Nonlinear Control of a DC Motor
Siti Hawa Abu-Bakar
School of Engineering and Built Environment,
Glasgow Caledonian University, United Kingdom
Universiti Kuala Lumpur – British Malaysian Institute,
Malaysia
E-mail: sabuba10@gcu.ac.uk/hawa012@gmail.com
Firdaus Muhammad-Sukki
School of Engineering and Built Environment,
Glasgow Caledonian University, United Kingdom
Faculty of Engineering,
Multimedia University, Malaysia
Khairil Anuar
Faculty of Engineering,
Multimedia University, Malaysia
Abdullahi Abubakar Mas’ud
School of Engineering and Built Environment,
Glasgow Caledonian University, United Kingdom
Abstract—This paper deals with a nonlinear system where an
online nonlinear control is applied to a DC motor. First,
nonlinear model of the motor is explained. Next, based on the
model, the nonlinearity is simulated using MATLAB. The
simulations were on two nonlinearities, ideal relay and relay with
hysteresis. To further investigate the relay with hysteresis, the
implementation of the LabVIEW program is presented to verify
the results obtained from the MATLAB simulations. It was found
that experiments show good agreements with the simulation
results.
Keywords— DC motor, nonlinear control, relay.
I. INTRODUCTION
For linear systems, there are many well-established control
techniques, for example root-locus, Bode plots, Nyquist
criterion, state-feedback, pole-placement etc. [1]-[4]. But for
nonlinear system, it is much more complex. The objective of
this project is to analyse and design a way of controlling the
motor using on-line nonlinear control. In this particular project,
a relay with hysteresis was chosen as the nonlinear controller.
II. NONLINEAR MODEL OF THE SYSTEMS
Consider theoretically the system in Fig. 1 having the relay
with hysteresis of gain 1 D and width 5 . 0 h as illustrated
in Fig. 2.
Fig. 1: Circuit using Relay with hysteresis as controller.
Fig. 2: Relay with hysteresis
The parameters of A, B and C in Fig. 1 were determined
from experiments [1] and the values are
389 . 3 0
1 0
A ,
656 . 20
0
B and 1405 . 0 0 C . The transfer function is
defined as:
B A sI C s G
L 1
) ( ) (
(1)
and based on the values obtained earlier, the transfer function
for the system is calculated to be
s s
s G
L
389 . 3
656 . 20
) (
2
.
To be able to analyse this system, a few assumptions are
made, which are:
No input, 0 ) ( t r
Linear part (motor) act as a low pass filter, that is,
higher order harmonics are damped.
Nonlinearity does not generate subharmonics
Nonlinearity is symmetric
Nonlinearity does not depend on frequency
Assume that the error signal, ) cos( ) ( t E t e
The authors would like to thank Yayasan TM and Majlis Amanah Rakyat
Malaysia for funding this study. Thanks are due to Dr JC Allwright for
supervising this project.
978-1-4673-4691-7/13/$31.00 ©2013 IEEE
2013 IEEE Conference on Sustainable Utilization and Development in Engineering and Technology
37