SCHOOL OF BASIC SCIENCES, TECHNOLOGIES, AND ENGINEERING

DIGITAL CONTROL

FINAL PROJECT

JHONATAN FLOREZ OBANDO

CC. 1098628702

PRESENTED TO THE TUTOR

Diego Fernando Sendoya Losada

NATIONAL OPEN AND DISTANCE UNIVERSITY (UNAD)

ENVIRONMENTAL ENGINEERING
CEAD BUCARAMANGA
2014
 SCHOOL OF BASIC SCIENCES TECHNOLOGIES AND ENGINEERING
DIGITAL CONTROL

INTRODUCTION

The analysis and design of such control systems make it necessary to

knowledge of tools such as the Z Transform, the transform of
Fourier Sampling and Stability the Digital Design Techniques based on the
frequency Techniques of digital design and Analysis in the state space. More
even if we add to this the great capability of analysis programs such as
Matlab, then through its study and proper use we can simulate
any type of system that is applicable to real life. The study and
investigation of digital control systems, based on developing
controllers, compensators and digital regulators within a system,
where to check the transient response of a system with
the purpose of reducing error through modeling processes
mathematician, using modeling tools such as MATLAB and SCILAB.
They have allowed students to graphically check the answers.
obtained mathematically. Thus, the purpose of this work is the
to make known through analytical operations and graphic checks,
the topics related to digital design techniques based on the
frequency.
This work is based on the knowledge acquired during the
development of the course Digital Control, Digital Controllers, within the
controllers we find a lead compensator can increase the
stability or the response speed of a system; a compensator in
Delay can reduce (but not eliminate) the steady-state error.
This work consists of a series of developed exercises.
in a practical and theoretical way in which we use the tool of
Labview and Matlab software for the purpose of graphing, simulating, and verifying the

results in an analytical and investigative manner.

Analysis and calculations will be carried out for each of the proposed exercises.
they will analyze transfer functions of a plant.
 SCHOOL OF BASIC SCIENCES TECHNOLOGIES AND ENGINEERING
DIGITAL CONTROL

Generally, the anticipatory, delay, and...

advance/delay are designed for a system in the form of a function of
transfer
The position error constants Kp will also be analyzed, the error in
steady state and the settling time for a function of
transfer of the plant.

OBJECTIVES

Design a digital PID controller for a closed-loop system with

some specific characteristics, such as maximum thrust and time
of establishment according to the parameters of the guide.

Use Labview software to analyze and understand the responses

of systems both to impulse and step inputs.

Calculate the error constant Kp, the steady-state error in the presence of
a unit step input and the settling time for the
transfer function of a discretized plant in closed loop.

Apply the knowledge acquired during the Digital Control course

and analyze the different closed-loop systems.
 SCHOOL OF BASIC SCIENCES TECHNOLOGIES AND ENGINEERING
DIGITAL CONTROL

DEVELOPMENT
Exercise 1: Assume that the transfer function of the plant is:

(a) Calculate the position error constant Kp, the steady-state error
stationary in response to a unit step input and the settling time
for the transfer function of the discretized plant without controller in
closed loop.
( )=Controlador
=Retenedor de orden cero
Plant
Entry
Exit

( )= ( ) ( ) ∗( ) ( )
( )= ( )− ( ) ( )
Now replacing (1) in (2)
E(s) = R(s) - Gp(s)Zoh(s)E*(s)
[E(s)=R(s)−Gp(s)Zoh(s)E∗(s)]∗
E*(s)=R*(s)−[Gp(s)Zoh(s)]*E*(s)
E*(s) + [Gp(s)Zoh(s)]*E*(s) = R*(s)
E*(s)[1+[Gp(s)Zoh(s)]*]=R*(s)
SCHOOL OF BASIC SCIENCES TECHNOLOGIES AND ENGINEERING
DIGITAL CONTROL
SCHOOL OF BASIC SCIENCES TECHNOLOGIES AND ENGINEERING
DIGITAL CONTROL
SCHOOL OF BASIC SCIENCES TECHNOLOGIES AND ENGINEERING
DIGITAL CONTROL
SCHOOL OF BASIC SCIENCES TECHNOLOGIES AND ENGINEERING
DIGITAL CONTROL
 ESCUELA DE CIENCIAS BÁSICAS TECNOLOGÍAS E INGENIERÍAS
DIGITAL CONTROL

OVERBOOST
SCHOOL OF BASIC SCIENCES TECHNOLOGIES AND ENGINEERING
DIGITAL CONTROL
SCHOOL OF BASIC SCIENCES, TECHNOLOGIES, AND ENGINEERING
DIGITAL CONTROL
 SCHOOL OF BASIC SCIENCES, TECHNOLOGIES, AND ENGINEERING
DIGITAL CONTROL

