International Journal of Computer Applications (0975 – 8887)
Volume 64– No.11, February 2013
A Sliding Mode Controller for a Three Phase
Induction Motor
Eman El-Gendy
Demonstrator at Computers and systems
engineering, Mansoura University, Egypt
Sabry F. Saraya
Assistant professor at Computers and systems
engineering, Mansoura University, Egypt
ABSTRACT
In this paper, a sliding mode controller (SMC) is designed to
control the speed of an induction motor fed by three phase
voltage source inverter based on space vector pulse width
modulation (SVPWM) technique. The sliding mode controller
is a nonlinear, high speed switching, feedback control strategy
that provides an effective and robust approach for controlling
nonlinear plants. The space vector pulse width modulation
technique is advanced, computation-intensive pulse width
modulation (PWM) technique and it is possibly the best
among all the PWM techniques for variable frequency drive
applications. The proposed scheme enables us to adjust the
speed of the motor by controlling the frequency and amplitude
of the stator voltage; the ratio of the stator voltage to the
frequency should be kept constant. It is introduced to maintain
a constant speed to when the load varies. Simulation results
show the validation of the proposed scheme.
Keywords
Sliding mode controller, induction motor, space vector pulse
width modulation.
1. INTRODUCTION
Induction motors are used in many applications such as
HVAC (heating, ventilation and air-conditioning), industrial
drives (motion control, robotics), automotive control (electric
vehicles), etc. [1, 2]. The squirrel-cage induction motor has
been widely used in many control systems and industrial
automatic systems. It has many advantages, such as simple
structure, firmness and low maintenance cost. However,
induction motors are difficult to control for several reasons:
their dynamics are intrinsically nonlinear and multivariable
[1].
Various high-performance control strategies have been
developed for the induction motor [3-9]. A well-known
method is field-oriented control (FOC) [3]. With field-
oriented techniques, the decoupling of the induction motor's
flux and torque can be controlled separately as an excited DC
motor. But FOC requires accurate and complex calculation of
the decoupling, so it is difficult to operate and easily
influenced by load disturbance and parameter uncertainties.
The conventional direct torque control (DTC) strategy is
useful for overcoming the disadvantages of FOC [4, 5]. DTC
doesn't require complex decoupling calculation and is easy to
implement due to its simple structure. However, DTC has
some drawbacks during induction motor operation, such as
large ripples in torque and flux at low speeds.
Abdelhameed F. Ibrahim
Lecturer at Computers and systems engineering,
Mansoura University, Egypt
Fayez. F. G. Areed
Professor at Computers and system engineering,
Mansoura University, Egypt
In order to reduce ripples in torque and flux, a discrete space
vector modulation (DSVM) has been proposed [6] by means
of a new switching table. Since its sampling period is
subdivided, this new switching table is more complex, with
the result that its DSVM takes more time to calculate and
requires a greater sampling time than that of the DTC does.
As an alternative, space vector modulation (SVM) is
incorporated with DTC for induction motor drives, as
described in [7-10], to provide higher control resolution and
help improve the drive's behavior. However, this simple
controller design based on SVM and DTC techniques is
sensitive to parameter variations and load disturbance.
The space vector pulse width modulation (SVPWM) method
is an advanced, computation-intensive PWM method and it is
possibly the best among all the PWM techniques for variable
frequency drive applications [11]. Space vector modulation is
based on the representation of three phase voltages as space
vectors. It exhibits the features of good dc-bus voltage
utilization and a low Total Harmonic Distortion (THD) when
compared to other PWM methods [12, 13].
Classical control systems like PI (proportional-integral)
control have been used for the speed control of induction
machines. The main drawbacks of classical PI controllers are
their large overshoot and excessive settling time. To face
these problems, sliding mode control has recently been
applied to the control of electrical drive systems. A sliding
mode controller (SMC) is a nonlinear, high speed switching,
feedback control strategy that provides an effective and robust
approach for controlling nonlinear plants [14-16]. However,
the SMC easily produces a chattering phenomenon due to its
discontinuous switching control.
In this paper, a sliding mode controller is designed to control
the speed of an induction motor fed by three phase voltage
source inverter based on space vector pulse width modulation
technique. Simulation results show the validation of the
proposed scheme.
The paper is organized as follows: Section 2, presents the
space vector pulse width modulation and the sliding mode
controller. The sliding mode controller based design is shown
in Section 3. In section 4, simulation results are given to
validate the proposed approach. Section 5 gives a comparative
study. This paper is concluded in Section 6.
33

