NOVATEUR PUBLICATIONS

International Journal of Research Publications in Engineering and Technology [IJRPET]

ISSN: 2454-7875
VOLUME 3, ISSUE 4, Apr.-2017
ROBUST CONTROL OF DC MOTOR USING SLIDING MODE CONTROL
APPROACH
JAGADALE SHRIKANT
PG department College of Engineering, Ambajogai, shrikant.jagdale@gamil.com

S. S. SANKESWARI
PG department College of Engineering, Ambajogai, sankeswari@gmail.com

ABSTRACT: di
L  u  Ri  λ 
Most of the industrial applications are based on the AC dt  (1)
motor drive because of the problems associated with d
the control of the DC motor drives. Authors have tried j  k t i  τl
dt
to address the problems associated with the DC drives. Where
Sliding mode control of the DC machine is the scope of i Armature current
the study of this paper. Authors have focused the scope ω Shaft speed
of the study to the DC drives used in industries. R Armature resistance
Reducing the disturbances is one of the advantages of λ0 Back emf constant
the experiment carried out. The method implemented τl Load torque
was found suitable to reducing dynamic chattering of u Terminal voltage
the drive. The simulation model is developed and it j Inertia of the motor rotor and load
was found suitable to improve the performance of the L Armature inductance
system. kt Torque constant
KEYWORDS: DC drives, control of drives, sliding mode
control, etc. Its motion is governed by second order equations (1) with
respect to armature current, i and shaft speed w with
I. INTRODUCTION: voltage u and load torque τ1. A low power-rating device
The technology implemented was firstly introduced in can use continuous control. High power rating system
1950. Variable structure control is useful for MIMO and needs discontinuous control. Continuously controlled
nonlinear systems [1], [2]. The VSC systems are also found voltage is difficult to generate while providing large
suitable for several robust control systems. The cluster of current.
research has been carried out for betterment of the
performance of the drives and various systems have been III. SLIDING MODE CONTROL DESIGN:
developed and implemented by the researchers in past few DC motors have been dominating the field of adjustable
years. All the design procedures will be carried out in the speed drives for a long time because of excellent
physical coordinates to make explanations as clear as operational properties and control characteristics. In this
possible. The system response is judged by the section different sliding mode control strategies are
performance index. The system is found useful for the high formulated for different objectives e.g. speed control,
frequency switching. torque control and position control.

II. DYNAMIC MODELING OF DC MACHINE: A. CURRENT CONTROL:

Fig.1 shows the model of DC motor with constant Let i* be reference current providing by outer control loop
excitation is given by and i be measured current. Consider a current control
problem, by defining switching function

s  i*  i (2)

Design a discontinuous control as

u  u  sign (s) (3)
Following state equations [1],[3],[7].
Where uo denotes the supplied armature voltage.

110 | P a g e
 NOVATEUR PUBLICATIONS
International Journal of Research Publications in Engineering and Technology [IJRPET]
ISSN: 2454-7875
VOLUME 3, ISSUE 4, Apr.-2017

 λ u 
JL
g(t) ,
di R 1
ss  s (  i  ) u s (4) If kt (9)
dt L L L  ss  0
Choice of control u0 as

di Then sliding mode will happen [6].