EXERCISE 2
 SCHOOL OF BASIC SCIENCES TECHNOLOGIES AND ENGINEERING
DIGITAL CONTROL

Suppose that the transfer function of the plant is:

1
G p ( s )=
s (s +1)
(a) Calculate the velocity error constant kv, the steady-state error
stationary in front of a unit step input for the function of
transfer of the discretized plant without controller in closed loop; and the
phase margin for the transfer function of the discretized plant without
open loop controller.
1
k v = lim ( z−1 ) HG ( z )
T z→1

[ ]
( )
1
s ( s +1 )
(
HG ( z=) 1−z
−1
z)
s

¿ ( 1−z−1 ) z
[ 1
+
1
s s +1s
2
−
1
]
−1 0.2z z z
HG ( z
) =(1−z )( ( 2
+
z−1 ) z−e−0.2
−
z−1
)

0.0187308 ( z +0.935525 )
¿
( z−1)(z−0.818731)

( )
1 0.0187308 ( z +0.935525 )
k v = limz−1( )
T z→1 ( z−1 ) ( z−0.818731 )
k v =0.04
Now the procedure is carried out to find the transfer function.
in closed loop and the position error in order to find the error in state
stationary in front of a step input.

( )
0.0187308 ( z +0.935525 )
( z−1 ) ( z−0.818731 )
k p =lim =1

(
0.0187308z+( 0.935525 )
)
z→1
1+
( z−1z−0.818731
)( )
1
e p= =0.5
1+ k p
Now the phase margin is given:
0.0187308 ( z +0.935525 )
HG ( z )=
( z−1 ) ( z−0.818731 )
 SCHOOL OF BASIC SCIENCES TECHNOLOGIES AND ENGINEERING
DIGITAL CONTROL

1+0.5Tw
z=
1−0.5Tw
1+0.5 ( 0.2 ) w
z=
1−0.5 ( 0.2 ) w
( 1+ 0.1w )
z=
1−0.1w

HG (w=)
0.0187308
( ( 1+ 0.1w )
1−0.1w
+ 0.935525 )
( (1−0.1w
1+ 0.1w )
−1 1−0.1w −0.818731 )
)( (1+0.1w )

0.0187308 ( 1+0.1w+0.935525−0.0935525w )
HG ( w )=
(1+ 0.1w−1+0.1w ) ( 1+0.1w−0.818731+0.818731w )
0.1813+0.1818w
0.2w(¿)
¿
0.0187308 ( 1,935525+0.065525w )
HG ( w )=
¿
0.00121w+0.03625
HG ( w )=
0.0363w2 +0.0362w

(b) Design a lead-lag digital compensator so that the system in

closed loop has a phase margin of 50º and the error constant of
speed is kv = 2. Assume that the sampling time is Ts = 0.2
seconds.
Now we proceed to find the transfer function of the plant.

G ( z )= ( 1−z−1 ) z
[ 2
k
s (s+1 ) ]
G ( z )= ( 1−z ) k (
−1
(
0.018273+0.01752z
( 1−z 1−0.81871−z
( −1 2
z ) −1−1

)
) −1
)
k ( 0.01873z+0.01752 )
G ( z )=
(z−1)(z−0.81871)
We perform a bilinear transformation.
1+0.5Tw1+ 0.1w
z= =
1−0.5Tw1−0.1w
 SCHOOL OF BASIC SCIENCES TECHNOLOGIES AND ENGINEERING
DIGITAL CONTROL

1+0.1w
0.01873 +0.01752
1−0.1w
G ( w)=k
(
1+ 0.1w
1−0.1w
−1
1+0.1w
1−0.1w )(
−0.81871 )
k(−0.000332653w2−0.096332w+0.996585)
G ( w )=
w2 +0.996805w
With Kv=2 we assume a transfer function for the digital controller.

