Scribd
Upload
14 views

Chapter 9: Design Via Root Locus Cont

This chapter discusses controller design via root locus analysis. It covers lead compensation to improve system response by adding poles and zeros. Lead compensation can be realized passively using networks. The chapter also discusses improving settling time and steady state error using PID controllers. PID design involves first using PD to meet response time specs, then PI to meet error specs. Physical realization of compensators involves approximating ideal components with practical passive and active circuits. Homework problems cover examples of designing lead, lag-lead compensators and implementing PID controllers with op-amps.

Uploaded by

ariful
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
14 views

Chapter 9: Design Via Root Locus Cont

This chapter discusses controller design via root locus analysis. It covers lead compensation to improve system response by adding poles and zeros. Lead compensation can be realized passively using networks. The chapter also discusses improving settling time and steady state error using PID controllers. PID design involves first using PD to meet response time specs, then PI to meet error specs. Physical realization of compensators involves approximating ideal components with practical passive and active circuits. Homework problems cover examples of designing lead, lag-lead compensators and implementing PID controllers with op-amps.

Uploaded by

ariful
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/ 16

Chapter 9: Design Via Root Locus

Cont………

1
2. Lead Compensation: Passive approximation of PD. Realized with a
passive network of TF,

Let us first look at the concept behind lead compensation design:

Solution: Many possible solution. So,


choose arbitrarily either
zC or pC. Then find the other.
2
Example:

Design lead compensation to achieve:


(i) 30% OS
(ii) improve settling time by a factor of 2.

Sol.:
30%OS    0.358
Uncompensated system:

CL poles: p  1.007  j 2.627


1, 2

Gain, K  63.21
And the third pole, p  7.98
3

So, 2nd-order approximation is OK.


4 4
We can use settling time formula, T S    3.972
real  part 1.007 3
Problem: keep %OS=30, make TS 
3.972
 1.986
2

P    jd
d

d

This gives, d  5.252 So, P  2.014  j5.252

Compensated system:
Assume, zc  5

angle contribution (without from pc)

4
So, an angle contribution of {-180˚ -(-172.69˚)}
= -7.31˚ is needed from the pc.
Thus,   7.31
2

5.252
 tan 7.31 
pc  2.014

 pc  42.96

The RL of the compensated system:

Dominant poles and gain:


p1, 2  2.014  j5.252
K  1423

Other poles: p3  43.8 and p4  -5.134

CL zero: z1  zc  5 5
2nd-order approximation is valid, since pole-zero are close and other pole is
more than 20 times the real part of the dominant pole.

HW: Design two other Lead Compensators for the same system with zc = -4 and
zc = -2
Comparison Table:

6
9.4: Improving SSE and TR
1. PID Controller:
Method: (i) first improve TR (using a PD)
(ii) then improve SSE (using a PI)
Reverse order is also possible but not to be covered here.

PID controller design:

 Two zeros and 1 pole at origin


Can be assembled as:
1 zero + 1 pole at origin  PI
the other zero  PD
Design steps:
(1) Evaluate performance of uncompensated system
(2) Design PD controller to meet the TR specs
(3) Design PI controller to yield SSE
(4) Determine the gains, K1, K2 and K3. 7
Ex 9.5:

Design a PID controller so that (i) peak time reduces to 2/3 of the
uncompensated system, (ii) %OS = 20, (iii) SSE = 0 for step input.

Sol.:

20%OS   0.456
Step 1: Uncompensated system (gain design)

CL dominant poles: p  5.415  j10.57


1, 2

gain, K  121.5

the third pole, p  8.169


3

CL zero, z1  8
2nd-order approximation is valid. 8
 
TPO    0.297
imaginary  part 10.57

121.5  8 1
K PO   5.4  eO     0.156
3  6 10 1  K PO

z PD
Step 2: PD controller design:
We want peak time,
2 2  
TPN  TPO   0.297   0.198   d 
3 3 d  15.87
0.198

So, the imaginary part of the required CL dominant pole is 15.87.

Using the trigonometry,


d
  8.13
So, the required CL dominant poles,
p1, 2  8.13  j15.87

Total angle contribution of the OL poles at -3, -6, -10 and zero at -8 to this point
(-8.13 + j15.87) is -198.37˚.
So, angle contribution required from the compensator zero is 18.37˚. 9
Thus the PD controller is,

The RL for the PD-compensated system:

Other parameters: gain, K  5.34, third pole, p3  8.079, CL zeros, z1  8 and z2  55.92

2nd-order approximation is OK. 10


TPN  0.198

5.34  8  55.92 1
K PO   13.27  eN     0.07
3  6 10 1  K PN

Step 3: PI controller design


Zero should be close to origin,

The RL for the complete system    0.456


The dominant CL poles,
p1, 2  7.516  j14.64

Gain, K  4.6

Other poles,
p3  8.099 and p4  0.468

CL zeros, z1  8, z2  55.92, and z3  0.5


2nd-order approximation is OK.
eF   0 Type 1 system 11
Determine K1, K2 and K3:

PID compensator: 4.6s  55.92s  0.5


GPID 
s



4.6 s 2  56.42s  27.96 
s

So, K1  259.5 K 2  128.6 K3  4.6

HW:
(1) Example 9.6  lag-lead compensator design
(2) Example 9.9  implementing a PID controller

12
9.6: Physical Realization of Compensator/Controller

13
14
Example 9.9: Implementing a PID Controller
Problem: Implement the PID controller of Example 9.5
Sol.:
The TF of the PID controller was: 

We know from the previous table:

TF of a PID controller =

So comparing the two equations:

4 unknows and 3 equations. So, assume a practical value for one parameter and
find the required values for other 3 parameters.
If,

15
HW:
(1) Exmple 9.10  realizing a lead compensator
(2) Case Studies, Antenna control

16

You might also like