Internati
o
nal
Journal of Ele
c
trical
and Computer
Engineering
(IJE
CE)
Vol
.
5
,
No
. 5, Oct
o
ber
2
0
1
5
,
pp
. 90
5~
91
7
I
S
SN
: 208
8-8
7
0
8
9
05
Jo
urn
a
l
h
o
me
pa
ge
: h
ttp
://iaesjo
u
r
na
l.com/
o
n
lin
e/ind
e
x.ph
p
/
IJECE
Dependability Evaluation and
Supervision in Thermal Power
Plants
Ma
rw
a
Ben Ha
mmo
u
da
, Mo
ha
med
Na
jeh La
khoua, Li
lia
El Amra
oui
U.R: M
ech
atron
i
c S
y
s
t
em
s
and S
i
gnals
,
ENI Car
t
h
a
ge,
Univers
i
t
y
of Carth
a
ge
,
Tun
i
s
i
a
Article Info
A
B
STRAC
T
Article histo
r
y:
Received Apr 7, 2015
Rev
i
sed
May 18
, 20
15
Accepte
d
J
u
n 2, 2015
In order to improve the productivity
a
nd the con
s
istency
of its maintenance
strateg
i
es, th
e industrial world
is ba
sed on different techn
i
ques
and tools
develop
e
d to ensure safe operati
on
and the supervision of productio
n
s
y
stems. In fact, dependabi
lit
y e
v
alua
tion is cru
c
ia
l to control
l
i
ng the risks
associat
ed with
s
y
stem
failur
e
, and for this
reason, it
is one of the
fundamental steps in au
tomated
s
y
stem
d
e
sign. In th
is paper
,
we presen
t
firstl
y th
e b
a
sic
concep
ts for th
e
stud
y
of
dep
e
nd
abili
t
y
as well
a
s
function
a
l
s
y
stems analy
s
is. Thus, we
present the method
SADT (Structured Analy
s
is
Design techniqu
e). Giv
e
n
the in
effectiv
en
ess of
methods that
are curr
ently
exploit
e
d ar
e n
o
t appropri
a
t
e
becaus
e
the
lev
e
l of
com
p
lexit
y
of s
u
c
h
industrial s
y
s
t
ems, we propo
se in
the first the Safe-SADT
method which
allows the expli
c
it form
ali
zat
ion
of functional in
tera
ction
,
the id
entifi
c
a
tion
of the
char
ac
te
ris
tic v
a
lu
es
aff
ect
ing com
p
lex
s
y
s
t
em
dep
e
nd
abili
t
y
,
th
e
quantifi
c
a
tion
of RAMS param
e
ters
(Reliabi
lit
y,
Availabil
i
t
y
,
Maintain
abi
lit
y,
and Safe
t
y
) f
o
r the s
y
s
t
em
’s operation
a
l
a
r
chit
ectur
e.
Secondly
,
a methodolog
y
for designing
supervisor
y
production sy
stems has
been pres
ented and has been a
pplied on an example of a SCADA
(supervisor
y
con
t
rol
and data acquisition)
s
y
s
t
e
m
of a
therm
a
l
power plan
t.
Finally
,
a model of operating saf
e
ty
and supervision of a production s
y
stem
is proposed.
Keyword:
Aut
o
m
a
ted syste
m
s
Dep
e
nd
ab
ility
Fun
c
tio
n
a
l an
alysis
Sup
e
rv
ision
Therm
a
l Powe
r
plants
Copyright ©
201
5 Institut
e
o
f
Ad
vanced
Engin
eer
ing and S
c
i
e
nce.
All rights re
se
rve
d
.
Co
rresp
ond
i
ng
Autho
r
:
M
a
rwa B
e
n
Ha
m
ouda
U.R: Mechat
ronic System
s
and Si
gnal
s
, E
N
I
C
art
h
age
,
Uni
v
ersi
t
y
o
f
C
a
rt
hage
, T
u
ni
si
a.
Em
ail: m
a
rwa.benham
m
ouda
@hotm
a
il.com
1.
INTRODUCTION
Tod
a
y, th
e d
e
p
e
nd
ab
ility o
f
p
r
od
u
c
tion
syste
m
s is
m
o
re an
d
m
o
re co
mp
lex
to
perform
,
n
o
t
o
n
l
y
because of the
num
ber of va
riables always
m
o
re num
erous to m
onitor
but also
beca
use of the
num
erous
in
terrelation
s
ex
istin
g b
e
t
w
een
th
em
, v
e
ry
d
i
fficu
lt to in
terpret wh
en th
e
pro
cess is
h
i
gh
ly au
to
m
a
ted
.
Ind
eed,
d
e
p
e
nd
ab
ility ch
aracterizes th
e p
e
rfo
r
m
a
n
ce o
f
a system i
n
th
at reflects its ab
ility
to
ach
iev
e
its missi
o
n
. It is
connected t
o
its capacity to resist
the
material, software a
n
d human failures
,
and t
h
e attacks of its
envi
ro
nm
ent
[1
]
.
The conce
p
ts of s
upe
rvision
are also esse
ntial. In
fact, the
supervision can
d
e
tect ano
m
alies with
out
n
ecessarily act d
i
rectly on
th
e
syste
m
in
o
r
d
e
r to op
timize it
s op
eration
and to
en
su
re t
h
e safety [2
].
In th
is
p
a
per,
we
p
r
esen
t i
n
a first
p
a
rt th
e m
a
in
d
e
fin
ition
s
and
co
n
c
epts o
f
d
e
p
e
n
d
abilit
y. In th
e
second
pa
rt,
we pre
s
ent t
h
e architecture a
n
d
the cha
r
acterist
i
cs o
f
sup
e
rv
isi
o
n. In
t
h
e th
ird p
a
rt, we presen
t th
e
m
e
thods a
n
d tools ens
u
ring the depe
ndability analysis fo
r
products syste
m
s
under
de
sign or re
desi
gn.
