TELKOM
NIKA
, Vol.13, No
.2, June 20
15
, pp. 421 ~ 4
3
1
ISSN: 1693-6
930,
accredited
A
by DIKTI, De
cree No: 58/DIK
T
I/Kep/2013
DOI
:
10.12928/TELKOMNIKA.v13i2.987
421
Re
cei
v
ed
No
vem
ber 1
1
, 2014; Re
vi
sed
March 16, 20
15; Accepted
April 4, 2015
Basal S
t
udy on Power Control Strategy for Fuel
Cell/Battery Hybrid Vehicle
Dingy
u
e Chen*
1
, Xia Li
2
, L
i
hao Chen
3
, Yonghui
Zha
ng
1
, L
i
Y
a
n
g
1
, Songson
g Li
1
1
School of Aut
o
mob
ile, Ch
an
g’a
n
Univ
ersit
y
, Xi’a
n, 710
06
4
,
P.R.China
2
Xi'
an Ji
aoto
n
g
Universit
y
, Xi'
an, P.R.Chin
a
3
Strasbourg U
n
iversit
y
, Stras
bour
g, F
r
ance
*Corres
p
o
ndi
n
g
author, e-ma
i
l
: cd
y868
@16
3
.
com
A
b
st
r
a
ct
In order
to
enh
ance
the
fuel
e
c
ono
my
of
hyb
r
id v
ehic
l
e an
d incre
a
se
th
e mi
lea
ge of
co
ntin
uatio
n
o
f
jour
ney, the p
o
w
er control strategy (P
CS) is
as sign
ifica
n
t as co
mp
one
nt si
z
i
ng
in
achi
e
v
ing
opti
m
a
l
fu
e
l
econ
o
m
y of the fuel ce
ll/batt
e
ry hy
bri
d
veh
i
cle (F
CBHV).
T
he mode
ls
of
F
C
BHV structure an
d opti
m
al
pow
er contro
l
strategy are
d
e
vel
ope
d by
e
l
ectric ve
hicl
e
simulati
on s
o
ftw
are ADVISOR w
h
ich us
es
a
hybri
d
backw
ar
d/forw
ard appr
oach. T
he res
u
lts de
monstr
ate that the pro
p
o
sed c
ont
rol st
rategy ca
n sati
sfy
the pow
er re
q
u
ire
m
e
n
t for tw
o stand
ard
driv
ing cyc
les
a
n
d
achi
eve th
e p
o
w
e
r distrib
u
tio
n
a
m
o
ng v
a
rio
u
s
pow
er sources
.
T
he compr
e
h
ensiv
e co
mpar
isons w
i
th t
he
pow
er tracking
controll
er (PT
C
) w
h
ich is w
i
d
e
ado
pted
in AD
VISOR verify t
hat the
prop
os
ed co
ntrol strat
egy h
a
s b
e
tte
r ration
ality a
nd
valid
ity in ter
m
s o
f
fuel eco
n
o
m
y
and dy
na
mic
p
r
operty in tw
o standar
d driv
i
n
g cycles. Ther
efore, the
pro
p
o
sed strategy
w
ill
provi
de a n
o
vel
appro
a
ch for the adv
anc
ed p
o
w
e
r control sy
stem of F
C
BH
V.
Ke
y
w
ords
: FCBHV, Power Control Strat
egy,
ADVISOR, Driving Cycl
es
1. Introduction
Hybrid
vehi
cl
es
are
vehi
cles th
at use t
w
o
o
r
mo
re
po
w
e
r
s
o
ur
ce
s fo
r
th
e d
r
ive
s
y
s
t
e
m
.
In contrast,
ordin
a
ry inte
rnal co
mbu
s
ti
on engi
ne (I
CE) vehi
cle
s
use a
sin
g
l
e
power
so
u
r
ce
con
s
i
s
ting of
reci
procatin
g engin
e
, typically fuel
ed with g
a
soline, to dri
v
e a com
p
l
e
x
transmission mechani
sm that is
then
coupled to the drive wheel
s [1],[2]. The disadvantages of
ICE vehicle
s
inclu
de lo
w e
nergy
efficien
cy, exce
ssive
harmful
ch
e
m
ical
emissio
n
s, hig
h
noi
se
level and
he
avy depen
de
nce
on
a si
n
g
le fuel
so
urce
. Hyb
r
id
el
ectri
c
vehi
cle
s
a
r
e o
ne
of the
solutio
n
s
pro
posed to ta
ckle the
perceived probl
e
m
s a
s
soci
ate
d
with the
e
nergy
cri
s
i
s
and
global
warming [3]. Hybri
d
vehicle
s
se
amlessly co
mbine two o
r
more po
we
r source
s into
one
drive sy
stem
. The fuel cell/battery hybrid v
ehi
cle
(FCB
HV) me
rge
s
hybri
d
vehicle a
nd
the
hydrog
en fue
l
cell techn
o
l
ogie
s
in ord
e
r
to r
epla
c
e the co
nventio
nal f
uel and
optimize the
fuel
con
s
um
ption.
Powe
r
contro
l strategy (P
CS)
and
co
mpone
nt
si
zi
ng affe
ct veh
i
cle
perfo
rma
n
ce
an
d
fuel eco
nomy
con
s
ide
r
ably
in FCBHV b
e
ca
use of
the multiple power source
s an
d differen
c
e
s
in
their ch
ar
act
e
risti
cs. Fu
rth
e
rmo
r
e, the
s
e two impo
rtant factors
are cou
p
led
—
d
i
fferent sele
ct
ion
of comp
one
nt sizi
ng shoul
d com
e
with
diffe
rent de
si
gn of po
we
r
control st
rate
gy.