2. BACKGROUND
2.1 Space Vector Pulse Width Modulation
For the AC voltage, drive application sinusoidal voltage
sources are usually not used [17]. They are replaced by six
power IGBTs that act as on/off switches to the rectified DC
bus voltage. Owing to the inductive nature of the phases, a
pseudo-sinusoidal current is created by modulating the duty-
cycle of the power switches. The basic power circuit topology
of a three-phase voltage source inverter supplying a star
connected three-phase load is given in Figure 1. A three
phase bridge inverter, from Figure 1, has eight permissible
switching states as shown in Figure 2.
Fig. 1: Three phase voltage source inverter.
Fig. 2: Eight switching states of VSI.
2.2 Sliding Mode Controller
The control problem is to get the motor speed to track
a specific time varying command in the presence of
model imprecision, load torque disturbances and measurement
noise. In sliding mode control, the system is controlled in
such a way that the tracking error and its rate
of change always move towards a sliding surface. The
sliding surface is defined in the state space by the scalar
equation [18]:
(1)
where: the sliding variable is:
(2)
where: is a positive constant that depends on the bandwidth
of the system.
The problem of tracking is equivalent to remaining on the
sliding surface for all the time, and the sliding variable is
kept at zero. For a second order system the switching surface
is a line. Control input is applied to drive the system state
onto the switching line, and once on it, the system is
constrained to remain on the line. The distance of the error
International Journal of Computer Applications (0975 – 8887)
Volume 64– No.11, February 2013
trajectory from the sliding surface and its rate of convergence
are used to decide the control input. The sign of the control
input must change at the intersection of tracking error
trajectory with the sliding surface. In this way, the error
trajectory is forced to move always towards the sliding
surface. Once it reaches the sliding surface, the system is
constrained to slide along this surface to the equilibrium
point. The condition of sliding mode is :
(3)
where: is a positive constant.
This equation is stricter than the general sliding condition:
and is equivalent to .
The q-axis stator voltage command is responsible for
changing torque. It is defined as:
(4)
where: is a positive constant, which is the gain of the
sliding mode controller. A suitable value of ensures sliding
to occur.
This controller gives unacceptable performance due to
high control activity, resulting in chattering of control variable
and system states. To reduce chattering, a boundary layer of
width is introduced on both sides of the switching line. This
amounts to reduction of the control gain inside the boundary
layer and results in a smooth control signal. Then the control
law is modified to:
(5)
where:
(6)
3. PROPOSED SLIDING MODE
CONTROLLER SCHEME
The flow chart of the sliding mode control process is
represented in Figure 3. It illustrates that the sliding mode
control is performed by reading the value of actual speed and
comparing it to the reference speed to generate an error signal.
This error signal and the derivative of error are used to
determine the switching surface and the state of the system
must remain on the sliding surface. The controller then
generates the 6 PWM signals for inverter and compares the
error value e to a set value . If the error is less than this
value, the error value is printed and the control process stops,
otherwise the process is repeated to an acceptable error value.
The block diagram of the induction motor using speed
controller is shown in Figure 4. In this Figure, the value of
the actual speed generated by the induction motor is
compared to the desired value to generate an error signal.
This error signal and its derivative are fed to the sliding mode
controller which generates a crisp value . This value is
added the motor speed which in turn forms the input to the
Voltage Source Inverter and V/f controller. The VSI (Voltage
Source Inverter) receives voltage signal from the V/f
34

