Journal of Physics: Conference

Series

PAPER • OPEN ACCESS You may also like

- Comparative Evaluation of Control
Assessment of Control Drive Technologies for Techniques of PMSM Drive in Automotive
Application
Induction Motor: Industrial Application to Electric Rakesh Shriwastava, Satayjit Deshmukh,
Ashwini Tidke et al.

Vehicle - Direct Torque Control of Asynchronous

Generator Based on Active Disturbance
Rejection Controller
To cite this article: A. Z. Ahmad Firdaus et al 2021 J. Phys.: Conf. Ser. 1878 012047 Wenhao Li and Haiping Liang

- Improvement of Direct Torque Control by

using a Space Vector Modulation Control
of Three-Level Inverter
A Achalhi, M Bezza, N Belbounaguia et al.
View the article online for updates and enhancements.

This content was downloaded from IP address 37.111.177.233 on 07/04/2024 at 07:46

ICE4CT 2020 IOP Publishing
Journal of Physics: Conference Series 1878 (2021) 012047 doi:10.1088/1742-6596/1878/1/012047

Assessment of Control Drive Technologies for Induction

Motor: Industrial Application to Electric Vehicle

A. Z. Ahmad Firdaus1, S. A. Azmi1, K. Kamarudin1, L.J. Hwai1, H. Ali1, M. S. Z.

Azalan1, and Zamri Che Mat Kasa2
1
Faculty of Electrical Engineering Technology, Universiti Malaysia Perlis
2
MSM Perlis Sdn Bhd, Perlis

ahmadfirdaus@unimap.edu.my

Abstract. Nowadays electric vehicle has increasingly gained much popularity indicated by
growing global share market targeted at 30% by 2030 after recording 7.2million global stock in
2019. Compared to Internal Combustion Engine (ICE) counterpart, Battery Electric Vehicles
(BEV) produce zero tailpipe emission which greatly reducing carbon footprints. Induction motor
has been widely used and its control technology has evolved from scalar type volt/hertz to recent
predictive control technology. This allows induction motor’s application to expand from being
the workhorse of industry to become prime mover in electric vehicle, where high performance
is expected. Among vector control scheme, Direct Torque Control (DTC) has gained interest
over Field Oriented Control (FOC) with simpler structure, better robustness and dynamics
performance yet suffer from high torque and flux ripple. In electric vehicle applications, high
ripple at low speed is highly undesirable, potentially causing torsional vibration. High
performance control requires speed sensor integration, which often increase complexity in the
design. The work aims to review the best control technology for induction motor in electric
vehicle application through performance parameter evaluation such as improvement on dynamic
response, torque and flux ripple reduction, and component optimization. Several arise issues in
motor control and possible methods to circumvent are highlighted in this work. In conclusion,
model predictive torque control (MPTC) is the most promising scheme for electric vehicle with
excellent dynamic response, good low speed performance, and 50% torque ripple reduction
compared to conventional DTC and potential integration with sliding mode observer for
sensorless solution.

Keywords: Electric vehicle, model predictive control, direct torque control, induction motor

1. Introduction
In 2019, electric vehicle global car stock was recorded at 7.2million, an increase by 40% on yearly basis.
Its global market share was targeted to increase to 30% by 2030 [1]. Compared to Internal Combustion
Engine (ICE) counterpart, Battery Electric Vehicle (BEV) which is propelled by electric motor (also
known as traction motor), using energy stored in rechargeable batteries produce zero tailpipe emission
[2], greatly reducing carbon footprints by half in 2019 [1]. Both DC motor and AC motor have been
used as traction motor. Series DC motors have been used as traction motor since 1900 due to high
starting torque capability and ease of control [3]. However, its limitation includes low efficiency,
tendency for heat build-up due to complications with brushes and commutator and lack of regenerative

Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution
of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd 1
ICE4CT 2020 IOP Publishing
Journal of Physics: Conference Series 1878 (2021) 012047 doi:10.1088/1742-6596/1878/1/012047

braking capability [2]. AC motor that has less rotating component and offer flexibility in term of power
recovering using regenerative braking. Three most popular types of AC traction motor used are
induction motor (IM), permanent magnet synchronous motor (PMSM), and switched reluctance motor
(SRM). Induction motor is preferred over PMSM due to lower material cost and over SRM due to higher
efficiency [4].
In industrial applications, 90% of the motors employed are three phase induction motor type [5].
Simplicity, ruggedness, low-cost and ease of maintenance are the factors contributing to its popular
selection in most industries ranging from palm oil to steel plant [6], [7]. They are primarily used to drive
machineries such as conveyors, blower fans, pumps, and compressors. Formerly, they were only
operated as fixed speed due to fixed supply nominal frequency. Compared to constant-speed drive, the
variable speed operation is superior with higher efficiency, better energy savings and significant cost
reduction through less frequent downtime [6], [8], [9]. In the sugar mills, speed control of centrifugal
pump has resulted 50% energy savings through 20% speed regulation [10]. Control technology has
evolved from scalar type – volt/hertz to recently developed predictive control. This has allowed
induction motor application to expand from being the workhorse of industry to become prime mover in
electric vehicle. Induction motor is used as prime mover in the current Tesla Model-S BEV[11].
This work is intended to provide some reference on technical aspect for researchers and industry
player in selecting the motor control technology that best suits the electric vehicle application, within
the purview relevant to induction motor. Through the review, parameters such as improvement on
dynamic response, torque and flux ripple reduction, as well as component optimization are discussed
and evaluated. This paper is organized in such arrangements: Section 2 highlights on the operational
requirements of electric vehicle, Section 3 introduces the available motor control strategies, Section 4
discusses the performance analysis of motor control strategies based on application in electric vehicle
and finally a conclusion.

