2017 International Conference on Communication, Control, Computing, and Electronics Engineering (ICCCCEE), Khartoum, Sudan
Modelling and Simulation of Electric Drive Vehicle
Based on Space Vector Modulation Technique and
Field Oriented Control Strategy
Saber M. E. Fadul*, Ishak Aris, Norhisam Misron,
Izhal A Halin
Department of Electrical and Electronics Engineering
Universiti Putra Malaysia, Malaysia
sabermefadul@ieee.org
Abstract—Recently, the electric vehicle has emerged as a
powerful platform for mitigating energy crisis and reducing
environmental pollution in the transportation sector. The major
drawbacks of electrically powered automobile that limits its
competitiveness with the internal combustion engine counterpart
are the vehicle driving range and battery energy capacity. Hence,
limited energy storage warrants the need for an effective and
efficient energy utilisation in the overall system. This paper uses
Field Oriented Control algorithm and Space Vector Modulation
technique to enhance and to optimise energy saving at the same
time improve the vehicle induction motor efficiency. A simple
electric vehicle drive with vehicle dynamic and tractive loads for
motor driving load were modelled and simulated using Matlab/
Simulink. Simulation results show that modelled vehicle speed
satisfied the acceleration index for the electric vehicle.
Keywords—Electric Vehicle; FOC; SVM; IM; PWM
I. INTRODUCTION
A change in the lifestyle has created more demand for travel,
especially in the urban areas where the economy continues to
grow. As a result, transportation becomes a major consumer
stakeholder of global energy consumption and greenhouse gas
emissions. Reducing the impact of transportation can be
achieved through enhancement of the energy efficiency,
especially for public and private vehicles.
An onboard-to-wheel is one step of energy conversion to
analyse energy efficiency of the passenger cars [1]. This step of
conversion onboard-to-wheel, the energy is converted by the
driving system to mechanical energy then converted into the
dynamic and potential energy required for the displacement.
However, this energy conversion process causes substantial
energy loss as onboard-to-wheel energy is influenced by vehicle
load resistances such as rolling resistance, aerodynamic drag
resistance, grading resistance, weight, and above that the driving
profiles followed by the vehicle. Two ways to improve the
onboard-to-wheel efficiency are the component level and the
system control level approaches.
For the above issues, the road transportation has been so far
the most popular field of research with the emergence of the
energy crisis. When the share of renewables is extended in the
electricity mixture, a reduction of CO2 emissions caused by the
978-1-5090-1809-3/17/$31.00 ©2017 IEEE
A.K.M. Parvez Iqbal
Department of Mechanical Engineering,
College of Engineering,
Universiti Tenaga Nasional, Malaysia
automobile field will be possible. Thus the usage of sustainable
energy resources becomes a vital factor for the vehicle industry.
Hybrid electric vehicles (HEVs) and Electric Vehicles (EVs)
seem to be promising solutions to decrease the emissions caused
by internal combustion engine vehicles. The main components
of EV are an electrical energy source, electrical motor, power
converters, vehicle interface and control unit supervises power
flow. An accelerator pedal is connected to the controller to
determine how much power from battery pack must be delivered
to the electric motor to overcome the vehicle load in different
conditions.
Modelling can help in understanding the behaviour of the
electric vehicle [2],[3], and a simulation is a key issue all through
different phases of the improvement of this kind of vehicle [4],
which requires modelling of various components and a number
of aspects [5],[6]. Simulation allows reduction of the expense
and length of the design cycle of EVs by testing the
configurations and energy management strategies before
prototype construction begins [7]. It can also be used to verify
the vehicle in different situations through driving cycle methods.
The type of model depends on the objectives of simulation [8],
[9].
This paper, therefore, investigates the designed key features
of EV propelled by three phase induction motor (IM) supplied
from a lithium-ion battery which acts as a power source for the
EV. The purpose is to create a simple EV model with low cost,
achievable requirement, performance and safety. The model can
be used to determine the required energy for the drive train and
speed, torque and power for the system to meet specific drive
cycle. The paper also describes the field oriented control (FOC)
combined with space vector modulation (SVM) technique to
improve the transient and steady state of the EV drive. Instead
of modelling an EV with its entire aspect a simple EV will be
used to achieve a model goals of the electric drive.
The paper is structured as follows: firstly SVM technique
alongside with FOC approach is discussed. Secondly, the model
