See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/310416585

DC MOTOR SPEED CONTROL USING PID

CONTROLLER

Conference Paper · June 2005

CITATIONS READS

0 872

1 author:

Fatiha Loucif
Université 8 mai 1945 - Guelma
1 PUBLICATION 0 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

i am working on adaptive control of robot manipulator View project

All content following this page was uploaded by Fatiha Loucif on 16 November 2016.

The user has requested enhancement of the downloaded file.

 ICCAS2005 June 2-5, KINTEX, Gyeonggi-Do, Korea

DC MOTOR SPEED CONTROL USING PID CONTROLLER

Fatiha Loucif
Department of Electrical Engineering and information, Hunan University, ChangSha, Hunan, China
(E-mail:[email protected])

Abstract. The PID controller design and choosing PID parameters according to system response are proposed in this paper. Here
PID controller is employed to control DC motor speed and Matlab program is used for calculation and simulation. Choosing PID
parameters are demonstrated by several contrast experiments and a way for setting PID parameters values is discussed.

Keywords: PID controller, DC motor, Matlab representation, P control, PI control, PID control.

1. INTRODUCTION error signal is the input to the PID controller.

The best-known controllers used in industrial control Before examining the input-output relationships and design
processes are proportional-integral-derivative (PID) methods for the PID controller, it is helpful to review typical
controllers because of their simple structure and robust
characteristics observed for the velocity response of a DC
performance in a wide range of operating conditions. The
design of such a controller requires specification of three motor to a step voltage input. Different characteristics of the
parameters: proportional gain, integral time constant and motor response (steady-state error, peak overshoot, rise time,
derivative time constant .so far, great effort has been devoted settling time, etc) are controlled by selection of the three gains
to develop methods to reduce the time spent on optimizing the that modify the PID controller dynamiclly.
choice of controller parameters [1].
This is discussed in detail below .the PID controller is defined
Setting the PID parameters are called tuning ,the classical
tuning method is to use the famous Ziegler-Nichols tuning by the following relationship between the controller input e (t)
formula. This tuning formula however, is not always effective and the controller output u (t) that is applied to the motor
in the sense that it may completely fail to tune the processes armature:
with, for example, relatively large dead time. t
de(t )
u (t ) = kp × e(t ) + ki × ∫ e(τ ) × dτ + kd ×
Moreover, its tuning will have to be supplemented with
purely experience-based fine-tuning to meet the specifications (1)
[2]. 0
dt
The parameters of the PID controller kp, ki and kd (or kp, Taking the Laplace transform of this equation gives the
Ti and Td) can be manipulated to produce various response transfer function G(s):
curves from a given process as we will see later.
u (t ) ki
G(s) = = (kp + + kd × s ) (2)
2. PID CONTROLLER e(t ) s
PID controllers are commonly used to regulate the
This transfer function clearly illustrates the proportional,
time-domain behavior of many different types of dynamic
plants. These controllers are extremely popular because they integral and derivative gains that make up the PID
can usually provide good closed-loop response characteristics, compensation. Select new definitions for the gain terms
can be tuned using relatively simple design rules, and are easy according to:
to construct using either analog or digital components.
Consider the feedback system architecture that is shown in Fig.
kp = k .
1 where it can be assumed that the plant is a DC motor whose k
speed must be accurately regulated. ki =
Ti
R (t) y (t) kd = k × Td
PID Plant k
Controller G ( s) = (k + + k × Td × s )
Ti × s
1
G ( s ) = k (1 + + Td × s )
Ti × s
Fig. 1 Feedback system architecture 1
G ( s ) = kp(1 + + Td × s ) (3)
The PID controller is placed in the forward path, so that its Ti × s
output becomes the voltage applied to the motor's armature The discrete-time equation expression is given as:
.the feedback signal is a velocity, measured by a tachometer n
kd
.the output velocity signal y (t) is summed with a reference or u (k ) = kp × e(k ) + ki × T s ∑ e(i ) + ∆e(k ) (4)
command signal R (t) to form the error signal e (t). Finally, the i =1 T s
 ICCAS2005 June 2-5, KINTEX, Gyeonggi-Do, Korea
Here, u (k) is the control signal, e (k) is the error, Ts is the by eliminating i(s) we can get the following open-loop transfer
sampling period for the controller, and function ,where the rotational speed is the output and the
∆e(k ) ≈ e(k ) − e(k − 1) voltage is the input .