2. Electric Vehicle Requirement

Figure 1 : Idealized torque / power – speed characteristics.[12]

Figure 1 shows the idealized torque / power – speed characteristics for electric vehicle. Electric vehicle
may operate in the constant torque region at low speed up to rated speed (also known as base speed). At
this range, optimum torque is generated where power increases in tandem with speed. Beyond rated
speed, power limit is capped and thus, generated torque decreases as speed is further increased. This is
due to flux generated-back electromotive force (EMF) exceeded maximum inverter output voltage and

2
ICE4CT 2020 IOP Publishing
Journal of Physics: Conference Series 1878 (2021) 012047 doi:10.1088/1742-6596/1878/1/012047

therefore, field weakening is required to mitigate the effect [13]. It is highly desirable for electric vehicle
to have high efficiency over wide operating speed range [12], [14] with a constant power operating range
of around 3 to 4 times the base speed. This, combines with good battery technology provides high
driving range which is essential requirement [3]. They are also expected to possess high torque density
and high overload capability without compromising the need to have high fault tolerance and robustness
appropriate to vehicle environment. High torque is essential for starting, at low speeds and hill climbing,
while high power is useful for high-speed cruising [12]. However, cost have to be kept low at affordable
range to tap the interest of middle-class consumers.
Motor ripple is also an issue in electric vehicle since it is the only active prime mover. Ripple poses
adverse effects on motor output characteristics during low-speed drive and when operating at high
torque. At high speed, the ripple effect would be cancelled by rotor inertia. It could cause torsional
vibration causing vehicle to shake, especially when the ripple frequency match the resonant frequency
of the system [15]. Motor ripple is partially caused by torque and flux ripple and could be due to side
effects of control scheme employed in the drive system such as conventional Direct Torque Control
[16]. Therefore, it is imperative that the ripple effect is minimized to ensure comfort of passenger and
prolong the lifespan of vehicle components.

3. AC Motor Control Strategies: Theoretical and Practical Consideration

For induction motor, control technologies can be classified to two main categories according to control
principle: scalar control and vector control.

3.1. Scalar Control

In scalar control, main parameter of motor is varied directly so that output parameter such as position,
speed and torque can be controlled. Two main types of AC induction motor are: wound rotor and squirrel
cage.
For wound rotor motor, the rotor winding is connected to slip rings which, in turn, connected to external
resistors. Variation of resistors have made speed control possible. However, the efficiency is relatively
low because of high voltage drop across the resistors. Furthermore, this control method seems
impractical with higher load conditions [3]. The alternative way to control is through implementation of
pulse width modulation (PWM). PWM control is possible through inverter circuit and is feasible with
operating nature of squirrel cage, which is why it is more popular than wound rotor.
In squirrel cage induction motor, variation of frequency of terminal voltage could be used to control
torque and speed of motor. However, controlling only stator frequency with disregard of keeping fixed
voltage/frequency ratio will deteriorate flux at the air gap. It resulted very high currents that would cause
motor damage at very low frequency while at high frequency motor may be stalled. Hence, both
frequency and voltage ratio need to be fixed in order to maintain constant torque when controlling speed.
This method is also known as volt/hertz method [3].
Variable frequency drive in industry utilizes both open-loop and closed-loop control. For non-critical
applications such as conveyors and centrifuge where low and medium speed operation at light and
medium load, open-loop control is sufficient. Typical open-loop motor control drive system involves
three phase inverter, driver circuit, the motor and connected load. The three phase AC supply is rectified
to DC prior to be fed to the inverter. The inverter supplies variable frequency three phase voltage and
current to the induction motor. The driver circuit consist of component that modulates the input signal
in the form of Pulse Width Modulation (PWM) signal that triggers the inverter circuit [5].
Figure 2 shows previous design utilizing volt-hertz concept for open-loop control. The reference
speed input combined with slip speed produce the command frequency. The command frequency is
converted to corresponding voltage command through a volt-hertz profile that maintains relationship at
approximately rated volt to frequency ratio [6]. For open loop, slip speed has to be estimated in the
absence of speed sensor but with adequate induction machine model parameters, it is possible [17], [18].

