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

net/publication/335464283

Design and Simulation of DC Motor Speed Controller using Pole Placement

Technique and MATLAB

Article · July 2016

CITATIONS READS

0 802

1 author:

Ahmed Abdelrahman A.
Karary University
16 PUBLICATIONS   64 CITATIONS   

SEE PROFILE

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

Robotics - Design and implementation View project

Biometrics Identification and Authentication View project

All content following this page was uploaded by Ahmed Abdelrahman A. on 10 December 2020.

The user has requested enhancement of the downloaded file.

 International Journal of Engineering Sciences Paradigms and Researches (IJESPR)
(Vol. 32, Issue 01) and (Publishing Month: July 2016)
(An Indexed, Referred and Impact Factor Journal)
ISSN: 2319-6564
www.ijesonline.com

Design and Simulation of DC MOTOR Speed Controller

using Pole Placement Technique and MATLAB

Ahmed Abdelrahman A.A.1, Asaad Musaab Ali Yousif2, Abubker Dahab3, Mohammed Ahmed A.A.4 and
Abdalbagy Ahmed M.5
1
Department of Electrical and Computer Engineering, Karary University, Omdurman, Sudan
mogwari2000@yahoo.com
2
Department of Electrical and Computer Engineering, Karary University, Omdurman, Sudan
abuasgad14@gmail.com
3
Department of Electrical and Computer Engineering, Karary University, Omdurman, Sudan
asaadm.mosap@gmail.com
4
College of Post Graduate Studies, Karary University Omdurman, Sudan
moh204b40@yahoo.com
5
College of Post Graduate Studies, Karary University Omdurman, Sudan
abdalbagya55@gmail.com

Publishing Date: July 27, 2016

Abstract High performance motor drives are necessary

Motor speed is very important in rotating machinery parts of industrial applications. A DC motor
applications. This thesis describes DC motor speed drive system with high performance has special
control based on pole placement feedback control characteristics such as good dynamic speed
technique. The pole assignment technique specified command tracking and load regulating response.
closed –loop poles such a system where the reference
input always zero is called regulator system. The
DC motors are well known for their excellent
problem of regulator poles (closed –loop poles) at control of speed for acceleration and deceleration
desired location is called pole assignment problem. [1].
And this can be done if the system is completely state
controllable. Controller's objective is to maintain the
speed of rotation of motor shaft with a particular step PID Controllers
response. By using pole placement technique specifies
state feedback and get good results are as enable PID Controllers used in large quantities for the
Overshoot less than 1.73%, Settling time less than choice when controller needed a closed loop.
1.16 sec, No steady state error and MATLAB 2013 PID provides the designer huge options if
software. designer want to change the dynamics of system.
Keywords: DC Motor, Speed Control, Pole Almost in any industry 90% controllers are PI
Placement Technique. controllers while the rest 10% are PID and other
controllers. If everything goes in right direction
then designer can get following advantages. P
Introduction part reduces step time to reach to designed
Control theory is an interdisciplinary branch position. This part removes the steady state error.
of engineering and mathematics that deals with I integral part removes the overshoot got by p. D
the behavior of dynamical systems. The usual part used to reach error but if system is noisy
objective of control theory is to determine then D part not used [2].
solutions for the proper corrective action from
the controller that result in system stability.
IJESPR
www.ijesonline.com
8
 International Journal of Engineering Sciences Paradigms and Researches (IJESPR)
(Vol. 32, Issue 01) and (Publishing Month: July 2016)
(An Indexed, Referred and Impact Factor Journal)
ISSN: 2319-6564
www.ijesonline.com