Finally,
we p
r
esen
t a g
e
n
e
ral m
o
d
e
l in
o
r
d
e
r to
stu
d
y
th
e d
e
p
e
nd
ab
ility an
d
the su
p
e
rv
isi
o
n
for an
ex
am
p
l
e o
f
an
in
du
strial in
stallatio
n
and
t
o
hav
e
a
safe i
n
dustrial syste
m
, a sup
e
rv
ision
m
e
th
od
was im
p
l
e
m
en
ted
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
I
J
ECE Vo
l. 5
,
N
o
. 5
,
O
c
tob
e
r
20
15
:
905
–
9
17
90
6
2.
DEPENDABILITY OF COMP
LE
X
AUT
O
MATED SYSTEMS
Dep
e
nd
ab
ility ai
m
s
to
estab
lish
tru
s
t
b
e
tween
th
e c
lien
t
an
d th
e
d
e
sign
er. Th
is is
wh
y
it is
m
o
v
i
ng
towa
rds
m
e
thodol
ogies e
n
s
u
ring e
ffici
ency, accura
cy and
reliability. To
ach
ieve t
h
ese
objectives
, functional
an
alysis is a
g
o
o
d
way t
o
d
e
si
g
n
, ev
alu
a
te and
m
a
n
a
g
e
the
dep
e
nd
ab
ility o
f
co
m
p
lex
au
t
o
mated
syste
m
s.
The n
o
t
i
o
n of
i
ndu
st
ri
al
pr
o
cess cove
rs a wi
de ra
n
g
e of
i
ndu
st
ri
al
sect
ors.
In
dee
d
, a
sy
st
em
i
s
defi
ned as a
set of interc
onne
c
t
ed elem
ents
, orien
t
ed towards ach
iev
i
ng
an
o
b
j
ectiv
e.
In
d
u
st
ri
al
sy
st
em
s are pr
od
uct
i
on t
o
ols that a
r
e subject to econom
i
c
and technical constraints. They
m
u
st
be abl
e
t
o
pr
od
uce
q
u
al
i
t
y
goo
ds
at
hi
gh
spe
e
ds
re
qu
i
r
ed.
They
a
r
e
al
so rel
a
t
e
d
t
o
m
odern a
n
d ef
fi
ci
ent
p
r
od
u
c
tion
tech
n
i
q
u
e
s th
at req
u
i
re a h
i
g
h
lev
e
l of kno
wled
g
e
and
sk
ills. In
fact, t
h
e co
m
p
lex
ity o
f
i
n
dustrial
p
r
o
cesses and
th
e v
a
riety o
f
co
nd
itio
ns fo
r t
h
eir op
eratio
ns are stead
ily in
creasing
.
Th
is
is wh
y th
ese syste
m
s
are often
aut
o
mated.
R
e
searche
r
s C
a
uf
fri
ez L. a
n
d al
. [
3
]
hav
e
prese
n
t
e
d a
m
e
t
hod
ol
o
g
y
fo
r desi
gni
n
g
com
p
l
e
x
au
to
m
a
ted
syste
m
s. Th
is m
e
t
h
odo
log
y
is
d
i
v
i
d
e
d
in
t
o
si
x
actig
rammes (Fig
ure
1
)
. Ind
e
ed
, actigrammes 1
and
2
co
n
c
ern
th
e fun
c
tion
a
l arch
itectu
r
e, actig
ramm
es 3
an
d 5 relate t
o
th
e h
a
rd
ware arch
itectu
r
e and
actigrammes 4
and 6 conce
r
n th
e ope
rational architecture.
Fig
u
re
1
.
A m
e
th
od
o
l
o
g
y
fo
r t
h
e
d
e
sign
o
f
intellig
en
t d
i
stribu
ted
con
t
ro
lled syste
m
[3
]
2.
1
Dependability Co
ncepts
Dep
e
nd
ab
ility is un
d
e
rstan
d
i
n
g
, assessin
g
,
p
r
ed
icting
,
m
easu
r
ing and
co
n
t
ro
l
failu
res
o
f
tech
no
log
i
cal syste
m
s an
d
human
failu
res [4
]. Dep
e
n
d
a
b
ility p
l
ays an
im
p
o
r
tan
t
ro
le in
ind
u
s
t
r
ial syste
m
s.
Ind
e
ed
, it is th
e scien
ce o
f
fai
l
u
r
e and
is ch
aracterized
b
y
th
e ab
ility o
f
an en
tity to
satisfy o
n
e
or m
o
re o
f
th
e
v
ital syste
m
fu
n
c
tio
ns und
er
g
i
v
e
n
co
nd
ition
s
.
Fou
r
in
terdep
end
e
n
t
p
a
rameters d
e
fi
n
e
th
e RAMS: Rel
i
ab
ility,
Av
ailab
ility,
Main
tain
ab
ility an
d Safety.
2.
1.
1
Relia
bility
Accord
ing
to
NF EN
13
306
, reliab
ility o
f
a syste
m
is th
e ab
ility fo
r it to p
e
rfo
rm
its fun
c
tio
n und
er
g
i
v
e
n
cond
itio
n
s
during
a giv
e
n
tim
e
in
terv
al [5
]. Th
e term
rel
i
ab
ili
ty is also
u
s
ed
to
d
e
term
in
e th
e
sy
st
em
/
e
qui
pm
ent
conce
r
ned
,
a guara
nt
eed
ope
rat
i
n
g t
i
m
e
before the oc
currence
of a failure. C
onsi
d
er the
ti
m
e
T o
f
o
c
curren
ce
o
f
t
h
e failu
re; th
is rando
m
v
a
riab
le to
d
e
fi
n
e
th
e concep
t of reliab
i
l
ity th
at is in
terp
reted
as th
e
p
r
ob
ab
il
ity th
at th
e en
t
ity d
o
n
o
t
crash
after a
g
i
ven ti
m
e
t fo
r
g
i
ven
co
nd
ition
s
.
Reliab
ilit
y is th
en
d
e
fi
n
e
d as th
e
p
r
ob
ab
ility asso
ciated
R
(t) t
o
th
is
n
o
tion
.