3
ICE4CT 2020 IOP Publishing
Journal of Physics: Conference Series 1878 (2021) 012047 doi:10.1088/1742-6596/1878/1/012047

Figure 2 : Volt/Hertz open-loop control[17]

Closed-loop control offers better dynamic performance and is more suitable for critical applications.
In closed-loop control, the system tracks the input pattern and automatically tune the parameter upon
load variation or disturbance. With sensor feedback, slip speed parameter is more accurate. In [19],
through comparison of reference and feedback speed, the error is fed to controller where the output is
developed torque that can be converted to slip speed. Figure 3 shows Volt/Hertz closed-loop control
using PI-controller.

Figure 3 : Volt/Hertz closed-loop control using PI-controller [19]

Control structure can be cascaded from only speed loop to add inner current loop, allowing more
accurate current and torque optimization [6] . PI-control has been implemented as controller for
volt/hertz system [19], [20] but alternatively Fuzzy logic control provides adopt with more
straightforward design approach based from open loop data without need of motor model parameter
knowledge for gain tuning [21].

3.2. Inverter Switching

With PWM used to switch inverter, power loss associated to voltage drop can be reduced as it decreases
average power delivered to a load through fast switching control [22]. Due to its variable duty cycle,
PWM is compatible with digital control applications. For motor control applications, using PWM at
high switching frequency works well as inertia load are resistant to discrete effects [6]. Two most widely
used in motor control are: Sinusoidal pulse width modulation (SPWM) and Space vector pulse width
modulation (SVM).
SPWM is popular with scalar control [18], [19] and can be produced using intersective method, by
comparing high frequency carrier signal with reference sinusoidal signal. Inverter output frequency is
determined by the frequency of the reference signal. High switching frequency is desirable to facilitate
harmonics filtering process. However as a drawback, inverter switching loss would also increase [23].
Techniques to eliminate unwanted harmonics include modification of carrier signal. Another drawback
of SPWM includes lower output phase voltage compared to the line supply[22].

4
ICE4CT 2020 IOP Publishing
Journal of Physics: Conference Series 1878 (2021) 012047 doi:10.1088/1742-6596/1878/1/012047

The objective of the SVM is to generate PWM load line voltages that are in average equal to reference
load line voltages [22]. SVM is a digital modulating technique that treats the inverter as single unit
where eight unique switching states (also referred as voltage vectors) can be driven to the unit. Eight
states of the VSI consist of six active vectors (V1 to V6) and two null vectors (V0 and V7) [24]. Output
voltage is created by repeatedly switching the adjacent vectors and the null vector. Reference signal is
sampled regularly using encoder or resolver [25]. The popular design is alternate-reversing sequence
where null vectors alternate for every sequence and the sequence reverses. Center-aligned PWM is a
SVM technique that combines both features for all sectors and has been adopted commercially in most
vector control applications[24].
In comparison to SPWM, SVM has proven to give superior performance in dynamic response and
current distortion when implemented for motor control application and also better DC bus utilization
[25]–[27] although the process of implementing SVM signal is much more complex and increased
computational effort [28]. The superior performance of SVM made them more suitable to be
implemented in vector control.

3.3. Vector Control

The scalar control is adequate when controlling the speed of compressor, fan or conveyor as the speed
change is intermittent. However, dynamic applications such as electric vehicle requires motor to rapidly
accelerate, reverse, stop, and start while responding to torques that may suddenly change. During such
transition, the voltages and currents are no longer sinusoidal, and computer generated waveshape change
instantaneously, rendering equivalent circuit-based model ineffective. Flux must be maintained both in
value and orientation to instantaneously develop required torque. This type of dynamic control is called
flux vector control [7]. Two most popular vector control types are field oriented control (FOC) and
direct torque control (DTC). Both FOC [29] and DTC [30] control schemes have been employed in
electric vehicle, obtained good dynamic performance and highly efficient over wide speed range.

3.3.1. Field Oriented Control

Figure 4 : Basic FOC scheme for three phase induction motor [31]

Figure 4 shows basic FOC scheme for induction motor. The concept was first introduced by F. Blaschke
in 1971 [16]. Motor signals consisting stator currents and rotor position were measured. Stator currents
were converted to dq-parameter using Clarke-Park transformation as feedback and were compared with
reference parameter to generate error signal which was independently regulated by controller (usually
PI-type). The controller sends correction signal in form of dq-parameter to be converted using Inverse
Park transformation into αβ-parameter and modulated to the inverter switching gates. Rotor flux position

5
ICE4CT 2020 IOP Publishing
Journal of Physics: Conference Series 1878 (2021) 012047 doi:10.1088/1742-6596/1878/1/012047

is needed to execute both Clarke-Park and inverse Park transformation and is usually measured from
motor speed sensor. Since then several improvement on the basic system has been introduced [32], [33].
The main advantages of FOC method are high dynamic performance, switching frequency, low
torque ripples and maximum fundamental component of stator current, but it suffers disadvantages such
as requirement of two coordinate transformations and current controllers, and high machine parameter
sensitivity [16].