Pole Placement Controller controller give better response, it means it can

used as a controller to the real time DC Motor
Pole Placement controller provides transient error [3].
excellent control as designer can adjust the time
to reach on the reference point by adjusting It presents a novel method for design and
poles. This controller use most straight forward simulation of a DC motor closed loop control
design. It starts with assumption that what system in Field Programmable Gate Array
controller wants to do. From the assumption device (FPGA). The digitally designed PID
symbolic characteristic equation is formed, and controller is more advanced, beneficial and
poles determined which leads to design therefore produces a better response, as
overshooting rise time. In this controller mostly compared to the analog PID with velocity
equation is 2nd order and most equation has measurement. It can be implemented on system-
more that 2 poles. on-chip devices easily. MATLAB Xilinx system
By this way equation can be formed and model generator toolbox based on Fixed-Point
is formulated as discrete transfer function. Arithmetic is used to design the digital PID
Disturbance model is not used when designer use controller using DSP architecture, plot the
pole placement design. Integrator design is used responses of the control system and generate the
to drive steady state to 0. In Pole Placement VHDL source code. The control system consists
poles are placed in closed loop transfer function of a digital PID, the model of a real DC motor, a
in reasonable position [2]. real and new incremental optical shaft encoder
model, encoder to rpm and position blocks.
Results show that the proposed system leads to
Fuzzy Controller lower the steady state error [4].

Fuzzy controller is an innovative technology that Many industrial applications require high
modifies the design of systems with engineering performance rotating electric drives. A proposed
expertise. Fuzzy logic use human knowledge to D C drive have precise speed control, stable
implement a system. It is more effective than operation in complete range of speed and good
PID controller as its reach to its reference level transient behavior with smooth and step less
in less time. It is mostly use in that no control. The purpose of developing a simulation
mathematical equations for handling system. using PI and PID control system is to get steady
Common sense, human thinking and judgment state and transient response of drive system.
are fuzzy rules. It helps engineers to solve non Once the type of controller has been decided
linear control problems. It mathematically then the design and analysis are done. It show
emulates human knowledge for intelligent focuses modeling of separately excited DC
control system and complex application [2]. motor for the analysis of machine under any
condition and compares the step response of
system with and without PI and PID controller
Previous Studies [5].
It presents an overview of Proportional Integral Presents the simulation and experimental
control (PI) and Artificial Intelligent control (AI) investigation into the development of PID
algorithms. AI and PI controller are analyzed controller using MATLAB/SIMULINK
using Matlab [Simulink] software motor. The software. The simulation development of the
main objective is illustrates how the overshoot PID controller with the mathematical model of
and steady state error of the DC motor can be DC motor is done using Ziegler–Nichols method
controlled using different controllers. The and trial and error method. The PID parameter is
simulation results demonstrate that the responses to be tested with an actual motor also with the
of DC motor with an AI control which is Fuzzy PID controller in MATLAB/SIMULINK
Logic Control shows satisfactory well damped software. In order to implement the PID
control performance. The results shows that controller from the software to the actual DC
Industrial DC Motor model develop using its motor data acquisition is used. From the
physical parameters and controlled with an AI
IJESPR
www.ijesonline.com
9
 International Journal of Engineering Sciences Paradigms and Researches (IJESPR)
(Vol. 32, Issue 01) and (Publishing Month: July 2016)
(An Indexed, Referred and Impact Factor Journal)
ISSN: 2319-6564
www.ijesonline.com