development parts for the vehicle design is presented, and
thirdly, the electrical drive vehicle model is described and finally
simulation results obtained are shown along with conclusions.
 II. PRINCIPLE OF FILED ORIENTED CONTROL BASED ON
SPACE VECTOR MODULATION
A. Space Vector Modulation Technique
It is a technique for the control of pulse width modulation
(PWM) [10]. This SVM algorithm can decrease the distortions
of the alternating current waveform generated from DC to save
energy [11]. At the same time, it can help in switching loss
reduction compared with standard PWMs such as sine PWM,
Third Harmonics PWM and 60o PWM [12],[13]. The ability to
apply the normal PWM units to execute SVM techniques can
make these techniques more applicable [14]. As shown in Fig.1,
the vectors (V1-V6) divide the plane into six sectors, every sector
is 60 degrees. The reference voltage (Vref) is generated by using
two adjacent voltage vectors (V1-V6) and one null-vectors (V0 or
V7). These quantities V1, V2, Vref and angle θ have to be
determined to realise Space Vector Pulse Width Modulation
(SVPWM).
Fig.1 Principle of space vector modulation (SVM)
Also, the associated time duration T1, T2, T0 of the voltages
V1, V2, Vref should be defined as well as the switching time of
each transistor of the inverter as depicted in Fig.2. For example,
in sector 1, V1 with an interval time T1 and vector V2 has interval
time T2 then vector V0 or V7 for interval time T0. While Vref
varies from zero to the maximum amplitude, Vref_max, before
nonlinear over modulation is reached.
To make sure, the summation of time for these states equals
to the reference vector period [15]. These interval times can be
calculated as follows:
V ref = V n .T1 + V n + 1 .T 2 + V n ull .T 0
T1 = T .m .sin ( 60 − Δθ ) , T2 = T .m .sin ( Δθ ) ,
T0 = T − T1 − T2
Where T is a period, m is the modulation factor (0-1), and ∆θ
is the angle between Vref and Vn.
Fig.2 Illustration of SVM switching period
Figure 3 shows eight possible switching vectors. These
vectors have two groups: a six possible switching voltage
vectors (V1-V6) and two possible null vectors (V0 &V7) for a
three-leg inverter utilizing SVM.
Fig.3 States of the inverter switches using SVM
B. Field Oriented Control Strategy
FOC or Vector control converts the measured motor
currents/voltages into two components the field current, and the
torque one. FOC has two types of vector control methods, the
direct and indirect methods. The direct vector control works with
measured flux in stator-rotor gap, while indirect method
estimates the magnetic flux from the applied voltages and
resultant currents through the model of the motor [16],[17]. Due
to the deviation in the air gap flux linkage values, scalar control
delivers poor dynamic response compared to FOC [18]. Also,
direct torque control produces ripple in flux and torque besides
it has a slow transient response to the step change in the torque
during start mode [19]. Therefore, FOC with SVM can provide
the optimal way to control the traction motor where torque
control is required. In EV application, FOC controls not only the
torque but also decreases the electric drive losses.
In FOC of IM, instead of using three current sensors only
two of them can be used to measure two of the three phase
instantaneous currents, ia and ib as shown in Fig.4, while the third
current is calculated as a summation of the three phase currents
ia, ib, and ic equal to zero. Using a Clarke transformation, these
three phase currents or voltages are converted into a variant two
phase system iα and iβ, (a,b,c) → (α, β).
(1)
isα = ia , isβ = (1/ 3) ( ia + 2ib ) (3)
(2) Then a two coordinate time invariant system is outputted by
using the Park transformation (α, β) → (d, q). The two-axis
currents id and iq is rotated to align with the rotor flux using rotor
flux position θ.
isq = isα .sinθ + is β .cosθ (4)
isd = isα .cosθ + is β .sinθ (5)
The commended currents id, iq and their measured values
each generate error signals. The id reference controls rotor
magnetizing flux while the iq reference manages induction motor
torque output. These error signals are amplified by the
proportional-integral (PI) controllers to create correction
voltages Vd and Vq. Using the inverse park transformation with 3 P Lm e e
a rotor flux position θ, the Vd and Vq are replaced back into the Te =
2 2 Lr
( e
iqsψ dr − ids e
ψ qr ) (11)
stationary frame to get quadrature voltage values, Vα and Vβ.
These quantities Vα and Vβ are then mathematically translated Where P is number of poles of IM, Lm is mutual inductance
back into three voltages value Va, Vb, and Vc using inverse between the stator and the rotor, Lr is the rotor inductance, iqs,
Clarke transformation. This transformation process determines iqr, ids and idr are q-axis stator and rotor currents and d-axis stator
the new PWM duty-cycle. The new estimated angle guides the and rotor currents respectively, is d-axis rotor flux and
FOC algorithm in placing the next voltage vector. is q-axis rotor flux.
In a VC drive and 0 . So, to simplify
equation (11) and if .