It i
s
d
e
fin
e
d
b
y
:
R (t) = prob
ab
ility o
f
a
n
o
n
-
d
e
fau
lting
en
tity o
n
th
e i
n
terv
al
[0, t], assu
m
i
n
g
it is
n
o
t
i
n
d
e
fau
lt at tim
e t=
0
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
Dep
e
nd
ab
ility Eva
l
ua
tion
a
nd Su
p
e
rvisio
n in Th
erma
l Po
wer Plan
ts
(
M
arw
a
Be
n
H
a
m
o
ud
a)
90
7
2.
1.
2
Ava
ila
bility
Availability, ra
ted A (t) is the
ability of a c
o
m
ponent
or sys
t
em
to be
up a
n
d runni
ng at a
give
n tim
e.
Mo
re
sp
eci
fically, av
ailab
ility
is th
e ab
ility to
respond
to the req
u
e
st in time. Th
e AFNO
R
X
6
0
-5
00
d
e
fi
n
e
s
av
ailab
ility as "th
e
ab
ility o
f
an
en
tity to
b
e
ab
le to
acco
m
p
lish
a requ
ired
fun
c
tion
u
n
d
e
r g
i
v
e
n
con
d
ition
at
a
gi
ve
n t
i
m
e or
d
u
ri
ng
a
gi
ve
n t
i
m
e
i
n
t
e
rval
,
as
sum
i
ng t
h
e
p
r
o
v
i
d
i
n
g t
h
e s
u
p
p
o
r
t
neede
d
m
a
i
n
t
e
nan
ce i
s
a
ssure
d
"[6].
Th
e term
av
ail
a
b
ility is
th
en
d
e
fi
n
e
d
as a prob
ab
ility
th
at an
en
tity
m
a
y
d
e
term
in
e, b
a
sed
on
the
reliab
ility o
f
the resu
lts, t
h
e
qu
ality o
f
serv
ice p
r
ov
id
ed
b
y
t
h
e system
/ eq
u
i
p
m
en
t, un
d
e
r g
i
v
e
n
co
nd
ition
s
at
a g
i
v
e
n
tim
e.
Th
e prob
ab
ilit
y asso
ciated
with
A (t) at ti
me t is
also
referred
to
av
ailabilit
y an
d
is ex
p
r
essed
by
:
A (t
) =
P
(no
-
defau
lting
E at ti
m
e
t)
2.
1.
3
Ma
inta
inability
Main
tain
ab
ility M (t) is th
e
ab
ility o
f
a co
m
p
on
en
t or
system
to
b
e
m
a
in
ta
in
ed
or
restored
co
nd
ition
.
Nex
t
AFNOR
stan
d
a
rd
is: "
u
n
d
e
r th
e g
i
v
e
n
cond
itio
n
s
of u
s
e, th
e ab
ility o
f
an
en
tity
to
b
e
m
a
in
tai
n
ed
o
r
restored in
a g
i
v
e
n ti
m
e
in
terval, in
a
state in wh
ich
it can
perfo
r
m
a requ
ired
fun
c
tion
when
th
e m
a
in
ten
a
n
c
e
i
s
per
f
o
rm
ed u
nde
r
gi
ve
n c
o
n
d
i
t
i
ons
wi
t
h
p
r
escri
b
e
d
pr
oce
d
u
r
es a
n
d
reso
urces
"[
7]
.
Main
tain
ab
ility o
f
a serv
iceab
l
e en
tity is characteri
zed b
y
a prob
ab
ility
M (t) t
h
e m
a
in
ten
a
n
c
e of an
en
tity E co
m
p
leted
in
g
i
v
e
n
co
nd
itio
ns, with p
r
o
cedures an
d
prescri
b
ed
m
ean
s, is co
m
p
leted
at ti
m
e
t ,
g
i
v
e
n
th
at E fails t
o
t
i
m
e
t = 0
[2
]. Th
is can
b
e
ex
pressed
b
y
: M (t
) = P
(E m
a
in
ten
a
n
c
e was co
mp
leted
at tim
e t
)
.
2.
1.
4
Sa
fet
y
(Sec
ur
ity)
Security is the
ability of a
product to m
eet
during
all
pha
ses of life, a
n
acceptable le
vel of safety
hazards likely
to cause
staff a
ssault or
a m
a
jor
de
gra
d
ation
of t
h
e product
or
its e
nvironment. Security is "the
ab
ility o
f
an
en
tity to
avo
i
d
ap
p
e
aring
i
n
g
i
v
e
n co
nd
itio
n
s
, critical or cat
astrop
h
i
c ev
ents» [7
].
Th
e role o
f
a
specialist SdF
is to re
duce
the business risk to its accep
ta
ble level by de
fi
ning: the
risk a
cceptability criteria,
security de
sign m
e
thods a
n
d residual
risk ass
e
ssm
ent
m
e
thods a
n
d
verifica
tion levels
acc
essibility.
2.
2
Met
h
o
d
ol
og
y for
Desi
gni
n
g Com
p
l
e
x
S
y
st
ems
To
stud
y th
e
d
e
p
e
nd
ab
ility o
f
co
m
p
lex
auto
m
a
ted
system
s, sev
e
ral
meth
od
s based
o
n
static and
dy
nam
i
c
m
odel
i
ng ha
ve
bee
n
p
u
t
i
n
pl
ace
. In
dee
d
,
fu
nc
t
i
onal
anal
y
s
i
s
i
s
based
pri
m
ari
l
y
on t
h
e
needs
analysis, funct
i
onal analysis necessa
ry
technical and func
tional analysis. F
eatures
suc
h
as a high
degree of
d
a
ta ou
tpu
t
tru
s
two
r
t
h
in
ess, serv
ice av
ailab
ility, a
n
d
protectio
n
are th
e resu
lt o
f
a
h
i
g
h
d
e
gree of serv
ice
d
e
p
e
nd
ab
ility wh
ich
is a
v
e
ry i
m
p
o
r
tan
t
p
r
o
p
e
rty in
th
e
clo
u
d
co
m
p
u
tin
g env
i
ro
n
m
en
t. Th
us it h
i
gh
lig
h
t
s
som
e
m
e
thods
, us
ually base
d on a gra
phi
cal repre
s
enta
t
i
on, i
n
orde
r
to facilitate the understanding of
co
m
p
lex
au
t
o
mated
syste
m
s an
d th
e
stud
y of th
eir d
e
p
e
n
d
ab
ility [8
].