u  L Riλ  Makes (4) The mechanical motion of a dc motor is normally much
 dt 
slower then electromagnetic dynamics.
ss  0 Which means that sliding can happen in s = 0 [4]. It means that L<<J in (1).
B. Speed control Following reduced order control methods proposed
below will solve chattering problem without measuring of
Let ω* be the reference shaft speed, then the second current and acceleration (x2).
order motion equation with respect to the error Speed tracking error is ωe=ω*-ω. The dc motor model
(e = ω*- ω) is of form. (1) in terms of ωe:
tate variable x1=e & x 2  e
 u  Ri  λ (    )
di
L
x 1  x 2 dt  e
(10)
(5) d
x 2  a1 x1  a 2 x 2  f(t)  b u j e   k t i  τ  j *
dt l
k λ R k Let L be equal to zero due to L << j.
a  t  , a  &b t Then (10) becomes with L=0
Where 1 JL 2 L JL λ k
are constant values. i    (    )  t u  τ  j * (11)
R e R 1
f(t)    a    a    R τ / JL  τ / J The sliding Substituting (11) into (10) results in.
2 1 1 a l
d kλ k
surface and discontinuous control are designed as
j e t  
(   )  t u  τ  j * 12)
s  c (     )  (    )
d dt R e R 1
dt (6) Equation (12) is a reduced order (first order) model of DC
u  u sgn (s) motor.
 The discontinuous control is designed as
This design makes the speed tracking error e converges
u  u sgn ( e) (13)
to zero exponentially after sliding mode occurs in s = 0, 
where c is a positive constant determining the and the existence condition for the sliding mode ωe=0 will
convergence rate .for implementation of control (6),angle be
of acceleration ( x 2  e ) is needed. τ R j R  *
u  λ (    )  1  (14)
The system motion is independent of parameters a 1, a2, b   e k k
and disturbances in g(t). t t
Combining (1) & (6) produces The principle advantage of the reduced order based
method is that the angle acceleration ( x 2  e ) is not
1 k k
s  c      ( λ i  τ )  τ l  t ( R i  k w )  t u
c
needed for designing sliding mode control [1].
J  l J JL t JL The unmodeled dynamics (1) may excite non-admissible
k chattering. Fig.3 shows the control structure based on
 g(t)  t u reduced order model and observer state.
JL (8)
Let us design an a asymptotic observer to estimate ωe [6 ]
where 
 kλ  k  
c 1 k : j e  t  (    e )  t u  τ 1 j * l (   )
d
g(t)  c  *   ( k i  τ )  τ  t ( R i  k  ) dt R R 1 e
J t l J l JL t

111 | P a g e
 NOVATEUR PUBLICATIONS
International Journal of Research Publications in Engineering and Technology [IJRPET]
ISSN: 2454-7875
VOLUME 3, ISSUE 4, Apr.-2017
d k λ k The torque control problem by defining switching
j e  t  (    )  t u  τ  j*
dt R e R 1 function

dτ 
l  l (   ) s  τ  τ (24)
(15)
dt 2 e As the error between the reference torque τ* and the real
Where torque τ developed by the motor.
 Design a discontinuous control as
we Estimated error u  u  sign (s) (25)
 Where u0 is high enough to enforce the sliding mode in s=0,
 e  e  e Speed tracking error which implies that the real torque τ tracks the reference
l1,l2 observer gain s  τ *  k i
The discontinuous control designed using estimate t
 k Ri λ k  k
state w e [ ] will be torque τ*.  τ   t   t  t u
 L L L
u  u sgn ( e ) (16) k

 f(t)  t u
the sliding mode will happen if L
 (26)
 τ R j R  * l1R 
u  λ (    e )  1   (   ) (17)
  e Where
k k k
t t t k Ri λ k 
  f(t)  τ *  t   t Depending on the reference
And   0 & τ  0 . L L
Chattering can be eliminated by using reduce observer L
states. The sliding mod occurs in the observer loop, which signal .for u   f(t)
kt
does not contain unmodelled dynamics.
C Position control k
To consider the position control issue, it is necessary to ss  sf(t)  t u  s  0 (27)
augment the motor equations (1) with
L
So sliding mode can be enforced in
dθ
 (18) s = 0.
dt
Where θ denotes the rotor position. IV. SIMULATION RESULTS:
The switching function s for the position control is selected To show the performance of the system the simulation
as result for the speed control of DC machine is depicted.
s  (θ * θ)  c1 (θ * θ )  c2 ( θ  θ) (19) Rated parameters of the dc motor used to verify the design
principle are
and the discontinuous control is
5 hp, 240V, R=0.5 Ω, L=1mH, j =0.001 kgm2, kt=
u  u sgn (s) (20)
0.008MmA -1