3.3.2. Direct Torque Control

Direct Torque Control was developed by Takahashi in 1989 [34]. Figure 5 shows conventional DTC
scheme for induction motor. In this scheme, stator signals were fed to flux and torque estimator, based
from motor flux equations. Electromagnetic torque and stator flux would be independently compared to
reference values, and their errors would be mitigated using two-level hysteresis controller for flux and
three-level hysteresis controller for the torque. Control outputs, together with flux sector selection signal
were fed to the switching table and the PWM output drives the inverter.

Figure 5 : Conventional DTC scheme for three phase induction motor [16]

Compared to FOC, DTC is simpler and requires less computational effort. Current controller and
coordinate transformations were not required. Despite its simplicity, DTC allows a good torque control
in steady-state and transient operating conditions, high dynamic response and robustness. In addition,
this controller is less sensitive to the parameters detuning [16].
However, conventional DTC produces high ripple in torque due to the non-linear hysteresis
controllers. This was due to the fixed width band of the controller resulting variable switching frequency.
The conventional switching table also produce only one voltage space vector for the entire sampling
period. Hence the motor torque may exceed the upper/lower torque limit even if the error is small [26],
[35]. This worsen in digital implementation where sampling time has significant effect and the ripple
may be higher than the hysteresis limits [35]. Limited selection of voltage vector also caused undesired
flux weakening phenomena at low speed operation [16].
One method to circumvent the problem of non-linearity of hysteresis controller in conventional DTC
was to employ variable band instead of using fixed width band. Fuzzy logic control was implemented
to vary the hysteresis band based from torque error to obtain constant switching frequency. Compared
to fixed band design, simulation results showed improvement of THD at stator current and reduced
torque ripple [36].
Space Vector Modulation was integrated in DTC (DTC-SVM) where more voltage vectors can be
utilized per sampling cycle. This allows more accurate switching according to rotor speed. With the
sampling frequency maintained constant, the torque ripple at low speed can be significantly improved,
produced better current harmonics [16] and minimized switching loss [26]. In another design, variable

6
ICE4CT 2020 IOP Publishing
Journal of Physics: Conference Series 1878 (2021) 012047 doi:10.1088/1742-6596/1878/1/012047

domain adaptive fuzzy control was implemented in speed loop to generate reference torque and PI-
controllers were used in place of hysteresis controllers. Compared to conventional DTC-SVM, faster
transient and lower ripple of both torque and flux were achieved [37]. Instead of two active states per
sampling cycle in SVM, more accurate voltage vectors can be generated using duty cycle modulation
between the two states. Adaptive fuzzy logic control was implemented as modulator to select duty cycle
in addition to conventional DTC scheme. Experimental results showed significant torque ripple
reduction by 58% while keeping flux ripple maintained [35].

3.3.3. Predictive Direct Torque Control

Prediction-based DTC was proposed through estimating parameters cycles ahead and including them in
comparison with reference values before feeding to independent hysteresis controllers [38]. This method
modified the input prior the controllers while retaining structure of conventional DTC. Forward Euler
method incorporated in discretized IM modelling managed to diminish effect of time delay caused by
data processing. Compared to conventional DTC, steady state torque and flux ripples reduced by half.
Model predictive control (MPC) uses a model to predict future plant output and solves an
optimization problem to select the optimal control. Unlike traditional PID controller, MPC can handle
multi input multi output systems that might have interaction between input and output. MPC is also
capable to preview outcomes and handle constraints, which justifies requirement of powerful and fast
processor with large memory. MPC controllers have been widely used in processing industries and
recently applied to healthcare, automotive and aerospace industries [39]. MPC can be classified to CS-
MPC (Continuous Set MPC) and FS-MPC (Finite Set MPC), where FS-MPC has limited possible state
and does not necessarily needs modulator, thus reducing computational effort [40].
MPC embedded into DTC schemes is also known as Model Predictive Torque Control (MPTC).
MPTC uses discretized IM model to estimate the current, flux and torque parameter. This eliminates the
need to transform to rotational frame [38] which reduces the computational effort. Stator flux and torque
are then predicted to be two cycles ( + 2) ahead. The speed is regulated at outer loop to generate torque
reference and usually needs a speed sensor [41]. Reference stator was set to be fixed. The predicted
parameters are then compared with reference parameter before cost function is used to evaluate
minimum value as basis for selection of modulating voltage vectors.
The cost function in the MPC compares the predicted and reference values [42]. In it, the weighting
factor was determined based on trade-off between the torque and flux tracking [43]. During each cycle,
stator voltage and all relevant parameters such as torque and flux were evaluated for 6 different sectors.
For each stator voltage vector, the cost function g is evaluated to determine the minimum value, upon
which it will be selected as modulating vector to drive the inverter [44].
Figure 6 shows predictive direct torque controller with fuzzy logic speed controller for IM drive. In
this scheme, torque reference was generated from speed error using fuzzy logic control and fed to cost
function together with reference stator flux and predicted parameters. Prediction horizon for torque and
stator flux was increased to two ( + 2) to account for time delay encountered during hardware
implementation [38], [41], [44].
MPTC offers lesser ripple in torque when compared with DTC. Experimental results shown
significant reduction of torque and current ripple by more than 50% although flux ripple is maintained
compared to DTC. Low speed operation was also improved [45]. In DTC, the hysteresis bounds were
responsible to generate a switching vector, implying indirect voltage vector selection process. Whereas
in MPTC the selection of the voltage vector is direct from optimization of cost function [46]. However,
the optimization of cost function is dependent on torque and flux weighting factors where tuning was
tedious [45]. Integration of fuzzy logic for speed control of MPTC improves the transient states and add
robustness against external load [41].

