Romanian Journal of Information Technology and Automatic Control, Vol. 33, No.

2, 135-146, 2023 135

Sensorless vector control of Dual Star induction Motor

Associated with Fuzzy Mutual–MRAS estimator in
electric vehicle
Ismail BENMILOUD, Katia KOUZI, Aissa AMEUR
Laboratory of Semiconductors and Functional Materials, Amar Telidji University of Laghouat, Algeria
is.benmiloud@lagh-univ.dz, k.kouzi@lagh-univ.dz, a.ameur@lagh-univ.dz

Abstract: Sensorless control has become an attractive and important topic from an industrial perspective.
The presence of sensors in electronic speed drives can reduce their robustness and reliability while increasing
costs and mounting complexity. Therefore, mechanical sensors will be replaced by a calculation algorithm
that provides speed estimation from the electrical terminal of the machine. It is imperative that the
performance required from this estimator be close to or similar to that provided by a physical sensor. In this
paper, different estimation and observation techniques will be presented. Then, Sensorless vector control
based on a mutual - MRAS estimator and Fuzzy mutual - MRAS estimator for Dual Star induction Motor
(DSIM) to estimate the speed, as well as the resistances for both the stator and rotor will be introduced.
Finally, the results of the simulation will be shown and discussed.
Keywords: Dual Star Induction Motor, Sensorless vector control, Fuzzy mutual-MRAS estimator, electric
vehicle.

1. Introduction
The use of dual-star induction motor drives in electric vehicles provides several benefits
compared to DC machines, such as reliability, smaller size, absence of brushes, lower cost and
reduced maintenance (Bhatt, Mehar & Sahajwani, 2019), (Pellegrino et al., 2011), (Hashemnia &
Asaei, 2008). However, conventional field-oriented control of an induction motor (IM) requires a
speed sensor to function properly. The use of this sensor necessitates additional electronic devices,
more space, increased wiring and precise installation, thereby reducing the reliability of the
controller system (Holmes, McGrath & Parker, 2011), (Jain et al., 2020), (Chakraborty & Hori,
2003). Moreover, at low power, the cost of the sensor is approximately the same as that of the
motor. To overcome these problems, many studies have focused on research techniques that
eliminate the need for speed sensors while maintaining a high level of performance. These
techniques, known as sensorless techniques (Verma et al., 2013), (Guzinski & Abu-Rub, 2013),
(Casadei et al., 2003), present both technical and financial challenges. However, they offer several
advantages, including lower cost, reduced maintenance requirements, reduced measurement noise,
elimination of the sensor cable, and improved reliability.

2. Field-oriented control for DSIM

To achieve high torque and speed performance in an induction motor (IM) drive, the field-
oriented control strategy proves to be highly efficient. The concept behind this technique is to
emulate the behaviour of separately excited DC motor drives using the Dual Star induction Motor,
which inherently exhibits decoupling between electromagnetic torque and magnetic flux. The state
equation of the proposed motor in the dq reference frame is utilized to eliminate the quadrature
component of the flux (Andriamalala, Razik & Sargos, 2008), (Sadouni & Meroufel, 2012),
(Merabet et al., 2007) by expressing that:
 dr = r*
 (1)
  dr = 0
It is obtained:

https://doi.org/10.33436/v33i2y202310
136 Revista Română de Informatică și Automatică, vol. 33, nr. 2, 135-146, 2023

Rr .Lm ( iqs1 + iqs 2 )

w*gl = . (2)
( Lm + Lr ) r*

. * ( i + i )
Lm
Tem* = p. (3)
( Lm + Lr ) r qs1 qs 2
To achieve robust control, it is important to know of the value w*gl however, consistently
obtaining this value can be challenging as it is dependent on the resistance, which can vary due to
temperature fluctuations. In the next section, a technique to estimate the motor resistance is
presented.

3. Sensorless speed control

Modern industrial applications that utilize asynchronous variable speed drives, particularly
electric vehicles demanding high dynamic and static performance, rely on orientation and/or flux
control techniques (Montanari et al., 2007). These techniques necessitate accurate knowledge of the
position, flux modulus, and/or motor speed. This information is typically obtained through
feedback control using direct electrical sensors (such as currents, voltages, and flux) or mechanical
sensors (such as rotational speed and angular position). However, these sensors sometimes require
specialized signal processing for the received physical signals.
Furthermore, the elimination of the mechanical speed sensor can bring economic benefits
and enhance operational reliability. Additionally, control without a speed sensor must deliver
performance that is comparable to that of a physical sensor, with minimal deviation.