2.
2.
1
Presentati
on
of the SADT Method
The S
A
D
T
m
e
t
h
o
d
i
s
a
m
e
t
hod
of
desi
g
n
a
nd a
n
al
y
s
i
s
of
com
p
l
e
x sy
st
em
s. It
was dev
e
l
ope
d by
D
T
R
o
ss i
n
t
h
e S
o
ft
ech com
p
any
i
n
t
h
e Uni
t
e
d
St
at
es arou
n
d
19
7
4
[
9
]
.
It
i
s
a
m
e
t
hod o
f
d
e
scen
di
n
g
fu
nc
t
i
onal
anal
y
s
i
s
t
h
at
st
art
s
f
r
om
t
h
e
gene
ral
t
o
t
h
e
part
i
c
ul
a
r
[
1
0]
.
I
ndee
d
,
t
ech
n
i
cal
sy
st
em
can
be
descri
bed
by
a
SA
DT m
odel
rep
r
ese
n
t
e
d
by
a di
a
g
ram
assem
b
l
y
(ac
t
i
g
ram
m
e and dat
a
gram
) hi
era
r
c
h
i
cal
l
y
or
dere
d.
The
m
e
t
hod
ol
o
g
y
i
s
ri
g
o
r
o
us,
wi
t
h
speci
fi
c sc
hem
a
sy
nt
ax r
u
l
e
s t
h
at
al
l
o
w a S
A
DT m
odel
t
o
c
o
m
m
uni
cat
e
conci
s
el
y
a
n
d
preci
sel
y
i
m
port
a
nt
i
n
fo
rm
at
ion
o
n
t
h
e
ope
r
a
tio
n
an
d th
e ex
tern
al in
terfaces of th
e system
[1
1
]
.
Processors or mechanism
s
are
natural and t
echnological ele
m
en
t
s
t
h
at
pe
rf
orm
t
h
e fu
nct
i
on
wh
ich
is
charact
e
r
i
zed
pri
m
ari
l
y
by
an act
i
on o
n
t
h
e i
nput
s
.
The
ent
r
y
co
nsi
s
t
s
of t
h
e i
n
c
o
m
i
ng pri
m
e
m
a
t
t
e
r. It
i
s
m
odi
fi
ed by
t
h
e f
unct
i
o
n
an
d
can
be c
o
nsi
d
e
r
ed:
p
r
o
d
u
ct
(mater
ial)
, en
erg
y
or
i
n
fo
r
m
ati
o
n. Th
e ou
tpu
t
show
s
th
e ou
tgo
i
ng
prim
e
m
a
terial
t
h
at is
m
a
in
ly
work
i
n
g
m
a
terials p
r
ov
id
ed
with
its ad
d
e
d v
a
lu
e.
On
e can
fi
nd
with
th
is
o
u
t
go
ing
prim
e
ma
terial fo
r
reports, en
erg
y
lo
ss
es
an
d
wast
e. C
ont
r
o
l
s
or
c
o
nst
r
ai
nt
s or st
eeri
n
g
cont
rol
are t
h
e
param
e
t
e
rs t
h
at
t
r
i
gger
o
r
m
odi
fy
t
h
e r
ealization of a function. T
h
ey
are classified int
o
four
categ
ories: th
e con
f
i
g
uration
p
a
ram
e
ters; tu
n
i
ng
p
a
ram
e
ters; op
erating
d
a
ta / o
p
erating
v
a
lu
e; setting
en
erg
y
or
prese
n
ce of prim
e
m
a
tter.
2.
2.
2
Present
a
ti
on o
f
S
a
fe
-S
ADT
Met
h
o
d
Meth
od
s fo
r assessin
g
th
e
d
e
p
e
nd
ab
ility cu
rren
tly op
erating
are
no
t appro
p
riate, g
i
v
e
n
th
e lev
e
l th
e
com
p
l
e
xi
t
y
of
i
n
d
u
st
ri
al
sy
st
e
m
s. Ineffi
ci
e
n
t
Exi
s
t
i
n
g m
e
t
hods
have
l
e
d t
o
t
h
e
de
vel
o
p
m
ent
and
pr
op
ose t
h
e
Safe- SADT meth
od
[12
]
, [13], [14
]
. Th
is meth
od
allows
t
h
e exp
licit fo
rm
alizatio
n
of functio
n
a
l in
teract
io
n
s
,
id
en
tificatio
n
o
f
ch
aracterist
i
c v
a
lu
es
th
at affect t
h
e
d
e
pen
d
a
b
ility o
f
co
m
p
lex
system
s, q
u
a
n
tificat
io
n
of
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
I
J
ECE Vo
l. 5
,
N
o
. 5
,
O
c
tob
e
r
20
15
:
905
–
9
17
90
8
p
a
ram
e
ters RAMS (Reliab
ility, Av
ailab
ility, Main
tain
ab
ility
an
d Safety)
of th
e op
eratin
g
syste
m
arch
itectu
r
e,
an
d
op
eration
a
l v
a
lid
atio
n
o
f
th
is arch
itecture in
term
s o
f
d
e
p
e
nd
ab
ility o
b
j
ectives and
co
n
s
t
r
ain
t
s set in
th
e
fun
c
tion
a
l sp
ecificatio
n
[14
]
.
Th
e m
a
in
pu
rpo
s
e
o
f
th
is m
e
t
h
od
is t
o
ev
aluate th
e d
e
p
e
n
d
ab
ility p
a
ram
e
t
e
rs
of
a g
l
ob
al system
d
u
r
ing
th
e desig
n
ph
ase takin
g
i
n
to
accoun
t its evo
l
u
tion in
tim
e (Fig
ure 2).