3 P L2m e e
Te = iqs ids (12)
2 2 Lr
Equation (12) represents the response of vector controlled
IM drive. is torque component and is flux component. If
the IM requires high speed, the direct current component
through the IM has to be controlled. To have a good torque
response, the flux component must be constant and the torque
component must be varied. That means idref is constant while iqref
is the desired variable torque signal as illustrated in Fig.4.

β -axis

ωe
b

is
ax
q-
Fig.4 Block diagram of a simplified indirect FOC ax

d-
is

bs
qs
In Fig.5, the reference frame dq-axis is rotating at speed ωe axis
iqs ωr rotor
iβ

ds
while the rotor which has ar, br and cr windings, rotates at speed qr
ids
θ
ωr. The hypothesis windings qr* and dr* are rotating at a speed *

*
dr
of the rotor flux vector because the reference frame is attached br ar θr a
with rotor flux vector. iα as α-axis

Therefore, FOC for an induction machine can be performed cr

based totally on a motor model in dq-axis fixed to the stator
stationary reference frame.
cs

dψ ds
vds = rs ids + − ωψ qs (6)
dt c
dψ qs Fig.5 The IM three phase with rotating and fixed reference frames
vqs = rs iqs + − ωψ ds (7)
dt
III. VEHICLE LOAD MODEL
dψ dr The analysis is the first step to provide the required size of
vdr = 0 = rr idr + − (ωe − ωr )ψ qr (8)
dt the induction motor and battery pack source. Therefore, the
vehicle motion can be completely determined by analysing the
dψ qr forces acting on it in the direction of motion. When the vehicle
vqr = 0 = rr iqr + − (ωe − ωr )ψ dr (9)
dt has mass mv and is moving at speed v uphill with an angle θ, it
faces a resistive force that tries to retard its motion as shown in
Since the rotor current cannot be accessed, rotor current iqr Fig.6.
must be replaced with stator current and the slip can expressed
as: These resistive forces are: rolling resistance, aerodynamic
drag resistance, and grading resistance. The equation of motion
e e
rr iqs L iqs along electric vehicle’s longitudinal axis is expressed by the
ωsl = ωe − ωr = Lm e = m e (10) following equation [20]:
Lr ψ dr Z r ψ dr
Ft = Frr + Fad + Fhc + Fac (13)
Zr is the motor time constant. The slip speed will be useful
when controlling the IM with FOC. Where Ft, Frr, Fae, Fhc, Fac and Fwa represent the total tractive
force, the road rolling resistance, aerodynamic resistance,
The equation of electromagnetic torque in the form of stator
current and rotor flux is given as follow in (11):
grading resistance, and acceleration resistance, respectively. The motor and Vmax is the maximum vehicle speed, the vehicle base
detail definitions are given in the following section. speed Vb in m/s and the motor angular velocity wmax, in rad/s,
can be written as:
A. Rolling resistance
The Frr is generated through the friction of the vehicle tire at ( )
Vb = Vmax / x = 2π nm,base rd / 60 N g (18)
the roadway contact surface and hardly relies on vehicle speed,
tire pressure and type and road surface characteristic. wmax = 2π nm , max / 60 (19)