controller and (frequency signal). The Voltage Source
Inverter uses these inputs to generate a three phase voltage
whose frequency and amplitude can be varied by the sliding
mode controller. The three phase voltage is fed to the
induction motor which then runs with a speed which tends to
follow the desired speed (reference speed, ).
In Figure 5, the block diagram of the sliding mode
controller using MATLAB Simulink is represented. The value
of the actual speed is subtracted from the value of the
reference speed to generate the error signal . Then the
error signal is differentiated to generate . The values and
are added and the result is passed
through the saturation function to generate . This value is
added to value of the actual speed to generate the frequency
and the voltage required by the voltage source inverter VSI.
Start
Value of speed , ratio of ripples varies as follows:
Error & 
derivative of error
Apply sliding mode
controller
Generate the required
voltage and frequency
Print e(t),
END
Fig. 3: Flow chart of the sliding mode control process.
International Journal of Computer Applications (0975 – 8887)
Volume 64– No.11, February 2013
4. SIMULATION RESULTS AND
ANALYSIS
The speed set point is 1000 rpm at time t = 0 s,. The speed
follows precisely the acceleration ramp. At t = 0.5 s, the full
load torque is applied to the motor shaft while the motor
speed is still ramping to its final value. This forces the
electromagnetic torque to increase to a high value and then to
stabilize at 11 N.m once the speed ramping is completed and
the motor has reached 1000 rpm. At t = 1 s, the speed set
point is changed to 1500 rpm and the electromagnetic torque
reaches again a high value so that the speed ramps precisely at
1800 rpm/s up to 1500 rpm under ull load. At t = 1.5 s, the
mechanical load passed from 11 N.m to -11 N.m, which
causes the electromagnetic torque to stabilize at
approximately at -11 N.m shortly after.
4.1 Using PI Controller
For speed, the rise time is 1.11 sec and the settling time is
1.514 sec. The maximum overshoot is 25 rpm which is less
than 2%. The peak time is 1.514 sec, and the delay time is 0.4
sec as shown in Figure 6. For torque shown in Figure 7, the
 From 0 sec to 0.5 sec: the ratio is 12%.
 From 0.5 sec to 1 sec: the ratio is 18%.
From 1 sec to 1.25 sec: the ratio is 3%.
 From 1.25 sec to 1.5 sec: the ratio is 3%.
 From 1.5 sec to 2.5 sec: the ratio is 24%.
So, when using PI controller, the speed and torque have
some ripples.
4.2 Using PID Controller
For speed, the rise time remains 1.11 sec and the settling
time is 1.53 sec. The maximum overshoot is decreased to 20
rpm which is less than 2%. The peak time is 1.88 sec, and the
delay time is 0.4 sec as shown in Figure 8. For torque shown
in Figure 9, the ratio of ripples varies as follows:
 From 0 sec to 0.5 sec: the ratio is 4%.
 From 0.5 sec to 1 sec: the ratio is 3.8%.
 From 1 sec to 1.25 sec: the ratio is 4%.
 From 1.25 sec to 1.5 sec: the ratio is 40%.
 From 1.5 sec to 2.5 sec: the ratio is 9%.
Thus, when using PID controller, the speed and torque
ripples are decreased.
4.3 Using sliding mode Controller
For speed, the rise time remains 1.11 sec and the settling
time is 1.1 sec which is less than PI or PID. The maximum
overshoot is 5 rpm which is very small and less than 2%. The
peak time is 2 sec, and the delay time is 0.4 sec as shown in
Figure 10. For torque shown in Figure 11, the ratio of ripples
varies as follows:
 From 0 sec to 0.5 sec: the ratio is 3.8%.
 From 0.5 sec to 1 sec: the ratio is 3.4%.
 From 1 sec to 1.25 sec: the ratio is 3%.
 From 1.25 sec to 1.5 sec: the ratio is 3.5%.
 From 1.5 sec to 2.5 sec: the ratio is 9.27%.
35
 International Journal of Computer Applications (0975 – 8887)
Volume 64– No.11, February 2013