Therefore
,
to
achi
eve maxi
mum fuel e
c
onomy for F
C
BHV, optim
al po
wer
co
ntrol an
d co
mpone
nt si
zi
ng
sho
u
ld b
e
d
e
termin
ed a
s
a co
mbin
ed
packa
ge. O
u
r resea
r
ch
has fo
rmul
ated an
d solve
d
a
power
control
probl
em of a FCBHV. Developme
n
t
o
f
the powe
r
control
strateg
y
is one of the
importa
nt tasks in
develo
p
i
ng hyb
r
id
ve
hicle
s
and
rel
a
tively many
literatures ca
n be
foun
d.
Y.
Gue
z
en
ne
c
et al. [4] sol
v
ed the sup
e
rviso
r
y cont
rol p
r
obl
em
of a FCB
H
V
as a
qua
si-static
optimizatio
n probl
em an
d found that
hybridi
z
ation
ca
n signifi
cantly
improve the
fuel eco
nomy
of
FCBHV.
Wa
n
g
Y et
al. [5]
use
d
the
eq
ui
valent con
s
u
m
ption mi
nim
i
zation
st
rate
gy to dete
r
mi
ne
an optimal p
o
we
r distri
bu
tion for a fuel cell/su
pe
rcapa
citor hybrid vehicle. The con
c
e
p
t of
equivalent fa
ctors in hybri
d
electri
c
veh
i
cle
s
has b
e
e
n
descri
bed b
y
Xiaolan Wu
et al.
[6]. In the
same
re
sea
r
ch, they also com
pared
their po
wer control re
sult to deterministic
dyn
a
mic
prog
ram
m
ing
result, which
can le
ad to a global o
p
tima
lity.
As a promi
s
i
ng technolo
g
y
fuel cells t
hat
co
nvert
chemi
c
al
ene
rgy of the fu
el into
electri
c
ity wit
hout combu
s
t
i
on are stu
d
i
ed worl
d
w
id
e
with an
aim
to improve th
e po
wer
outp
u
t,
lowe
r the co
st and extend
the life
of o
peratio
n for
wide
sp
rea
d
a
pplication
s
. The fuel cell
s
are
gene
rally vie
w
ed
as a d
e
p
enda
ble p
o
wer
sou
r
ce for
many ap
plica
t
ions, such a
s
hyb
r
id vehi
cle,
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 13, No. 2, June 20
15 : 421 – 43
1
422
distrib
u
ted p
o
we
r
g
ene
ra
tion,
and po
rtable po
wer so
urce [7]-[9]. Due to
slo
w
dyn
a
m
ic
cha
r
a
c
teri
stic of fuel cell,
a FCBHV ha
s bee
n
pro
p
o
se
d. The p
o
we
r co
ntrol
of hybrid veh
i
cle
whi
c
h determines the power assignment bet
ween
the fuel
cell system
and
auxiliary energy
stora
ge devi
c
es is a
n
impo
rtant tech
niqu
e.
In re
ce
nt ye
ars,
a va
riety
of
control
stra
tegie
s
for
power contro
l have
bee
n
use
d
to
hybrid vehi
cl
e. Thounth
o
n
g
et al. [10] had u
s
e
d
an
innovative control la
w b
a
sed
on flatne
ss
prop
ertie
s
for fuel cell/sup
e
rcapa
cito
r h
y
brid po
we
r source. Paladi
ni et al
. [11] had presented
an
optimal control strate
gy to power
a vehicl
e with
b
o
th fuel cell
and batte
ry to redu
ce f
uel
con
s
um
ption.
However, in
these
works the
prop
ose
d
cont
rol st
rategie
s
had
not adeq
uate
l
y
con
s
id
ere
d
the bala
n
ce b
e
twee
n fuel eco
nomy
an
d dynami
c
property of hybrid vehi
cle [
12].
Furthe
rmo
r
e,
an app
rop
r
i
a
te intelligent
control strategy had n
o
t been p
r
o
p
o
s
ed fo
r a hy
brid
vehicle. In
this
pap
er, a
se
co
nda
ry
developm
ent
for el
ect
r
ic vehicl
e
sim
u
lation
software
ADVISOR is i
m
pleme
n
ted
based on the
system a
r
chitecture of FCBHV.
In orde
r to e
nhan
ce the f
uel economy
of hy
brid vehicle a
nd in
crea
se the mil
eage of
contin
uation
of journey, the PCS i
s
a
s
signifi
cant
a
s
com
pon
ent sizing i
n
a
c
hie
v
ing optimal f
u
e
l
eco
nomy of
the FCB
H
V. Control strat
egie
s
we
re l
a
rgely b
a
sed
on heu
risti
c
rule
s, whi
c
h
is
usu
a
lly far from true
-opti
m
ality. This study
pre
s
ent
s a combi
ned
power
control
of FCBHV, the
power control
algorithm
wa
s develo
ped f
r
om st
o
c
h
a
sti
c
dynami
c
progra
mming
(SDP) motivated
basi
s
fun
c
tio
n
s. According
to standard driving cy
cl
e
con
d
ition
s
, the prop
osed control st
rateg
y
is
contraste
d
wi
th
the power tracki
ng
control st
rategy
which
is wi
de
a
dopted
in A
D
VISOR in
terms
of the indexe
s
of fuel eco
n
o
my and dyn
a
mic p
r
op
erty.