3.1. Speed estimation using MRAS technique

This technique is based on an adaptive system that utilizes an MRAS (Model Reference
Adaptive System) model (El Ouanjli et al., 2022), which consists of two flux estimators as
illustrated in Figure 1. One of these estimators, known as the reference model (voltage model),
does not include speed. The other estimator is the adjustable model (current model), which takes into
account the speed. The error resulting from the disparity between the outputs of the two estimators
(flux, emf, reactive power), along with the estimated speed generated by an adaptation algorithm, is
addressed in previous studies (Singh, Kachhwaha & Fulwani, 2022), (Cardenas et al., 2005).
Nevertheless, the speed estimate-based MRAS technique has a significant drawback, which
is its high sensitivity to machine parameters. To address this issue, several studies have proposed
online adaptation techniques utilizing either the stator resistance or the rotor resistance. However,
the current model is also sensitive to the rotor time constant, and adapting it simultaneously with
speed estimation poses challenges (Yang & Chin, 1993) (Mitronikas, Safacas & Tatakis, 2001).

Figure 1. MRAS Structure

http://www.rria.ici.ro
Romanian Journal of Information Technology and Automatic Control, Vol. 33, No. 2, 135-146, 2023 137

3.2. Speed estimation using mutual-MRAS technique

In the mutual MRAS approach, two rotor flux estimators are compared. This approach
enables the estimation of both the rotational speed and stator resistance of the DSIM. Figure 2
illustrates the structure of the Mutual MRAS, which consists of a reference model (voltage model),
an adjustable model (current model), and two adaptation mechanisms.

3.2.1. Reference model

Regarding the voltage model, the stator equation is described as a reference model in the
stationary reference frame (𝛼, 𝛽). It generates the reference value for the rotor flux component
(Sahraoui, Ameur & Kouzi, 2018). The following equations represent the two reference rotor flux
components determined from the reference model:
d rv Lr + Lm  dis1 Lr . Lm dis 
dt
=
Lm
v − r .i
 s1 s s1
− (
L + L . )
s m dt − L + L . dt 

(4)
 m r 

d rv Lr + Lm  di L . L di 
dt
=
Lm  s1 s s1 (
s m dt )
v − r .i − L + L . s1 − r m . s 
Lm + Lr dt 
(5)


3.2.2. Adaptive model

The current model utilizes the speed value and current input signals to compute the rotor flux
components. This model describes the DSIM rotor voltage equation in the stator reference frame
(Sahraoui, Kouzi & Ameur, 2017) as follows:

)
d r i
(
Lm 1
= is1 + is 2 − . r i − r .ri (6)
dt Tr Tr

d  r i Lm   1
=  i +i − . ri − r .r i
dt Tr  s1 s 2  Tr (7)

3.2.3. Adaptation mechanism

Two adaptation mechanisms take the error signals e r and eRs as input, with significant
differences between the two flux estimators (Touam et al., 2021).

e r =  rv .  r i −  r v .  ri (8)

eRs = is ( rv ) (

− ri + is rv − ri ) (9)

To minimize errors e r and eRs , the adaptation mechanisms deliver both estimated speed and
stator resistance by reinjecting them into their respective adjustable models.
The adaption rule, as described (Rsn et al., 2016) can be expressed as follows:

r = k pr .e r + kir . e r .dt (10)

Rs = k p Rs .eRs + kiRs . eRs .dt (11)

Where: k pr , kir ,and k pRs , k iRs : are respectively the gains of the two PI controllers.

http://www.rria.ici.ro
138 Revista Română de Informatică și Automatică, vol. 33, nr. 2, 135-146, 2023

Figure 2. The suggested mutual-MRAS estimator structure

The rotor resistance can then be determined by the following relation:

Rrn
Rr = Rr . (12)
Rns

3.3. Speed estimation using fuzzy mutual-MRAS Technique

The performance of the suggested fuzzy mutual-MRAS estimator is presented. The Figure 3
illustrates the structure of the proposed estimator as described by (Rsn et al., 2016).

Figure 3. Structure of fuzzy Mutual MRAS estimator

The concept of this approach involves modifying the estimator structure by replacing the
traditional PI regulators and adaptation mechanisms with intelligent PI fuzzy controllers (Touam et
al., 2021). This modification aims to enhance the performance of the estimator.

4. Vehicle dynamics
When considering the dynamics of the vehicle, it is observed that the tractive force needs to
overcome various resistive forces, including grading resistance (𝐹g), aerodynamic drag (𝐹𝑤),
rolling resistance (𝐹r), and inertial effects (𝐹i) (Ehsani et al., 2008). These forces can be represented
by the following equation and are illustrated in Figure 4:
Ftractive = Fw + Fi + Fg + Fr (13)

http://www.rria.ici.ro
Romanian Journal of Information Technology and Automatic Control, Vol. 33, No. 2, 135-146, 2023 139

Aerodynamic drag can be characterized by air density  , vehicle body form, vehicle frontal
area A f , and vehicle speed V (Lian et al., 2016).

