Abu Dhabi University
EEN 365 - Control Systems
Project Report
Designing Position Control System
Author:
Muhammad Obaidullah 1030313 Supervisor:
Ali Raza 1012542 Dr. Riad Kanan
Mohammed Ali Sadi 1003857
Section 1
June 3, 2013
Contents
1 Specifications 3
2 Introduction 3
3 Simplifying the Block Diagram 4
4 Procedure for Lead-Lag Compensator Design 5
5 Designing Lead Compensator CLead 6
5.1 Finding the Desired Pole and Angle of Deficiency . . . . . . . . . . . . . . . . . . . . 6
5.2 Calculating lead pole, zero, and K . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
6 Designing Lag Compensator CLag 8
6.1 Calculating ratio of pole and zero . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
6.2 Assume a zero and find according pole . . . . . . . . . . . . . . . . . . . . . . . . . . 8
6.3 Final Transfer Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
7 List of Software Used 8
8 Testing using MATLAB 9
8.1 Writing the Code . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
8.2 Plotting root locus and finding the step response . . . . . . . . . . . . . . . . . . . . 9
8.3 Final Circuit Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
9 Conclusion 10
10 Presentation 11
11 Team Dynamics 12
1
List of Figures
1 The insides of a typical servo motor which uses a similar control circuit and technique.
[1] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2 This is the root locus plot of the transfer function, as we can see the graph has a
line where the percentage overshoot becomes 10%. We have to select a point on this
line in order to get a correct step response. . . . . . . . . . . . . . . . . . . . . . . . 9
3 This is the final Step response after selecting a point on the 10% overshoot line. As
we can see the Overshoot is about 10% and the settling time is approximately 1s. . . 10
2
Abstract
This project report explains in great detail how to design and manufacture a stable Wien
Bridge Oscillator and how to improve the design by using amplitude controlling non-linear
circuit.
1 Specifications
A position control is to be designed with a 10% overshoot, a settling time of 1 second, and Kv
= 1000.
This is the list of available components:
2 Introduction
This project was a compulsory course project for the EEN365
Control Systems. The aim of this project was basically to design a
Consolidated Position Control System with the aim of properly and
accurately running a DC Motor. In order to do this, fundamental
control systems analysis were employed which involved strategies
including some arithmetic procedures. In the basic form, a raw
transfer function is given to students and with that are certain
specifications are also provided. These specifications are then uti-
lized to modify the transfer function and mould it and compensate
it for the extra things and the things that lagged. It follows that
the raw transfer function has to be made better for these things by
designing two compensators namely the lead and lag compensators
to enhance the process of moulding it. This moulding results in the
formation of a new and improved ultimate transfer function for the
Figure 1: The insides of a typ-
DC Motor control system. This new Transfer function takes into
ical servo motor which uses
account the lead and lag compensator design calculations that were
a similar control circuit and
inserted in the old transfer function. Using this transfer function,
technique. [1]
the stability and success of the new control system for the motor
was tested and simulated using Matlab and required software. [2]
3
3 Simplifying the Block Diagram
In this project, a transfer function was given to us. The design of the lead and lag compensator
required us to find the angle deficiency between the basic transfer function pole and zero given and
the pole and zero that we will later go on to compute. So for us to do that, we used the transfer
function given to us. The transfer function for DC motor is multiplied with the one of the power
amplifier to give us relatively more accurate results. This set of calculations and multiplying is
done in the Building Block stage where we design the initial system. In this system the power
amplifier AMOLIFIES the initial function which in essence means that it gets multiplied to it and
the pre-amplifier in this case is our gain. And the gain is also embedded in the building blocks.
This gain, with the amplifier and together with the transfer function give us the SIMPLIFIED
transfer function in the building blocks stage.
In order to design the passive compensator we need to first simplify the block diagram into unity
feedback loop.
COMPLETE BLOCK DIAGRAM: This is the complete clock diagram of the system before
designing the passive compensator for it. First we have a Position to voltage converter which con-
verts the input position from the user to the meaningful voltage and then we have a error calculator
(summation block). The position to voltage controller is usually a potentiometer connected to a
voltage supply. As the potentiometer is rotated by the user, the resistance changes in turn changing
the voltage drop across it. Thus a voltage corresponding to the voltage is sent into the system. The
second process is to calculate the error by subtracting the input voltage from the output voltage.
This error then passes through Pre-Amplifier, Power Amplifier and DC Motor and through the
gearbox to actual position of shaft.
STEP 1: In this step, we multiply the Pre-amplifier, Power Amplifier, DC Motor, and Gear Box
transfer functions to make a single transfer function because they are cascaded together.
4
STEP 2: In this step, we use moving blocks method to move both Position to Voltage converters
inside. Therefore, we end up with a unity feedback loop with the transfer function G(s) as following.
STEP 3: This is the overall transfer function of the system before designing the compensator.
4 Procedure for Lead-Lag Compensator Design
A lead compensator can be designed using a mathematical technique called root locus.
