TELKOM
NIKA Indonesia
n
Journal of
Electrical En
gineering
Vol. 12, No. 9, September
2014, pp. 66
0
2
~ 661
1
DOI: 10.115
9
1
/telkomni
ka.
v
12i9.614
4
6602
Re
cei
v
ed Ap
ril 26, 2014; Revi
sed
Jun
e
23, 2014; Accepted July 2
0
,
2014
Determination of Powe
r Consumption of Electric
Vehicle for Assorted Pavemen
t Gradient Using
MATLAB/SIMULINK
Md. Ruhul Amin*, Rajib
Bara
n Ro
y
Dep
a
rtment of Electrical
and Electron
ic
Engi
neer
ing,
Univers
i
t
y
of Information T
e
chnol
og
y a
nd Sci
ences (UIT
S)
Dhak
a-12
12, B
ang
lad
e
sh
*Corres
p
o
n
id
n
g
author, e-ma
i
l
: ruhul.am
i
n@
uits.edu.b
d
A
b
st
r
a
ct
Due to l
e
ss po
llutio
n
, sile
nce
and h
i
g
h
effici
ency el
ectric v
ehicl
es are fas
c
inati
ng to the
citi
z
e
n
s
w
i
dely. T
her
e
are n
u
m
b
e
r of
physic
a
l
and
el
ectrical
as
p
e
cts are to
be
co
n
s
ider
ed
duri
ng
vehicl
e
mode
li
ng
.
W
hen e
l
ectric
vehicl
e ru
ns
on
the co
ncrete s
u
rface ro
lli
ng r
e
sistanc
e res
i
s
t
the veh
i
cl
e tir
e
to
mov
e
fast. O
n
the oth
e
r h
a
n
d
w
hen th
e verti
c
al a
n
g
l
e
of ve
hicle
dr
iv
e is
i
n
creas
ed
or d
e
creas
ed
by s
o
mew
hat d
egr
ee
s
the forw
ard fo
rce is als
o
inc
r
ease
d
or d
e
c
r
ease
d
resp
ec
tively. In this
pap
er, a typic
a
l el
ectric ve
hi
cle
dyna
mic is mo
del
ed to mech
anic
a
l par
a
m
et
er after conv
er
sion its mech
a
n
ical tor
ques, the el
ectrical p
o
w
er
is obtai
ne
d by
requ
ired
mech
anic
a
l torq
ue a
nd a
ngu
lar ve
l
o
city of prop
el
as w
e
ll as w
h
e
e
l. T
he per
man
ent
m
a
gnet DC
motor is
used to simu
late the
vehicle speed
and t
o
rque
. MATLAB created surface or r
oad
grad
ients ar
e
used to
inv
e
sti
gate torq
ue
in
different
u
p
s
and
dow
ns at
different a
n
g
l
e.
After analy
z
i
n
g
battery error t
he pr
oporti
on
a
l
inte
gral c
ontr
o
ller
ga
in
is
d
e
rive
d for ach
i
evin
g des
ire
d
output fro
m
th
e
vehicl
e.
Ke
y
w
ords
: sp
eed, gra
d
i
ent, torqu
e
, motor, c
ontrol
l
er, MAT
L
AB, force
Copy
right
©
2014 In
stitu
t
e o
f
Ad
van
ced
En
g
i
n
eerin
g and
Scien
ce. All
rig
h
t
s reser
ve
d
.
1. Introduc
tion
The oil re
se
rves are not ine
x
haustibl
e, it is
theref
ore n
e
ce
ssary to find additio
nal
energy
sou
r
ces. T
h
is
chan
ge
ha
s
also
b
e
com
e
necessa
ry
due
to
the
fact
that
the
burni
ng
of
oil
gives
a
n
e
g
a
tive enviro
n
m
ental imp
a
ct [1]. Nowa
da
ys
mo
st
cars are usi
ng
inte
rnal com
b
u
s
tion
engin
e
s but
manufa
c
turers are
con
s
id
ering u
s
in
g di
ferent kin
d
s
of energy so
ruce that re
du
ce
the pollution
[2]. The
energy
carrie
r
which
seem
s
be
the
most
usable
is
the
electrical
energy [3]. The ele
c
tri
c
v
ehicl
e de
sig
n
i
ng is co
mpl
e
x pro
c
e
s
s i
n
clu
d
ing
ene
rgy di
stributi
on,
vehicle t
r
an
smissi
on, road
cha
r
a
c
teri
sti
c
s and
so on
[4]. In this p
aper,
s
se
rie
s
of
spe
ed a
n
d
slop
e of angl
e are fo
rmed
in MATLAB workpla
c
e th
e
n
the car
sp
eci
f
ication
s
are assume
d to find
the req
u
ire
d
torqu
e
in ea
ch up an
d do
wn [5]. There
are
several term
s are con
c
erned
with t
he
mech
ani
cs
which
are
use
d
in MATLA
B
to state
th
e variabl
es
durin
g the d
e
sig
n
ing
electric
vehicle. After that wheel t
o
rqu
e
i
s
con
v
erted
to
mo
tor torque. In
this a
r
ticl
e t
he Pe
rman
e
n
t
Magnet
DC (PMDC) moto
r is u
s
ed to
simulate th
e torque
-spe
ed
cha
r
a
c
teri
stics of the m
o
tor.