Figu
re
2.
S
A
F
E
-S
ADT
f
o
rm
alism
To m
eet
t
h
e object
i
v
es speci
fi
ed i
n
a s
p
eci
f
i
cat
i
on, t
h
e S
a
fe-S
A
D
T m
e
t
hod
de
fi
ne
d t
h
e
ope
rat
i
o
nal
architecture of a syste
m
that
is obtai
ne
d
fr
o
m
t
h
e pro
j
ect
i
o
n o
f
a f
unct
i
o
n
a
l
archi
t
ect
ure
on a set
of re
so
urce
s
.
Gi
ve
n t
h
at
a
n
y
fu
nct
i
o
n
per
f
o
r
m
e
d by
so
ft
w
a
re em
bedde
d i
n
t
h
e m
a
t
e
ri
al
is defi
ne
d by
a
structure
d
seri
es, the
reliab
ility
m
u
s
t
b
e
adv
a
n
ced acro
s
s th
e sy
ste
m
[1
2
]
.
A
Safe-SADT
b
l
o
c
k
is t
o
th
e
sp
ecification
of the
reliab
ility targ
et for st
u
d
y
ing
t
h
e
fun
c
tion
s
and
q
u
a
n
tification
o
f
RAMS
p
a
ram
e
ters o
f
functio
n
s
.
The i
n
put
fo
r t
h
e Safe
- S
A
D
T
bl
oc
k co
nt
ai
ns t
h
e
di
ffe
re
n
t
funct
i
ons t
h
at
t
h
e sy
st
em
can r
un
(s
pee
d
co
n
t
ro
l, tem
p
eratu
r
e con
t
ro
l, etc.). Th
e obj
ectiv
e d
e
fi
n
e
d
reliab
ilities fo
r th
ese serv
ices
are often
ind
i
cated
in
t
h
e s
p
eci
fi
cat
i
o
ns
of
t
h
e
f
unct
i
onal
re
qui
rem
e
nt
s a
n
d
R
A
M
S
pa
ram
e
t
e
r vect
or as
soci
at
i
o
n.
The
out
p
u
t
f
o
r Sa
fe-S
A
D
T
bl
oc
k s
h
o
w
s t
h
e e
vol
ut
i
o
n
of t
h
e R
A
M
S
param
e
t
e
rs o
f
f
u
nct
i
ons
per
f
o
r
m
e
d, t
h
e
resul
t
s
obt
ai
n
e
d by
t
h
e p
r
o
j
ect
i
o
n
/
de
sired
allo
cation
,
as well as th
e respon
se time and
co
n
t
ro
llab
l
e and
un
con
t
ro
llable ev
en
ts. Th
e Safe-SADT fo
rm
alis
m allo
ws th
e d
e
signer to
d
e
fin
e
,
fo
r each
subsystem
a
t
each
hierarc
h
ica
l
level, som
e
param
e
ters
that will be a
g
gregated to
determ
i
n
e the
val
u
e re
sulting
fr
om
t
h
e gl
ob
al
set
t
i
ng su
b-
sy
st
em
funct
i
ons
o
r
m
odes
of
o
p
erat
i
o
n.
The re
sp
o
n
se
t
i
m
e
s and co
st
s are
con
v
e
n
i
e
nt
t
o
eval
uat
e
. T
h
e
resp
o
n
se t
i
m
e
of eac
h s
u
b
-
s
y
st
em
depen
d
s
on t
h
e t
ech
no
l
ogy
an
d
ope
r
a
t
i
ng
syste
m
[13].
A
t
th
e to
p
of
th
e Saf
e
-SADT b
l
o
c
k
,
th
ere ar
e n
o
n
-
con
t
ro
llab
l
e ev
en
ts su
ch
as r
a
ndom h
a
r
d
w
a
r
e
fai
l
u
res
or s
o
ft
ware a
nd c
o
nt
rol
l
a
bl
e eve
n
t
s
such c
o
r
r
ect
i
v
e act
i
ons
, w
h
i
c
h bri
ng t
h
e sy
st
em
t
o
a wo
rki
n
g
co
nd
itio
n. Constrain
t
s and
cri
t
eria (eg, reliab
ility, ch
aracterizatio
n
systems for
p
e
rfo
rm
a
n
ce, sub
s
ystems) are
al
so c
onsi
d
ere
d
wi
t
h
t
h
e
co
nt
rol
s
o
n
t
o
p
o
f
a
Safe
- S
A
DT
b
l
ock.
At th
e b
o
tto
m
o
f
th
e b
l
o
c
k, the eq
u
i
p
m
en
t an
d
th
eir d
e
p
e
ndab
ility ch
aracteristics are k
n
o
wn
fro
m
th
e
expe
ri
m
e
nt
al
resul
t
s
, l
a
b
o
rat
o
ry
t
e
st
or m
a
nu
fact
u
r
er
dat
a
.
The form
alism
used
to
m
o
d
e
l th
e
red
und
an
cy
execut
i
o
n m
e
di
um
t
o
st
udy
t
h
e im
pact
of
re
d
u
n
d
a
n
cy
o
n
R
A
M
S
param
e
t
e
rs [
1
2]
.
The
use
o
f
T
o
p-
Do
w
n
a
p
p
r
o
ach
fo
r t
h
e
de
com
posi
t
i
on
o
f
t
h
e
sy
st
em
, st
art
i
ng i
n
gl
o
b
a
l
l
e
vel
A
0
u
n
til th
e elem
e
n
tary fun
c
tion
s
of lev
e
l
An
.