2. FCBHV S
t
ruc
t
ure a
nd Optim
a
l Po
w
e
r Con
t
rol St
rategy
With the
adv
ancement i
n
the technol
og
y of fuel
cell
s, there i
s
a
n
i
n
crea
sing
int
e
re
st in
usin
g fuel
cel
l
s for
hybrid
vehicle [1
3]. The F
C
BHV is
a
p
opul
ar hybrid stru
ct
ure as Figu
re
1
sho
w
n. In
thi
s
stru
cture, a
fuel
cell
sy
stem d
e
sig
ned
for ve
hicular propul
sion
a
pplication m
u
st
have a po
we
r den
sity, a startup, a
nd
a tran
sient
resp
on
se si
mi
lar to presen
t-day ICE-b
a
s
ed
vehicle
s
. A battery is gen
erally co
nne
cted acro
ss
th
e fuel cell sy
stem to provi
de su
pplem
e
n
tal
power fo
r
st
arting th
e
system. Th
e fu
el cell sy
ste
m
and
invert
er
con
n
e
c
t b
y
a unidi
re
cti
onal
DC/
DC conv
erter for
mat
c
hin
g
voltage
cla
s
s. Th
e
a
d
vantage
s of
this
stru
ctu
r
e are lo
w p
o
w
er
and tran
sie
n
t resp
on
se d
e
mand fro
m
the fuel
cell system an
d conveni
ent
braki
ng ene
rgy
recovery.
Figure 1 sho
w
s th
e FCB
H
V stru
cture
a
nd key
cont
ro
l signal
s for
p
o
we
r control.
FCBHV
c
o
ns
is
ts
of several subs
ys
tems
: driver, fuel
cell sy
st
em (F
CS
),
bat
t
e
ry
,
DC
/
DC co
nv
ert
e
r,
electri
c
d
r
ive
,
and vehicl
e dynamics.
Con
s
ide
r
ing
various veh
i
cle state
s
–
such as p
o
w
er
deman
d, batt
e
ry state
of charg
e
(S
OC), and ve
hicl
e
spe
ed –
t
he power co
ntrol
system
(P
CS)
sen
d
s the fu
el cell
current
req
u
e
s
t to th
e DC/DC
con
v
erter;
sen
d
s the moto
r to
rque
re
que
st
to
the ele
c
tri
c
d
r
ive; controls t
he
rege
nerative bra
k
in
g ra
tio. In orde
r t
o
gen
erate th
e motor torqu
e
requ
este
d fro
m
the PCS, the inverter d
r
aws cu
rre
nt from the ele
c
tric DC bu
s where the batt
e
ry
and th
e
DC/
DC conve
r
ter are
con
n
e
c
ted in
pa
ra
ll
e
l
.
The DC/DC conve
r
ter can
control
t
he
c
u
rr
en
t flo
w
in
to
th
e DC
bu
s
,
w
h
er
ea
s th
e
ba
tte
ry h
e
r
e i
s
“p
assive
ly” conn
ecte
d
to the
DC
bu
s–
the differe
nce bet
wee
n
t
he
curre
n
t draw f
r
om th
e
inverter an
d
the current
o
u
tflow fro
m
t
he
DC/
DC
conv
erter
will be
compensated by the passi
ve
battery. Therefore,
the power split ratio
betwe
en the
battery and t
he fuel cell
system is a
c
hi
eved by the PCS sen
d
ing
the fuel cell net
c
u
rrent reques
t to the DC/
DC
c
onverter.
The g
oal of p
o
we
r
control i
n
FBHV i
s
to
minimize fue
l
con
s
u
m
ptio
n whil
e maint
a
ining
the battery S
O
C
by sendi
ng a
dequ
ate
curre
n
t re
que
st
comma
nd to
the DC/
DC conve
r
ter. To
achi
eve thi
s
goal, o
p
timal
power
co
ntrol
strategy n
e
e
d
s to
be
de
si
gned
for th
e
PCS to b
a
lan
c
e
the FCS power and the b
a
ttery powe
r
.
Many powe
r
control al
gorithms in
tech
nical literatures
were de
sig
n
e
d
by rule
-ba
s
ed or
heu
risti
c
metho
d
s. T
hose rul
e
-b
a
s
ed m
e
thod
s are
simple
a
nd
easy to un
de
rstan
d
be
ca
u
s
e they com
e
from en
gin
eerin
g intuitio
n. However,
they often lack
optimality or
cycle
-
be
ating
.
Ideally, minimization
of fuel con
s
um
ption of hyb
r
id
vehicle
s
can
b
e
achi
eved
onl
y whe
n
the
driving
sce
nario
is kno
w
n
a p
r
io
ri. The
dete
r
ministic dyn
a
mic
prog
ram
m
ing
techniq
ue ca
n accompli
sh
this global
o
p
timum. The
n
again, the result cann
ot be
reali
z
ed a
s
a
powe
r
co
ntrol scheme b
e
ca
use it is
not possible
to predi
ct the future drivi
ng
scena
rio.
