(𝑡)=lim𝑠.𝑌()=12‬‬
‫‪𝑡 → ∞ 𝑠 →0‬‬

‫‪𝑒 𝑠 =1 − . 2 =0‬‬

‫با کنترل فیدبک توانستیم مکان قطب سیستم را جابجا کرده و آنرا در محلی که‬
‫خواسته طراحی را ارضا میکند قرار دهیم‪ .‬این ایده اصلی ترین روش طراحی‬
‫کنترل کننده است‪ ،‬که به آن ‪ Pole Placement‬میگوئیم‪.‬‬
 Influence for Pole Location
• For Stable Second Order System:

The second order system cannot be

divided to two first order systems
 Influence for Pole Location
• For Stable Second Order System:

The second order system can be

divided to two first order systems

If one of the poles is far away (to the left)

from the other pole (10 times), its effect on
settling time can be ignored
 High Order System
• If a stable system of order n has r real negative poles and 2m
complex poles, then
• The system can be divided to r first order
and m second order subsystems
• If a pole is far left (10 times) of
θ
other poles, its effect can be ignored
 Idea of Pole Placement
• The differential equation between control
force, u(t), and the velocity of a system, v(t), is
as follows:

a disturbance force, d(t), is also applied to the

system.

Draw the block diagram of the system:

Is the system stable?
 Idea of Pole Placement
D(s)
V(s)
U(s) 10
𝑠 − 10

The system is unstable

 D(s) Y(s)
R(s)
+
𝐾𝑝 + U(s) + 10 V(s) 1
- + 𝑠 − 10 𝑠
-
𝐾𝑑

To stabilize the system, and to control the position of the system,

the position is measured by a sensor and fed back to the control.
In the control, the reference position is compared with the
measured position and error is calculated. The control force is
then computed by the following equation (PD control)
𝑢 ( 𝑡 ) =𝐾𝑝. 𝑒 (𝑡 ) − 𝐾 𝑑. 𝑣 (𝑡)
kp and kd are control gains, which has to be tuned by the designer
of the controller.
 Design Control Problem
• Design the controller so that:
Settling time is 0.2 s
Percent overshoot is 10%
⋯
¿ 2 2
𝑠 +2 𝜉 𝜔𝑛 𝑠+ 𝜔𝑛

10𝐾𝑑−10=40→𝐾𝑑=5

20 2
𝜔𝑛= →10.𝐾𝑝=𝜔𝑛→𝐾𝑝=1
0 .6
If the reference command is a unit step, and the disturbance is a
step with the magnitude of 0.2, calculate the steady state error:

Without disturbance, the system has no

steady sate error.

¿¿