Thus, when using sliding mode controller, the speed and voltage source inverter based on space vector pulse width
torque have some ripples with magnitude less than either modulation technique. The proposed scheme enabled us to
using PI or PID controllers. These results are summarized in adjust the speed of the motor by controlling the frequency and
Table 1 and Table 2. amplitude of the stator voltage; the ratio of the stator voltage to
the frequency should be kept constant. It is introduced to
maintain a constant speed to when the load varies. Simulation
5. COMPARATIVE ANALYSIS results showed the validation of the proposed scheme. As a
conclusion, the sliding mode controller gives lower overshoot
In this section, we compare our results with A R. than PI and fuzzy PI controllers. As a future work, more
Arulmozhiyaly and K. Baskaran [17] using different speeds analysis is needed to face the sliding mode controller torque
and load torques. Figure 12 shows the speed using the ripples which are not completely eliminated.
proposed sliding mode controller and Figure 13 shows the
speed using PI and fuzzy PI controller at reference speed of
800 rpm and no load. Figure 14 shows the speed using the
proposed sliding mode controller and Figure 15 shows the
7. REFERENCES
speed using PI and fuzzy PI controller at reference speed of [1] Bimal K. Bose, “Modern Power Electronics and AC
800 rpm and load of 5 N.m. Figure 16 shows the speed using Drives,” Pearson education, Prentice Hall; 1 edition,
the proposed sliding mode controller and Figure 17 shows the 2001
speed using PI and fuzzy PI controller at reference speed of
[2] Werner Leonhard, “Control of Electrical Drives,”
1200 rpm and no load. Figure 18 shows the speed using the
Springer Verlag, 3nd edition 2001 .
proposed sliding mode controller and Figure 19 shows the
speed using PI and fuzzy PI controller at reference speed of [3] S. K. Sul and T. A. Lipo, "Field oriented control of an
1200 rpm and load of 5 N.m. induction machine in a high frequency link power
From the results shown in the last figures, it is clear that the system", IEEE Transactions on Power Electronics, vol.5,
sliding mode controller gives better response and lower no.4, pp.436-445, 1990.
overshoot than PI and fuzzy PI controllers. Table 3 shows the
settling time using the proposed sliding mode controller and [4] I. Takahashi and T. Noguchi, "A new quick-response and
fuzzy PI controllers. The results are approximately the same. high-efficiency control strategy of an induction motor",
IEEE Transactions on Industry Applications, vol.A-22,
no.5, pp.820-827, 1986.
Table 1: Comparison of speed between PI, PID, and SMC. [5] J.-K. Kang and S. K. Sul, "New direct torque control of
Speed PI PID Sliding mode induction motor for minimum torque ripple and constant
switching frequency", IEEE Transactions on Industry
Rise time (sec) 1.11 1.11 1.11 Applications, vol.35, no.5, pp.1076-1082, 1999.
Settling time (sec) 1.514 1.53 1.1 [6] D. Casadei, G. Serra and A. Tani, "Implementation of a
Maximum overshoot 25 20 5 direct torque control algorithm for induction motors
based on discrete space vector modulation", IEEE
Peak time (sec) 1.514 1.88 2 Transactions on Power Electronics, vol.15, no.4, pp.769-
Delay time (sec) 0.4 0.4 0.422 777, 2000.
[7] N. R. N. Idris and A. H. M. Yatim, "Direct torque control
Table 2: Comparison of torque between PI, PID, and of induction machines with constant switching frequency
SMC. and reduced torque ripple", IEEE Transactions on
Torque PI PID Sliding Mode Industrial Electronics, vol.51, no.4, pp.758-767, 2004.
0 - 0.5 12 % 4.0 % 3.8 % [8] C. Lascu, I. Boldea and F. Blaabjerg, "A modified direct
0.5 – 1 18 % 3.5 % 3.4 % torque control for induction motor sensorless drive",
Ratio of 1 – 1.25 3.0 % 4.0 % 3.0 % IEEE Transactions on Industry Applications, vol.36,
ripples
1.25 – 1.5 3.0 % 40 % 3.5 % no.1, pp.122-130, 2000.
1.5 – 2.5 24 % 9.0 % 9.27 % [9] Y.-S. Lai and J.-H. Chen, "A new approach to direct
torque control of induction motor drives for constant
Table 3: Comparison of settling time in sec between fuzzy inverter switching frequency and torque ripple
PI and SMC. reduction", IEEE Transactions on Energy Conversion,
vol.16, no.3, pp.220-227, 2001.
Load Condition Fuzzy PI Sliding mode
[10] D. Sun, J. G. Zhu and Y. K. He, "A space vector
1000 rpm with no load 0.7 0.71 modulation direct torque control for permanent magnet
synchronous motor drive systems", Proc. of the Fifth
1000 rpm with load 0.79 0.71
International Conference on Power Electronics and Drive
1200 rpm with no load 0.79 0.858 Systems, vol.1, pp.692-697, 2003.
1200 rpm with load 0.85 0.85 [11] Wei-Feng Zhang and Yue-Hui Yu, “Comparison of
Three SVPWM Strategies,” Journal of electronic science
and technology of china, No.3, pp. 283-287, Sep. 2007.
6. CONCLUSION [12] Iqbal.A, “Analysis of space vector pulse width
This paper proposed a sliding mode controller design to modulation for a five phase voltage source inverter,” IE
control the speed of an induction motor fed by three phase (I) journal-EL, Vol. 89, Issue 3, pp.8-15, Sep. 2008.