Th
is ap
pro
a
ch
allo
ws t
h
e ex
p
l
i
c
it
m
o
d
e
lin
g of th
e ch
aracteri
s
tic
s
of t
h
e o
p
e
r
at
i
onal
arc
h
i
t
ect
ure an
d i
d
e
n
t
i
f
i
cat
i
on o
f
i
t
s
depe
nde
nci
e
s
i
n
orde
r t
o
bet
t
e
r u
nde
rst
a
nd t
h
e
d
e
p
e
nd
ab
ility o
f
th
e ov
erall arch
itecture. On
ce th
e lo
west
lev
e
l o
f
th
e Safe- SADT
b
l
o
c
k
is reach
e
d, it is
pos
si
bl
e t
o
assess t
h
e R
A
M
S
param
e
t
e
rs of t
h
e sy
st
em
usi
ng t
h
e B
O
T
T
OM
-
U
P ap
pr
o
ach.
During
this stag
e
of a
g
g
r
eg
at
i
o
n
,
desi
g
n
ers ca
n veri
fy
that the
specifications and c
o
nstrai
nts
im
posed by the entire syste
m
are
satisfied. T
h
e
designer ca
n al
so at this
stag
e
to
qu
an
tify th
e
respon
se tim
e
an
d
esti
m
a
te th
e co
st of alternativ
es
f
o
r
th
e op
er
a
tio
n
a
l ar
ch
ite
c
t
u
r
e
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
Dep
e
nd
ab
ility Eva
l
ua
tion
a
nd Su
p
e
rvisio
n in Th
erma
l Po
wer Plan
ts
(
M
arw
a
Be
n
H
a
m
o
ud
a)
90
9
3.
THE IN
DU
ST
RIAL
S
U
PER
V
ISI
O
N
Su
pe
rvi
s
i
o
n c
onsi
s
t
s
of c
o
m
m
a
ndi
n
g
a p
r
o
cess an
d su
pe
r
v
i
s
i
n
g i
t
s
wo
rk
i
ng [
1
5]
, [
16]
.
To ac
hi
ev
e
t
h
i
s
g
o
al
, t
h
e s
upe
r
v
i
s
o
r
y
sy
st
em
of a p
r
oce
ss m
u
st
col
l
ect, su
per
v
i
s
e a
n
d rec
o
r
d
i
m
por
t
a
nt
so
urces
o
f
dat
a
lin
k
e
d
to
th
e
pro
cess, to
d
e
tect th
e p
o
ssib
l
e lo
ss o
f
fun
c
tio
n
s
an
d
alert th
e hu
m
a
n
o
p
e
rato
r. In
add
ition
,
th
e
su
perv
ision
of in
du
strial p
r
ocesses allo
ws
th
e con
t
r
o
l
an
d t
h
e c
o
m
m
a
nd o
f
aut
o
m
a
t
e
d p
r
oce
ss i
n
o
r
de
r t
o
o
p
tim
ize its o
p
e
ratio
n and
t
o
en
su
re t
o
secu
rity. Th
e m
a
in
parts of a
supe
rvised sy
stem
are [17]
, [
18]
, [
1
9]
:
•
The Ma
n-Mac
h
ine
Interface
s
(MMI),
d
i
sp
layin
g
info
rm
ati
o
n th
an
ks to the in
fo
rm
atio
n
syn
t
h
e
sis system
.
•
Th
e sup
e
rv
isory to
o
l
s,
supp
lyin
g
serv
ices than
ks to
t
h
e a
u
t
o
m
a
tic supe
rvi
s
ory system
and the
decision
support system
s.
•
The c
o
ntrol/c
ommand pa
rt, interface
betwee
n the
MMI, the
supervis
ory tools a
n
d the
process.
•
Th
e
pr
o
cess is
also
called pr
od
u
c
tion system or
o
p
e
r
a
tiv
e par
t
,
p
e
rf
or
m
i
n
g
th
e
ph
ysical wo
rk
on
the input
pr
o
duct
fl
o
w
.
Th
ese task
s are p
e
rfo
r
m
e
d
by hu
m
a
n
o
p
e
rato
rs wh
o m
a
k
e
d
ecision
s after ev
al
u
a
tin
g th
e situ
atio
n
of
t
h
e p
r
ocess
fr
o
m
t
h
e o
b
ser
v
e
d
va
ri
abl
e
s,
usi
n
g e
xpe
rt
kn
o
w
l
e
dge
t
o
sol
v
e c
o
m
p
l
e
x si
t
u
at
i
ons
[
2
0]
, [
2
1]
.
Sup
e
rv
isory co
n
t
ro
l and
d
a
ta acq
u
i
sitio
n syste
m
s
(SCADA) are wi
d
e
ly u
s
ed
in
in
du
stry fo
r
su
perv
isory
con
t
ro
l
and
d
a
ta acq
u
i
sition
o
f
i
n
du
strial p
r
o
c
esses. Th
e
pro
c
ess can b
e
indu
strial, infrastru
cture
o
r
facility [1
5
-
1
6
]
. SC
ADA syste
m
is u
s
ed
to
ob
serv
e an
d su
p
e
rv
ise th
e
sh
op
floo
r eq
uip
m
en
ts in
v
a
rio
u
s
i
n
d
u
st
ri
al
aut
o
m
a
t
i
on appl
i
c
a
t
i
ons [
1
7-
18]
.
A SC
A
D
A
sy
st
em
i
s
used t
o
m
oni
t
o
r an
d
supe
r
v
i
s
e an
ove
ral
l
p
r
o
cess
b
e
ing
im
p
l
e
m
en
ted
by in
d
i
v
i
du
al auto
m
a
ted
syste
m
s. It
typ
i
call
y
co
n
s
ists of th
e fo
llowing
four ite
ms
[2
2]
, [2
3]
, [2
4]
:
A m
a
ster term
i
n
al
u
n
it
(MTU),
wh
ich
is th
e
cen
tral
se
rve
r
whe
r
e i
n
f
o
rm
at
i
on a
b
out
t
h
e
ove
ral
l
p
r
ocess
i
s
collected.
This
allows
a ce
nt
ralized
work
statio
n
op
erator
to
m
o
n
itor,
a
n
alyze, a
n
d control the
entire
pr
ocess
f
r
om
a rem
o
t
e
l
o
cat
i
on.