1
Fw = . Af .CD (V − VW )
2

2 (14)
Where VW , V is a measure of the wind speed and C D is the aerodynamic drag coefficient
that describes the form of the vehicle body.
Acceleration Force represents the dynamic of the vehicle in terms of acceleration or braking
(Ehsani et al, 2008). It can be calculated using the following equation:
dV
Facc = M = M . (15)
dt
γ: vehicle acceleration.
As a result of the elastic material of the tire, a resistance force against rotational movement is
produced in front of the wheel center. The following equation is used to calculate the tire rolling
resistance:
Fr = M .g.Cr (16)

Cr
Where: M and represent the vehicle mass, rolling resistance coefficient.
If a vehicle moves on a sloping road, gravity creates a resistive force (Husain, and Islam,
1999). which can be represented by the gradient of path degree  as follows:
Fg = M .g .sin() (17)

Figure 4. Representation of dynamic vehicles

Where Table 1 and Table 2 represent the studied EV used in simulation, and motor
parameters respectively.
Table 1. Main vehicular parameters

Name Value Unit

vehicle mass 820 kg
Vehicle frontal area 2.7 m2
Aerodynamic drag 0.25 -
coefficient
Air density 1.23 kg.m-3
Wheel radius 0.3 m
Gravity 9.81 m/s2
rolling resistance 0.01 -
coefficient

http://www.rria.ici.ro
140 Revista Română de Informatică și Automatică, vol. 33, nr. 2, 135-146, 2023

Table 2. Parameters of dual star motor used in the simulation

Nominal Voltage (𝑉): 𝑉𝑛 = 220/380

Nominal Current (𝐴): 𝐼𝑛 = 6.5
Nominal Speed (𝑟𝑑/𝑠): 314
Stator Resistance (Ω): 𝑅𝑆1 = 𝑅𝑆2 = 3.72
Rotor Resistance (Ω): 𝑅𝑟 = 2.12
Stator Inductance (𝐻): 𝐿𝑆1 = 𝐿𝑆2 = 0.022
Rotor Inductance (𝐻): 𝐿𝑟 = 0.006
Mutual Inductance (𝐻): 𝐿𝑚 = 0.3672
Moment of Inertia (𝐾𝑔.𝑚2 ): 𝐽 = 0. 0625
Coefficient of Friction (𝑁𝑚𝑠/𝑟𝑑): 𝐾𝑓 = 0.001
Nominal Frequency (𝐻𝑧): 𝑓 = 50
Number of Pairs Poles: 𝑃=1

5. Simulation results
To evaluate the performance and stability of the proposed Fuzzy Mutual-MRAS estimator,
several tests were conducted under various dynamic operating conditions. These tests included
scenarios with abrupt changes in command speed. The purpose of these tests was to assess the
effectiveness of the estimator and ensure the stability of the overall system.

Figure 5. The rotation speed motor

http://www.rria.ici.ro
Romanian Journal of Information Technology and Automatic Control, Vol. 33, No. 2, 135-146, 2023 141

Figure 6. The rotor error speed estimation

Figure 7. The total flux

Figure 8. The motor electromagnetic torque

http://www.rria.ici.ro
142 Revista Română de Informatică și Automatică, vol. 33, nr. 2, 135-146, 2023

The estimated speed demonstrates rapid convergence, indicating the capability to maintain
accurate estimation across different speed levels with an acceptable error. Additionally, the
efficiency and accuracy of the suggested method estimator in resistance estimation were evaluated
using two scenarios.
In the first scenario, Figures 9 and 10 depict the variation of stator and rotor resistance. The
resistance values increase linearly up to 100% between 0 and 1 second, then remain constant at
twice their nominal value between 1 and 3 (s). Subsequently, the resistance decreases linearly to
-50% of its nominal value from 3 to 5 (s) and finally remains constant at -50% of its nominal value
after 5 seconds.

