Control Systems Lab Lab # 8
Control Systems LAB # 8
Design Via Root Locus
What is Root Locus … ?
Root Locus technique produces a plot that shows the locations of poles of a
closed loop system on the S-Plane as K varies and from this plot we can choose the
suitable K that meet our specification conditions.
3( s + 2)( s + 3)
Example : Find the root locus for the T.F G (s) =
s ( s +1)( s + 4)( s + 6)
Matlab code :
clc
clear all
close all
num=3*[conv([1 2],[1 3])]
x=[conv([1 0],[1 1])]
y=[conv([1 4],[1 6])]
den=[conv(x,y)]
g=tf(num,den)
rlocus(g)
sgrid
Result :
1
Control Systems Lab Lab # 8
Example : Using Matlab plot the root locus for the T.F and design K
controller to get 10% O.S
3
G( s) =
s ( s +8)
At 10 % O.S we find that ζ
= 0.591
Matlab Code :
num=[3]
den=[1 8 0]
g=tf(num,den)
rlocus(g)
sgrid(.591,0)
From the above plot of root locus we find that if K = 15.3 then the O.S will be 10 %
num=[3]
den=[1 8 0]
g=tf(num,den)
rlocus(g)
sgrid(.591,0)
sgrid
k=15.3
gf=feedback(k*g,1)
figure
step (gf)
2
Control Systems Lab Lab # 8
If we put K = 15.3 we find that its satisfy our need (10 % O.S)
Stability By root Locus :
a. Example : for the following T.F Plot the root locus of this system and
obtain the value range of the proportional gain to make the system: 1. Stable
2. Unstable
3. Margin
s +2
G ( s) =
s ( s + 0.6 s + 4)
2
3
Control Systems Lab Lab # 8
From the Root Locus we find that :
1- The system is stable for K<1.77
2- The system is Margin stable for K = 1.77
3- The system is unstable for K >1.77
Extra About root locus :
We knew that root locus plot the poles for an closed loop transfer function that
will be controlled by Proportional controller and the system must have unity
feedback path as in figure below.
Root locus will take this
transfer function and use it
to plot the poles as K varies
E(s) G(s)
X(s) +- K Y(s)
(Plant)
Note that Unity feedback
But if the system does not has unity feedback as in figure below, so we should
convert it to another has unity feedback by the solution below.
E(s) G(s)
X(s) +- K Y(s)
(Plant)
)F(s
NON-Unity feedback
4
Control Systems Lab Lab # 8
E(s) G(s)
X(s) +- K H ( s) = Y(s)
1 + G ( s) F ( s)
Unity feedback
Note that K is out of our converting because we consider that we added it after
conversion.
Matlab Command :
rlocus (sys) : This command
used to plot the root locus of the open loop transfer function which will
be controlled by K controller.
sgrid : Generate an s-plane grid.
sgrid (z,wn) : Generate an s-
plane grid of constant damping factors and natural frequencies.
*********************************************************************
LAB WORK:
1. (Stability Proportional Controller) s +1
G ( s) =
By Matlab, For the open loop transfer function s ( s + 0.5s +1)
2
b. Plot the open loop step response and comment on it.
c. Plot the closed loop transfer function and comment on it.
d. Plot the root locus of this system and obtain the value range of the
proportional gain to make the system: 1. Stable
2. Unstable
3. Margin
2. (Design Problem) Manufacturing Robotics
Robotics has revolutionized the manufacturing industry, particularly in the
manufacture of automobiles. The image below shows the GMFanuc Robotics
Corporation Model P-150, six-axis articulated arm, electric-servo-driven robot
painting an automobile*. The six axes are simultaneously controlled, and driven by
state-of-the-art compact ac servomotors. The result is a responsive system with fast
acceleration and deceleration, precision painting, and requiring no brush maintenance.
Let us assume that we have to design a robot such as this one, and that one of the axes
can be represented by the block diagram as shown below:
5
Control Systems Lab Lab # 8
Also assume that for the purpose of this application, a critically damped system is desirable.
Using MATLAB do the following:
1. Find the closed-loop transfer function T(s) for the system.
2. Display the step response for unity amplifier gain and step input.
3. Obtain the percent overshoot, settling time, rise time, DC gain and steady state error.
4. Calculate the amplifier gain Ka which will result in a critically damped system (the fastest
response without overshooting).