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
930
Basal Stud
y on Powe
r Co
ntrol Strateg
y
for Fuel Cell/
Battery Hyb
r
i
d
Vehicle
(Di
ngyue Chen
)
423
Figure 1. Driv
e stru
cture of FCBHV
The p
o
we
r
control
strate
gy desi
gne
d
by
the SDP approa
ch
c
an overcom
e
these
limitations
of
existing al
go
rithms [14]. T
he ide
a
of
th
e infinite h
o
ri
zon
SDP i
s
t
hat if the ove
r
all
power de
ma
nd is mod
e
le
d as a sto
c
h
a
stic p
r
o
c
e
s
s, an optimal controlle
r ca
n be de
sign
ed
based o
n
the
sto
c
ha
stic
m
odel. First, th
e drive
r
p
o
wer d
e
man
d
i
s
model
ed a
s
a discrete-tim
e
stocha
stic dynamic p
r
o
c
e
s
s by
using a
Markov chai
n
model, which is con
s
tru
c
t
ed from stan
dard
driving
cycle
s
. In other
words, the p
o
wer deman
d fro
m
the drive
a
t
the next time step
dep
en
ds
on the cu
rren
t power d
e
ma
nd and vehi
cl
e spe
ed:
l
wh
wh
i
dem
dem
j
dem
r
j
il
P
P
P
P
P
,
,
,
fo
r
N
l
N
j
i
p
,
2
,
1
,
,
,
2
,
1
,
(1)
whe
r
e th
e p
o
w
er de
man
d
dem
P
and th
e
whe
e
l
sp
eed
wh
are q
uantized i
n
to
grid
s of
p
N
and
N
,
respe
c
tively.
Then, for the
discretized state vector,
x
(SOC
,
wh
,
dem
P
), co
rre
spo
ndin
g
optimal
fuel cell curre
n
t req
u
e
s
t co
mmand,
reg
net
fc
I
u
,
,
, is
determi
ned t
o
minimi
ze th
e expe
cted
cost of
hydrog
en con
s
umptio
n and
battery ener
gy usag
e ove
r
infinite hori
z
on:
SOC
rct
H
N
k
k
k
N
W
W
E
J
,
1
0
2
lim
(2)
Whe
r
e
1
0
the di
scount i
s
fa
ctor,
rct
H
W
,
2
the re
act
ed hyd
r
og
en
mass, an
d
SOC
W
penali
z
e
s
th
e
battery
ene
rgy use b
a
se
d on
the S
O
C valu
e. Thi
s
SDP p
r
o
b
le
m can
be
either
solved by a
policy iterati
on or valu
e iteration
p
r
o
c
ess. The resulting SDP control
strateg
y
gene
rate
s op
timal fuel cel
l
current re
q
uest a
s
a function of batt
e
ry SOC, wh
eel sp
eed, a
nd
power
dema
n
d
. The
cont
ro
l strate
gy ach
i
eves hi
gh
fu
el econo
my while
su
cce
s
sfully maintai
n
ing
battery SOC.
3. ADVISOR MODEL OF FCBHV
ADVISOR was create
d
in
the
MAT
L
A
B
/Simulink e
n
vironm
ent. The pro
g
ram
uses
a
n
iterative calculation
sche
me to g
ene
rate outp
u
ts
o
f
a vehi
cle’s
velocity an
d
energy u
s
e
a
t
all
times du
rin
g
a given
sim
u
lation [15]-[
18]. The u
s
e
r
mani
pulate
s
a
se
ries of
Gra
phical User
Interface (G
UI) scree
n
s to
input vario
u
s vehi
cle pa
ra
meters and
d
r
ive cycl
e re
q
u
irem
ents a
n
d
monitor thei
r i
m
pact o
n
veh
i
cle pe
rforma
nce, f
uel e
c
o
nomy, and e
m
issi
on
s. Th
e three m
a
in
GUI
scree
n
s in A
D
VISOR
are
the vehi
cle in
put scr
een, t
he
simulatio
n
paramete
r
s
scree
n
a
nd t
he
results scree
n
.
Example
s
of
thes
e
scre
ens are
sho
w
n in Fi
gure 2
–4. In the ve
hicle
input
screen
(Figu
r
e 2
)
, th
e user b
u
ild
s
a vehicl
e of i
n
tere
st
by selecting
option
s
from
a seri
es of d
r
op
-d
o
w
n
menu
s. Each
list inclu
d
e
s
several prep
rogra
mm
ed
p
a
rts fo
r use i
n
the vehicl
e
.
The user m
a
y
also create custom com
p
onent
s
by
e
d
iting the p
r
opertie
s
of
e
a
ch
part. Thi
s
featu
r
e ma
ke
s
ADVISOR co
nvenient for i
nnovativ
e vehicle d
e
si
gn
and si
mulatio
n
. In the simu
lation pa
rame
ters
scree
n
(Fi
gure 3), the use
r
define
s
the
drive
cycl
e para
m
eters f
o
r the
event
over which t
he
vehicle i
s
to be sim
u
lated.
Vehicle p
e
rf
orma
nc
e ca
n
be revie
w
ed
in the re
sult
s screen
(Fig
ure
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 13, No. 2, June 20
15 : 421 – 43
1
424
4), where fuel
eco
nomy a
n
d
emi
ssi
ons
are
displa
yed
along
sid
e
de
tailed plot
s of
time-de
pen
d
ent
outputs. Th
e use
r
ca
n sel
e
ct from a
w
i
de array
of output option
s
related to sp
eed an
d torq
ue,
fuel co
nsum
ption, emissi
ons, b
a
ttery
cha
r
ge l
e
vel, etc., an
d display u
p
to four p
l
ots
simultan
eou
sl
y.