7
ICE4CT 2020 IOP Publishing
Journal of Physics: Conference Series 1878 (2021) 012047 doi:10.1088/1742-6596/1878/1/012047

Figure 6 : MPTC with fuzzy logic scheme [41]

In Fuzzy-predictive-DTC (FPTC) scheme [47], fuzzy logic control structure is integrated to MPTC
algorithm, allowing adaptive variation and dynamic rules. After cost function evaluation, fuzzy logic
further process torque and flux error for tracking based on developed axis of constant torque and constant
stator flux variation. The fuzzy output is then modulated using SVM for triggering. Experimental results
shown significant reduction of torque and current ripple by 50% and flux ripple by 20% compared to
MPTC. The design was tested to be satisfactory performance at low speed and robust to 50% variation
of main parameter. However, on the downside, computational burden was increased with deployment
of SVM modulator and complex fuzzy inference engine even though the design approach adopted
minimal membership functions and Takagi-Sugeno inference system.

3.4. Sensorless Design Consideration

Most high performance control needs speed sensor measurement for accurate positioning of rotor angle
in coordinate transformation, sector selection in SVM [26] or determining developed torque in speed
loop [17]. Sensor coupling, inertia, torsional resonance, and the maximum sensor speed are among issues
to be considered for sensor selection [6]. In electric vehicle, space and weight are among the main
constraint. Hence, sensorless system might be the alternative solution.

Figure 7 : Sensorless three-level DTC drive [48]

Speed adaptive flux observer was used for rotor speed measurement and was implemented on
conventional DTC [48]. Figure 7 shows block diagram of sensorless DTC using speed adaptive flux
observer. Estimates of electromagnetic torque and stator flux were regulated independently by hysteresis

8
ICE4CT 2020 IOP Publishing
Journal of Physics: Conference Series 1878 (2021) 012047 doi:10.1088/1742-6596/1878/1/012047

controllers to generate corrective control action through switching table together with rotor angle for
sector selection. Experimental results show that estimated and the actual speed matched at very low
speed. Sliding mode observer was also used with vector control in sensorless design with better torque
ripple and robust performance[49], [50].
Extended Kalman Filter (EKF) has also been proposed in sensorless design [51]. Even though EKF
has good performances with noises, but the filter is unsteady with gross external disturbance and internal
modeling estimation error. Setting the attenuation level to threshold value improves its robustness and
lessen the error effect. [52] However, estimation error still persist even though motor dynamics control
in wide speed range was effective [14].

4. Performance analysis of motor control strategies

Table 1 summarizes induction motor control technologies. Only volt/hertz variable frequency drive is
scalar type control while other methods were vector types. From perspective of electric vehicle
application, operational features such as dynamic response, ripple performance, low speed performance,
sensorless solution and cost reduction are important features to be looked into.
Table 1. Induction Motor Control Technology
Literature Method Findings
[17], [18], [20] Volt/hertz variable Scalar control. Commonly used in industrial
frequency drive applications and driven by SPWM. Problems at low
speed and not suitable with high dynamic applications.
[16], [32], [33] Conventional FOC Vector control. High dynamic performance but also
schemes. high computational effort due to complexity and high
parameter sensitivity.
[16], [26], [36] DTC schemes. Simpler implementation with using switching table
and hysteresis controller but at higher ripple for
conventional DTC. Performance improved with
integration with SVM, variable hysteresis band and
additional speed control loop. However,
computational time also increased.
[38], [41], [44], Prediction based DTC Uses cost function as alternative to using SVM and
[45] using MPC (MPTC). coordinate transformation has improved the torque
ripple and good low speed performance at reduced
computational effort.
Fuzzy-predictive torque Integrates fuzzy structure after cost function and
control (FPTC) output modulated using SVM. Improved dynamic rule
base designed. Significant torque and flux ripple
[27]
reduction of compared to MPTC. Robust and
satisfactory low speed performance. Downside is high
computational burden.
Observer for sensorless Excellent actual speed matching down to very low
speed on vector control speed for discrete SVM despite trade-off at high load
[48]–[50]
system. with sliding mode observer provides better ripple
performance.
EKF for sensorless Alternative to observer design. However internal
[14], [52] speed control. speed estimation error persisted and susceptible to
gross disturbance.

