0% found this document useful (0 votes)
31 views9 pages

Answer HW02

This document discusses advanced automatic control systems. It first presents two examples of transforming transfer functions into Jordan canonical forms and state-space representations. The second part derives the equations of motion and state-space model for an inverted pendulum system. It provides the transfer function and computes the controllable and observable forms. Simulation files are referenced to verify the analytical results.

Uploaded by

pouya mansouri
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
31 views9 pages

Answer HW02

This document discusses advanced automatic control systems. It first presents two examples of transforming transfer functions into Jordan canonical forms and state-space representations. The second part derives the equations of motion and state-space model for an inverted pendulum system. It provides the transfer function and computes the controllable and observable forms. Simulation files are referenced to verify the analytical results.

Uploaded by

pouya mansouri
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 9

‫کنترل خودکار پیشرفته‬

‫امید عاقب‬

‫مهیار نراقی‬
1. Companion + Jordan forms:
a.
𝑠+2 Eq.1.1
𝐻(𝑠) =
𝑠 3 +2𝑠 2 +5𝑠
0 1 0 0
̇
=> 𝑐𝑜𝑛𝑡𝑟𝑜𝑙𝑙𝑎𝑏𝑙𝑒 𝑓𝑜𝑟𝑚: 𝑥⃗ = [0 0 1 ] 𝑥⃗ + [0] 𝑢
Eq.1.2
0 −5 −2 1
𝑦 = [2 1 0]𝑥⃗
0 0 0 2
̇
=> 𝑜𝑏𝑠𝑒𝑟𝑣𝑎𝑏𝑙𝑒 𝑓𝑜𝑟𝑚: 𝑥⃗ = [1 0 −5] 𝑥⃗ + [1] 𝑢
Eq.1.3
0 1 −2 0
[
𝑦= 0 0 1 ⃗ ]𝑥
Jordan form:
𝑠+2 𝑟1 𝜆+𝑗𝛾 𝜆−𝑗𝛾 𝑎𝑓𝑡𝑒𝑟 𝑓𝑖𝑛𝑑𝑖𝑛𝑔 𝑟1 ,𝜆 𝑎𝑛𝑑 𝛾
𝐻(𝑠) = = + + ⇒ 𝑟1 = 0.4, 𝜆 = −0.2, 𝛾 =
𝑠((𝑠+1)2 +4) 𝑠 𝑠+1+𝑗2 𝑠+1−𝑗2 Eq.1.4
0.15, 𝜎 = 1, 𝜔 = −2
In which 𝜎 and 𝜔 are used as follows:
0 0 0 1 0 0 0 1
̇
=> 𝑥⃗ = [0 0 1 ] 𝑥⃗ + [0] 𝑢 = [0 0 1 ] 𝑥⃗ + [0] 𝑢
2 2) Eq.1.5
0 −(𝜎 + 𝜔 −2𝜎 1 0 −5 −2 1
𝑦 = [𝑟1 2(𝜆𝜎 − 𝛾𝜔) 2𝜆]𝑥⃗ = [0.4 0.2 −0.4]𝑥⃗
b.
𝑠+3 𝑟1 𝑟21 𝑟
22
𝑎𝑓𝑡𝑒𝑟 𝑓𝑖𝑛𝑑𝑖𝑛𝑔 𝑟1 ,𝑟21 𝑎𝑛𝑑 𝑟22 Eq.1.6
𝐻(𝑠) = (𝑠+1)2(𝑠+2) = + + (𝑠+1) 2
⇒ 𝑟1 = 1, 𝑟21 = −1 , 𝑟22 = 2
𝑠+2 𝑠+1
1 −1 2
=> 𝐻(𝑠) = + + (𝑠+1)2 Eq.1.7
𝑠+2 𝑠+1
−2 0 0 1
̇
=> 𝐽𝑜𝑟𝑑𝑎𝑛 𝑓𝑜𝑟𝑚: 𝑥⃗ = [ 0 −1 0 ] 𝑥⃗ + [0] 𝑢 ; 𝑦 = [1 −1 2]𝑥⃗ Eq.1.8
0 1 −1 0

2. According to figure1:

figure 1 inverted pendulum system.


So the equations of motion of this 2 DOF system can be easily derived using Lagrange's
method:
1 1
𝑇 = 𝑀𝑥̇ 2 + 𝑚(𝑥̇ 2 + 𝑙 2 𝜃̇ 2 + 2𝑙𝑥̇ 𝜃̇ cos(𝜃)) Eq.2.1
2 2

𝑃 = 𝑚𝑔𝑙𝑐𝑜𝑠(𝜃) Eq.2.2

ℒ =𝑇−𝑃 Eq.2.3
𝑑 𝜕ℒ 𝜕ℒ 𝑘𝑒 𝑘2𝜔 𝑘𝑒 𝑘 2𝜔
( ) − 𝜕𝑥 = 𝑓 = 𝑅𝑟 − => (𝑀 + 𝑚)𝑥̈ + 𝑚𝑙𝜃̈ cos(𝜃) − 𝑚𝑙𝜃̇ 2 sin(𝜃) = − Eq.2.4
𝑑𝑡 𝜕𝑥̇ 𝑅𝑟 𝑅𝑟 𝑅𝑟
𝑑 𝜕ℒ 𝜕ℒ
( ) − = 0 => 𝑚𝑙𝜃̈ + 𝑚𝑥̈ cos(𝜃) 𝑚𝑔𝑠𝑖𝑛(𝜃) = 0 Eq.2.5
𝑑𝑡 𝜕𝜃̇ 𝜕𝜃
If the pendulum is stabilized, eqs.2.4,5 can be simplified more to finally deal with a pair of
linear ODEs. On the other hand:
𝑥̇ = 𝑟𝜔 Eq.2.6