Figure 2. ADVISOR vehicl
e input scre
e
n
Figure 3. ADVISOR Simulation paramet
ers
scre
en
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
930
Basal Stud
y on Powe
r Co
ntrol Strateg
y
for Fuel Cell/
Battery Hyb
r
i
d
Vehicle
(Di
ngyue Chen
)
425
Figure 4. ADVISOR Re
sul
t
s scre
en
Figure 5.
Block di
agram of
the FCBHV configuration
In this pap
er,
a
se
con
d
a
r
y develo
p
men
t
for A
D
VISOR i
s
im
pleme
n
ted b
a
sed
o
n
th
e
system
archit
ecture of F
C
BHV. Figure
5 sh
ow
th
e Simulink
syste
m
s for
FCB
H
V configu
r
ati
ons,
respe
c
tively.
These blo
ck
diagram
s rep
r
esent ho
w ADVISOR ap
pl
ies the drive
cycle an
d veh
i
cle
prop
ertie
s
to
analyze the
power flo
w
.
ADVISOR
a
p
p
lies a dyn
a
m
ic g
a
in to
d
e
termin
e
whe
t
her
the de
sired
power flo
w
can b
e
p
r
ovid
ed to
ea
ch
e
l
ement
rep
r
e
s
ente
d
in
the
blo
c
k dia
g
ra
m.
Thro
ugh di
screte time ste
p
solutio
n
m
e
thod
s, Simulink i
s
abl
e
to solve th
e cha
r
a
c
teri
stic
differential e
q
uation
s
of the
system. A
D
VISOR en
abl
e
s
the
use
r
to
modify many
variable
s
in
the
FCBHV. Ea
ch majo
r
com
pone
nt in th
e FCB
H
V
ca
n be
ch
ang
e
d
inde
pen
de
ntly to simul
a
te
different confi
guratio
ns. O
u
r simul
a
tion
wa
s ba
sed
on
d
e
s
ig
ns
fr
om p
r
e
v
io
us
C
i
ty H
y
b
r
id
bu
s
.
The po
we
r requireme
nt calcul
ation
s
were comp
are
d
to earlie
r drivetrai
n
s to
determin
e
the
approp
riate e
ngine
s an
d m
o
tors fo
r the
simulatio
n
. O
u
r obj
ective
wa
s to co
nstruct a compa
r
i
s
on
in optimal f
uel e
c
ono
my of the FCBHV, and
t
herefo
r
e th
e
values fo
r
some
non
-critical
comp
one
nts were
h
e
ld co
nstant at
defa
u
lts
a
c
ross
th
e PCS. A fuel
cell
syste
m
model
ba
sed
on
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 13, No. 2, June 20
15 : 421 – 43
1
426
50 kW po
we
r-effici
en
cy model of Internatio
nal
Fu
el Cell Co
m
pany is ado
pted as Fig
u
r
e 6
sho
w
n. A
re
sistan
ce–
c
a
p
a
c
itan
ce
(RC)
equivalent
ci
rcuit m
odel
is used to
dev
elop th
e b
a
ttery
model a
s
Fig
u
re 7
sh
own. The validity of thes
e m
o
d
e
ls
wa
s verifi
ed by plentif
ul experim
en
ts
[19]-[22].
Figure 6. Simulink mo
del o
f
fuel cell syst
em
Figure 7. Simulink mo
del o
f
batter
4. DRIVING CYCLES A
N
D
RESU
LTS
The UDDS
a
nd HWFET
were
the
sta
nding driv
in
g
cycl
es
used
throu
gho
ut this
study
(Figu
r
e 8).
T
he
Urb
an Dynamom
eter Driving
Sch
e
dule (UDDS
),
or “the city
test”, whi
c
h
h
a
s
a
total
length
o
f
7.45
mil
e
s and an avera
ge spe
ed of
19.59 mph, wa
s
u
s
e
d
to rep
r
e
s
ent
typ
i
cal
driving
c
on
ditions of lig
ht d
u
ty vehicle
s
i
n
t
he
city. Th
e Hi
gh
way F
uel E
c
ono
my
Driving
Sche
dule
(HWFET
), with a high
er av
erag
e spee
d of 48.3 m
ph
and 10.2
6
mil
e
s in total len
g
th, was
use
d
to
rep
r
e
s
ent hig
h
way driving
con
d
ition
s
.