PLC
s
a
nd/
or R
T
Us,
w
h
i
c
h
co
nt
r
o
l
t
h
e
fi
el
d e
qui
pm
ent
doi
n
g
t
h
e
p
r
ocess.
C
o
m
m
uni
cat
i
o
ns e
q
ui
pm
ent
for
t
r
an
sfe
rri
n
g
dat
a
bet
w
ee
n t
h
e P
L
C
s
/
R
TUs
an
d t
h
e
M
T
U
.
Hum
a
n m
achine inte
rface
(HMI) s
o
ft
ware
,
whic
h e
n
able
s th
e
on-screen
o
p
e
ration
o
f
the
in
pu
ts (wit
h
o
u
t
phy
si
cal
l
y
t
ouc
hi
n
g
t
h
em
) and
di
spl
a
y
s
t
h
e st
at
us of t
h
e
out
put
s
.
The HM
I
i
s
i
n
st
al
l
e
d on al
l
wor
k
st
at
i
o
n
s
(in
c
lud
i
ng
t
h
e
MTU), allowing
th
e op
erato
r
s to
h
a
v
e
easy an
d in
tu
itiv
e con
t
ro
l ov
er th
e syste
m
s.
4.
PROP
OS
AL FOR
DEPE
N
DABILIT
Y M
O
DEL AN
D S
U
PER
V
IS
IO
N
Th
e
d
e
v
e
l
o
p
m
en
t of a m
o
d
e
l h
e
lp
s
u
s
to
facilitate
th
e an
alysis, d
e
scrip
t
i
o
n
and
und
erst
an
d
i
n
g
of th
e
in
du
strial system
[2
5
]
. Fo
rm
alis
m
,
co
n
c
ep
ts
an
d rep
r
esen
tat
i
o
n
t
o
o
l
s th
at
we
will u
s
e
wi
ll h
e
lp
t
o
d
eepen
th
e
st
udy
sy
st
em
in t
h
e desi
gn st
age. O
u
r g
o
al
i
s
t
h
e ri
go
ro
us
use of a m
e
t
hod
ol
o
g
i
cal
ap
p
r
oac
h
t
o
f
u
n
c
t
i
onal
an
alysis an
d
su
p
e
rv
isio
n
o
f
an
ind
u
s
t
r
ial facility. Th
is
meth
od
o
l
o
g
y
will b
e
ad
op
ted
t
o
an
alyze produ
ction
sy
st
em
s t
o
pr
ovi
de a t
o
ol
f
o
r c
o
m
m
uni
cat
i
on
bet
w
ee
n t
h
e
vari
o
u
s
use
r
s o
f
t
h
e sy
st
em
. W
e
q
u
o
t
e
as a
n
exam
ple, thermal powe
r
ge
ne
ration of elect
rical ene
r
gy.
Fig
u
re 3
shows th
e three
essen
tial p
h
a
ses th
at are cen
t
ral to
th
e
dev
e
lop
m
en
t o
f
th
e m
o
d
e
l
:
fun
c
tion
a
l an
al
ysis o
f
t
h
e syst
e
m
; an
alysis o
f
th
e
o
p
e
ration
a
l arch
itecture;
su
perv
ision
o
f
th
e ind
u
strial plan
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
I
J
ECE Vo
l. 5
,
N
o
. 5
,
O
c
tob
e
r
20
15
:
905
–
9
17
91
0
Fi
gu
re
3.
M
o
d
e
l
of
anal
y
s
i
s
a
n
d
su
pe
rvi
s
i
o
n
of
an
i
n
dust
r
i
a
l
p
r
oces
s
In t
h
e
fi
rst
p
h
a
se, we
pr
oce
e
d wi
t
h
a
n
an
al
y
s
i
s
and f
u
n
c
t
i
onal
m
odel
i
ng
of t
h
e i
n
du
st
ri
al
sy
st
em
base
d on t
h
e S
ADT m
e
t
hod.
The st
ru
ct
ure
of t
h
i
s
m
e
t
hod
t
o
m
a
st
er t
h
e
com
p
l
e
xi
t
y
of t
h
e pr
ocess wi
t
h
i
t
s
t
o
p
-
d
o
w
n
an
d
m
odul
ar a
n
al
y
s
i
s
.
In the
first
pla
ce, we
pre
p
are
a SADT m
o
del to
de
fine
the objecti
v
es a
n
d the
view for whic
h the
m
odel is crea
ted. The
n
we
create the A-0
diagram
re
p
r
esen
ting
th
e
m
a
in
fu
n
c
tion
fo
r an
alysis in
an
actig
ramme. Th
e
d
i
graph
A0
an
d A1
d
e
v
e
lop
m
en
t is th
e third
step
in th
is
p
h
a
se.
Th
is an
alysis allo
ws
u
s
to ob
t
a
in
a fun
c
tio
n
a
l ar
chitecture
a
n
d ha
rdwa
re ar
chitecture
of the system
.
An
exam
pl
e o
f
m
odel
i
ng
of
a
t
h
erm
a
l
powe
r
plant
of electric powe
r is
underway.
Once t
h
e funct
i
onal arc
h
itecture a
nd
hardware arc
h
itect
ur
e have
been i
d
ent
i
f
i
e
d, t
h
e ne
xt
st
ep i
s
t
o
d
e
term
in
e th
e
o
p
e
ration
a
l arch
itectu
r
e
of an
in
du
strial pro
c
ess u
s
i
n
g th
e new Safe- SADT fo
rm
alis
m
.
The fi
rst
st
ep o
f
suc
h
a f
u
nct
i
onal
a
n
al
y
s
i
s
app
r
oach
by
Saf
e
- SA
DT m
e
t
hod i
s
t
o
i
d
ent
i
f
y
t
h
e sy
st
em
and t
h
e f
u
nct
i
ons
t
o
be
per
f
o
rm
ed an
d t
h
e
m
a
t
e
ri
al
s us
ed m
e
dia. The
n
the second
st
ep
is to
d
e
fi
ne th
e
o
p
e
ration
a
l arch
itectu
r
e b
y
th
e
proj
ection
(allo
cation
)
of t
h
e
funct
i
onal a
r
chitect
ure
of t
h
e
ha
rdware
architecture. Finally, we
proceed
w
ith the
devel
opm
ent of Sa
fe-
SADT
m
odel.