Combining (1) (18 ) (19) λ0= 0.001vs rad-1 and τl=Bω where b=0.01 Nms rad-1
k fig.3.1,3.2,3.3,&3.4 depicts the simulation result of the
s  h(t)  t u (21) Where reduced order speed control with measured speed system
JL response ,error, squared error and integral square
c 1 k error(ISE).The high frequency chatter is due to neglecting
h(t)   * c  * c   1 (k i  τ )  c   τ  t (Ri  k ) (22) the fast dynamics i.e. dynamics of the electric of the
1 2 J t l 2 J l JL t
electric part. fig.3.1,3.2,3.3,&3.4 depicts the simulation
JL
Choice of u0 as u 0  h(t) (23) result of the reduced order speed control with observed
kt speed system response ,error, squared error and integral
 square error(ISE) using(15). In order to reduce the
Makes s s  0 which means that sliding mode can happen weighting of the large initial error & to penalise small error
s=0 with properly chosen c1 & c2 .We can make velocity occurring later in response move heavily, the performance
tracking error e = w* - w converges to zero. index is the integral square error (ISE) .
D. Torque Control

112 | P a g e
 NOVATEUR PUBLICATIONS
International Journal of Research Publications in Engineering and Technology [IJRPET]
ISSN: 2454-7875
VOLUME 3, ISSUE 4, Apr.-2017

Fig. 3.1 Reduced order speed control with observer speed

Fig 3.6 Switching surface for reduced order speed control

with observer state.

Fig.3.2 Reduced orders speed control with measured State

Fig 3.7 Switching surface for reduced order Speed control

with Measured State.

Fig 3.3 speed error of reduced order with observer state

Fig.3.8 Square error for reduced order speed control with

observer state.

Fig 3.4 Speed error of reduced order with measured state

Fig.3.5 Square error of reduced order with Observer & Fig.310.Armature voltage for reduced order speed
Measured State. Control with observer state.
113 | P a g e
 NOVATEUR PUBLICATIONS
International Journal of Research Publications in Engineering and Technology [IJRPET]
ISSN: 2454-7875
VOLUME 3, ISSUE 4, Apr.-2017
and AC Electrical Machines,“ in Proc.of
theInternationalIECON’94, pp581-586.
5) P. C. Krause,O.Wasynczuk,Scottd D.Sudhoff, “ Analysis
of electric Machinery and Drive systems,”IEEE Press
John Wiley & Sons Inc.Publication, II edition 2004.
6) V.Utkin, JurgenGuldner, J.Shi,“Sliding mode control in
Electromechanical Systems, “Taylor and Francis
London, 1999.
7) Katsuhiko Ogata, “Modern Control Engineering”,
Prentice Hall of India Pvt. Ltd. New Delhi, II edition,
Fig.3.11Armature voltage for reduced order speed
1995, pp 295-303.
Control with measured state.

Fig 3.9. Integral square error of reduced order speed

control with observer and Measured state

V. CONCLUSION:
The SMC approach to speed control of dc machines is
discussed. Both theoretical and implementation result
speed control based on reduced order with measured
speed and reduced order with observer speed, using
simulation are conducted. Besides, reduced order observer
deals with the chattering problem, encounter often in
sliding mode. Control area Selection of the control
variable (angular position, speed, torque) leaves basic
control structures unchanged. The system is proven to be
robust to the parameters variations, order reduction.

REFERENCES:
1) V.I.Utkin,“Sliding mode control Design Principles and
applications to Electric drives,”IEEE Trans. On
Industrial. Electronics, vol.40, no.1, pp.23- 36, feburary
1993.
2) De Carlo, Zak S. H. Mathews G.P., Variable structure
control of Nonlinear multivariable systems: A
Tutorial,”Proc.of IEEE, Vol.76, No.3, pp.212-232, March
1988.
3) A.Sabanovic,KenzoWada,FarukBilalovic,MilanVujovic,“
Sliding Modes in Electrical machines Control systems,”
Proc. Of IEEE, vol., No., pp.73-78, 1993
4) N.Sabanovic,A.Sabanovic,k. Jezernik O.M. Vujovic,“
Current Control in three phase switching Converters

114 | P a g e