So after linearization and substituting eq.2.6 in eqs.2.5,6:


𝑘2 𝑚𝑔 𝑘
𝑥̈ + 𝑥̇ + 𝜃 = 𝑒 *
𝑀𝑟 2 𝑅 𝑀 𝑀𝑅𝑟
𝑀+𝑚 𝑘2 𝑘
̈𝜃 − ( ) 𝑔𝜃 − 𝑥̇ = − 𝑒 **
𝑀𝑙 𝑀𝑟 2 𝑅𝑙 𝑀𝑅𝑟𝑙
a. State-Space representation:
𝑥1 = 𝑥 𝑥1 0 1 0 0 0
𝑥̇ 1 = 𝑥2 = 𝑥̇ 𝑥2
=> 𝑥
⃗ = [ ] => 𝑥⃗ ̇ = [0 −12.5 −0.49 0
] 𝑥⃗ + [
0.25
]𝑢
𝑥3 = 𝜃 𝑥3 0 0 0 1 0
𝑥̇ 3 = 𝑥4 = 𝜃 ̇ 𝑥 4 0 12.5 10.29 0 −0.25
𝑥 𝑥1 1 0 0 0 0
𝑦 = [ ] = [𝑥 ] = [ ] 𝑥⃗ + [ ] 𝑢 Eq.2.7
𝜃 3 0 0 1 0 0
0 1 0 0 0
0 −12.5 −0.49 0 0.25 1 0 0 0 0
=> 𝐴 = [ ] & 𝐵= [ ] & 𝐶 = [ ] & 𝐷=[ ]
0 0 0 1 0 0 0 1 0 0
0 12.5 10.29 0 −0.25
b. Block diagram of the system is shown in figure2:

figure 2 block diagram of the system.


c. Transfer Function:
𝑌1 (𝑠) 0.25𝑠 2 − 2.45
𝑈(𝑠) 4 3 2 − 122.5𝑠
= 𝐻(𝑠) = 𝐶(𝑠𝐼 − 𝐴)−1 𝐵 + 𝐷 = 𝑠 + 12.5𝑠 − 10.29𝑠 2 Eq.2.8
𝑌2 (𝑠) 0.25𝑠

[ 𝑠 4 + 12.5𝑠 3 − 10.29𝑠 2 − 122.5𝑠]
[ 𝑈(𝑠)]
d. Controllable and Observable forms (Companion forms):

𝑌1 (𝑠) 0.25𝑠 2 − 2.45


= Eq.2.9
𝑈(𝑠) 𝑠 4 + 12.5𝑠 3 − 10.29𝑠 2 − 122.5𝑠
0 1 0 0 0
0 0 1 0 0
− 𝑐𝑜𝑛𝑡𝑟𝑜𝑙𝑙𝑎𝑏𝑙𝑒 𝑓𝑜𝑟𝑚: 𝐴1 = [ ] & 𝐵1 = [ ]
0 0 0 1 0
0 122.5 10.29 −12.5 1

𝐶1 = [−2.45 0 0.25 0] & 𝐷1 = [0]

0 0 0 Eqs.2.10
0
1 0 0 122.5
− 𝑜𝑏𝑠𝑒𝑟𝑣𝑎𝑏𝑙𝑒 𝑓𝑜𝑟𝑚: 𝐴2 = [ ] & 𝐵2 = [−2.45 0 0.25 0]′
0 1 0 10.29
0 0 1 −12.5

𝐶2 = [0 0 0 1] & 𝐷2 = [0]

𝑌2 (𝑠) 0.25𝑠 2
=− 4
𝑈(𝑠) 𝑠 + 12.5𝑠 3 − 10.29𝑠 2 − 122.5𝑠

0 1 0 0 0
0 0 1 0 0
− 𝑐𝑜𝑛𝑡𝑟𝑜𝑙𝑙𝑎𝑏𝑙𝑒 𝑓𝑜𝑟𝑚: 𝐴1 = [ ] & 𝐵1 = [ ]
0 0 0 1 0
0 122.5 10.29 −12.5 1
Eqs.2.11
𝐶1 = [0 0 0.25 0] & 𝐷1 = [0]

0 0 0 0 0
1 0 0 122.5 0
− 𝑜𝑏𝑠𝑒𝑟𝑣𝑎𝑏𝑙𝑒 𝑓𝑜𝑟𝑚: 𝐴2 = [ ] & 𝐵2 = [ ] & 𝐶2 = [0 0 0 1] & 𝐷2
0 1 0 10.29 0.25
0 0 1 −12.5 0
= [0]

PART.2: SIMULATION
1. Refer to the first problem of PART.1: (file: Part2_p1.m)
Which results in the same answer as PART.1.

2. Using the 'state-space' block in Simulink, the system can be easily shown as:

figure 3 Created model in Simulink.


To check the result of Part.1.Q2.c:


Which confirms the answer in c.
And for the observable and controllable forms, the code written below leads to the same answer as part d:

o Files: 'Part2_P2.slx' and 'Part2_p2.m'

3.
o Files: 'Part2_p3.mlx'

You might also like