See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/361406550

Development of a PMSM motor field-oriented control algorithm for electrical

vehicles

Article in Materials Today Proceedings · June 2022

DOI: 10.1016/j.matpr.2022.06.080

CITATIONS READS

14 1,552

5 authors, including:

Ramesh Pushparajan Umavathi M.

CMR Institute of Technology B.M.S. College of Engineering Bengaluru. e-mail:[email protected]
25 PUBLICATIONS 96 CITATIONS 13 PUBLICATIONS 19 CITATIONS

SEE PROFILE SEE PROFILE

C. Bharatiraja Ramanathan .G
SRM Institute of Science and Technology SRM Institute of Science and Technology
305 PUBLICATIONS 3,860 CITATIONS 19 PUBLICATIONS 88 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Ramanathan .G on 05 July 2022.

The user has requested enhancement of the downloaded file.

 Materials Today: Proceedings xxx (xxxx) xxx

Contents lists available at ScienceDirect

Materials Today: Proceedings

journal homepage: www.elsevier.com/locate/matpr

Development of a PMSM motor field-oriented control algorithm for

electrical vehicles
P. Ramesh a, M. Umavathi b, C. Bharatiraja c,⇑, G. Ramanathan c, Sivaprasad Athikkal c
a
Department of Electrical and Electronics Engineering, CMR Institute of Technology, Bengaluru, Karnataka 560037, India
b
Department of Electrical and Electronics Engineering, B.M.S College of Engineering, Bengaluru, Karnataka 560019, India
c
Department of Electrical and Electronics Engineering, SRM Institute of Science and Technology, Kattankulathur, Tamil Nadu 603203, India

a r t i c l e i n f o a b s t r a c t

Article history: Recently, with issues such as global warming, melting of glaciers and climate change, the need for sus-
Available online xxxx tainable methods of living has seen a drastic increase. Combustion cars that run on petrol and diesel
are slowly being replaced with electrical vehicles (EVs). Currently, EVs face challenges in terms of range
Keywords: of kilometers offered, availability of charging systems, infrastructure for their development, etc. As the
Permanent Magnet Synchronous Motor need for electrical vehicles increase, the usages of Permanent Magnet Synchronous Motor (PMSM) too
Single Chip Controller have seen a growing trend. Motor control in an electrical vehicle is extremely crucial as a vehicle on road
Field Oriented Control
is subject to different terrains and hence, different loads. Throughout all these different scenarios, it is
Electrical Vehicles
Hardware in the Loop simulation
essential that the motor operates at its optimum efficiency in order to provide optimum vehicle perfor-
mance for the driver and passengers. Currently, several methods are available for the control of electrical
vehicles, nevertheless the field-oriented control (FOC) method for permanent magnet motors is widely
used due to its efficient nature. This paper discusses with the design of modified FOC control algorithm
for control the PMSM motor eliminates an initial jerk seen during the starting condition of a motor
thereby providing a smoother start. As the entire simulation is done using Hardware in the Loop model,
the values obtained have real life constraints taken into account, as opposed to a software only simula-
tion. Moreover, the transient time taken to reach a steady state value in speed and to also change direc-
tion of rotation have been improved as well. The simulation has been designed entirely on the Altair
Embed software.
Copyright Ó 2022 Elsevier Ltd. All rights reserved.
Selection and peer-review under responsibility of scientific committee of the International Conference on
Advanced Materials and Mechanical Characterization (ICAMMC 2021).

1. Introduction analysis of the comparison between PMSM motors and Switched

Reluctance Motors has been detailed in [2]. Owing to their reduced
The shift towards renewable sources of energy has largely been size, robust nature, efficiency, high power to weight ratio, PMSMs
driven due to issues such as global warming and climate change. have seen a growing presence in the electrical vehicle industry.
The transportation sector has slowly seen the steady increase in These PMSMs have been, for a long time, used in tools for machin-
the usage of Hybrid Electrical Vehicles (HEVs) and Electrical Vehi- ing, actuators and drives that require servo motors for robotic pur-
cles (EVs) due to their major advantage of zero emissions. Tradi- poses. Advances in the field of PMSM motors have largely been due
tionally, the three main types of motors used for EVs and HEVs to the introduction of recent magnetic substances such as Samar-
are Induction Motors (IMs), Permanent Magnet Synchronous ium Cobalt and Neodymium Iron Boron. These magnetizing sub-
Motors (PMSMs) and Switched Reluctance Motors (SRMs) [1]. A stances possess a significant amount of high resistance for
comparison of these three motors across various parameters at a demagnetisation and density of energy. Earlier, presence of com-
speed of 1500 rpm has been depicted in Table 1 below. A further pounds such as AlNiCo or Aluminium Nickel Cobalt made motors
larger in size and prone to demagnetisation if motor currents were
incorrect. Development in the flux characteristics of a PMSM motor
⇑ Corresponding author.
E-mail address: [email protected] (C. Bharatiraja).

https://doi.org/10.1016/j.matpr.2022.06.080
2214-7853/Copyright Ó 2022 Elsevier Ltd. All rights reserved.
Selection and peer-review under responsibility of scientific committee of the International Conference on Advanced Materials and Mechanical Characterization (ICAMMC
2021).