2.1 Physical setup and system equation:

DC motors have speed-control capability, which means that θ k
speed, torque and even direction of rotation can be changed at = .
( J × s + b) × ( L × s + R ) + k
2
V
any time to meet new conditions [3] .The electric circuit of the
armature and the free body diagram of the rotor are shown in
the following figure:
3. DESIGN REQUIREMENTS:
First, our uncompensated motor can only rotate at 0.1 rad/sec
with an input voltage of 1 volt (this will be demonstrated later
when the open-loop response is simulated. since the most
basic requirement of a motor is that it should rotate
At the desired speed, the steady-state error of the motor speed
should be less than 1% .the other performance requirement is
that the motor must accelerate to its steady-state speed as soon
as it turn on. In this case, we want it to have a settling time of
Fig. 2 electric circuit and free body diagram less than 2 seconds. Since a speed faster than the reference
may damage the equipment, we want to have an overshoot of
For this example, we will assume the following values for the less than 5%.
physical parameters. If we simulate the reference input(R) by a unit step input, then
Moment of inertia of the rotor (J) =0.01 kg.m^2/s^2. the motor speed output should have:
Damping ratio of the mechanical system (b)= 0.1 nms. Settling time less than 2seconds.
Electromotive force constant (k=ke=kt)=0.01 nm/amp. Overshoot less than 5%.
Electric resistance (R) =1 ohm. Steady-state error less than 1%.
Electric inductance (L)= 0.5 H.
Input (v): source voltage. 4- MATLAB REPRESENTATION AND
SYSTEM RESPONSE:
Output (theta): position of shaft.
The rotor and shaft are assumed to be rigid. 4.1. Open-loop response:
The motor torque T, is related to the armature current i, by a
constant factor kt. The back emf, is related to the rotational 4.1.1. Transfer function: we can represent the above transfer
velocity by the following equations: function into Matlab by the numerator and denominator

e = emf = ke × θ& .
matrices as follows
T = kt × i .
num =k
den :=(J*s +b)*(L*s+R) +k^ 2
In SI units (which we will use),kt (armature constant) is equal
*The m-file is as follow (program 1):
to ke (motor constant) .
J=0.01;
From the figure above we can write the following equations
based on Newton’s law combined with kirchhoff`s law: b=0.1;
k=0.01;
J × θ&& + b × θ& = k × i .
R=1;
di
L × + R × i = V − k × θ&. L=0. 5;
dt num=k;
den=[(J*L) ((J*R)+(L*b)) ((b*R) + k^2)];
2.2 Transfer function: Step (num, den, 0:0.1:3)
Title (`step response for the open- loop system `)
Using Laplace transforms, the above modeling equations can
be expressed in terms of s.
S ( J × s + b) × θ ( s ) = k × i ( s ).
( L × s + R) × i ( s) = V − k × s × θ ( s).
 ICCAS2005 June 2-5, KINTEX, Gyeonggi-Do, Korea
calculation results was assumed in Table 1, and the
corresponding plots was gathered in Fig. 4:

Table 1

Maximum
Steady
Rise overshoot Peak
kp State
time(sec) (%) amplitude
error
10 0.488 / 0.5 0.5
0 0.301 / 0.334 0.681
50 0.158 / 0.167 0.94
70 0.125 4 0.125 1.04
100 0.097 14 0.091 1.14
200 0.064 32 0.048 1.32
300 0.051 41 0.032 1.41

Fig. 3 open loop response.

We see that the proportional controller kp have the effect of
From the plot we see that when 1 volt applied to the system, reducing the rise time; from (0.48sec) for kp=10 to (0.051sec)
the motor can only achieve a maximum speed of 0.1 rad/sec, for kp=300; and will reduce and never eliminate the steady-
ten times smaller than our desired speed. Also, it takes the state error from (0.5) for kp=10 to (0.032) for kp=300. so the
motor 3 seconds to reach its steady-state speed; this does not response becomes more and more faster by increasing the gain
satisfy our 2 seconds settling time criterion. kp. But will increase the maximum overshoot; from (4% to
41%).
4.2. Closed-loop response:
The closed-loop program looks like (program 2):
J=0. 01;
b=0.1;
k=0.01;
R=1;
L=0. 5;
num=k;
den=[ (J*L) ((J*R)+(L*b)) ((b*R) + k^2)] ;
%%%%% Different values of kp , ki and kd
%%%%%%%%%
kp= ;
ki= ;
kd= ;
numc=[kd,kp,ki];
denc=[0 1];
numa=conv(num,numc); Fig. 4 closed loop response with different values of kp .
dena=conv(den,denc);
6- PROPORTIONAL INTEGRAL CONTROL
[numac,denac]=cloop(numa,dena);
(PI):
step(numac,denac) The settling is time still too long, we try now to increase the
xlabel('time(sec)'), ylabel('velocity(rad/sec)') gain ki with kp=100, all the results are shown in the Table 2.
Title (`pid control `) And the corresponding plot is gathered in Fig. 5.
Grid;