The PMDC motor ha
s in
verse
cha
r
a
c
teristics bet
ween the torq
ue and spee
d. In this point o
f
view when th
e vehicle i
s
d
r
iven up i
n
two o
r
thre
e
d
egre
e
with
h
o
rizontal axi
s
it requires l
a
rge
torque
wh
ere
less spee
d i
s
cond
ucte
d
[6]. Subs
eq
u
ently motor e
l
ectri
c
al p
o
wer requi
rem
e
nts
are fulfilled
b
y
adding
DC battery sto
r
ed in v
ehi
cle
[7]. During t
he large u
p
s and do
wn
s
of
vehicle move
ment there a
r
e differe
nt requireme
nt of current to meet the de
mand of po
wer
need
ed to
achieve torque
to re
ach t
he
p
eak of the val
l
ey. In this
ca
se th
e PI cont
rolle
r is u
s
ed
to
monitor volta
ge and
curre
n
t level with the dema
nd [8
].
2. Electric Vehicle Driv
e
Train Oper
ation
In a typi
cal
g
a
soli
ne
po
we
red
vehi
cle th
e ga
s tan
k
i
s
not a
pa
rt of t
he d
e
si
gn
mo
del [9].
There is no
way to put the gasoline to
the tank
wh
en vehicl
e on
ly consume
s
the gasoline f
r
om
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Determ
inatio
n of Power
Consum
ption of Electr
ic Vehi
cle for Asso
rt
ed… (M
d. Ru
hul Am
in)
6603
the tank [10]. A standard shift from internal com
b
u
s
tion engi
ne
s to electri
c
vehi
cle
s
is that in
an
electri
c
vehi
cl
e the battery is part of the
d
r
ive train a
s
shown belo
w
in Figure 1.
Figure 1. Electri
c
Vehicl
e Driv
e Train wi
th vehicle inte
rface
The d
r
ive tra
i
n con
s
ume
s
ene
rgy from
the
b
a
ttery
durin
g m
o
tori
ng a
c
tion. T
he d
r
ive
train can al
so
fed ch
arg
e
b
a
ck to the b
a
ttery
when th
e motor i
s
o
p
e
rated
as
a g
enerator
duri
ng
rege
ne
rator
action [11]. This can be
occurre
d
duri
ng bra
k
in
g o
r
duri
ng the vehicle is b
e
i
ng
powere
d
by
an Internal
Combu
s
tion E
ngine
(ICE
).
In the di
agra
m
, the batte
ri
es
are
ge
nerally
con
s
tru
c
ted
o
f
Lithium Ion
cell
s an
d
sup
p
lies
300
volts (+ve
) an
d h
a
ve high
current capa
city to
the po
we
r el
ectro
n
ics. A
battery controller
m
onito
rs key
b
a
ttery
paramete
r
s and co
ntrol
s
the
battery pa
ck. The p
o
wer
electroni
cs u
n
it invert
s th
e DC b
a
ttery voltage into
three
-
ph
ase
AC
voltage at the
app
rop
r
iate f
r
equ
en
cy an
d
voltage fo
r the moto
r to
meet the
req
u
ired
spee
d
and
torque [12]. The AC moto
r is typically a high effi
cie
n
cy AC Indu
ction Motor (I
M) or Perm
a
nent
Magnet Syn
c
hron
ou
s Moto
r (PMSM
)
. Th
ese
motors
supply eithe
r
a
c
celeration to
rque
or
braki
ng
torque fo
r b
o
th dire
ction
s
o
f
rotation. When t
he ve
hicle’s b
r
a
k
e
s
are applie
d the
motor o
perate
s
in re
gen
erati
on mo
de th
us
reversin
g
both the
current di
re
ction an
d torq
ue di
re
ction.
The
reverse
d
torq
ue dire
ction
offers vehi
cle
braki
ng torq
ue while h
e
l
p
ing to re
cha
r
ge the batte
ry.
The Vehi
cle
Interface
correspon
ds
with the
Battery Controll
er and M
o
to
r Cont
rolle
r, and
provide
s
an i
n
terface with
the vehicl
e
-
le
vel control
s
a
nd se
nsors [13].