Please cite this article as: P. Ramesh, M. Umavathi, C. Bharatiraja et al., Development of a PMSM motor field-oriented control algorithm for electrical vehi-
cles, Materials Today: Proceedings, https://doi.org/10.1016/j.matpr.2022.06.080
P. Ramesh, M. Umavathi, C. Bharatiraja et al. Materials Today: Proceedings xxx (xxxx) xxx

Table 1 the usage of expensive elements for computing. To combat this

Comparison of three main types of motors used for electrical vehicles. problem, a hardware prototyping of an intelligent dq PI controller
Type of Motor torque Copper Loss Iron Loss Efficiency was proposed by Mohd. Marufuzzaman et al. [18]. Jianxiong and H.
motor [Nm] [W] [W] [%] Gu proposed a sliding mode control to improve the control perfor-
SRM 303 7653 404 85.2 mance and robustness of the PMSM [19]. A particular method of
IM 297 8591 148 83.1 field-oriented control with an MRAS adaptive estimator for speed
PMSM 303 4328 198 91.3 measurement has been proposed by Pradeep Kumar, Naveen
Kaushik et al. in [20].
The rest of the paper is organized as follows. Proposed embed-
where flux weakening aided the motor in increasing the range of ding and extraction algorithms are explained in Sections 2 and 3.
its base speed has been proposed in [3]. Experimental results are presented in Section 4. Concluding
Different regions of operation in the PMSM produce varying tra- remarks are given in Section 5.
jectories of characteristic natures and subsequent torque ripples. A
study of the same has been presented in [4] where torque ripples
2. Proposed FOC theory
under different operating conditions have been analysed, and a
comparison of the two types of PMSM control – mainly the FOC
Field oriented control is by far, the most important form of con-
and DTC or direct torque control, have been detailed in [5].
trol for the permanent magnet synchronous motor. It is known
Research in the areas of motor control for the permanent magnet
that, for a given amperage of current, maximum torque can be
synchronous motor has largely been focused on the design and
obtained only when the stator flux is orthogonal or at 90° to the
development of different methods of field-oriented control. Many
rotor flux. Alternate names given to these two fluxes are the ‘cur-
of the following methods, and the research conducted in this field,
rents along the quadrature axis and direct axes’. When the angle
seek to combat a variety of problems. They also aim to improve the
between them is 0°, the motor does not spin. As the angle
controlling performance of the motors while taking care to not
increases, so does the torque and at 90°, there is maximum torque
increase complexity. The literature survey reports a variety of stud-
for a given value of current. Field oriented control seeks to main-
ies and research that has taken place so far in the control of perma-
tain this angle between the rotor flux and stator flux through its
nent magnet synchronous motors. Z. Tang and B. Akin proposed a
control methods, which will be explained here. Shown in Fig. 1 is
new Least Mean Square or LMS algorithm to suppress the dead
the basic block diagram that depicts the working of the field-
time present in FOC based PMSM drives [6]. A novel hybrid control
oriented control algorithm for a PMSM motor. The field-oriented
that utilized both active and virtual vectors for the FOC control
control algorithm works as follows. Two quantities from the PMSM
algorithm was proposed by S. Wang and C. Xia in [7]. Research in
motor are measured, namely the 3 phase stator currents and the
the investigation of two novel half switching frequency signal
angle of the rotor. The angle of the rotor is usually measured using
injects or HSFSI demodulation algorithms that were used for a
a sensor. The three phase stator currents are converted into a two-
PMSM in EV propulsion systems was studied in [8]. For the pur-
phase vector frame in the stationary state, and these currents are
poses of monitoring torque at a fast rate and to also achieve high
termed as Ialpha and Ibeta . These two currents denote the original
performance of a PMSM drive, a novel current controller was pro-
three phase stator currents in the alpha and beta, stationary refer-
posed by A. D. Alexandrou, N. K. Adamopoulos et al. in [9]. Niu Li,
ence axis. The transfer from three phase currents to alpha and beta
Yang Ming et al. wrote a technological review that detailed a list of
currents is achieved by using the Clarke transformation. In order to
high-performance current control strategies for the PMSM motor
ease calculations and simplify the mathematical process, the alpha
[10]. The review discussed the various current control strategies
and beta forms of the stator currents need to undergo another
present in the field. Problems present in the automatic speed track-
transformation to transition from a stationary frame to rotating
ing control of EVs that are propelled by PMSM motors was solved
frame since the rotor is rotating compared to the stationary frame
by using a scheme that used reconfiguration and a HOSM or Higher
of the stator currents. This transform is done achieved using the
Order Sliding Mode observer. This was proposed in the event of a
Park transformation. The currents in this new, rotating frame of
failure of sensors to enable proper performance of the drive and
reference are referred to as the direct axis current, Id and the
the corresponding research has been proposed by Suneel K Kom-
quadrature axis current, Iq . In order to achieve maximum output
muri, Michael Defoort et al. in [11].
torque for a given value of current, we must orient the quadrature
Two independent technology reviews, one that focused on anti-
axis current to be perpendicular or 90° to the direct axis current.
lock brake systems and traction systems for EVs driven by motors
The axis along the rotor frame is usually said to be the direct axis.
that are individually controlled, and two, that focused on control
In order to do this, the value of the direct axis current, Id is set to as
strategies for grid integration of electrical vehicles has been pro-
low a value as possible, ideally 0 A. However, in practice, we have
posed in [12-14]. A similar research paper along the lines of anti-
given a very minimum value in order to not have completely no
lock braking system had been proposed in [15] by Zhongshi Zhang
direct axis current. Subsequently, by increasing the quadrature axis
et al. and it discussed a novel PMSM control. Recently, the control
component, the net stator current vector slowly starts to shift
algorithms developed for better control of PMSM motors has seen
towards the quadrature axis, until it is perpendicular to the rotor
the inclusion of artificial intelligence, machine learning and the use
axis. Once this is achieved, the corresponding correction voltages
of neural networks or NNs. In 2018, an advanced method of con-
that corresponds to the value of current for which it is perpendic-
trolling speed using ANNs or Artificial Neural Networks in combi-
ular to the direct axis is noted, and these voltages undergo a series
nation with the existing FOC algorithm was proposed by Ho Pham
of reverse Park and reverse Clark transforms in order to arrive at
Huy Anh et al. [16]. The SVM is mostly used for two-level and mul-
the corrected, three phase voltages, which are then modulated
tilevel inverters drive [21–26]. Hakan Celik and Tevfik Yigit pro-
onto the inverter as is shown in the block diagram (Fig. 1).
posed a new type of FOC algorithm that used a proportional
integral controller with two degrees of freedom. The parameters
for the PI controller were arrived at by means of Particle Swarm 2.1. Equivalent circuits and mathematical modelling
Optimization, otherwise known as PSO and this research has been
detailed in [17]. The performance of a drive that uses PMSM motor The equivalent circuit for the PMSM motor along the two axes,
depends on the circuits that drive the motor and also necessitates namely the direct axis and quadrature axis are shown in Figs. 2 and
2
P. Ramesh, M. Umavathi, C. Bharatiraja et al. Materials Today: Proceedings xxx (xxxx) xxx