simulation and the experiment, the result System Equation

performance of the PID controller is compared in
term of response and the assessment is presented The Motor Torque
[6].
The motor torque is related to armature current (ia) by
When the impulsive modes of the system are constant factor (kt) is given by:
both controllable and observable, it is shown in (1)
this paper that these modes can be eliminated by
almost any constant output feedback. For a The back emf is related to the rotational velocity
strongly controllable and observable generalized is given by:
system, after its impulsive modes are eliminated, (2)
the poles of the system can be assigned In SI unit( which we will use) kt (armature
arbitrarily by state feedback. Even if the states constant) is equal kw (motor constant).
are not available, the paper adopts an efficient
method to design a compensator for the system From the Figure (1) we can write the following
[7]. differential equations based on Newton's law
combined with Kirchhoff's law
The problem of designing a compensator to (3)
obtain arbitrary pole placement in the system
consisting of the plant and compensator in (4)
cascade is considered. The design uses only
those state variables which can be measured. It is (5)
shown that for a controllable observable plant a
compensator of order β=min⁡(v_c-1,v_o-1) is (6)
sufficient to achieve this result. Here v_c and
v_o is the controllability and observability index The state equation can be written as:-
of the plant. This result is obtained by first
showing that any multi-input multi-output linear (7)
time-invariant system may be made controllable (8)
(observable) from a single input (output) using
only output feedback. The main result is then From the both equations (5) and (6) above, it can
proved in a constructive manner which explicitly be representing in vector form in other word state
relates the compensator parameters to the space equation
coefficients of the desired characteristic
polynomial [8].
(9)
ABDUL LATIF AIZAT presents an overview of
motor speed control to develop the pole
placement controller to control the speed of DC
motor and implement the controller into ladder Y= + (10)
diagram which connected to plc. Finally analysis
of response is made after the pole placement is In order to make the DC motor modeling , the
implemented into the system [9]. values of the physical parameters should be assumed.
these values of parameter are derived

(J) Moment of inertia of the rotor 0.02215

kg.m^2
(b) Motor viscous friction constant 0.002953
N.m.s
(Ke) electromotive force constant 1028
Figure 1: Free Body Diagram of DC- Motor V/rad/sec
Speed
IJESPR
www.ijesonline.com
10
 International Journal of Engineering Sciences Paradigms and Researches (IJESPR)
(Vol. 32, Issue 01) and (Publishing Month: July 2016)
(An Indexed, Referred and Impact Factor Journal)
ISSN: 2319-6564
www.ijesonline.com

(Kt) Motor torque constant 1.28 After create a state space and output
N.m/Amp equation , the substituting the parameters into
(R) Electric resistance 11.2 Ohm vector form:
(L) Electric inductance 0.1215 H

Modern control design is especially usefully in

multivariable systems. One approach in modern
control systems accomplished by the use of state
feedback is known as pole-placement design .the
pole-placement design allows all roots of the Y= +
system characteristic equation to be placed in
desired locations. This results in a regulator with
constant gain vector k.

Table 1.1: Parameter of DC Motor Plant (Subsystem)

A, n × n constant matrix [-92.1818 -10.535;57.7878 -0.1333]

B, n×1 constant matrix [250; 0]

C, 1×n constant matrix [0 1]

D, feed forward [0]

Before pole placement can be applied, we shall First method to find the matrix gain, k is by
choose the control signal to be using the MATLAB command (prompt). In
U=-k x MATLAB prompt, use the coding
The control signal u is determined by an
instantaneous state such a scheme is called state k=place(A,B,P) to obtain the gain values.
feedback. The 1×n matrix, k is called the state k=place (A,B,P) computes feedback gain matrix,
feedback gain matrix, In MATLAB place k that achieves the desired closed loop pole
command calculates the appropriate gains to locations p, assuming all the inputs of the plant
place the pole at desired position in the Laplace are control inputs. The length of p must match
s-domain Recall that pole placement control use the row size of A matrix.
state feed back to generate the proper command Other method is by determining the gain matrix,
input, k by using mathematical calculation. To obtain
the k value, first is finding the poles, p value by
varying the percentage of overshoot (%OS) and
And a vector p of desired self-conjugate closed – settling time (TS). But to vary the values, must
loop pole locations. Place computes gain matrix refer to the Bench mark of motor. For the Clifton
k such that the state feedback, U=-k x places the precision servo motor model, the bench mark is
closed –loop poles at the location p. referring to table below

IJESPR
www.ijesonline.com
11
 International Journal of Engineering Sciences Paradigms and Researches (IJESPR)
(Vol. 32, Issue 01) and (Publishing Month: July 2016)
(An Indexed, Referred and Impact Factor Journal)
ISSN: 2319-6564
www.ijesonline.com