• In this transfer function, the magnitude of p0 is more than that of Z0.. A lead essentially
shifts the acquired root locus to the left of the plane. The consequence of this is that the
stability pf the system is improved and the speed is well found too.
• A soon as the lead compensator becomes a part of the system, the value of the zeroes where
it intersects will turn out to be a bigger negative number than before. The final number and
5
the difference of the poles and zeroes stays the same but the recently added pole is relatively
a larger negative value than the recently added zero. And this is why the compensator moves
the locus to the left of the complex plane, As a result, the stability and the response speed is
greatly improved.
A lag compensator can also be designed using the root locus , the form of transfer function using
the same root locus method
• This is in close proximity to the lead compensator but there is a slight change in this. That
change is that the magnitude of p0 is less than the magnitude of z0/ Following this, we say
that the root locus in this case is shifted to the right of the complex plane. To omit this
unwanted change, the zero and poles of a lag are huddled close so that it doesnt have a
substantial adverse impact on the systems stability.
• Once again, the lag compensator works on the same principles as described before which
involved the asymptotes. The Intersections of root locus shits it to the right of the planeHow
does the lag controller shift the root locus to the right? Below is repeated the equation for
finding where the asymptotes of the root locus intersect along the real axis.
• As a consequence of the addition of a lag compensator, this intersection turns out to be a
smaller negative number than the previous one. The final net is the same but the the renewed
pole is a small negative number than the renewed zero. That is why the intersection is moved
to the right in the complex plane.
5 Designing Lead Compensator CLead
5.1 Finding the Desired Pole and Angle of Deficiency
After getting transfer function, we
get the damping ratio after taking
into consideration the damping ratio
which we substitute in the formula for
the Percentage Overshoot. It is per-
tinent to mention here that as men-
tioned before, there are some specifi-
cations provided to us for us to design
the compensators, the % Overshoot
according the specification given is
10%. When we equate the Overshoot
equation with 10% overshoot, we will
be able to get the Damping Ratio.
Following this, we substitute the damping ratio we got in the formula for the settling time. The
settling time according to the specification was 1s. So we similarly equate the formula with 1 in
order to get the Natural frequency because the natural frequency is one of the terms in this formula
with the damping ratio. Once we get this both, we substitute both of them in the formula given to
us to calculate the desired pole and thus we get the desired pole. This desired pole has both the
real part and the Imaginary part which s the j part. This pole as a whole (Taking both the real
and imaginary parts) is substituted in the simplified transfer function that we got. We replace the
s term in it with the desired pole which gives us [−5.425 × 10−7 + j2.445 × 10−6 ].
6
5.2 Calculating lead pole, zero, and K
We choose the zero as -0.505 for our
lead compensator because it makes
it easy for us to simplify further
the transfer function since the trans-
fer function contains the term ”(s +
0.505)” and it also satisfies the con-
dition that the zero of the lead com-
pensator should be very close to zero.
We find the pole by making a angle
of 77.5 (Angle of deficiency with our
zero and we got the pole at -9.60. We
used the software GeoGebra for cal-
culating this angle accurately.
Now our transfer function looks something like the following:
(s + 0.505)
CLead = K (1)
(s + 9.60)
As we can see that we still have to find the K in our equation and replace this K with the Pre-
amplifier K. So we put a magnitude condition on G(s)Clead to find K:
0.004
G(s)CLead = K| |=1 (2)
s(s + 9.60)(s + 40)
K = 481639 (3)
Therefore the final CLead is:
(s + 0.505)
CLead = 481639 (4)
(s + 9.60)
7
6 Designing Lag Compensator CLag
6.1 Calculating ratio of pole and zero
The first step is calculating the zero and pole ratio which comes from satisfying the following
equation.
Kv = lims→0 sG(s)CLead (s)CLag (s) (5)
Substituting the value of Kv = 1000.
0.004 × 481639 Z
× = 1000 (6)
9.60 × 40 P
Therefore,
Z
= 199.3 (7)
P
6.2 Assume a zero and find according pole
Let Z(zero) = −0.1
0.1
= 199.3 (8)
P
P = −0.0005 (9)
Therefore,
(s + 0.1)
CLag = (10)
(s + 0.0005)
6.3 Final Transfer Function
481639(s + 0.1)
T (s) = (11)
s(s + 0.0005)(s + 9.60)(s + 40)
7 List of Software Used
• SimuLink Software.
• MATLAB.
• GeoGebra.
• LaTeX.