Fig. 1. Block diagram depicting working of the field-oriented control algorithm.

3 respectively. The equivalent models are derived by transforming

the voltage of the motor drive into parameters ‘dq’ and providing a
connection to their respective ‘dq’ circuit. Using these circuits, an
equation for speed of the motor can be obtained as follows:
np dwrot
ðT E
T L Þ ¼ J ð1Þ
2 dt
where, np is the number of poles, wrot is the rotational angular
speed, J is the inertia of the motor and load, and T L is the load
torque.
The mathematical modeling of a permanent magnet syn-
chronous motor mainly deals with respect to the stator as the rotor
consist of permanent magnets with no electrical magnetization. An Fig. 3. Quadrature axis equivalent circuit for PMSM motor.
equation for the stator voltage in a reference name that is station-
ary in a space vector representation is as shown below:
dkst
V st ¼ Rst Ist þ ð2Þ where as and a2s are operators that are spatial and required for the
dx
orientation of stator windings, and their values are as ¼ ej2p=3 and
where, V st ; Rst , Ist , and kst are respectively, the voltage, resistance, a2s ¼ ej4p=3 . V sta ; V stb ; V stc represent the instantaneous values of the
current and flux linkages of the three phase stator windings, stator voltages. Ista ; Istb ; Istc are the instantaneous values of the three
expressed in a reference frame that is stationary and fixed with phase stator currents and the stator flux linkages are represented by
respect to the stator. ksta ; kstb ; kstc where the stator flux linkages are derived respectively
The quantities voltage, resistance, current and flux linkages are from their mutual and self-inductances.
respectively defined by the following equations as shown below:
V st ¼ ½V sta ðt Þ þ as V stb ðt Þ þ a2s V stc ðtÞ ð3Þ 2.2. Forward Clarke and Park transforms

Ist ¼ ½Ista ðt Þ þ as Istb ðtÞ þ a2s Istc ðt Þ ð4Þ The Forward Clarke transformation or simply, the Clarke Trans-
formation is used to convert a three-phase quantity in a stationary
kst ¼ ½ksta ðt Þ þ as kstb ðt Þ þ a2s kstc ðtÞ ð5Þ frame to a two-phase quantity, with perpendicular axes, in that
same stationary frame. Usually, the two new quantities in the sta-
tionary frame are called as the ‘alpha’ and ‘beta’ quantities, respec-
tively denoted by the symbols, a and b. In the case of the field-
oriented control, the Clarke transformation is used to convert the
three-phase stator current values into two-phase stator currents,
Ialpha and Ibeta .
The equations for the Forward Clarke transformation are as fol-
lows, and the resulting vector representation for the same is shown
in Fig. 4.
3
Ialpha ¼ Ia ð6Þ
2
p p
3 3
Ibeta ¼ Ib
Ic ð7Þ
2 2
Fig. 2. Direct axis equivalent circuit for PMSM motors.