Table 1.2: The Bench Mark for the Clifton Necessary and Sufficient Condition
Precision Servo for Arbitrary Pole Placement
the percentage of overshoot
<1.73% Consider a control system defined by Equation
(%OS) (7). Let the control signal selected be .
Substituting control signal into Equation (7), we
have:
(11)
settling time (TS) <1.16s
The solution of the above equation is given as:

The poles, p equation:

The eigenvalues of the matrix are
P1=-α+ j ω and p2=-α- j ω (11) nothing but desired closed loop poles. The
α=ξ ωn (12) necessary and sufficient condition for arbitrary
w=ωn (13) pole placement is that the system is completely
state controllable. If suppose the system is not
The damping ratio is one the compares the completely state controllable then there are
exponential decay frequency of the envelope to eigenvalues of a matrix that cannot be
the natural frequency . We define the damping controlled by state feedback. The eigenvalues of
ratio, ξ to be: matrix are called regulator poles.
These regulator poles when placed in left half of
ξ= s-plane then approaches zero as time t
approaches infinity. The problem of placing the
But as we vary the %OS, the inverse %OS closed-loop poles at the desired location is called
equation allows one to obtain the damping ratio, a pole placement problem [10].
ξ. The inverse is giving by:-

ξ= =0.7906 (14)

The natural frequency, wn of second order

system is the frequency of oscillation of the
system without damping we can obtain the
natural frequency value by the settling time, TS
equation

Wn=4/ξ TS=3.1624 (15) Figure 2: Control System Design via Pole

After solving the term of damping ratio and Placement Controller
natural frequency which is equation 14 and 15,
substitute the values into equation 12 and 13 There are many methods to obtain k in
when α and w are solve. Next is substituting α MATLAB code in Appendices used
and w into equation 11 which is the poles Ackermann's method
equation. in the
code write parameters A,B,C and D, damping
α=-2.5002, ω=1.9364 ratio, ξ=0.7906, the natural frequency value ω
=3.1624 after that get poles from second order
p1=-2.5002+1.9364j, p2=-2.5002-1.9364j equation by using Ackermann's method to obtain
k after that apply state feedback(closed –loop)
equation and transfer function in the time from 0
to 10 step 0.1 run m file to plot step response of
compensated system
IJESPR
www.ijesonline.com
12
 International Journal of Engineering Sciences Paradigms and Researches (IJESPR)
(Vol. 32, Issue 01) and (Publishing Month: July 2016)
(An Indexed, Referred and Impact Factor Journal)
ISSN: 2319-6564
www.ijesonline.com

Figure 3: Simulink of Pole Placement Design

Methodology
In this research the mathematical model of the
DC motor is presented so as to control the speed
of the DC motor using pole placement techniques
by obtain the state feedback and using
MATLAM to evaluate them and simulation.

Figure 5: Uncompensated Step Response

Simulation Results
Adding a Pre Compensator
The simulink state space model can be used to
obtain the response. A simulink model is From this plot we see that the steady-state error
constructed as shown in figure (1) the state space is too large. One approach for eliminating the
description diag box is open. And A, B, C and D steady-state error is to simply scale the input so
constants are entered in the appropriate box in that the output in turn is scaled to the desired
MATLAB matrix notation. The simulation level. This is little challenging in our example
parameters are set to appropriate values. because we have two states to consider.
Therefore, we need to compute what the steady-
state values of both states should be, multiply
them by the chosen gain k, and use the result as
our ''reference'' for computing the input u. This
can be done in one step by adding constant gain
pre compensator after the reference as shown in
the following schematic.

Figure 4: Simulation Block Diagram

The file is opened and is run in the simulink
window. The simulation results is response
shown in fig (4) below
Figure 6: Adding Constant Gain Pre Compensator

We can find ne in MATLAB code from apply

state feed back in the system we get step
response which satisfy the specifications
mentioned above.