T
he stand
ar
d
cycles were ex
tended
to
25,
50, 75,
10
0 a
nd
150
mil
e
s
by repli
c
ation. Driving cycle
s
f
r
om a
calib
rat
ed a
nd vali
da
ted si
mulatio
n
net
work we
re
als
o
us
ed in this
s
t
udy to v
e
rify t
he resul
t
s achi
eved from stand
ard cycle
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
930
Basal Stud
y on Powe
r Co
ntrol Strateg
y
for Fuel Cell/
Battery Hyb
r
i
d
Vehicle
(Di
ngyue Chen
)
427
Figure 8. Co
mpari
s
o
n
s b
e
t
ween hi
gh
wa
y (above)/
c
ity (belo
w
)
cycle
In ord
e
r to
compa
r
e the
i
ndexe
s
of fu
el econo
my and dyn
a
mi
c pro
perty of
hybrid
vehicle, the
PCS are d
e
s
ign
ed fo
r F
C
BHV. In
a
ddition, a
c
co
rding
to two
stan
dard
cycle
con
d
ition
s
, the pro
p
o
s
ed
control
strat
egie
s
ar
e
co
ntraste
d
with
the powe
r
followin
g
co
ntrol
strategy which is wide a
d
o
p
ted in ADVISOR for
FCB
H
V. The spe
ed cu
rves of
PCS for FCB
H
V,
and p
o
wer t
r
acking
co
ntro
ller (PT
C
) for FCB
H
V ca
n
match
with t
he requi
red
speed
cu
rve
s
in
two cycle
co
ndition
s. The
r
efore, PCS desi
gne
d ca
n satisfy the speed
requi
reme
nts for two
stand
ard
cy
cl
e
co
ndition
s. The spe
ed cu
rves are sh
o
w
n in
Figu
re
8. Unli
ke
som
e
strategie
s
t
hat
deplete
s
o
r
o
v
erch
arge th
e battery, ou
r cont
rolle
r de
monst
r
ate
s
th
at it can mai
n
tain the b
a
ttery
SOC within li
mited ope
rati
ng ran
ge. In Figure 8,
the optimization
result in time horizon of ci
ty
and hi
gh
way
cycl
es
wa
s sh
own. Sim
ilar to th
e o
r
iginal
SDP
controlle
r, the p
s
eud
o-S
D
P
controlle
r
spli
t the requi
red
motor p
o
wer
to the fu
el
cel
l
and
the
batt
e
ry a
nd
maint
a
ins the
batte
ry
SOC.
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 13, No. 2, June 20
15 : 421 – 43
1
428
Figure 9. Co
mpari
s
o
n
of optimal re
sults fo
r 200 s of (a) city cycl
e (b) high
way cy
cle
Figure 10. Op
timized PCS
and vehi
cle st
ate traj
ecto
rie
s
(a
) city cycl
e (b) hi
gh
way
cycle
In Figure 9, result
s for city
and high
way
cycl
e
s
are compa
r
ed. Th
e city cycle o
f
Figure
9(a
)
has mo
re accele
ratio
n
s/de
cel
e
rati
ons so t
he vehicle can capt
ure mo
re re
g
enerative bra
k
ing
energy. The
r
efore, th
e o
p
timized
sen
s
itivity slope
of
t
he city
cy
cle is relatively
fl
at
co
mpa
r
ed
to
that of the highway cycle,
i.e.,
highway
city
x
x
,
,
(Fig
ure
10). Fig
u
re
9
(
b)
sh
ows th
e re
sults fo
r
the first
200
s of hi
gh
way
cycle, in
which the v
ehi
cle is
lau
n
ching and
the
n
crui
sing at
50 m
ph.
Whe
n
the ve
hicle first lau
n
ch
ed, the po
wer
dema
nd
sud
denly in
creases a
nd th
e battery help
s
to
assist
po
we
r
for the
F
C
S,
of whi
c
h
the
net po
we
r
rat
e
is limited.
Whe
n
the
ve
hicle
crui
se
s, the
pse
udo
-SDP
controlle
r run
s
the FCS
“sl
o
w an
d st
ea
d
y
” while the b
a
ttery operates a
s
an e
n
e
r
gy
buffer to cove
r the fast dyn
a
mics of power dem
and [2
3]-[25].
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
930
Basal Stud
y on Powe
r Co
ntrol Strateg
y
for Fuel Cell/
Battery Hyb
r
i
d
Vehicle
(Di
ngyue Chen
)
429
Figure 11. Op
timized (a) fu
el cell an
d (b
)
battery c
h
arac
teris
t
ics
for city c
y
c
l
e
Figure 12. Effect of DO
H o
n
fuel ec
onomy for c
i
ty c
y
c
l
e
The optimi
z
a
t
ion pro
c
e
s
s down
s
i
z
e
s
the co
mpressor and i
n
cre
a
se
s the de
gree o
f
hybridi
z
ation
(DOH), the DOH is
th
e rati
o of the com
b
ustion e
ngin
e
powe
r
to the
total powe
r
tra
i
n
power. Thu
s
, the FCS effici
ency in
cre
a
ses in the
lo
wer net po
we
r rang
e from 0
to 26 kW, wh
ere
the optimi
z
ed
fuel cell
(F
C) en
gine
prim
arily ope
rate
s (Figu
r
e
11).
The maxim
u
m efficien
cy
of
the optimi
z
ed
FC
engi
ne i
s
aro
und
56%.
Although
the
do
wn
sized
compresso
r
h
e
re
re
du
ce
s t
he
maximum ne
t powe
r
of the FCS, the o
p
timized
p
s
e
udo-S
D
P con
t
roller
su
ccessfully run
s
th
e
FCS
within t
he redu
ce
d
maximum n
e
t
power lim
it.
Figu
re 1
1
(b) sh
ows that
even tho
ugh
the
increased DOH reduces the
batte
ry
size, the optimized
battery
design can still capture the
majority of re
gene
rative brakin
g ene
rgy within its re
du
ced po
we
r limit. If fuel cell vehicles g
o
into
production in the near future,
thei
r
degree of hybri
d
ization
will
signifi
cantly i
m
pact the vehicle
price du
e to
high ma
nuf
acturi
ng a
n
d
material
co
sts of fuel
cells a
nd batt
e
rie
s
[26]-[2
9
].
Therefore, by
examining th
e effect of DOH o
n
fuel e
c
on
omy, ca
r
manufa
c
turers can d
e
term
ine
the trade
-off betwe
en fuel
saving
s an
d manufa
c
turi
n
g
co
sts.