3
P. Ramesh, M. Umavathi, C. Bharatiraja et al. Materials Today: Proceedings xxx (xxxx) xxx

Fig. 5. Vector representation of forward Park transformation.

Fig. 4. Vector representation of Forward Clarke Transformation.

where Ialpha and Ibeta are the stator current values after the Clarke
Transformation, and Ia , Ib and Ic are the original stator current
values.
The Forward Park transformation or simply, the Park transfor-
mation is a set of mathematical equations used to convert a two-
phase quantity in the stationary frame, to a two-phase quantity,
in the rotating frame. The two new quantities in the new frame
of reference are represented by the letter ‘D’ and ‘Q’. The two-
phase stator currents, Ialpha and Ibeta in the stationary frame to ID
or the direct axis current and IQ or the quadrature axis current in
the rotating frame. The equations for the forward Park Transforma-
tion are as follows, and the vector representation of the same has
been shown in Fig. 5.

ID ¼ Ialpha cos £D þ Ibeta sin £D ð8Þ

IQ ¼
Ialpha sin £D þ Ibeta cos £D ð9Þ

where, ID and IQ is the direct axis current and quadrature axis cur-
rent respectively and £D is the angle between the rotor axis, also
known as the direct axis, and the ‘A’ phase of the three-phase stator
current. Fig. 6. Vector representation of the reverse Park transformation.

2.3. Reverse Park and reverse Clarke transformations

The Park reverse transformation is used to convert the direct

axis and quadrature axis quantities back to the alpha and beta val-
ues. The Park reverse transformation transforms the ‘D’ and ‘Q’ val-
ues in a rotating frame of reference back to the ‘alpha’ and ‘beta’
values in the stationary reference frame. In this case, the correction
voltages V D and V Q undergo a Park reverse transformation to
obtain the corrected, V alpha and V beta in the stationary frame. The
equations for the reverse Park transformation are as follows, and
its corresponding vector representation has been depicted in
Fig. 6.Fig. 7..

V alpha ¼ V D cos £D
V Q cos £D ð10Þ

V beta ¼ V D sin £D þ V Q cos £D ð11Þ

where V D and V Q are the correction voltages in the rotating refer-

ence frame and V alpha and V beta are the alpha and beta values of
the stator voltages in the stationary reference frame. The net volt-
age vector is represented by V s and the angle between the rotor axis
and the ‘A’ phase is represented by £D .
The reverse Clark transformation is a mathematical equation
that converts the ‘alpha’ and ‘beta’ values of the correction voltages
back to the original three phase form of the stator voltages in the Fig. 7. Vector representation of reverse Clarke transformation.

4
P. Ramesh, M. Umavathi, C. Bharatiraja et al. Materials Today: Proceedings xxx (xxxx) xxx

stationary reference frame. The newly generated three phase volt- avoid the accidents. The FOC techniques provides this faster
ages are the corrected voltages that are later on modulated onto response by comparing the actual speed with the reference speed
the inverter. This transformation forms the penultimate stage of and the obtained error is given to the PI controller block which
the entire FOC algorithm with the final stage being the modulation modifies the pulse pattern that is required to drive the motor at
of these new voltages onto the inverter board. The vector represen- defined speed. The simulation for the FOC control is done and
tation for the reverse Clark Transformation is shown in figure 4.11 the response are discussed in the upcoming section.
and the corresponding mathematical equations are:

2 3. Simulation model and design

VA ¼ V alpha ð12Þ
3
The simulation for the modified FOC was done entirely using
1 1 the Altair Embed software. The software enables a user to connect
V B ¼
V alpha þ p V beta ð13Þ actual hardware components and control them with the simula-
3 3
tion using USB and UART channels that interface the CPU with
the C2000 Launchpad. Our model aimed at improving the existing
1 1
V C ¼
V alpha
p V beta ð14Þ FOC algorithm by improving the transient conditions of the motor,
3 3
as will be discussed in Section 3. The components used for testing
where, V A , V B , V C are the three phase stator voltages in the regular the simulation were a 7.1 A, 6000 rpm rated PMSM, a TI C2000
coordinate system and a stationary reference frame. Launchpad and a TI DRV 8301 BOOSTXL motor driver, along with
The proposed FOC controls the speed and torque of the motor at a DC supply with an input voltage of 20 V. The simulation diagram
quick time. The EV requires faster control response in real time to consists of a main controller block, and scopes to observe transient
changes in motor speed. The values of direct axis (D-axis) current,
quadrature axis (Q-axis) current and values for the PI controller are
Table 2 fed as inputs to the controller. There is also a pre scaling block that
Simulation parameters for the modified field-oriented control algorithm. has been added to accommodate different data types so that all
Input parameter Value kinds of input values can be read by the C2000 microcontroller
DC voltage 20 V
chip. As per FOC theory (refer Fig. 1 for the block diagram), in order
Kp value 4.7 for the torque produced per amp of current to be maximum, the
Ki value 55.8 direct axis current must theoretically be zero, which is why for
D-axis current 0.1 A our simulation, the D-axis current has been limited to just 0.1 A.
Q-axis current 1.7 A
The controller block contains mathematical modules for the
Clarke and Park transformations that are crucial in implementing

Fig. 8. Simulation diagram for field-oriented control of a PMSM.