1+tw
G d ( w )=
1−αw

1+tw k(−0.000332653w2−0.096332w+0.996585)
k v =lim
w→0 ( 1−αw
( ) w2 +0.996805w
)=2

2
k= =2
0.999779
Now we proceed to design a controller that has an angle of 50° without
affect the constant k.
Thus 50-31.6=18.4, a margin between 8 and 12 degrees is recommended considering
the shift of the gain at the crossover frequency will be assumed to be 28°
maximum phase advance angle. To calculate the attenuation factor.
1−α
sinϕm=
1+α
We replace it with the angle
1−α
sen28=
1+α
( 1+α ) sen28=1−α
a=0.361
We find the point where the magnitude is not balanced.
we replace by w=jv in G(w).
1
−20 log =−4.425
√ 0.361

(√ ( ) √ ( ) )
2 2
2 v v
1+ 1+
300 10
vertG ( jv )=
v ❑√ 1+ v 2
The frequency is equal to 1.7
The crossing frequency is given by:
 SCHOOL OF BASIC SCIENCES TECHNOLOGIES AND ENGINEERING
DIGITAL CONTROL

1
v c= =1.7
√ ατ
τ =0.9790ατ=0.3534

The compensator would be as follows:

1+τw1+0.9790w
G d ( w )= =
1−αw1+0.3534w

( ( −0.000332653w−0.096332w+0.996585
)(
2
1+0.9790w
)
)
G D ( w ) G ( w )= 2
w + 0.996805w 1+0.3534w
We convert the transfer function of the compensator to z terms by
bilinear transformation medium where T=0.2:
z−1
w=10
z +1

G ( w )=
d
( z+1 )
z −1
1+0.9790 10

1+0.3534 (10
z+ 1 )
z−1

2.3798z−1.93869
G D ( z )=
z−0.558888
In MatLab we have:
 SCHOOL OF BASIC SCIENCES, TECHNOLOGIES, AND ENGINEERING
DIGITAL CONTROL

2. PRACTICAL ACTIVITY
ACTIVITY 1
First, we must find the response of the plant in open loop:
Open loop plant response
p1=-1
p2=-2
k=10
Gps=zpk([],[p1 p2],k)
step(Gps);
 SCHOOL OF BASIC SCIENCES, TECHNOLOGIES, AND ENGINEERING
DIGITAL CONTROL

CLOSED LOOP
Glcsc=tf(feedback(Gps,1))
[numsc, densc] = tfdata(Glcsc, 'v');
Wn=sqrt(densc(3))
zeta=(densc(2))/(2*Wn)
step(Glcsc,5);

DESIGN CONDITIONS
ts=1.9;
SI=19;
zita=-(log(SI/100))/(sqrt(pi^2+(log(SI/100))^2))
wn=4/(zita*ts)
name
0.4673
wn =
4.5047
Gcs=zpk([p1 p2],[0 -2*zita*wn],wn^2/k)

Gcs =

2.0293 (s+1) (s+2)

------------------
 SCHOOL OF BASIC SCIENCES, TECHNOLOGIES AND ENGINEERING
DIGITAL CONTROL

s (s+4.211)

Continuous-time zero/pole/gain model.

Glccc=minreral(feedback(Gcs*Gps,1))

Glccc =

20.293
----------------------
(s^2 + 4.211s + 20.29)

Continuous-time zero/pole/gain model.

step(Glccc,3)

Test
T=0.1;
Gz = zpk(c2d(Gps, T, 'zoh'))

Gz =

0.04528 (z+0.9048)
---------------------
 SCHOOL OF BASIC SCIENCES TECHNOLOGIES AND ENGINEERING
DIGITAL CONTROL

(z-0.9048) (z-0.8187)

Sample time: 0.1 seconds

Discrete-time zero/pole/gain model.
Gcz=zpk(c2d(Gcs,T,'tustin'))
Gcz =

1.9362 (z-0.9048) (z-0.8182)

----------------------------
(z-1) (z-0.6522)

Sample time: 0.1 seconds

Discrete-time zero/pole/gain model.
Glcz = minreal(feedback(Gcz * Gz, 1))

Glcz =

0.087669 (z+0.9048) (z-0.9048) (z-0.8182)

--------------------------------------------
(z-0.818) (z-0.9047) (z^2 – 1.565z + 0.7321)

Sample time: 0.1 seconds

Discrete-time zero/pole/gain model.
step(Glcz)
 SCHOOL OF BASIC SCIENCES TECHNOLOGIES AND ENGINEERING
DIGITAL CONTROL

Point 2
numDz = [0.01873 0.01752];
denDz = [(1-1) (1-0.81871)];
Kv = 2
sys = Kv*tf(numDz, denDz, 0.2)
margin(sys)
sisotool(sys)

This is the Uncompensated system

 SCHOOL OF BASIC SCIENCES TECHNOLOGIES AND ENGINEERING
DIGITAL CONTROL

This is the compensated system

SCHOOL OF BASIC SCIENCES TECHNOLOGIES AND ENGINEERING
DIGITAL CONTROL
SCHOOL OF BASIC SCIENCES, TECHNOLOGIES, AND ENGINEERING
DIGITAL CONTROL