2. Modeling Process a
n
d
Ke
y
Equatio
n
s
The m
odel
de
velopment
proce
s
s
con
s
i
s
ts of
1)
Determining
ho
w t
he m
odel
will
be u
s
e
d
,
2) Identifying
the key eq
u
a
tions i
n
clu
d
i
ng me
cha
n
ical term
s, parameters,
an
d
assumptio
n
s, 3)
Building a
nd
refining th
e m
odel, and th
e
n
4) T
he a
c
tu
al model
appl
ication a
nd e
v
aluation. T
h
e
model
can b
e
use
d
to evaluate the e
nergy fl
o
w
of a DC m
o
tor drive train from mechani
cal
para
m
eters,
and to d
e
termine the a
b
ili
ty of the
system to meet
specif
i
c
drive cycle sp
eed and
torque
req
u
irements. Th
e major
comp
o
nents of the
model a
r
e inp
u
t road torqu
e
at different road
gradi
ent
con
d
i
tion incl
udin
g
real
time ai
r
drag
resi
st
an
ce
of environ
ment, frontal
area
of vehi
cl
e
,
rolling resi
sta
n
ce of vehicl
e
even air den
sity, dem
and road
spe
ed, motor mod
e
l, motor co
ntroll
er
model, battery model, and
PI controller. A block dia
g
ram of the model is p
r
e
s
ente
d
belo
w
in
Figure 2. In the model the
required ro
a
d
s
pee
d and
road gradie
n
ts are in
put
s and the m
a
jor
model
blo
c
ks are
the
req
u
ired
forwa
r
d
force, wheel
torqu
e
a
s
well a
s
moto
r
torque,
moto
r
model,
contro
ller mo
del, ba
ttery model,
PI controll
er
model, an
d fe
edba
ck from t
he PI Co
ntroll
er
to the main p
o
we
r co
ntroll
er. The feed
b
a
ck incl
ude
s
a one-sa
mple
delay with a
n
initial con
d
ition
to prevent an
algeb
rai
c
loo
p
in the Simulink mod
e
l.
3.1. Vehicle Kinetic
s
A simplifie
d
model
of the
roa
d
vehi
cle
dynami
c
s ca
n be
u
s
ed
to
estimate
the
tra
c
tive
requi
rem
ent of the vehicle
drive-trai
n, from whi
c
h the
individual co
mpone
nt spe
c
ificatio
ns
ca
n be
rated with-re
gard
-
to
thei
r pea
k
and co
n
t
inuou
s
dutie
s.
As it is see
n
in equ
ation
belo
w
the mo
tor
force
req
u
ire
d
is dire
ctly prop
ortio
nal to the we
ight
of the vehicle, front
al are
a
of vehicle and
squ
a
re
p
r
op
o
r
tional t
o
the
linea
r velo
cit
y
of vehicl
e.
From
the fig
u
r
e
2
whe
n
th
e si
ne
angl
e
is
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 9, September 20
14: 66
02 – 661
1
6604
increa
sed
the
force i
s
i
n
creased.
Con
s
i
derin
g th
e
a
b
o
ve a
s
pe
cts t
he
req
u
ire
d
f
o
rce
of vehi
cle
can b
e
writte
n as bel
ow:
F
F
m
g
s
i
n
θ
F
m
(
1
)
For rolling resistan
ce,
F
k
mg
cos
θ
(
2
)
Aerodyn
a
mic resi
stan
ce
o
r
drag
fo
rce,
F
ρ
C
A
v
(
3
)
And the tran
sient force
req
u
ired a
c
cele
rating or retarding the vehi
cle,
Whe
r
e:
k
The
rolling resi
stance
coeffi
cient which includes tire loss
and is approxi
m
ated to be
indep
ende
nt of spee
d and
prop
ortio
nal to the vehicle
norm
a
l rea
c
ti
on force
m
The vehicle
and paylo
ad
mass
θ
The roa
d
gra
d
ient
g
The gravitati
onal con
s
tant
ρ
The den
sity of air
C
The dra
g
force coefficie
n
t
A
The vehicle f
r
ontal a
r
ea a
nd
v
The vehicle li
near velo
city
Figure 2. Force
s
Acting on
Vehicle
3.2. Wheel a
nd Trac
tion Machine Tor
ques
Determinatio
n the fo
rces active
upon
t
he vehi
cle
the road
wheel to
rqu
e
can
b
e
cal
c
ulate
d
fro
m
the equatio
n of motion merely.
T
J
ω
d
r
F
(
4
)
Whe
r
e
J
,
ω
and
are the whe
e
l inertia
,
angular ve
locity and mean radiu
s
respe
c
tively and
is a factor p
r
o
portio
n
ing torque
distrib
u
tion o
n
the re
ar ax
le. By way of
example fo
r a
dire
ct re
ar
wheel d
r
ive scenari
o
, it
is a
s
sumed th
at there i
s
a
n
eq
ual share of the
requi
re
d tract
i
ve force bet
wee
n
each wheel drive ma
chin
e (i.e.
0
.
5
). For an on-boa
rd drive
machi
ne o
p
tion, a gea
r st
age is i
n
cl
ud
ed in the d
r
iv
e-trai
n, thus t
he output
to
rque of the tra
c
tion
machi
ne is related to the road whe
e
l torque by the total transmissi
on gea
r
, transmissio
n
ef
f
i
cien
cy
η
an
d the ma
chin
e rotor i
nerti
a
J
. Incorporating these into
the equatio
n
of motion
yields a ge
ne
ral expressio
n
for tractio
n
machi
ne torq
ue:
T
J
ω
η
T
(
5
)
Expressin
g
the wheel a
n
d
tractio
n
ma
chin
e an
gula
r
velocitie
s
in
terms
of the
vehicle
linear velo
city yields.