5
P. Ramesh, M. Umavathi, C. Bharatiraja et al. Materials Today: Proceedings xxx (xxxx) xxx

Fig.9. Controller block for the FOC algorithm.

the FOC control algorithm. Also given within the control module modified and modified algorithms respectively, during the starting
are blocks that prevent any sudden and drastic changes to the condition of the motor.
input values. This is mainly done in order to ensure that there is The quantity on the X axis denotes time in seconds and the Y
no damage to the motor. The entire simulation gets compiled axis represents the speed of the motor in rpm. As seen in Fig. 10,
and converted into C code which is then loaded onto the C2000 there is a slight jerk that is present during the starting condition,
microcontroller chip. Table 2 details the simulation parameters of where for an extremely brief moment the motor stutters a bit
the model, and the simulation circuit that incorporates the modi- before picking up the input and achieving a steady state speed.
fied FOC algorithm has been shown in Fig. 8. The simulation circuit Moreover, the time taken to reach this steady speed after the initial
that incorporates the modified FOC algorithm is shown below.Fig9. start condition is approximately 0.5 s, as can be calculated from
Fig. 9.. determining the X axis values at the respective intervals. However,
The inputs to the controller are the values of direct axis current,
quadrature axis current, and the Kp and Ki values for the propor-
tional integral (PI) controller that is present within the main con-
troller block of the simulation diagram. Our modifications and
improvements aim at fine tuning the control algorithm by deter-
mining new values for the PI controller that will better the
response time of the motor. The simulation file contains two parts
– a compile file where the simulation is designed, and a debug file,
where the output can be obtained and where its corresponding
input values can be changed in real time. The controller block for
the field-oriented control algorithm is shown below.

4. Simulation results and discussion

The motor, which was run using the improved and modified
FOC simulation, was tested and compared with the non-modified
control algorithm in two cases – 1) response time and nature of Fig. 11. Modified algorithm.
the motor during the starting condition and 2) the transient condi-
tions during change in rotation from an anticlockwise to clockwise
direction. Figs. 10 and 11 show the output waveforms for the non-

Fig.10. Non-modified algorithm. Fig. 12. Anticlockwise to clockwise transient for the non-modified algorithm.

6
P. Ramesh, M. Umavathi, C. Bharatiraja et al. Materials Today: Proceedings xxx (xxxx) xxx

Fig. 13. Anticlockwise to clockwise transient for the modified algorithm.

Table 3 in the case of the modified algorithm, shown in Fig. 11, there is no
Comparison between the non-modified and modified algorithm. initial stuttering present. The motor begins picking up speed at
Area of comparison Time taken Time taken around 3.6 s, and quickly reaches its steady state speed at 3.9 s,
by by indicating a quicker response time of 0.3 s when compared to
non-modified modified the non-modified algorithm which takes 0.5 s. The second compar-
algorithm algorithm
ison between the modified and non-modified algorithms dealt
Starting to steady state condition 3.9 s 3.6 s with the transient condition of the motor when the direction of
Clockwise to anticlockwise transient 3.25 s 1.25 s
rotation is changed from an anticlockwise direction to a clockwise
condition
direction. This change in direction is achieved by controlling the
value of the quadrature axis current. A negative value of quadra-
Table 4 ture axis current causes the motor to rotate anti-clockwise, and a
Values of torque for different values of quadrature axis current. positive value of quadrature axis current causes clockwise rotation.
Quadrature axis current (A) Motor torque (mNm) The comparison of the two transient conditions between the non-
modified and modified algorithms is shown respectively in Figs. 12
0 0
0.2 22
and 13.Fig. 12.
0.6 64 Similar to the graphs shown before, the graphs in Figs. 12 and
0.8 88 13 have the X axis represent time in seconds and the Y axis denote
1.0 110 motor speed in rpm. The modified algorithm took around 1.25 s to
1.2 140
make the motor rotate from an anticlockwise to clockwise direc-

Fig.14. Experimental setup for implementation of the FOC algorithm for the PMSM motor.

7
P. Ramesh, M. Umavathi, C. Bharatiraja et al. Materials Today: Proceedings xxx (xxxx) xxx

Fig.15. Stator voltage (above) and stator current (below) waveforms.

Fig. 16. Line to line voltage (phase A, B).

tion, which is around 2 s faster than the non-modified algorithm to-line voltage between phases A and B were measured using the
that takes around 3.25 s to do the same. Table 3 details the com- oscilloscope and has been shown in Fig. 16. Moreover, the input
parison of the modified and non-modified FOC algorithm and the DC voltage of 20 V was measured using the oscilloscope, and the
respective time taken under different conditions.Table 4. waveform of the same has been shown in Fig. 17.Fig. 18.
In order to experimentally verify the accuracy of the algorithm,
a mixed signal oscilloscope was used to obtain waveforms of the 4.1. Relation between torque and quadrature axis current.
stator voltage and current using voltage and current probes. The
experimental setup along with the stator voltage and current According to the field-oriented control theory, the output tor-
waveforms are shown respectively in Figs. 14 and 15. The line- que of the motor is dependent on the value of the quadrature axis
8
P. Ramesh, M. Umavathi, C. Bharatiraja et al. Materials Today: Proceedings xxx (xxxx) xxx