Figure 9. Stator resistance estimation using Fuzzy mutual-MRAS estimator in linear variation

Figure 10. Rotor resistance estimation using Fuzzy mutual-MRAS estimator in linear variation

In the second scenario presented in Figures 11 and 12, the resistance variation is modelled
exponentially, which provides a closer representation of real conditions.

http://www.rria.ici.ro
Romanian Journal of Information Technology and Automatic Control, Vol. 33, No. 2, 135-146, 2023 143

Figure 11. Stator resistance estimation using Fuzzy mutual-MRAS estimator in exponential variation

Figure 12. Rotor resistance estimation using Fuzzy mutual-MRAS estimator in exponential variation

From the obtained results depicted in Figures 9 to 12, it is evident that the estimated stator
and rotor resistance closely track their real values. Figures 7 and 8 illustrate the decoupling
between torque and flux, even in the presence of variable resistance. These findings highlight the
effectiveness and robustness of the proposed approach in achieving accurate control and
maintaining performance under varying conditions.

6. Conclusion
In this research, the implementation of a parallel fuzzy Mutual-MRAS estimator was utilized
to simultaneously estimate the resistances of the stator, rotor, and motor speed. The main objective
of this study was to establish a robust sensorless Field-Oriented Control for Dual Star Induction
Motors in electric vehicles. The result of the research has effectively accomplished this objective
and demonstrated the successful achievement of a reliable and accurate control system for DSIM in
electric vehicles. These findings contribute to the advancement of electric vehicle technology and
the optimization of motor control systems.

http://www.rria.ici.ro
144 Revista Română de Informatică și Automatică, vol. 33, nr. 2, 135-146, 2023

REFERENCES
Andriamalala, R.N., Razik, H. & Sargos, F. M. (November 2008) Indirect-rotor-field-oriented-
control of a double-star induction machine using the rst controller. In 2008 34th Annual Conference
of IEEE Industrial Electronics. pp. 3108-3113. IEEE.
Bhatt, P., Mehar, H. & Sahajwani, M. (2019) Electrical motors for electric vehicle–a comparative
study. Proceedings of Recent Advances in Interdisciplinary Trends in Engineering & Applications
(RAITEA).
Cardenas, R., Pena, R., Proboste, J., Asher, G. & Clare, J. (2005) MRAS observer for sensorless
control of standalone doubly fed induction generators. IEEE Transactions on Energy conversion.
20(4), 710-718.
Casadei, D., Serra, G., Tani, A., Zarri, L. & Profumo, F. (2003) Performance analysis of a speed-
sensorless induction motor drive based on a constant-switching-frequency DTC scheme. IEEE
Transactions on Industry Applications. 39(2), 476-484.
Chakraborty, C. & Hori, Y. (2003) Fast efficiency optimization techniques for the indirect vector-
controlled induction motor drives. IEEE Transactions on Industry Applications. 39(4), 1070-1076.
Ehsani, M., Gao, Y., Longo, S. & Ebrahimi, K. M. (2018) Modern electric, hybrid electric, and
fuel cell vehicles. CRC press.
El Ouanjli, N., Mahfoud, S., Derouich, A., El Daoudi, S. & El Mahfoud, M. (2022) Speed
Sensorless Fuzzy Direct Torque Control of Induction Motor Based MRAS Method. In Digital
Technologies and Applications: Proceedings of ICDTA’22, Fez, Morocco, Volume 2. pp. 779-790.
Cham: Springer International Publishing.
Guzinski, J. & Abu-Rub, H. (2012) Speed sensorless induction motor drive with predictive current
controller. IEEE Transactions on Industrial Electronics. 60(2). pp. 699-709.
Hashemnia, N. & Asaei, B. (September 2008) Comparative study of using different electric motors
in the electric vehicles. In 2008 18th International Conference on Electrical Machines. IEEE.
pp. 1-5.
Holmes, D. G., McGrath, B. P. & Parker, S. G. (2011) Current regulation strategies for vector-
controlled induction motor drives. IEEE Transactions on Industrial Electronics. 59(10). 3680-3689.
Husain, I. & Islam, M. S. (1999) Design, modeling and simulation of an electric vehicle system.
SAE transactions, 2168-2176.
Jain, J. K., Ghosh, S. & Maity, S. (2020) Concurrent PI controller design for indirect vector
controlled induction motor. Asian Journal of Control. 22(1), 130-142.
Jin, X., Yin, G. & Chen, N. (2019) Advanced estimation techniques for vehicle system dynamic
state: A survey. Sensors. 19(19), 4289.
Lian, Y., Zhao, Y., Hu, L. & Tian, Y. (2015) Longitudinal collision avoidance control of electric
vehicles based on a new safety distance model and constrained-regenerative-braking-strength-
continuity braking force distribution strategy. IEEE Transactions on Vehicular Technology. 65(6),
pp. 4079-4094.
Merabet, E., Abdessemed, R., Amimeur, H., Hamoudi, F. & Boukhlouf, R.C.M.E.H. (2007) Field-
oriented control of a dual star induction machine using fuzzy regulators. Algorithms. 6, 12.