Whe
e
l ang
ula
r
velocity in terms of radiu
s
of wheel,
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Determ
inatio
n of Power
Consum
ption of Electr
ic Vehi
cle for Asso
rt
ed… (M
d. Ru
hul Am
in)
6605
ω
(
6
)
Motor angular veloc
i
ty in terms
of
total gear differential ratio (
),
ω
n
(
7
)
From
whi
c
h t
he ma
chin
e torqu
e
eq
uati
on c
an b
e
e
x
presse
d in t
e
rm
s of the
vehicle
linear velo
city by substitutin
g
Equation (1
), (2), (6
) an
d (7) into Equ
a
tion (3
).
T
η
η
η
k
cos
θ
s
i
n
θ
mg
ρ
C
A
v
(8)
Mech
ani
cal p
o
we
r is torqu
e
multiplied b
y
mechani
cal
spe
ed:
P
T
ω
(
9
)
As
noted earlier, Battery Elec
tric
Vehic
l
es
(BEV) and Hybrid Elec
tric
Vehic
l
es
(HEV)
freque
ntly use spe
c
ial, hi
g
h
efficie
n
cy
Perm
an
ent M
agnet Syn
c
h
r
onou
s M
o
tors
(PMSM). T
h
is
type of motor may be refe
rred to a
s
a
brushl
ess
DC
motor b
e
cau
s
e it run
s
from
DC voltag
e
but
doe
s n
o
t hav
e b
r
u
s
he
s. P
M
SM moto
rs actu
ally u
s
e
AC volta
ge t
hat is suppli
e
d by the
Mot
o
r
Controlle
r. The motor con
t
roller inve
rts
the DC
volta
ge to produ
ce an AC voltage at the proper
voltage and freque
ncy. Th
e motor volta
ge is freq
uent
ly a 10-20K
Hz Pulse
Widt
h Modulate
d
AC
voltage whe
r
e the voltage and freque
n
c
y are adju
s
t
ed to provide
the prope
r motor sp
eed
and
magneti
c
fiel
d values. A
DC p
e
rm
ane
nt magnet m
o
tor was
used in the
si
mulation mo
del
pre
s
ente
d
bel
ow for inve
sti
gating the ch
ara
c
teri
st
ics
of vehicle d
r
i
v
e syst
em. T
h
is type of motor
is not appropriate for BEV or HEV i
n
appli
c
atio
ns practi
cally having
we
ight and effici
ency
con
s
id
eratio
n
s
. The moto
r
model in
clud
es some te
rms an
d para
m
eters for po
wer l
o
ss an
d time
lag
while
oth
e
r te
rm
s
were omitted
fro
m
the
model.
The
mod
e
l a
c
count
s fo
r p
o
we
r lo
ss i
n
t
h
e
windi
ng
re
sist
ance a
nd tim
e
lag
du
e to t
he e
nergy
st
orag
e in
the
magneti
c
fiel
d of the
wi
ndi
ng
indu
ctan
ce. There i
s
no fiel
d power lo
ss
bec
au
se it is a perm
ane
nt magnet field.
The develo
p
e
d
torque
T
is proportio
nal to the arm
a
ture
curre
n
t
I
,
T
N
m
K
I
(
1
0
)
Whe
r
e
moto
r
phy
sical
consta
nt
is d
epen
ds on physi
cal
p
a
rameter of motor
c
o
ns
truc
tion.
On the other hand the developed voltage
V
is propo
rtional to the armature sp
eed
W
during the g
e
nerato
r
a
c
tion
.
V
V
olt
W
r
ad/sec/K
(
1
1
)
Motor armatu
re input or te
rminal voltage
is equal to the sum of de
veloped volta
ge plus
resi
stan
ce
an
d indu
ctan
ce
voltage drop
s. The
g
ene
ral equatio
n o
f
voltage at high sid
e
of th
e
motor controll
er ca
n be writ
ten as:
V
Vo
l
t
I
Amp
R
Ohm
L
Henry
A/s
V
V
olt
(
1
2
)
3.3. Motor Controller
The moto
r
controlle
r i
s
a
s
sumed t
o
b
e
an id
eal
co
ntrolle
r with
no po
we
r lo
ss an
d no
time lag.
Th
e controller simply rai
s
e
s
t
he b
a
ttery vo
ltage to
meet
the
highe
r v
o
ltage
nee
ds of
the motor. The dimen
s
io
nless co
nsta
nt gain or
K ratio of the input and output voltages i
s
determi
ned i
n
ord
e
r to m
e
et the moto
r’
s ne
ed
s. The
sam
e
K ratio is u
s
e
d
to a
d
just the
current
so that input
and outp
u
t po
wer valu
es a
r
e equal.