In terms of dynamic response, all vector control strategy has superior performance over scalar control
due to advanced induction motor modelling that maintains flux in dynamic condition. For vector control
strategy, advantages of DTC over FOC are elimination of current regulators, coordinate transformations,

9
ICE4CT 2020 IOP Publishing
Journal of Physics: Conference Series 1878 (2021) 012047 doi:10.1088/1742-6596/1878/1/012047

less sensitivity to the parameter detuning, simple implementation, high dynamic response and
robustness. However, the drawbacks include high current, flux and torque ripple, variable switching
frequency and lack of direct current control. Implementing SVM to substitute switching table adds more
voltage vectors, generate constant switching frequency which reduces current and torque ripple by 50%
to be in par with FOC. However, the problem persists at low speed operation. This was circumvented in
MPTC which uses predictive algorithm and cost function to substitute SVM to further improves current
and torque ripple by 50% at reduced computational burden. Low speed performance was also improved.
FPTC integrates dynamic fuzzy rules after cost function and SVM. Compared to MPTC, FPTC produced
20% lower flux ripple and 50% lower current and torque ripple. However, FPTC requires high
performance computing due to complex fuzzy inference engine and modulator deployment.
FPTC design may be robust and produce best ripple performance, but requires high performance
computing. MPTC still produce excellent dynamic performance even at low speed with relatively higher
ripple but was significantly improved over DTC and FOC design. For sensorless solution, EKF and
observer has been used as speed estimator in the system. Sliding mode observer has been widely used
in DTC with improved accuracy compared to EKF which suffers from estimation error and lack
robustness. MPTC with its predictive features may be used together with sliding mode observer to
provide sensorless control solution for electric vehicle.

5. Conclusion
This paper has reviewed the control technologies associated for induction motor in industrial application
to electric vehicle. As prime mover in electric vehicle, expected high performance includes excellent
dynamic performance, low torque and flux ripple particularly at low speed and sensorless solution for
less complexity. MPTC with its predictive algorithm features may be used together with sliding mode
observer to provide sensorless control for electric vehicle due to excellent dynamic performance even
at low speed while producing low torque and flux ripple. The features achieved at low computational
burden with relatively simpler structure bode well with low cost requirement for electric vehicle.

Acknowledgement
The authors would like to acknowledge Faculty of Electrical Engineering Technology, Universiti
Malaysia Perlis for the funding and continuous support for this research.

References
[1] “Global EV Outlook 2020 – Analysis - IEA.” [Online]. Available:
https://www.iea.org/reports/global-ev-outlook-2020. [Accessed: 15-Nov-2020].
[2] J. Erjavec, Hybrid, Electric & Fuel-Cell Vehicles, 2nd ed. Delmar Cengage Learning, 2013.
[3] K. T. Chau, Electric Vehicle Machines and Drives. Singapore: John Wiley & Sons, Singapore
Pte. Ltd, 2015.
[4] D. G. Dorrell, A. M. Knight, L. Evans, and D. A. Staton, “Comparison of Different Motor
Design Drives for Hybrid Electric Vehicles,” pp. 3352–3359, 2010.
[5] N. Mohan, Advanced Electric Drives: Analysis, Control, and Modeling Using
MATLAB/Simulink. 2014.
[6] N. Mohan, T. M. Undeland, and W. P. Robbins, Power electronics : converters, applications,
and design, 3rd ed. John Wiley & Sons, 2003.
[7] T. Wildi, Electrical machines, drives, and power systems, 6th ed. Pearson New International
Edition, 2013.
[8] B. Wu, High-Power Converters and ac Drives. Hoboken, NJ, USA: John Wiley & Sons, Inc.,
2006.
[9] R. Saidur, S. Mekhilef, M. B. Ali, A. Safari, and H. A. Mohammed, “Applications of variable
speed drive (VSD) in electrical motors energy savings,” Renew. Sustain. Energy Rev., vol. 16,
no. 1, pp. 543–550, 2012.
[10] ESKOM, “Variable Speed Drives : Reducing energy costs in the sugar industry,” 2016. .