5- PROPORTIONAL CONTROL:

Let's first try using a proportional controller; so we set ki=0,

kd=0 in program (2) And we take different values of kp the
 ICCAS2005 June 2-5, KINTEX, Gyeonggi-Do, Korea
Table 2

Maxi-
Rise Peak
mum Steady
settling time amplitu-
ki oversh state
time(sec) (sec) de
-oot error
(%)
1 / 0.111 13.6 0.088 1.14
10 / 0.11 14.4 0.067 1.14
20 / 0.11 15.3 0.05 1.15
50 2.69 0.107 17.9 0.017 1.18
100 1.03 0.104 22.2 0.001 1.22
200 0.774 0.098 30.5 0 1.3

Fig. 6 closed loop response with different values of kd.

So now we know that if we use a pid controller with:

kp=100,ki=200 and kd=10; all our design requirements will be
satisfied and the response looks like:

Fig. 5 closed loop response with different values of ki.

Now we see that the response is much faster than before; the
settling time becomes (0.774 sec) for ki=200, and the
steady-state error becomes very small and eliminated for
ki=200.

7-PROPORTIONAL-INTEGRAL-DERIVATIV
E CONTROL (PID):
Now, we increase the gain kd, with kp=100;ki=200. All
results are illustrated in the Table 3 and the corresponding
plots are shown in Fig. 6. Fig. 7 system closed loop response with PID control.
From the results we see that the settling time reduced from
(0.58 sec to 0.257sec) for (kd=1 to kd=10) and the overshoot
from (23% to 1.03%) and there is small change in the rise Effects of PID controllers parameters kp, ki and kd on a closed
time. loop system are summarized in the table bellow.
Table 3

Rise Maxi-
Steady Peak
settling time mum
kd state amplitude
time(sec) (sec) oversh
error
oot(%)
1 0.588 0.106 23 0 1.23
3 0.403 0.116 12.6 0 1.13
5 0.419 0.121 6.09 0 1.06
8 0.201 0.127 1.43 0 1.01
10 0.257 0.132 1.03 0 1.01
 ICCAS2005 June 2-5, KINTEX, Gyeonggi-Do, Korea

Table 4

Maximum settling
closed Rise Steady-
Overshoot
loop time State
(%) time
response (sec) error
(sec)
as the decrease increase Small
increase change decrease
of kp
as the decrease increase increase
increase eliminate
of ki
as the Small decrease decreas-
increase change e Small
of kd change

CONCLUSION:
From the experimental results we know that the proportional
controller (kp) will have the effect of reducing the rise time
and will reduce; but never eliminate the steady-state error, an
integral control (ki) will have the effect of eliminating the
steady-state error, but it may make the transient response
worse .a derivative control (kd) will have the effect of
increasing the stability of the system, reducing the overshoot,
and improving the transient response.

REFERENCES

[1] Zhen-yu zhao ,Masayoshi Tomizuka ,and

Satoru Isaka , “ Fuzzy gain of PID controller,
” Members,IEEE 1392p .

[2] S.Z.HE1 S.H.TAN and F.L.Xu2 P.Z.Wang3

department of electrical engineering, NUS10
kent Ridge crescent singapore0511 “PID self-
tuning control using a fuzzy adaptive
mechanism,” ,p708.

[3] Christopher T.kilian “Modern control

technology: components and systems,” 2end
edition, p295 .

[4] Robert H.Bishop, “Modern control system

Analysis and Design Using MATLAB,”

[5] Chapter 1: plant process characterization

and PID see website
<http://entc462.tamu.edu/ClosedLoopControlS
ystem.PDF>.

View publication stats