High side
volt
age
V
is equal t
o
K times the low si
de volta
g
e
V
,
V
K
V
(
1
3
)
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 9, September 20
14: 66
02 – 661
1
6606
High side current
I
is equal to 1/K times the low si
de vol
t
age
V
,
I
1
K
V
(
1
4
)
3.4. Batter
y
The b
a
ttery i
s
mo
dele
d
a
s
a
voltage
sou
r
ce
with
an inte
rnal
resi
stan
ce. T
he mo
del
accou
n
ts fo
r i
n
ternal
po
we
r loss i
n
the
re
sista
n
ce of th
e battery. T
h
ere i
s
no tim
e
lag
com
pon
e
n
t
in the
mod
e
l
.
The
battery
is a
s
sumed
to h
a
ve a
consta
nt inte
rnal
voltage,
E
B
. The battery
terminal volta
ge, V
B
, is equ
al to the sum
of the internal
voltage and
resi
stan
ce voltage d
r
op. Th
e
battery voltage and battery
curre
n
t are e
qual to the co
ntrolle
r low
si
de voltage an
d curre
n
t.
V
Vo
l
t
I
Amp
R
Ohm
E
V
olt
(
1
5
)
A
ssu
ming
V
V
and
thus the Equation (15
)
can be written as:
V
Vo
l
t
I
Amp
R
Ohm
E
V
olt
(
1
6
)
The battery
model u
s
e
s
the cu
rrent an
d volt
age information fro
m
the Motor Co
ntrolle
r to
cal
c
ulate
the
re
quired
bat
tery’s i
n
tern
al
voltage.
T
h
i
s
voltag
e i
s
comp
ared
wit
h
the
a
c
tual
E
B
value to creat
e a batte
ry voltage e
r
ror,
B
EER
, and that erro
r i
s
u
s
e
d
by the PI controlle
r mo
d
e
l to
adju
s
t the loo
p
gain.
B
E
act
u
al
E
c
alcula
t
e
d
(
1
7
)
3.5. Proporti
onal Integral
(PI) Controller
The PI co
ntroller a
c
cept
s
the B
ERR
sign
al from the
b
a
ttery model
and u
s
e
s
p
r
o
portion
al
(K
p
) and inte
g
r
al (K
i
) to calculate the gain
K value that
is used by the
motor co
ntrol
l
er.
K
K
s
K
B
(
1
8
)
4. Simulation Model Blo
cks
4.1. For
w
a
rd
Force Mode
l
The sim
u
lati
on blo
c
k for the forward
forces i
n
cl
u
de Equatio
n
(1)–(3
) fo
r th
e wh
eel
torque. Th
e b
l
ock is sho
w
n
below in Fig
u
re 3.
Figure 3. Effects of Sele
cting Differe
nt Switchi
ng un
de
r Dynami
c
Co
ndition
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Determ
inatio
n of Power
Consum
ption of Electr
ic Vehi
cle for Asso
rt
ed… (M
d. Ru
hul Am
in)
6607
4.2. Wheel T
o
rque
The si
mulati
on blo
ck fo
r the whe
e
l torqu
e
in
clud
es Equ
a
tion
(4)–(5
) for th
e motor
torque. The
block is
sho
w
n belo
w
in Figure 4.
.
Figure 4. Wh
eel Torque M
odel Block
4.3. PI Controller Model
The
block m
odel i
n
cl
ude
s Equation
18
for th
e
cont
rolle
r. The
G
a
in
(K)
of th
e Moto
r
Controlle
r is determi
ned b
y
the output of the PI C
ontroller mod
e
l. The gain ha
s a
n
initial starti
ng
value of 0.1. This value wa
s pre
s
et
within t
he co
ntrolle
r’s inte
gration bl
ock to minimize
the
possibility of
a Simulin
k
si
mulation
erro
r du
e to
an al
gebraic loo
p
.
An alge
brai
c
loop i
s
b
a
si
cally
a divide by zero o
peration
when the
si
mulation is
trying to solve
the set of line
a
r eq
uation
s
. The
PI Controller
che
c
ks to se
e that the
out
put is
not
zero. If the outp
u
t is
ze
ro the
n
the
cont
roll
er
outputs a
sm
all value (0.0
01). Thi
s
is d
one to prev
e
n
t model anal
ysis failure d
ue to dividing
by
zero when
solving the lin
ear e
quatio
n
s
. The
cont
roller al
so in
clude
s a gai
n
limiting block to
prevent exce
ss fee
dba
ck signal
s. The bl
ock is sho
w
n
belo
w
in Figu
re 5.
Figure 5. PI
Controlle
r Mo
del Block
4.4. Driv
e
Sy
stem Mod
e
l
The Speed a
nd Torq
ue va
lues were wri
tten to the MATLAB Workspa
c
e, and th
e values
were th
en
re
ad into
the
m
odel
sp
eed
a
nd to
rqu
e
lo
o
k
-u
p ta
ble
s
.
