Scribd
Upload
31 views4 pages

CSD_RL_steps

Uploaded by

Abinok Selvin
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
31 views4 pages

CSD_RL_steps

Uploaded by

Abinok Selvin
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/ 4

CSD - Root Locus

General

1. Check for feedback for the system and obtain correct transfer func on.
2. Get the no. of poles and zero and also their loca n
3. Find No.of assymptodes = P-Z (only real values)
4. Find centroid (angle of assymptodes are taken in this point, all
asymptodes intersect at this point)
5. Find angle of assymptodes. (subs tute only required no.of n)
6. If there is a complex pole , consider that as dominant pole and find the
angles of other poles & zeros to that dominant pole.
7. Find angle of departure or arrival by w=180 –(pole angle) + (zero angle)
{this is valid only is complex poles are present}
8. Check for existance for breakaway or breakin points.
9. If exist, find that points by 1 + G(s)H(s) = 0. In this form , keep K one side
and differen ate and equate to zero. {dK/ds=0}
10.The obtained order of equa on gives the details of no. of break points
and the loca on
11.To find the natural frequency & gain K value, form characteris c
equa on 1 + G(s)H(s)=0. Now by subs tu ng s=jw and separa ng real &
imaginary term and equa ng to zero, we get Wn & K.
12. Now with all the values plot the root locus in rough graph.

Gain Compensa on

Propor onal P
 Improves Overall Speed

1. Find damping ra o by the clue given in ques on.


2. What ever the order of the system, but considering that as second order
system,Write the general form of dominant pole of second order .
3. Subs tute the damping ra o value in that Sd, we will get in func on of
Wn.
4. To find Wn, find all angles caused by poles and zeros to dominant
pole.(use tan’ formula).
5. Summa on of all angles should be equal to mul ple of 180.
6. While subs tu ng put -ve sign for pole angle & +ve angle for zero angle.
7. To that obtained equa on take tan on both sides and solve for Wn.
8. A er ge ng Wn , find the exact dominant pole Sd.
9. If needed now find, peak me(Tp), rise me(Tr), se ling me(Ts) , peak
overshoot.
10.To find gain K, subs tute the value of dominant pole Sd in 1/|G(s)H(s)|.
11.Even if they give any ini al gain(real number in numerator), we should
find the gain(Pcontroller) of the changed system (we considered as
second order system)
12.Now by subs tu ng gain K value, we get exact T.F.
13.To find ESS , find Kp value . Kp = lim s0 G(s)H(s).
14.If input is ramp, Kv = lim s0 sG(s)H(s).
15.Now find ESS = 1/(1+Kp).
16.Draw the root locus by rough graph.

Propor onal Integral PI


 Improves steady state responses (ESS =0)

1. Follow all the steps(1-16) of propor onal controller.


2. To make steady state error = 0, add pole at origin(integrator).
3. Add a zero also such that the order of the system is not changed.
4. To maintain the root locus pass through the dominant pole , place the
zero very near to the origin by finding the angle defficiengy.
5. Now again all the angles caused by P & Z should be mul ple of 180.
6. Find new gain K for new T.F (will be same as the old gain).
7. Now find new Kp a er & ESS. That are Kp is infinity and ESS = 0.
8. Sketch the root locus of the new T.F.
9. If SSE is said to be reduced by a factor, don’t place the compensator pole
in origin.
10.Find the new SSE that is needed, and with that find new Kp.
11.Now take the ra o of new Kp & old Kp. {Kp*/Kp}. Which must be equal
to the ra o of compensated zero & pole (Zc & Pc).
12.Now assume any value of pole near the origin and with this assumed
value of pole and the Kp ra o, we can find the value of Zc.
13.For LEAD compensator assume Zc is closer to origin.
14.For LAG compensator assume Pc near the origin.
15.Now find the new gain K by subs tu ng Sd in new G(s)H(s), by which the
error men oned is required.

 The angular contribu on of PI compensator zero & pole cancel out.


 The dominant pole is s ll on root locus & the system type has increased.
 The gain before and a er compensa on is more or less same.
 The ra o of lengths from compensator pole & zero is approx unity.
 The SSE is improved without disturbing the Transfer behaviour

Propor onal Deriva ve PD


 Improves Transient Responses ( %OS & Wd & Ts  decreases )

1. Follow all the steps(1-16) of propor onal controller.


2. Find the damping ra o from clue in ques on. Damping ra o will not
change in PD controller.
3. If % decrease or any condi on on me is given. First find se ling or peak
me and reduce it by the given factor.
4. For the new Ts; damping ra o should be constant and natural frequency
Wn should be changed.
5. This new Wn changes the dominant pole Sd new & this will not pass
through the root locus and certain angle of deficiency is introduced.
6. The angle deficiency has to be contributed by the PD compensator zero
Zc.
7. And by adding Zc the T.F also changes which will also change the gain K.
8. Find new gain Kp(new) by subs tu ng Sd in new G(s)H(s).
9. To make system less complex. Just add both Zc & Pc. But take Zc such
that it cancels a pole. And the Pc will compensate the deficient
angle(applicable only for some cases).
10. Sketch the systems root locus.
 The new Wn changes the dominant pole and root locus does not pass
through new Sd.
 So by adding Zc ,the root locus is forced to pass through the new Sd.
 This changes the gain K and also the SSE.

Propor onal Integral Deriva ve PID


 Improves Response Speed, SSE, Transient Response

 First do PD compensa on and next PI compensa on


 Because PI compensator makes SSE=0 and does not changes gain K.
 If PD compensa on is done last, it will change the SSE=0 (made by PI
compensator at first)

Lead & Lag Compensator

1. Follow all the steps(1-16) of propor onal controller.


2. Check for the me condi on in ques on.
3. Add Pc & Zc to make the design sa sfying the condi on.
4. Wn will change & damping is constant. This changes the Sd.
5. Add Zc such that it cancels a pole in G(s)H(s). and add Pc to make root
locus pass through new Sd.
6. Find gain K for new T.F by K = lim sSd 1/|G(s)H(s)|
7. Plot on graph.

 By canceling a pole by Zc and adding a Pc, the system order remains


same.
 For lead compansator Zc lies before Pc and vice versa.

You might also like