10
ICE4CT 2020 IOP Publishing
Journal of Physics: Conference Series 1878 (2021) 012047 doi:10.1088/1742-6596/1878/1/012047

[11] “Model S | Tesla.” [Online]. Available: https://www.tesla.com/models. [Accessed: 29-Jan-

2020].
[12] Z. Q. Zhu and D. Howe, “Electrical machines and drives for electric, hybrid, and fuel cell
vehicles,” Proc. IEEE, vol. 95, no. 4, pp. 746–765, 2007.
[13] H. Zheng, B. Wang, and W. Luo, A novel field weakening control strategy with variable
reference voltage for asynchronous motor, vol. 3, no. PART 1. IFAC, 2013.
[14] A. M. Trzynadlowski, M. Farasat, and M. S. Fadali, “Efficiency improved sensorless control
scheme for electric vehicle induction motors,” IET Electr. Syst. Transp., vol. 4, no. 4, pp. 122–
131, 2014.
[15] W. Sibo, Z. Huichao, L. Zhiyu, and W. Xiaoxu, “A new torque ripple test method based on
PMSM torque s ripple analysis for electric vehicle,” pp. 1–10, 2015.
[16] D. Casadei, F. Profumo, G. Serra, and A. Tani, “FOC and DTC: Two viable schemes for
induction motors torque control,” IEEE Trans. Power Electron., vol. 17, no. 5, pp. 779–787,
2002.
[17] O. E. S. Mohammed Youssef, “A New Open-Loop Volts/Hertz Control Method for Induction
Motors,” 2018 20th Int. Middle East Power Syst. Conf. MEPCON 2018 - Proc., pp. 266–270,
2019.
[18] Z. Zhang, Y. Liu, and A. M. Bazzi, “An improved high-performance open-loop V/f control
method for induction machines,” Conf. Proc. - IEEE Appl. Power Electron. Conf. Expo. -
APEC, pp. 615–619, 2017.
[19] S. Srilad, S. Tunyasrirut, and T. Suksri, “Implementation of a scalar controlled induction motor
drives,” 2006 SICE-ICASE Int. Jt. Conf., pp. 3605–3610, 2006.
[20] H. M. D. Habbi, H. J. Ajeel, and I. I. Ali, “Speed Control of Induction Motor using PI and V /
F Scalar Vector Controllers,” vol. 151, no. 7, pp. 36–43, 2016.
[21] A. Z. A. Firdaus, S. A. Azmi, N. I. Norazli, and K. Kamarudin, “Design and simulation of
fuzzy logic controller for variable frequency drive of three phase ac induction motor,” J. Adv.
Res. Dyn. Control Syst., vol. 11, no. 12 Special Issue, pp. 834–842, 2019.
[22] M. H. Rashid, Power Electronics: Circuits, Devices, and Applications. Pearson, 2009.
[23] M. E. ASKER and H. KILIÇ, “Modulation Index and Switching Frequency Effect on
Symmetric Regular Sampled Spwm,” Eur. J. Tech., vol. 7, no. 2, pp. 102–109, 2017.
[24] Infineon Technologies AG, “Different PWM Waveforms Generation for 3-Phase AC Induction
Motor with XC164CS,” no. July, pp. 1–23, 2006.
[25] Y. Ma, L. Fan, and Z. Miao, “Realizing space vector modulation in MATLAB/Simulink and
PSCAD,” in 45th North American Power Symposium, NAPS 2013, 2013.
[26] V. T, S. Srinivasa, N. I. T. Warangal, and A. Machine, “Reduction of torque ripple in direct
torque control of induction motor using constant switching frequency operation,” Energy
Convers., vol. ൯, no. November, pp. 1–6, 2010.
[27] A. Berzoy, J. Rengifo, and O. Mohammed, “Fuzzy Predictive DTC of Induction Machines with
Reduced Torque Ripple and High-Performance Operation,” IEEE Trans. Power Electron.,
2018.
[28] M. A. Hannan, J. A. Ali, A. Mohamed, and A. Hussain, “Optimization techniques to enhance
the performance of induction motor drives: A review,” Renew. Sustain. Energy Rev., vol. 81,
no. September 2016, pp. 1611–1626, 2018.
[29] H. Liu et al., “Induction Motor Drive Based on Vector Control for Electric Vehicles,” in 2005
International Conference on Electrical Machines and Systems, 2005, pp. 2–6.
[30] Bhim Singh, Pradeep Jain, A. P. Mittal, and J. R. P. Gupta, “Direct torque control: a practical
approach to electric vehicle,” in 2006 IEEE Power India Conference, 2006, p. 4 pp.
[31] M. B. Bilal Akin, “Sensored Field Oriented Control of 3 phase Induction Motor.,” 2013.
[32] M. Ichiro, N. Osamu, and N. Sakae, “A NEW SIMPLIFIED CURRENT CONTROL
METHOD FOR FIELD ORIENTED INDUCTION MOTOR DRIVES,” in Conference Record
of the IEEE Industry Applications Society Annual Meeting, 1989, no. 6, pp. 390–395.