The
Clo
c
k in
put to th
e lo
o
k
-u
p
tables u
s
ed t
he followin
g
time base values that we
re setup in the model pa
ram
e
ters tabl
e: Tmin
= 0, T
s
tep
=
0.01, Tsto
p
= 110
se
co
nd
s. The di
sp
l
a
yed Scope
val
ues were
al
so written to t
h
e
MATLAB Wo
rkspa
c
e
as Structures with
Time. A
MAT
L
AB script
wa
s u
s
e
d
to p
r
e
-
load th
e spee
d
and torq
ue d
a
ta in the Worksp
ace, Ru
n the Si
mulation, obtain the key data from the Sco
pe
Structu
r
e
s
, a
nd plot th
e d
a
ta. The
co
mplete
fo
rce
model
an
d motor drive model are
shown
belo
w
in Figu
re 6.
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 9, September 20
14: 66
02 – 661
1
6608
Figure 6. Forward Force M
odel an
d Motor De
rive System
5. Results a
nd Simulations
The Moto
r Powe
r plot in
Figure 7 sho
w
s
b
o
th Moto
ring a
nd Reg
eneration. When both
curre
n
t an
d v
o
ltage
are
po
sitive value
s
t
hen th
e
DC
Motor i
s
provi
d
ing to
rqu
e
in
the di
re
ction
of
rotation and
power
i
s
bein
g
tran
sfe
rre
d to
the
lo
ad.
T
h
is i
s
normal
motorin
g
o
p
e
r
ation.
Ho
wev
e
r,
whe
n
the mot
o
r current i
s
in the opp
osit
e pola
r
it
y of the voltage, th
en the moto
r
is bein
g
pu
sh
ed
and a
c
ting a
s
a ge
ne
rato
r with
curre
n
t
flow back i
n
to the batte
ry. In this ca
se the vehi
cl
e
gradi
ent
wa
s ze
ro in
othe
r word
s th
e
spe
ed a
nd t
o
rqu
e
wa
s fe
d by MATLA
B
workspa
c
e
to
investigate th
e motorin
g
an
d rege
ne
ratio
n
action.
Figure 7. Roa
d
spe
ed, Torque an
d Power
Cu
rve
s
Showin
g Motori
ng and
Reg
e
neratio
n
0
10
20
30
40
50
60
70
80
90
100
0
50
10
0
15
0
20
0
Ti
m
e
(
S
e
c
)
S
p
ee
d(
K
m
H
)
R
e
q
u
i
r
ed
R
o
ad
S
p
ee
d
0
10
20
30
40
50
60
70
80
90
100
-2
0
0
0
20
0
R
e
qu
i
r
ed
R
o
ad
T
o
r
que
Ti
m
e
(
S
e
c
)
T
o
rq
u
e
(N
-m
)
0
10
20
30
40
50
60
70
80
90
100
-1
0
0
0
-5
0
0
0
50
0
10
00
(
+
T
)
*
(
+S
)
=
+
P
=
1
s
t Q
u
a
d
r
a
n
t
=
M
o
to
r
i
n
g
(
-
T
)
*
(
+
S
)
=
-
P
=
4
t
h Q
u
ad
r
a
nt
=
R
e
ge
ner
at
i
o
n
R
e
q
u
i
r
ed
R
o
ad
P
o
w
e
r
Ti
m
e
(
S
e
c
)
Ro
ad P
o
w
e
r
(
W
a
t
t
)
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Determ
inatio
n of Power
Consum
ption of Electr
ic Vehi
cle for Asso
rt
ed… (M
d. Ru
hul Am
in)
6609
5.1. Batter
y
Voltage Erro
r
The si
mulati
on mod
e
l a
d
just
s the
controlle
r g
a
in (K) to m
eet drive to
rque a
nd
rege
ne
ration
requi
rem
ents.
The
sim
u
lati
on
comp
ar
ed
the n
o
minal
battery inte
rn
al voltage, VB
=20
0
volts
or VBatt (actu
a
l
), with a
calculat
ed b
a
ttery voltage ba
sed on
the m
o
tor voltage
a
n
d
c
u
rrent values
to get VBatt
(c
alc
u
lated).
The di
ffere
nce, VBerr, wa
s used a
s
an e
rro
r sig
nal inp
u
t
to the Prop
ortional Integral
(PI) Controll
er. Thi
s
VBerr si
gnal
wa
s
plotted over the range
of the
simulatio
n
op
eration. T
h
is
plot is sho
w
n
belo
w
in
Figu
re 8. The m
a
ximum error
of -200
occu
rs at
the very be
g
i
nning
of the
simul
a
tion.