36

[13] Mondal, S.K.; Bose, B.K.; Oleschuk, V.; Pinto, J.O.P,
“Space vector pulse width modulation of three-level
inverter extending operation into overmodulation
region,” IEEE Transactions on Power Electronics, Vol.
18, Issue 2, pp. 604 – 611, Mar. 2003.
[14] J.-J. E. Slotine and W. Li, "Applied nonlinear control",
Englewood Cliffs, Prentice-Hall, New Jersey,1991.
[15] X. Liu and W. Wang, "High order sliding mode and its
application on the tracking control of piezoelectric
systems", International Journal of Innovative Computing,
Information and Control, vol.4, no.3, pp.697-704, 2008.
[16] X.-Z. Zhong, H.-L. Xing and K. Fujimoto, "Sliding mode
variable structure control for uncertain stochastic
Speed in rpm
Fig. 6: Reference vs. actual speed using PI controller.
International Journal of Computer Applications (0975 – 8887)
Volume 64– No.11, February 2013
systems", International Journal of Innovative Computing,
Information and Control, vol.3, no.2, pp.397-406, 2007.
[17] R. Arulmozhiyaly and K. Baskaran, "Implementation of
a Fuzzy PI Controller for Speed Control of Induction
Motors Using FPGA", Journal of Power Electronics,
Vol. 10, No. 1, January 2010.
[18] Kanungo Barada Mohanty, "Sensorless Sliding Mode
Control of Induction Motor Drives" , Electrical
Engineering Department, National Institute of
Technology, Rourkela-769008, India, May 18, 2009.
Time in sec
Fig. 8: Reference vs. actual speed using PID controller.
37
 International Journal of Computer Applications (0975 – 8887)
Speed in rpm Volume 64– No.11, February 2013

Time in sec
Fig. 10: Reference vs. actual speed using SMC.

Fig. 13: Reference vs. actual speed using fuzzy PI

controller.
Torque in N.m

2) Speed is 1000 rpm with load of 5 N.m:

Time in sec

Fig. 7: Electromagnetic torque using PI controller.

Speed in rpm
Torque in N.m

Time in sec

Fig. 14: Reference vs. actual speed using SMC.

Time in sec
Fig. 9: Electromagnetic torque using PID controller.
Torque in N.m

Fig. 15: Reference vs. actual speed using fuzzy PI

Time in sec
controller.
Fig. 11: Electromagnetic torque using SMC.

1) Speed is 1000 rpm with no load: 3) Speed is 1200 rpm with no load:
Speed in rpm
Speed in rpm

Time in sec
Time in sec
Fig. 16: Reference vs. actual speed using SMC.
Fig. 12: Reference vs. actual speed using SMC.

38
 International Journal of Computer Applications (0975 – 8887)
Volume 64– No.11, February 2013

4) Speed is 1200 rpm with load of 5 N.m:

Speed in rpm
Fig. 17: Reference vs. actual speed using fuzzy PI
controller. Time in sec
Fig. 18: Reference vs. actual speed using SMC.

Fig. 19: Reference vs. actual speed using fuzzy PI

controller.

39