11
ICE4CT 2020 IOP Publishing
Journal of Physics: Conference Series 1878 (2021) 012047 doi:10.1088/1742-6596/1878/1/012047

[33] H. Rasmussen, “Adaptive Field oriented control of induction motors,” ECC 1997 - Eur.
Control Conf., no. July, pp. 349–354, 1997.
[34] I. Takahashi and Y. Ohmori, “High-performance direct torque control of an induction motor,”
IEEE Trans. Ind. Appl., vol. 25, no. 2, pp. 257–264, 1989.
[35] L. Romeral, A. Arias, E. Aldabas, and M. G. Jayne, “Novel direct torque control (DTC)
scheme with fuzzy adaptive torque-ripple reduction,” IEEE Trans. Ind. Electron., vol. 50, no.
3, pp. 487–492, 2003.
[36] N. Farah, M. H. N. Talib, Z. Ibrahim, S. N. M. Isa, and J. M. Lazi, “Variable hysteresis current
controller with fuzzy logic controller based induction motor drives,” 2017 7th IEEE Int. Conf.
Syst. Eng. Technol. ICSET 2017 - Proc., no. October, pp. 122–127, 2017.
[37] Y. Pan, Y. Zhang, and Z. Wang, “A novel variable domain adaptive fuzzy control of direct
torque control for induction motor based on space vector control,” in 2010 Seventh
International Conference on Fuzzy Systems and Knowledge Discovery, 2010, pp. 639–643.
[38] J. Beerten, J. Verveckken, and J. Driesen, “Predictive Direct Torque Control for Flux and
Torque Ripple Reduction,” IEEE Trans. Ind. Electron., vol. 57, no. 1, pp. 404–412, 2010.
[39] J. Han, Y. Hu, and S. Dian, “The State-of-the-art of Model Predictive Control in Recent
Years,” IOP Conf. Ser. Mater. Sci. Eng., vol. 428, no. 1, 2018.
[40] K. Wróbel, P. Serkies, and K. Szabat, “Model predictive base direct speed control of induction
motor drive—continuous and finite set approaches,” Energies, vol. 13, no. 5, 2020.
[41] A. Ammar, B. Talbi, T. Amied, Y. Azzoug, and A. Kerrachr, “Predictive Direct Torque
Control with Reduced Ripples for Induction Motor Drive Based on T-S Fuzzy Speed
Controller,” Asian J. Control Editor. Off., 2017.
[42] S. Vazquez, J. Rodriguez, M. Rivera, L. G. Franquelo, and M. Norambuena, “Model Predictive
Control for Power Converters and Drives: Advances and Trends,” IEEE Trans. Ind. Electron.,
2017.
[43] P. Cortés et al., “Guidelines for weighting factors design in model predictive control of power
converters and drives,” in Proceedings of the IEEE International Conference on Industrial
Technology, 2009.
[44] H. Miranda, P. Cortes, J. I. Yuz, and J. Rodriguez, “Predictive Torque Control of Induction
Machines Based on State-Space Models,” IEEE Trans. Ind. Electron., vol. 56, no. 6, pp. 1–4,
2009.
[45] Y. Zhang, H. Yang, and B. Xia, “Model predictive torque control of induction motor drives
with reduced torque ripple,” IET Electr. Power Appl., vol. 9, no. 9, pp. 595–604, 2015.
[46] K. V. Praveen Kumar and T. Vinay Kumar, “Predictive torque control of open-end winding
induction motor drive fed with multilevel inversion using two two-level inverters,” IET Electr.
Power Appl., vol. 12, no. 1, pp. 54–62, 2018.
[47] A. Berzoy, J. Rengifo, and O. Mohammed, “Fuzzy Predictive DTC of Induction Machines with
Reduced Torque Ripple and High-Performance Operation,” IEEE Trans. Power Electron., vol.
33, no. 3, 2018.
[48] Y. Zhang et al., “An Improved Direct Torque Control for Three-Level Sensorless drive,” IEEE
Trans. Power Electron., vol. 27, no. 3, pp. 1502–1513, 2012.
[49] P. T. Doan, T. L. Bui, H. K. Kim, and S. B. Kim, “Sliding-mode observer design for sensorless
vector control of AC induction motor,” 2013 9th Asian Control Conf. ASCC 2013, 2013.
[50] V. Shah and N. Krishna Prakash, “FPGA Implementation of Sensorless Field Oriented Current
Control of Induction Machine,” 2017 IEEE Int. Conf. Comput. Intell. Comput. Res. ICCIC
2017, pp. 1–5, 2018.
[51] K. L. Shi, T. F. Chan, Y. K. Wong, and S. L. Ho, “Speed estimation of an induction motor
drive using an optimized extended Kalman filter,” IEEE Trans. Ind. Electron., 2002.
[52] W. Yang, X. Cai, and J. Jiang, “Speed sensorless vector control of induction motor based on
robust extended Kalman filter,” Proc. Int. Conf. Power Electron. Drive Syst., vol. 1, pp. 423–
426, 2012.

12