This l
a
rg
e
e
r
ror i
s
a natu
r
al respon
se
to startin
g
t
h
e
simulatio
n
. T
he si
mulatio
n
quickly reco
vers
and
hol
ds a
n
e
r
ror
o
f
about
+76
durin
g the i
n
i
t
ia
l
starting
of th
e moto
r. It is no
rmal
to h
a
ve
a
high
er erro
r h
e
re
b
e
ca
use the
motor
develo
p
e
d
voltage, VD(Volt) = WD
(ra
d/se
c)/Km,
is low du
rin
g
startu
p, especi
a
lly whe
n
the cu
rre
nt is
increa
sing. T
he ne
gative e
rro
r o
c
curs d
u
ring
re
gen
eration. By revi
ewin
g the m
o
tor voltage
drop
equatio
n, VL (Volt)
= IL(A
mp) *
R
A(O
h
m) + EB
(Vol
t), the cha
nge i
n
cu
rrent pol
a
r
ity will ca
use
the
reverse
pol
ari
t
y of the IL(A
mp) *
R
A(Oh
m) te
rm. T
h
i
s
voltag
e
cha
nge
will im
pa
ct the
magnit
ude
of the input and outp
u
t of the PI controlle
r be
cau
s
e of the red
u
ce
d differen
c
e bet
wee
n
the
cal
c
ulate
d
an
d actual volta
ge in the erro
r equatio
n, B
ERR
= E
B
(actual) - E
B
(calcu
lated).
Figure 8. Battery Voltage Erro
r (B
Err
).
5.2. Controller Gain
The G
a
in
(K) of the M
o
tor Co
ntrolle
r i
s
det
ermined
by the outp
u
t of the PI
Co
ntrolle
r
model. A plot
of the value
of the Controller G
a
in (K
) duri
ng the
simulatio
n
is
sho
w
n b
e
lo
w in
Figure 9. Th
e co
ntrolle
r g
a
in in
cre
a
ses duri
ng t
he ti
me when th
e
motor
spe
ed i
s
in
crea
sing,
and
decrea
s
e
s
when th
e m
o
to
r
spe
ed i
s
de
cre
a
si
ng. T
h
e gai
n h
a
s a
n
initial
sta
r
ting valu
e of
0
.
1.
This
wa
s p
r
eset
within t
he controll
er in the 1/
s i
n
tegratio
n bl
ock. Thi
s
val
ue is
set in
th
e
simulatio
n
by openin
g
up
the 1/s blo
c
k. The additio
n
of the initial conditio
n
o
n
the integration
block h
e
lps t
o
minimi
ze th
e po
ssi
bility of a Simulink simulatio
n
e
rro
r du
e to a
n
alge
brai
c l
oop.
An alge
brai
c
loop i
s
b
a
si
cally a divide
by ze
ro o
p
e
r
ation when
th
e sim
u
lation i
s
trying
to so
lve
the set of line
a
r eq
uation
s
. Figure
10
sho
w
s that vehi
cl
e gra
d
ient in
different spee
d with varia
b
l
e
spe
ed. It is
observed tha
t
when vehicl
e need
s more torque to o
v
erco
me the gradi
ent moto
r
draws
more power from the battery.
0
10
20
30
40
50
60
70
80
90
100
0
100
200
Ti
m
e
(
S
e
c
)
S
pee
d(
K
m
H
)
R
equ
i
r
e
d
R
oad
S
p
eed
0
10
20
30
40
50
60
70
80
90
100
-
200
0
200
R
e
qu
i
r
ed
R
o
ad
T
o
r
q
u
e
Ti
m
e
(
S
e
c
)
T
o
rq
u
e
(N
-m
)
0
10
20
30
40
50
60
70
80
90
100
-
200
0
200
B
a
t
t
ery
V
o
l
t
ag
e
error
Ti
m
e
(
S
e
c
)
B
a
t
t
er
y
V
o
l
t
a
ge e
r
r
o
r
(
V
o
l
t
)
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 9, September 20
14: 66
02 – 661
1
6610
Figure 9. Con
t
roller G
a
in K Value with Ba
ttery Voltage
Figure 9. Vehicle Lin
ear Sp
eed, Gradient
, Torque a
nd
Powe
r
0
10
20
30
40
50
60
70
80
90
10
0
0
100
200
Ti
m
e
(
S
e
c
)
S
p
ee
d(
K
m
H
)
Requ
i
r
ed Road
S
pee
d
0
10
20
30
40
50
60
70
80
90
10
0
-200
0
200
M
o
t
o
r V
o
l
t
ag
e (V
ol
t
)
Ti
m
e
(
S
e
c
)
Vo
l
t
a
g
e
0
10
20
30
40
50
60
70
80
90
10
0
0
1
2
Co
n
t
ro
l
l
e
r
G
a
i
n
Ti
m
e
(
S
e
c
)
Ga
i
n
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Determ
inatio
n of Power
Consum
ption of Electr
ic Vehi
cle for Asso
rt
ed… (M
d. Ru
hul Am
in)
6611
5. Conclusio
n
A simulation
model for
a
electri
c
vehi
cle
with vect
or controlled
perm
anent
magnet
traction
moto
r is develo
p
e
d
and
al
so ve
rified. In
this
pape
r a
batte
ry ele
c
tric ve
hicle
have
be
en
modele
d
and
design
ed. After analyzi
ng
the tor
que
sp
eed ch
aracte
ristics of mot
o
r and
wheel
, it
is con
c
lu
ded t
hat vehicle ki
netic ene
rgy durin
g run
n
in
g in negative slop
e or in ze
ro clut
ch can be
conve
r
ted to
electri
c
al
po
wer
by mean
s of reg
ene
ration sche
me
of motor. O
n
the other
h
and
whe
n
moto
r requires l
a
rge
torqu
e
for
p
o
sitive sl
ope
it deman
ds
more
cu
rrent
as
well
as
more
power to motor. Motor con
t
roller can u
s
ed to m
onitor the condition
of speed in term
s of voltage
and the
dem
ande
d torq
ue
in term
s of current to pr
ove the in
stru
ction to be fulfill
ed the requi
red
value by valu
e of gai
n obta
i
ned fo
r PI co
ntrolle
r. T
he
battery of the
electri
c
ve
hicl
e is
de
signe
d
in
such a way that both the
power an
d energy requirements are fulfille
d for a given
driving
cycle.