IJESPR
www.ijesonline.com
13
 International Journal of Engineering Sciences Paradigms and Researches (IJESPR)
(Vol. 32, Issue 01) and (Publishing Month: July 2016)
(An Indexed, Referred and Impact Factor Journal)
ISSN: 2319-6564
www.ijesonline.com

%OS - Percentage overshoot

Ts - Settling time
ξ - Damping ratio
wn –Natural frequency

Code system
os=input('enter over shoot')
ts=input('inter settling time')
zeta=-log(os)/sqrt(pi^2+(log(os)^2))
wn=4/(zeta*ts)
A=[-92.1811 -10.5350;57.7878 -0.1333];
B=[8.2305;0];
C=[0 1];
P=roots([1 2*zeta*wn wn^2])
Figure 7: The Compensate Step Response num=[wn^2]
den=[[1 2*zeta*wn wn^2]]
From the figure above we have:
[A2,B2,C2,D2]=tf2ss(num,den)
sys1=tf(num,den)
1. Settling time less than1.16 second.
k=acker(A,B,P)
2. Overshoot less than 1.73 .
A1=A-B*k;
3. No steady state error. ne=-(([0 1]*([A1]^-1)*[8.2305;0]))^-1
[n d]=ss2tf(A1,B*ne,C,0)
sys=tf(n,d)
Conclusion t=0:0.01:10;
[x y]=step(sys,t);
As we mentioned previously the control
plot(y,x)
speed of DC motor is really problem in
figure
engineering field. However deciding to control it
step(sys,t)
by using pole placement technique specifies state
feedback and get good results are as enable
Overshoot less than 5% , Settling time less than
References
2 sec, No steady state error
[1] D. A. staton, M. I. McGilp and T. J. E.
Miller, “DC machine teaching experiment,”
List of Symbols in proceedings of the European Power
Electronics Association, 1993
J -Moment of inertia of the motor
[2] Ali Junaid Ashraf , "DC motor Control via
B -Damping ratio of the mechanical system
Fuzzy /Pole Placement /P I Controller " JUN
K –Electromotive force constant
2010
R –Electric resistance
[3] Abdulrahman A.A.Emhemed, Modeling and
L - Electric inductance
simulation of DC motor control system
ia - Electric current
using intelligent control ,International
W r - Speed /velocity
Journal of control,2012.
x – State vector (x – vector)
[4] Masoud Mansouryar, Modeling and
U -Control signal (input)
simulation of DC motor control system with
A –n × n constant matrix, system matrix
digital PID controller and encoder in FBCA
B - n × n constant matrix, input matrix
using Xilinx system generator, International
C - 1 × n constant matrix, output matrix
Journal of control,2011 .
D - 1 × 1 constant matrix, feed forward matrix
[5] Ms.Manjusha Patial, Modeling and
K - Gain matrix (for the controller)
simulation of DC drive using PI and PID
P - Poles
controller, International Journal of
IJESPR
www.ijesonline.com
14
 International Journal of Engineering Sciences Paradigms and Researches (IJESPR)
(Vol. 32, Issue 01) and (Publishing Month: July 2016)
(An Indexed, Referred and Impact Factor Journal)
ISSN: 2319-6564
www.ijesonline.com

innovative research in Electrical, Electronic,

Instrumentation and Control Engineering,12
/12/2014.
[6] Mohamed Farid, Mohamed Faruq (2008)
PID controller design for DC motor using
matlab application. Faculty of Electrical &
Electronic Engineering, Universiti Malaysia
Pahang.
[7] Yue-Yun Wang, Song-Jaio SHI, Zhong-Jun
ZHANG, “Pole placement and Compensator
Design of Generalized Systems”,
International Journal of Control, Aug. 1986.
[8] Frederick M. Brasch, James B. Pearson,”
Pole Placement Using Dyna-mic
Compensators”, International Journal of
Control, Feb. 1980.
[9] Abdul Latlf , "Speed Control Of DC Motor
Using Pole Placement Controller
Implementation With PLC " NOV 2008
[10] U. A. Bakshi, M. V. Bakshi, “Modern
Control Theory”, Technical Publications
Pune, 2008.

IJESPR
www.ijesonline.com
15

View publication stats