Where x is speed factor which is defined as the ratio of the

maximum speed to the base speed. This speed factor x is such
that for IM is around 4, for permanent magnet synchronous
machine is about 2 to 4, for switch reluctance machine is about
6 [20].
TABLE I. DYNAMIC PARAMETERS
Parameter Value Unit
Mass of the vehicle (Mv) 1200 kg
Coefficient of aerodynamic drag (Cad) 0.26 -
Tire rolling resistance coefficient (Crr) 0.010 -
Wheel radius (rd) 0.250 m
Front area (Af) 2.200 m2
Density of air (pa) 1.205 kg/m3
Effective rolling radius 0.25 m
Gear ratio (Ng) 7.2 -
Fig.6 Force acting on a moving vehicle Gear efficiency (ηg) 100 %
Horizontal distance from CG to front axle (hdf) 1.4 m
Horizontal distance from CG to rear axle (hdr) 1.4 m
When Crr is a tire rolling resistance coefficient, the Frr can be CG height from ground (h) 0.5 m
written as:
Frr = C rr M v . g .cos (α ) (14) A traction force Tt can written based on wheel torque Tw and
wheel radius rd, as:
B. Aerodynamic Drag Resistance
Ft = Tw / rd (20)
The Fad occurs while air flows over the vehicle. The force is
a function of the frontal area, the shape, etc. This drag resistance Also the traction force that can act on the vehicle can be
can be written as: written as:

Fad = (1 / 2 ) pa .cad . A f ( vv + vw )
2
(15) ( )
Ft = 60 N gη g Pm / 2π nm , max rd (21)

Where ρa is air density; Cad aerodynamic drag coefficient; Af The Newton's second law of motion and principles of vehicle
is vehicle frontal area; vv is vehicle speed, and vw is headwind design can be applied therefore, the vehicle acceleration can be
velocity. expressed as:
C. Grading Resistance dv
a= = ( Ft − ( Frr + Fad + Fhc + Fac ) ) / M v (22)
When the vehicle weight component acts along the slope, the dt
grading is a required force to drive the vehicle up a slope. This
grading resistance force Fhc opposes the forward motion uphill Where v is the vehicle speed. The balanced power equation
and helps downward motion. It can be expressed as: can be written as:

Fhc = M v .g .sin (α ) (16) Pt (t ) = ( Frr + Fad + Fhc + Fac ) .v (t ) (23)

D. Acceleration Resistance IV. ELECTRIC MOTOR AND BATTERY FOR EV

The force needed when the speed of the vehicle is varied to
deliver the linear acceleration of the vehicle. This force is
additionally required to accelerate uphill or decelerate downhill.
Fac = M v .a (17)
Where a: vehicle acceleration.
Table 1 was used to calculate the total force which leads to
determining the motor size and power needed to propel the EV.
To design the parameters of the vehicle and electric motor
velocities and speeds, when nm,max is the maximum rpm of the
A. Electric Motor Drive
Due to their torque-speed characteristics, electric motors are
capable of delivering a high starting torque. The selected drive
motor rated speed, and maximum speed have to conform to the
requirements of the torque-speed characteristic of the motor.
When the vehicle starts off, the motor speed is low and works in
the constant torque state, as long as speed is higher than the rated
speed it works in the constant power state.
For its reliable operation, ruggedness, low maintenance
requirements, and high efficiency, IM motor drives will continue
to be employed in EVs. However, to achieve precise control to depicted in Fig. 10 and Fig.11, respectively. While the required
the induction motor is not easy due to its complicated nonlinear torque follows the reference torque. This torque is around 120
system and the electric rotor variables are not measured directly. Nm.
So controlling an induction motor becomes tedious task and a
critical issue, especially for the EV applications where both fast
transient responses and excellent steady state speed performance
are required.
B. Battery Choice
The first step to design and size a battery is to understand the
battery parameters [21], [22]. For EV propulsion, the battery
must produce voltage and current according to the motor Fig.8 Pedal accelerator steps signal applied to control unit
requirements and peak power requirement of the vehicle in the
typical driving cycle [23], [24]. These parameters such as energy
density and specific energy, maximum sustainable discharge
rate, life cycle, operating temperature, and capacity and state of
charge (SOC) can be monitored to achieve a long life of battery
operating conditions.
V. SIMULATION RESULTS
The developed model was done in Matlab-Simulink. Figure
7 shows the EV model which consists of five blocks: a three Fig.9 Battery SOC (%) during 16 seconds
phase induction motor, battery, a converter, vehicle load, and