In
the simulatio
n
, linear mo
d
e
ls for the e
n
e
rgy sour
ce
s are u
s
ed. Th
e future work will inco
rpo
r
ate
more a
c
curate model
s of the ene
rgy so
urces in the
simulation.
Referen
ces
[1]
GRENIER
Mathie
u. Desi
gn
of an on-
bo
ard charg
e
r for
plug-
in h
y
b
r
i
d
electric
al ve
hicle (P
HEV).
Chalm
e
rs Un
iv
ersit
y
of T
e
chn
o
lo
g
y
Göte
bor
g, S
w
e
d
e
n
; 20
09.
[2]
Gao DW, Mi C
,
Emadi A.
Modeli
ng a
nd si
mulati
on of elect
r
ic and hybr
id
vehicl
es
. Proc
eed
ings of the
IEEE. 2007; 95
(4): 729 – 7
45.
[3]
Mape
lli
F
L
, T
a
rsitano
D, Ma
ur
i M. Plu
g
-i
n h
y
brid
el
ectric ve
hicle:
Mod
e
li
ng
, protot
ype
re
al
izatio
n, an
d
inverter l
o
sses
reducti
on a
nal
ysis.
IEEE Transactions on Industrial Electronics.
2010; 5
7
(2
): 598-60
7.
[4]
Mape
lli
F
L
, T
a
rsitano
D, Ma
ur
i M. Plu
g
-i
n h
y
brid
el
ectric ve
hicle:
Mod
e
li
ng
, protot
ype
re
al
izatio
n, an
d
inverter l
o
sses
reducti
on a
nal
ysis
. IEEE Transactions on Industrial Electronics.
2010; 5
7
(2
): 598-60
7.
[5]
Saura
bh. Mo
del-B
ased
De
sign for
H
y
b
r
id El
ectric
Vehic
l
e S
y
ste
m
s, Paper 2
008-
01-0
0
8
5
,
www
.
m
ath
w
o
r
k
s
.com.
[6]
Rizkal
l
a. A
ppl
i
c
ations
of
Co
mputer-Bas
ed
Po
w
e
r El
ectronics
to E
l
ect
r
ic Ve
hicl
e T
e
chno
log
y
,
A
n
Interdisci
pli
nar
y
Se
nior C
ours
e
, ASEE AC. 2000.
[7
]
C
h
an
C
C
,
Bouscay
r
o
l
A, Ch
en
K. Electric, h
y
br
id, a
nd fu
el-
c
ell ve
hic
l
es: A
r
chitectures
an
d mod
e
li
ng,
IEEE Transactions on Vehic
u
lar Technology.
2010; 5
9
(2): 5
89-5
98.
[8]
Norman S
Nis
e. Contro
l S
y
stems Eng
i
ne
eri
ng, Si
xt
h Ed
iti
on, Ca
liforn
i
a
Stat
e Pol
y
tech
nic Un
iversit
y
,
Pomon
a
, Calif
orni
a, USA. 2011: 547-
59
7.
[9]
Fundam
enta
l
s of Vehicl
e D
y
n
a
mics, Gilles
p
i
e
, SAE. 110-11
5.
[10]
H
y
brid El
ectric
Vehic
l
es, Mi, Masrur, Gao, W
ile
y
.
4
2
-45.
[11]
Electric an
d H
y
brid Ve
hicl
es D
e
sig
n
F
und
amentals 2
nd Ed.
Iqbal H
u
sai
n
, CEC Press.
[12]
Z
hang
Li
w
e
i, W
ang Xi
ao, Z
hang
Yuj
i
n
DC Po
w
e
r S
ystem of Elect
r
ic Veh
i
cle.
TE
L
K
OM
N
I
KA
Indon
esi
an Jou
r
nal of Electric
al Eng
i
ne
eri
n
g
.
2013; 1
1
(11):
665
1-66
56.
[13]
Lei L
i
n, Yu
ank
ai Li
u, W
ang Pi
ng, F
ang
Hon
g
.
T
he Electric
Vehic
l
e L
i
thium
Batter
y
Mo
nito
ring S
y
stem
.
T
E
LKOMNIKA Indon
esi
an Jou
r
nal of Electric
al Eng
i
ne
eri
n
g
.
2013; 1
1
(4): 2
247-
225
2.
Evaluation Warning : The document was created with Spire.PDF for Python.