http://www.rria.ici.ro
Romanian Journal of Information Technology and Automatic Control, Vol. 33, No. 2, 135-146, 2023 145

Mitronikas, E. D., Safacas, A. N. & Tatakis, E. C. (2001) A new stator resistance tuning method for
stator-flux-oriented vector-controlled induction motor drive. IEEE Transactions on Industrial
Electronics. 48(6), 1148-1157.
Montanari, M., Peresada, S. M., Rossi, C. & Tilli, A. (2007) Speed sensorless control of induction
motors based on a reduced-order adaptive observer. IEEE Transactions on Control Systems
Technology. 15(6), 1049-1064.
Pellegrino, G., Vagati, A., Guglielmi, P. & Boazzo, B. (2011) Performance comparison between
surface-mounted and interior PM motor drives for electric vehicle application. IEEE Transactions
on Industrial Electronics. 59(2), 803-811.
Rsn, R., Rs, R., Ls, L., Cr, C. & Kf, J. (2016) Performances des structures mutuelles adaptatives
MRAS et MRAS-Flou appliquées au contrôle sans capteur de vitesse pour actionneurs asynchrones.
Communication Science &technology. 16, 12-22.
Sadouni, R. and Meroufel, A., 2012. Indirect rotor field-oriented control (IRFOC) of a dual star
induction machine (DSIM) using a fuzzy controller. Acta Polytechnica Hungarica. 9(4), 177-192.
Sahraoui, K., Ameur, A. & Kouzi, K. (November 2018) Performance enhancement of sensorless
speed control of DSIM using MRAS and EKF optimized by genetic algorithm. In 2018
International Conference on Applied Smart Systems (ICASS). IEEE. pp. 1-8.
Sahraoui, K., Kouzi, K. & Ameur, A. (2017) Optimization of MRAS based Speed Estimation for
Speed Sensorless Control of DSIM via Genetic Algorithm. Electrotehnica, Electronica,
Automatica. 65(3), 156-162.
Singh, R. Kachhwaha, M. & Fulwani, D. M. (2022) MRAS-based Integral Sliding Mode Control of
Electric Vehicles under Speed and Load Torque Fluctuations. IEEE International Conference on
Power Electronics, Drives and Energy Systems (PEDES), Jaipur, Indi. pp. 1-6.
Touam, M., Chenafa, M. & Chekroun, S. (2021) Sensorless nonlinear sliding mode control of the
induction machine at very low speed using FM-MRAS observer. International Journal of Power
Electronics and Drive Systems. 12(4), 1987.
Verma, V., Chakraborty, C., Maiti, S. & Hori, Y. (2013) Speed sensorless vector controlled
induction motor drive using single current sensor. IEEE Transactions on Energy Conversion. 28(4),
938-950.
Yang, G. & Chin, T. H. (1993) Adaptive-speed identification scheme for a vector-controlled speed
sensorless inverter-induction motor drive. IEEE Transactions on Industry Applications. 29(4),
820-825.

http://www.rria.ici.ro
146 Revista Română de Informatică și Automatică, vol. 33, nr. 2, 135-146, 2023

Ismail BENMILOUD was born in Laghouat, Algeria, in 1996. He received his M.Sc. from
Amar Telidji University of Laghouat in 2020. Currently, he is a Ph.D. student at the same
University. His current research interests include Advanced control of Electric vehicles and power
converters.

Katia KOUZI was born in Algeria. She obtained her Engineer and Master’s Degrees in
Electrical Engineering in 1998 and 2002. She received her Ph.D. degree in Electrical and Computer
Engineering from the University of Batna, in 2008. Her research interests are focused on the
Advanced Control of AC Drives, including Vector, Sensorless, Intelligent Artificial Control,
Electric vehicles (EVs) control, Renewable Energy Systems Control, Managements and Storage
Systems. She is a researcher in the Semiconductors and Functional Materials Laboratory, as well as
a full professor at the Electrical Engineering Institute at Laghouat University, Algeria.

Aissa AMEUR was born in Laghouat, Algeria, in 1971. He obtained his Magister and Ph.D.
degrees in Electrical Engineering in 2005 and 2012, from Batna University, Algeria. In 2005, he
joined the Electrical Engineering Department of Laghouat University, Algeria as Assistant Lecturer.
Since May 2012, Dr Ameur is an Assistant Professor in the same department. He is a researcher in
the LeDMaScD laboratory, at Laghouat University, Algeria. His main research interests include
Modelling of Electrical Machines, Electrical Drives Control, Fault Diagnosis, Artificial
Intelligence and Renewable Energy Systems Control.

http://www.rria.ici.ro