Figure 12 illustrates the ef
f
e
ct of the DOH on fuel
economy
for the
city cycle. To
obtain
each point of
the gra
ph, the DOH valu
e is firs
t set,
and then
other five de
si
gn varia
b
le
s are
optimize
d
to
get the maxi
mum fuel e
c
onomy for th
e sp
ecifi
c
val
ue of DOH.
The results
show
that the o
p
timal DOH i
s
arou
nd
0.653
. Com
pared t
o
the
ba
selin
e de
sig
n
, the
numb
e
r of f
uel
cell
s wa
s increa
sed from 381 to 498, whe
r
ea
s
the
battery cap
a
c
ity could be
decrea
s
ed from
7.035 to
4.87
Ah. As the
DO
H in
crea
ses f
r
om
0.
2 t
o
0.6, the
fue
l
economy
im
prove
s
becau
se
the fuel
cell
efficien
cy in
crea
se
s. Whe
n
t
he
DO
H g
oes beyon
d
0.75,
the fu
el
economy
drops
becau
se de
creased battery
capa
city fails to
capture the rege
ne
rative bra
k
ing e
n
e
r
gy.
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 13, No. 2, June 20
15 : 421 – 43
1
430
5. Conclusion
In this p
ape
r, the PCS m
e
thod
whi
c
h
is impl
ement
ed in A
D
VISOR
enviro
n
m
ent i
s
utilized to design relevant
energ
y control strategies
for FCB
H
V
for the improvement of fuel
eco
nomy a
n
d
milea
ge
of co
ntinuatio
n
of jou
r
ney.
A se
con
d
a
r
y develo
p
men
t
for ADVISO
R i
s
impleme
n
ted based on
the
system archit
ecture
of
FCBHV. We
sug
geste
d a
co
m
p
reh
e
n
s
ive a
nd
system
atic framework t
hat
make
s it po
ssi
ble to opti
m
ize po
we
r
control and
compon
ent si
zin
g
simultan
eou
sl
y for the
de
si
gn of
FCB
H
V
.
The
re
sult
s i
ndicate that t
he p
r
op
osed
control
strate
gy
can
sati
sfy the po
we
r requ
ireme
n
t for t
w
o
stand
ar
d
driving
cycl
es. In two
cycle
con
d
ition
s
, the
PCS for
FCBHV ha
s
sm
aller
co
nsum
ption tha
n
th
e PTC for
F
C
BHV.
Hen
c
e, the p
r
op
o
s
ed
strategy
will give a novel appro
a
ch for the
advan
ce
d energy cont
ro
l system of FCBHV.
Referen
ces
[1]
Ding
y
u
e
C, L
i
feng W
,
Lih
ao
C, Yu S, Jianc
hao
B. T
he De
sign Meth
od of
Exten
d
e
d
Ra
nge El
ectric
Vehic
l
es.
Adva
nced Mater
i
als
Rese
arch
, 20
1
4
; 827: 61-
65.
[2]
Ramos PCA,
Romero A, Gir
a
l R, Ca
lvente
J, Mart
inez-Sal
amero L. Math
ematica
l
an
al
ysis of h
y
bri
d
topol
ogi
es effi
cienc
y for PE
M fuel cel
l
po
w
e
r s
y
stems d
e
sig
n
.
J Electr Power Energy Syst.
2010;
32(5): 10
49-
10
61.
[3]
Hajiz
ad
eh
A,
Golkar MA.
C
ontrol
of
h
y
bri
d
fue
l
c
e
ll/e
n
e
r
g
y
stora
g
e
di
stributed
g
ene
ration
s
y
ste
m
aga
inst volta
g
e
sag.
J Electr Pow
e
r Energy S
yst.
2010; 32(2
)
: 488-49
7.
[4]
Y Guezen
nec,
T
Choi, G Pagan
ell
i
, G Rizz
oni. Pr
oc
eed
in
gs of the Ame
r
ic
an C
ontro
l
Confer
ence.
Denv
er, CO. 2003.
[5]
W
ang Y, Choi
S, Lee E. Effici
ent and ri
ppl
e-
mitigati
ng dc-d
c converter for resid
entia
l fuel
cell s
y
stem.
J Electr Pow
e
r Energy Syst.
2009; 32(
1): 43-
49.
[6]
Xi
ao
lan
W
,
Bingg
an
g C,
Xu
e
y
an
L, Ju
n
X, Xi
aol
on
g R.
Comp
one
nt si
zing
optimiz
ati
on of
plu
g
-i
n
h
y
bri
d
electric
vehicl
es.
Appl
i
ed Ener
gy.
201
1; 88(3): 79
9-8
04.
[7]
Che
n
D, Wu H,
Bao J
e
t
a
l
.
T
h
e Secur
i
ty T
e
ch
nol
ogy
and T
e
n
d
ency
of New
E
n
ergy V
e
h
i
cle
in
the F
u
ture
.
201
3 F
i
fth Inter
natio
nal
C
o
nfer
ence
on
Me
asu
r
ing
T
e
chnol
o
g
y
an
d Mec
hatro
n
i
cs Autom
a
tion.
Ho
ng K
o
n
g
.
2
0
1
3
;
3
9
:
122
7-
122
9.