Fig.7 Matlab/Simulink model for electric vehicle drive train system

control unit.
The motor which propels the EV load is supplied by a Between t = 1 - 4 s, the vehicle and motor speed reached 40
lithium-ion battery which acts as the main power source for the km/h and 3000 rpm respectively, while torque continued to
EV. A converter inverts the DC voltage from the battery to three nearly 200 Nm during this motor based speed. At t = 4 – 15 s,
phase AC fed to the motor. The control unit block has a FOC the measured torque shown in Fig. 12 is around 50 Nm which
algorithm, as uses SVM technique to control the gates of the follows the reference torque with clear difference due to vehicle
switches for the converter which control the motor dynamics. A parameters such as vehicle weight, vehicle tyre size and
torque actuator from Simulink is used to couple the torque transmission systems. While the motor power is around 22 kW
developed by the motor shaft with the vehicle gearbox splitting as shown in Fig.13. During the same time intervals, battery
to the wheels through the differential. power responds exactly to the required motor power. Also, the
battery current which responds to the vehicle load current as
The motor torque for the EV can be determined if the vehicle shown in Fig.14 is around 85 A.
load and acceleration are known. Therefore, pedal accelerator
signal reference, shown in Fig.8, is fed for 16 seconds to the
control unit block that determines the torque reference.
Furthermore, vehicle parameters can be modified for electric
motor to produce the required torque.
As the battery SOC depletes with 0.5% over the 16-second
acceleration as depicted in Fig.9, the motor power and torque
were responding to overcome the vehicle load at this interval.
As the acceleration starting from initial t = 0.0s to t = 1s, the
vehicle and motor speed reached 10 km/h and 900rpm as Fig.10 Vehicle speed in (Km/h) during 16 seconds
 When the vehicle slows down or commanded acceleration is
negative both motor and vehicle speeds decreased respectively.
The simulation result of acceleration time when the speed is 0-
60 km/h during 10 sec, can extremely meet the acceleration
index for EV.
Fig.11 Motor speed in (rpm) during 16 seconds
Fig.12 Motor drive torque in (Nm) during 16 seconds
Fig.13 Motor and battery drive power in (kW) during 16 seconds
Fig.14 EV battery current in (A) during 16 seconds
VI. CONCLUSIONS
This paper modelled and simulated an electric vehicle and
induction motor controlled using FOC and SVM technique in
Matlab/Simulink environment. It also simulated and analysed
the vehicle dynamic output speed and torque which show the
vehicle speed reached 60 km/h and the motor torque reached
around 50 Nm in less than 10 seconds. This speed satisfies the
acceleration index for the EV. The simulation results also show
that the battery state of charge depletes with 0.5% during the
discharge state, with an average battery power of 22 kW and
current of 85A. Simulation based on the acceleration signal
shows that it is simple to determine how much battery capacity
required by the EV. The EV performance and speed are
determined by the produced induction motor torque which is
controlled by the FOC.
ACKNOWLEDGMENT
This work was financially supported by Department of
Electrical and Electronics Engineering, Faculty of Engineering,
Universiti Putra Malaysia.
REFERENCES
[1] A. Sciarretta, G. De Nunzio, and L. L. Ojeda, “Optimal Ecodriving
Control: Energy-Efficient Driving of Road Vehicles as an Optimal
Control Problem,” IEEE Control Syst., vol. 35, no. 5, pp. 71–90, 2015.
[2] K. Chen, a. Bouscayrol, a. Berthon, P. Delarue, D. Hissel, and R. Trigui,
“Global modeling of different vehicles,” IEEE Veh. Technol. Mag., vol.
4, no. 2, pp. 80–89, 2009.
[3] C. C. Chan, A. Bouscayrol, and K. Chen, “Electric, hybrid, and fuel-cell