Fig. 17. Input DC voltage depicting a value of 20 V.

rent based on the simulation that we ran. The input voltage pro-
vided was 20 V. The graph shown below depicts the relation
between the two quantities.
The theoretical formula for the PMSM motor torque under FOC
is:

3
Te ¼ k np IQ ð15Þ
2

where, k is the stator inductance, np is the number of pole pairs and

IQ is the quadrature axis current value. For a value of IQ = 0.6 A, and
substituting the values for each quantity, where k = 0.18mH, np = 4,
Fig. 18. Quadrature axis current (Iq) vs shaft torque of PMSM motor. we get T e = 32 ð0:18 x 10
3 Þð4Þð0:6Þ giving us 0.0648 Nm or 64 Nm
which matches our experimental value. The behaviour of the motor
in this situation can be attributed to the case where a driver pushes
current and by increasing the quadrature axis current, we can on the accelerator pedal to command more torque and speed from
obtain higher values of torque. The table below shows the different the wheels. In this case, pressing down on the accelerator pedal
values of torque obtained at various values of quadrature axis cur- would mean increasing the value of the quadrature axis current,

Fig. 19. Transient change in motor speed due to change in quadrature axis current.

9
P. Ramesh, M. Umavathi, C. Bharatiraja et al. Materials Today: Proceedings xxx (xxxx) xxx

Fig.20. Braking scenario of PMSM motor.

Fig.21. (a) XY plot of output current under a loaded condition, (b) Vector plot of output voltage under a loaded condition.

Fig. 22. (a) XY plot of output current in the unloaded condition, (b) Vector plot of output current in the unloaded condition.

which as we can see from the graph, causes an increase in motor tion for the motor as well as the removal of initial noises during the
torque. starting condition. The motor undergoes a rapid yet smooth decel-
The corresponding increase in torque results in the increase of eration from 2750 rpm to 0 rpm with minimal noise. The transient
motor speed as depicted by the graph below. By increasing the braking scenario has been shown in Fig. 20.
value of Iq (quadrature axis current) from 0.8 A to 1.0 A, the shaft
speed of the motor increases from 2685 rpm to around 2750 rpm 4.3. Output current waveform in loaded and unloaded conditions
as shown in Fig. 19 and the transient time is 0.02sec
approximately. The PMSM motor, under field-oriented control was tested under
loaded and unloaded conditions and the following output wave-
4.2. Braking scenario for the PMSM motor forms were obtained. A load of around 1 kg was applied to the shaft
of the motor. Upon adding the load to the shaft of the motor, the
In order to understand the braking characteristics of our PMSM amplitude of the output current rose significantly.
motor while under the influence of the field-oriented control algo-
rithm, we rapidly brought the motor to a stop from a speed of 4.3.1. Loaded condition
around 2750 rpm. The values used for Kp and Ki to test these char- The unloaded condition refers to the absence of any load pre-
acteristics were the optimal values of Kp = 4.7 and Ki = 55.8 that we sent on the shaft of the motor. In this state, the motor is free to
had arrived at for improving the anticlockwise to clockwise transi- run without any resistance on its shaft. Due to this reason, the cur-
10
P. Ramesh, M. Umavathi, C. Bharatiraja et al. Materials Today: Proceedings xxx (xxxx) xxx