[8]
T
hounthon
g P
,
Pierfed
e
rici
S, Martin J-P
,
Hin
a
j
e M,
Davat B. Mo
deli
ng
an
d co
ntrol of fu
e
l
cell/su
perca
pa
citor h
y
b
r
id
so
urce
base
d
on
differe
ntial
fla
t
nesscontro
l.
IEEE Trans V
e
h Technol t.
201
2; 59(6): 27
00-2
710.
[9]
Pala
din
i
V, Do
nateo T
,
de Risi A,
Laforgi
a
D. Control strateg
y
o
p
ti
miza
tion of a fuel-
c
ell e
l
ectri
c
vehicl
e.
J Fuel Cell Sci Tec
h
n
o
l
. 200
8; 5 (1): 12-1
9
.
[10]
Dursu
n
E, Kil
i
c
O. Compar
ativ
e eva
l
u
a
tion
of
different po
w
e
r
mana
gem
ent st
rategies
of a
stand-a
l
on
e
PV/W
i
nd/PEMF
C
h
y
br
id p
o
w
e
r s
y
stem.
J Electr Pow
e
r Energy Syst.
201
2; 34(1): 81-8
9
.
[11]
Emadi A, L
e
e
YJ, Rajas
hek
ara Ka
ushik.
Po
w
e
r
electro
n
ics a
nd mot
o
r drives i
n
e
l
e
c
tric, h
y
br
i
d
electric, an
d pl
ug-i
n
h
y
bri
d
el
ectric vehic
l
es.
IEEE Trans In
dust Electron.
200
8; 55(6): 22
37-2
245.
[12]
Neeta K, Pritp
a
l S.
Model
ing
and optim
izati
on of a h
y
bri
d
po
w
e
r s
y
stem
for an unman
ned surfac
e
vehicl
e.
Journ
a
l
of Pow
e
r Sources
. 2012; 1
9
8
(4): 368-
37
7.
[13]
Narasim
ha B,
Vija
ya
n S. A k
n
o
w
l
e
d
ge-
base
d
ob
ject mod
e
l
i
ng
advis
or for
deve
l
op
in
g qu
alit
y o
b
jec
t
mode
ls.
Expert Systems w
i
th Appl
icatio
ns.
2
012; 39(
6): 289
3-29
06.
[14]
JIN J, CHEN
X, ZHANG
L.
M
ode
lli
ng
an
d
simulati
on
of r
ear
w
h
e
e
l
driv
e EMCVT
vehi
cle b
a
se
d o
n
ADVISOR.
Journal of She
n
ya
ng Un
iversity o
f
T
e
chnolo
g
y.
201
2; 34(2): 66
0-66
5.
[15]
Jin
K, Ruan X
,
Yang
M, Xu M.
A h
y
br
id fu
el c
e
ll
po
w
e
r s
y
stem.
IEEE T
r
ans Indust E
l
ectron.
20
09
;
56(4): 12
12-
12
22.
[16]
Dai C, C
h
e
n
W
,
Chen
g Z
,
Li Q
,
Jiang Z
,
Jia J.
Seeker optimi
z
ation alg
o
rith
m
for
glo
bal op
timizatio
n
: a
case stud
y
on
optima
l
mo
de
lli
ng of
pr
oton
ex
change mem
b
rane fuel c
e
ll (PEMFC).
J E
l
ectr Power
Energy Syst.
2011; 33(
1): 369
-376.
[17]
Li Q, Ch
en W
,
W
ang Y,
Liu
S, Jia J. Par
a
mete
r id
entific
ation
for PEM
fuel c
e
ll m
e
ch
anism m
o
d
e
l
base
d
on effec
t
ive informe
d adaptiv
e particl
e
s
w
arm o
p
timiz
a
tion.
IEEE Trans Indust Ele
c
tron.
2011
;
58(6): 24
10-
24
19.
[18]
Li Q, Chen W
,
W
ang Y, Jia
J, Han M. Non
line
a
r rob
u
st control of pr
oto
n
e
x
ch
ang
e m
e
mbra
ne fue
l
cell b
y
state fe
edb
ack e
x
act li
near
izatio
n.
J Power Sour.
2009; 19
4(1): 33
8-34
8.
[19]
Che
n
DY, Ch
e
n
LH, Ba
o JC,
Guo Z
,
T
i
an F
.
Ke
y
T
e
chni
ques
an
d Perf
ormanc
e Ana
l
ysis of Sol
a
r
Electric Automobile.
Advanced Materials Research.
20
14;
846-
847: 1
39-1
43.
[20]
Jia J, Li Q,
Wang Y, Cham
YT
, Han M. Mode
lin
g an
d d
y
namic char
act
e
ristic simul
a
ti
on of proton
ex
c
h
ange membrane fuel cell.
IEEE Trans Energy Conver
s.
2009; 24(
1): 283-
291.
[21]
Z
hao H, B
u
rke
Andre
w
F
.
Opti
mizatio
n
of fu
el
ce
ll
s
y
stem o
p
e
ratin
g
co
nditi
o
n
s for fue
l
ce
ll
vehicl
es.
J
Power Sour
. 2009; 18
6(3): 40
8-41
6.
[22]
Palma L, T
odorovic MH, Enje
ti PN. Anal
y
s
is
of common-mode vo
ltage i
n
ut
ilit
y
-
interacti
v
e fuel cel
l
po
w
e
r cond
itio
ners.
IEEE Trans Indust Elect
r
on
. 200
9; 56(1
)
: 20-27.
Evaluation Warning : The document was created with Spire.PDF for Python.