8
8 Testing using MATLAB
8.1 Writing the Code
Before designing the schematics of the Lag-Lead compensator, we test our transfer function using
MATLAB. The first step was to write the code and then obtain the root locus graph. Each line of
code is explained as a comment entered next to the line
1 numenator= 4 8 1 6 3 9 ∗ [ 1 0 . 1 ] ; % Numenator
denominator=p o l y ( [ 0 −40 −9.60 − . 0 0 0 5 ] ) ; % Denominator
3 G=t f ( numenator , denominator ) ; % C r e a t e t h e f u n c t i o n named G
Gzpk=zpk (G) ; % D e f i n e z e r o p o l e g a i n
5 r l o c u s (G) % Find r o o t l o c u s o f t h i s f u n c t i o n
pos =10; % P e r c e n t a g e Overshoot
7 z=−l o g ( pos / 1 0 0 ) / s q r t ( p i ˆ2+[ l o g ( pos / 1 0 0 ) ] ˆ 2 ) ;
s g r i d ( z , 0 ) ; %P l o t a g r i d f o r t h i s
9 t i t l e ( [ ’ Root Locus with ’ , num2str ( pos ) , ’ P e r c e n t Overshoot L i n e ’ ] )
[ K, p]= r l o c f i n d (G) ; % Puts r u l e r f o r u s e r f o r e a s i n e s s
11 pause
T=f e e d b a c k (K∗G, 1 ) ;
13 s t e p (T) ; % Give t h e i n p u t o f Step t o t h e system
t i t l e ( [ ’ Step Response f o r Design o f ’ , num2str ( pos ) , ’ P e r c e n t ’ ] )
8.2 Plotting root locus and finding the step response
Figure 2: This is the root locus plot of the transfer function, as we can see the graph has a line
where the percentage overshoot becomes 10%. We have to select a point on this line in order to
get a correct step response.
9
Figure 3: This is the final Step response after selecting a point on the 10% overshoot line. As we
can see the Overshoot is about 10% and the settling time is approximately 1s.
8.3 Final Circuit Diagram
9 Conclusion
• This Control Systems project was a course compulsory one and it was aimed at designing a
position control system for the movement of a DC Motor, not only this, but based on a a
certain set of specifications, this whole position control system was supposed to be designed
by us. The specifications given to us were that related to transient time response. That is,
the percentage overshoot and the settling time values were given to us and then based on
this, we had to make this system work.
10
• A main transfer function for the motor is given to us and we had to reduce it to a simplified
form. The reduced function is what we manipulate and manouvre with. Following the same
concept, we tried to design it. Using the transient response formulas which were laid out
during the beginning lectures of this course, we substituted the percentage over shoot andf
made damping ratio the subject of the formula in order to get the damping ratio. Following
this, we substituted this damping ratio value in the formula for the settling time. The latter
involves the natural frequency and making this the subject of the formula, we get the damping
ratio.
• After getting these two values, we replace the variables by the actual values that we got in
order to get the the intersections and subsequently the desired poles, Once the desired poles
are known, we see that this desired pole has real as well as an imaginary part. This pole is
then plotted on an axis and this is where the work of compensators actually arises.
• Since the pole is mid-air because of the presence of trhe complex part, the lead and lag
compensators are designed. The lead compensator shifts the root locus to the left. We need
this compensator in order to compensate for the instability of the system which arises because
of the desired pole. This compensator has a transfer function which requires us to locate new
poles and zeroes using a protractor.
• The angle of deficiency is found by substituting the desired pole in the reduced transfer
function, taking the tan of this we get the angle of deficiency, once we have it, we keep the
protractor on the complex point of the desired pole whch is both on the X and Y axis and
locate the angle of deficiency, wherever the lines from the located angle of the protractor
intersect, that will be our new pole and zero which we got from the lead compensator.
• The lag compensator is found by the limit of the transfer function when the variable S in it
approaches 0. In this case the specicification requires us to use a specific Kv.
• When we have both the compensator transfer functions, we simply cascade it with the reduced
basic transfer function we obtained earlier and end up with the final transfer function. We
then plot it on a suitable software and the step response with exactly 10% overshoot can
be easily visible. The end result is that we finally obtain an accurate transfer function for
the DC motor which has all the proper compensations against the raw transfer function we
started out with. The refined transfer function for the motor which employs both the lead
and the lag, ensures a much better stability and response speed and ultimately proves to be
a help in positioning the control system.
10 Presentation
You can access our presentation from the following link:
https://www.dropbox.com/s/g61exza6m2fckqp/Control%20Systems.pptx
11
11 Team Dynamics
Part and Member Weight Grade Obaidullah Ali Raza Ali Sadi
Abstract 10% 33.3% 33.3% 33.3%
Introduction 15% 33.3% 33.3% 33.3%
Procedure Part 1 15% 33.3% 33.3% 33.3%
Procedure Part 2 15% 33.3% 33.3% 33.3%
Results Part 1 15% 33.3% 33.3% 33.3%
Results Part 2 15% 33.3% 33.3% 33.3%
Conclusion 15% 33.3% 33.3% 33.3%
Total 100% 33.3% 33.3% 33.3%
References
[1] Physical ComputingServo Motors ”A inside of a typical servo motor ”, Internet:http:
//lizarum.com/assignments/physical_computing/images/basics/components/servo/
insides.png, 1999 [3 June 2013].
[2] Control Tutorials for MATLAB ”Designing Lead and Lag Compensators”, Internet:http:
//ctms.engin.umich.edu/CTMS/index.php?aux=Extras_Leadlag, [3 June 2013].
12