vehicles: Architectures and modeling,” IEEE Trans. Veh. Technol., vol.
59, no. 2, pp. 589–598, Feb. 2010.
[4] M. Amrhein and P. T. Krein, “Dynamic simulation for analysis of hybrid
electric vehicle system and subsystem interactions, including power
electronics,” Veh. Technol. IEEE Trans., vol. 54, no. 3, pp. 825–836,
2005.
[5] F. L. Mapelli, D. Tarsitano, F. L. Mapelli, and D. Tarsitano, “Modeling
of Full Electric and Hybrid Electric Vehicles Modeling of Full Electric
and Hybrid Electric Vehicles,” in new generation of electric vehicles,
2012, p. 384.
[6] D. W. Gao, C. Mi, and A. Emadi, “Modeling and Simulation of Electric
and Hybrid Vehicles,” Proc. IEEE, vol. 95, no. 4, pp. 729–745, Apr.
2007.
[7] T. A. T. Mohd, M. K. Hassan, and W. A. Aziz, “Mathematical Modeling
and Simulation of an Electric Vehicle,” J. Mech. Eng. Sci., vol. 8, no.
June, pp. 1312–1321, 2015.
[8] M. Juraj, “Mathematical modeling of the electric drive train of the sports
car,” pp. 1–10, 2012.
[9] T. A. T. Mohd, M. K. Hassan, I. Aris, A. C. Soh, B. S. K. K. Ibrahim,
and M. K. Hat, “Simulation Based Study of Electric Vehicle
Parameters,” vol. 10, no. 19, pp. 8541–8546, 2015.
[10] E. R. C. Da Silva, E. C. Dos Santos, and C. B. Jacobina, “Pulsewidth
modulation strategies,” IEEE Ind. Electron. Mag., vol. 5, no. 2, pp. 37–
45, 2011.
[11] A. Mehrizi-Sani and S. Filizadeh, “An optimized space vector
modulation sequence for improved harmonic performance,” IEEE Trans.
Ind. Electron., vol. 56, no. 8, pp. 2894–2903, 2009.
[12] J. Hobraiche, J. P. Vilain, P. Macret, and N. Patin, “A new PWM strategy
to reduce the inverter input current ripples,” IEEE Trans. Power
Electron., vol. 24, no. 1, pp. 172–180, 2009.
[13] G. Nisha, S. Ushakumari, and Z. Lakaparampil, “Harmonic elimination
of space vector modulated three phase inverter,” Proc. …, vol. II, 2012.
[14] K. V. Kumar, P. A. Michael, J. P. John, and S. S. Kumar, “Simulation
and Comparison of Spwm and Svpwm Control for Three Phase
Inverter,” ARPN J. Eng. Appl. Sci., vol. 5, no. 7, pp. 61–74, 2010.
[15] a Iqbal and E. Levi, “Space vector modulation schemes for a five-phase
voltage source inverter,” Power Electron. Appl. 2005 Eur. Conf., p. 12
pp.-pp.P.12, 2005.
[16] D. M. Reed and H. F. Hofmann, “Direct field-oriented control of an
induction machine using an adaptive rotor resistance estimator,” in 2010
IEEE Energy Conversion Congress and Exposition, ECCE 2010 -
Proceedings, 2010, pp. 1158–1165.
[17] P. Shrawane, “Indirect Field - Oriented Control of induction motor,” in
12th International Power Electronics Congress ({CIEP)}, 2010, pp.
102–105.
[18] G. Kohlrusz and D. Fodor, “Comparison of Scalar and Vector Control
Strategies of Induction Motors,” Hungarian J. Ind. Chem. Veszprém, vol.
39, no. 2, pp. 265–270, 2011.
[19] G. Sheng-wei and C. Yan, “Research on Torque Ripple Minimization
 Strategy for Direct Torque Control of Induction Motors,” Comput. Appl.
Syst. Model. (ICCASM), 2010 Int. Conf., vol. 9, no. Iccasm, pp. 278–281,
2010.
[20] M. Eshani, Y. Gao, S. Gay, and A. Emadi, Modern electric, hybrid
electric and fuel cell vehicles 2nd. Edition. 2010.
[21] A. Affanni, A. Bellini, G. Franceschini, P. Guglielmi, and C. Tassoni,
“Battery choice and management for new-generation electric vehicles,”
IEEE Trans. Ind. Electron., vol. 52, no. 5, pp. 1343–1349, 2005.
[22] J. Schmalstieg, S. Käbitz, M. Ecker, and D. U. Sauer, “From Accelerated
Aging Tests to a Lifetime Prediction Model : Analyzing Lithium-Ion
Batteries,” 27th World Electr. Veh. Symp. Exhib. EVS 2014; Barcelona;
Spain, pp. 1–12, 2014.
[23] R. C. Kroeze and P. T. Krein, “Electrical battery model for use in
dynamic electric vehicle simulations,” 2008 IEEE Power Electron. Spec.
Conf., pp. 1336–1342, 2008.
[24] O. Tremblay and L. A. Dessaint, “Experimental validation of a battery
dynamic model for EV applications,” World Electr. Veh. J., vol. 3, no. 1,
2009.