rent drawn by the motor is minimal, and the peak amplitude of the Declaration of Competing Interest
output current is much lower when compared to the loaded state.
From the plotted graphs in Fig. 21, we can conclude that under The authors declare that they have no known competing finan-
loaded condition, the motor draws more current, hence there is cial interests or personal relationships that could have appeared
an increase in the amplitude of the output current. Under loading, to influence the work reported in this paper.
the motor draws a maximum current of 0.175 A per unit value, and
in the unloaded state, the maximum current drawn is 0.05 A per
unit value. The testing under loaded condition and unloaded con- Acknowledgement
ditions were carried out in order to simulate the real-life experi-
ence of different kinds of terrain when travelling in an electric car. This work is supported by Science and Engineering Research
Board (SERB), Department of Science and Technology (DST) - Pro-
ject File name: TAR/2019/000287 Dated; 25-11-2019. Project PI;
Dr.C.Bharatiraja and PI students authors are Thanks to the SERB
4.3.2. Unloaded condition
and DST.
The loaded condition can be seen as moving up a hill, where
more torque is required to keep the car running, thus drawing
more current. Whereas, the unloaded condition can be seen as References
travelling on level surface where a lesser torque is enough to keep
the car in motion. The Fig. 22 shown the current obtained during [1] A.M. Lulhe, T.N. Date, A technology review paper for drives used in electrical
vehicle (EV) & hybrid electrical vehicles (HEV), 2015 International Conference
the unloaded condition. As no load is connected, the current draws on Control, Instrumentation, Communication and Computational Technologies
in the motor is very less that is reflected in XY plot and vector plot (ICCICCT), 2015, pp. 632–636, doi:10.1109/ICCICCT.2015.7475355.
of output current unloaded condition. [2] R.M. Pindoriya, B.S. Rajpurohit, R. Kumar, K.N. Srivastava, Comparative analysis
of permanent magnet motors and switched reluctance motors capabilities for
electric and hybrid electric vehicles, IEEMA Engineer Infinite Conf. (eTechNxT)
2018 (2018) 1–5, https://doi.org/10.1109/ETECHNXT.2018.8385282.
[3] K. Baoquan, L. Chunyan, C. Shukang, Flux-Weakening-Characteristic Analysis
5. Conclusion of a New Permanent-Magnet Synchronous Motor Used for Electric Vehicles,
IEEE Trans. Plasma Sci. 39 (1) (Jan. 2011) 511–515, https://doi.org/10.1109/
TPS.2010.2076355.
The design and implementation of a modified field-oriented
[4] J. Lara, J. Xu, A. Chandra, Effects of Rotor Position Error in the Performance of
control algorithm was carried out. The modified algorithm was Field-Oriented-Controlled PMSM Drives for Electric Vehicle Traction
compared with the non-modified algorithm in the starting condi- Applications, IEEE Trans. Ind. Electron. 63 (8) (Aug. 2016) 4738–4751,
tion and also the anticlockwise to clockwise transient condition. https://doi.org/10.1109/TIE.2016.2549983.
[5] M. Abassi, A. Khlaief, O. Saadaoui, A. Chaari, M. Boussak, Performance analysis
It was found that the modified algorithm improved upon the orig- of FOC and DTC for PMSM drives using SVPWM technique, in: 2015 16th
inal algorithm in terms of possessing an improved response time in International Conference on Sciences and Techniques of Automatic Control and
order to reach a steady state and also reduction in the time taken to Computer Engineering (STA), 2015, pp. 228–233, https://doi.org/10.1109/
STA.2015.7505167.
change direction of rotation. Experimental verification in order to [6] Z. Tang and B. Akin, ‘‘A New LMS Algorithm Based Deadtime Compensation
obtain waveforms for the stator voltage and current using a mixed Method for PMSM FOC Drives,” in: IEEE Transactions on Industry Applications,
signal oscilloscope was carried out as well. It was found out that vol. 54, no. 6, pp. 6472-6484, Nov.-Dec. 2018, doi: 10.1109/TIA.2018.2853045.
[7] Y. Yan, S. Wang, C. Xia, H. Wang, T. Shi, Hybrid Control Set-Model Predictive
the new modified algorithm significantly improved upon the orig- Control for Field-Oriented Control of VSI-PMSM, IEEE Trans. Energy Convers.
inal field-oriented control algorithm. Other case studies such as the 31 (4) (Dec. 2016) 1622–1633, https://doi.org/10.1109/TEC.2016.2598154.
relation between quadrature current and motor torque, quadrature [8] J. Lara, A. Chandra, Performance Investigation of Two Novel HSFSI
Demodulation Algorithms for Encoder-less FOC of PMSMs Intended for EV
current and motor speed, and braking scenario, loaded and Propulsion, IEEE Trans. Ind. Electron. 65 (2) (2018) 1074–1083.
unloaded conditions were undertaken and their results were [9] A.D. Alexandrou, N.K. Adamopoulos, A.G. Kladas, Development of a Constant
reported as well. Switching Frequency Deadbeat Predictive Control Technique for Field-
Oriented Synchronous PermanentMagnet Motor Drive, IEEE Trans. Ind.
Electron. 63 (8) (Aug. 2016) 5167–5175, https://doi.org/10.1109/
TIE.2016.2559419.
[10] N. Li, Y. Ming, G. Xianguo, X. Dianguo, A comparative study of model predictive
CRediT authorship contribution statement
current control and FOC for PMSM, in: 2014 17th International Conference on
Electrical Machines and Systems (ICEMS), 2014, pp. 3143–3147, https://doi.
P. Ramesh: Conceptualization, Methodology, Writing – original org/10.1109/ICEMS.2014.7014034.
draft, Data curation, Formal analysis, Funding acquisition, Investi- [11] S.K. Kommuri, M. Defoort, H.R. Karimi, K.C. Veluvolu, A Robust Observer-Based
Sensor Fault-Tolerant Control for PMSM in Electric Vehicles, IEEE Trans. Ind.
gation, Project administration, Resources, Software, Supervision, Electron. 63 (12) (Dec. 2016) 7671–7681, https://doi.org/10.1109/
Validation, Visualization, Writing – review & editing. M. Uma- TIE.2016.2590993.
vathi: Conceptualization, Methodology, Writing – original draft, [12] V. Ivanov, D. Savitski, B. Shyrokau, A Survey of Traction Control and Antilock
Braking Systems of Full Electric Vehicles with Individually Controlled Electric
Data curation, Formal analysis, Funding acquisition, Investigation, Motors, IEEE Trans. Veh. Technol. 64 (9) (Sept. 2015) 3878–3896, https://doi.
Project administration, Resources, Software, Supervision, Valida- org/10.1109/TVT.2014.2361860.
tion, Visualization, Writing – review & editing. C. Bharatiraja: Con- [13] H. Fujimoto, J. Amada, K. Maeda, Review of traction and braking control for
electric vehicle, IEEE Veh. Power Propul. Conf. 2012 (2012) 1292–1299,
ceptualization, Methodology, Writing – original draft, Data https://doi.org/10.1109/VPPC.2012.6422491.
curation, Formal analysis, Funding acquisition, Investigation, Pro- [14] C. Bharatiraja, A. Shaikh, H. Katare, A. Modi, Investigation on Simultaneous
ject administration, Resources, Software, Supervision, Validation, Operation of Two Hub BLDC Motors for Electric Vehicles, J. Appl. Sci. Eng. 24
(2) (April 2021) 233–238.
Visualization, Writing – review & editing. G. Ramanathan: Con- [15] Z. Zhang, R. Ma, L. Wang, J. Zhang, Novel PMSM Control for Anti-Lock Braking
ceptualization, Methodology, Writing – original draft, Data cura- Considering Transmission Properties of the Electric Vehicle, IEEE Trans. Veh.
tion, Formal analysis, Funding acquisition, Investigation, Project Technol. 67 (11) (Nov. 2018) 10378–10386, https://doi.org/10.1109/
TVT.2018.2866828.
administration, Resources, Software, Supervision, Validation, Visu-
[16] H.P.H. Anh, C.V. Kien, T.T. Huan, P.Q. Khanh, Advanced Speed Control of PMSM
alization, Writing – review & editing. Sivaprasad Athikkal: Con- Motor Using Neural FOC Method, in: 2018 4th International Conference on
ceptualization, Methodology, Writing – original draft, Data Green Technology and Sustainable Development (GTSD), 2018, pp. 696–701,
curation, Formal analysis, Funding acquisition, Investigation, Pro- https://doi.org/10.1109/GTSD.2018.8595688.
[17] H. Celik, T. Yigit, Field-Oriented Control of the PMSM with 2-DOF PI Controller
ject administration, Resources, Software, Supervision, Validation, Tuned by Using PSO, Int. Conf. Artif. Intell. Data Process. (IDAP) 2018 (2018) 1–
Visualization, Writing – review & editing. 4, https://doi.org/10.1109/IDAP.2018.8620902.

11
P. Ramesh, M. Umavathi, C. Bharatiraja et al. Materials Today: Proceedings xxx (xxxx) xxx

[18] Marufuzzaman, M.B.I. Reaz, M.S. Rahman, M.A.M. Ali, ‘‘Hardware prototyping [23] M. Kamalesh, N. Senthilnathan, C. Bharatiraja, Design of a Novel Boomerang
of an intelligent current dq PI controller for FOC PMSM drive,” Int. Conf. Electr. Trajectory for Sliding Mode Controller, Int. J. Control Autom. Syst. 18 (11) (May
Comput. Eng. (ICECE 2010), 2010, pp. 86-88, doi: 10.1109/ 2020) 2917–2928.
ICELCE.2010.5700559. [24] C. Bharatiraja, S. Jeevananthan, R. Latha, V. Mohan, Vector selection approach-
[19] J. Xiong, H. Gu, Research on a Sliding Mode Variable Structure Control FOC of based hexagonal hysteresis space vector current controller for a three-phase
PMSM for Electric Vehicles, in: 2018 IEEE 9th International Conference on diode clamped MLI with capacitor voltage balancing, IET Power Electron. 9 (7)
Software Engineering and Service Science (ICSESS), 2018, pp. 1088–1091, (2016) 1350–1361, https://doi.org/10.1049/iet-pel.2015.0184S.
https://doi.org/10.1109/ICSESS.2018.8663843. [25] C. Bharatiraja, S. Jeevananthan, J.L. Munda, A Timing Correction Algorithm-
[20] P. Kumar, N. Kaushik, S. Dahiya, Simulation of sensor less speed control of Based Extended SVM for Three-Level Neutral-Point-Clamped MLI in Over
PMSM based on FOC method with MRAS adaptive speed estimator, 2013 Modulation Zone, IEEE J. Emerg. Select. Top. Power Electron. 6 (1) (2018) 233–
Annual IEEE. 245, https://doi.org/10.1109/JESTPE.2017.2723518.
[21] C. Bharatiraja, S. Jeevananthan, J.L. Munda, R. Latha, Improved SVPWM vector [26] M. Selvaperumal, D. Kirubakaran, C. Bharatiraja, A hybrid space vector
selection approaches in OVM region to reduce common-mode voltage for modulation for the near zero common mode voltage and common-mode
three-level neutral point clamped inverter, Int. J. Electr. Power Energy Syst. 79 current mitigation in diode clamped multilevel inverter fed induction motor
(1) (Oct. 2016) 285–297. drive, Int. Trans. Electr. Energy Syst. 30 (10) (2020), https://doi.org/10.1002/
[22] S. Balasubramanian, A. Panchanathan, B. Chokkalingam, S. Padmanaban, Z. 2050-7038.12535.
Leonowicz, Module Based Floorplanning Methodology to Satisfy Voltage Island
and Fixed Outline Constraints, Electronics 7 (11) (Nov. 2018) 325.

12

View publication stats