When we a
r
e faced
with
a
SA
DT
Safe
-
b
l
ock c
o
m
p
l
e
x,
we
use t
h
e sa
m
e
decom
posi
t
i
on
pr
ocess
o
f
t
h
e
SA
DT m
e
t
h
o
d
.
The
b
o
t
t
o
m
-
up
ap
pro
ach is th
en
u
s
ed
to agg
r
eg
ate th
e
resu
ltin
g b
a
sic fu
n
c
ti
o
n
s
of t
h
e
p
r
evio
u
s
dec
o
m
pos
i
t
i
on.
Th
is seco
nd
ph
ase
o
f
t
h
e m
o
d
e
l prov
ides an
o
p
e
ratio
nal architecture of t
h
e i
ndu
str
i
al p
l
an
t stud
ied.
An
exam
ple of a
water station of a the
r
mal power pl
a
n
t for
produci
n
g electric energy has
been c
h
ose
n
i
n
o
r
d
e
r to
v
a
lid
at
e th
is seco
nd
ph
ase.
After t
h
e two
p
h
a
ses of an
al
ysis o
f
th
e functio
n
a
l and operational arc
h
i
t
ecture of the
syste
m
, the
third
phase
of t
h
e m
odel is to
im
ple
m
ent a
monitoring
tool as well as
m
a
n
-
m
achin
e interfaces for m
onitori
ng
in
du
strial pro
c
ess. Ind
eed, sup
e
rv
ision
allows m
o
n
ito
ri
n
g
and c
o
nt
r
o
l
o
f
ope
rat
i
o
n o
f
a
n
i
n
st
al
l
a
t
i
on i
n
o
r
der
to
stay with
in
th
e n
o
r
m
a
l o
p
e
rating
ran
g
e
reg
a
rd
less
of external disturba
nces
.
Usin
g a SCAD
A s
y
stem
(SCADA)
fo
r
ex
am
p
l
e, all ano
m
al
ies id
en
tified
in a
water
s
t
at
i
on
of a
t
h
e
r
m
a
l
pow
er
pl
a
n
t
can
be
det
ect
ed i
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
Dep
e
nd
ab
ility Eva
l
ua
tion
a
nd Su
p
e
rvisio
n in Th
erma
l Po
wer Plan
ts
(
M
arw
a
Be
n
H
a
m
o
ud
a)
91
1
real tim
e
, from alar
m
s
and ca
n ide
n
tif
y
t
h
ei
r
causes.T
h
us s
upe
r
v
i
s
i
on i
s
a
refere
nce t
o
ol
fo
r t
h
e o
p
e
r
at
or a
n
d
it also
allo
ws hi
m
to
in
teract directly
with
a co
n
t
ro
l and
su
perv
ision
system
.
5.
STUD
Y OF D
E
PEND
ABIL
ITY AN
D SU
PERVI
SIO
N
OF REVE
RS
E
OS
MOSI
S STATIO
N
The Production Centre Ra
de
s Electricity
is a ther
m
a
l po
wer
pl
ant
(T
P
P
) t
h
at
co
nsi
s
t
s
on a sy
stem
p
r
od
u
c
i
n
g th
e
electricity wh
ile u
s
i
n
g dry water steam
to
d
r
ag
th
e altern
at
o
r
in
ro
tation
(Fig
ure
4
)
. Th
is steam
is ge
nerate
d i
n
a
furnace t
h
at trans
f
orm
s
the c
h
em
ical
e
n
ergy of
the fuel
(natural
ga
s, heavy fuel-oil)
in
calorific e
n
ergy.
Fig
u
re
4
.
Fu
n
c
t
i
o
n
a
lity o
f
a therm
a
l p
o
w
er p
l
an
t
Thi
s
i
s
o
n
e
of
t
h
e m
o
st
im
por
t
a
nt
p
r
o
d
u
ct
i
o
n
s
o
f
el
ect
ri
c
p
o
w
er
st
at
i
ons
i
n
Tu
ni
si
a (
3
7%
of
nat
i
o
nal
p
r
od
u
c
tion
)
. In fact, a TPP is
a p
o
wer p
l
an
t in
wh
ich
th
e pri
m
e
m
o
v
e
r is st
ea
m
d
r
iv
en.
Water is h
eated
, tu
rn
s
in
to
steam
an
d sp
in
s a steam
tu
rb
in
e
whic
h
dri
v
es an electrical gene
rator.
After it p
a
sses th
ro
ugh
th
e t
u
rb
i
n
e,
t
h
e st
eam
i
s
conde
nse
d
i
n
a con
d
e
n
ser
.
The
great
est
va
ri
at
i
on i
n
t
h
e d
e
si
g
n
of TP
Ps i
s
du
e t
o
t
h
e di
ffere
nt
fuel
sources. Som
e
pre
f
er to use
t
h
e term
ene
r
gy
center
b
ecause suc
h
facilities convert
forms
of heat energy into
electrical energy [26].
In
TPPs
, m
echanical powe
r i
s
produced by
a heat
e
n
gi
ne
w
h
i
c
h t
r
a
n
s
f
o
r
m
s
t
h
erm
a
l
ener
gy
,
oft
e
n
fr
om
co
m
bust
i
on
o
f
a f
u
el
, i
n
t
o
rot
a
t
i
o
nal
ener
gy
. M
o
st
t
h
erm
a
l
powe
r
pl
ant
s
pr
od
uce
st
eam
,
and t
h
ese are
som
e
times called steam
power pla
n
ts. TI
Ps
are classified
by the type of
fuel
a
nd t
h
e t
y
pe o
f
p
r
i
m
e
m
over
in
stalled
[26
]
.
5.
1
Presentati
on
of the Studied
Sys
t
em
The re
verse
os
m
o
si
s (R
.O) st
at
i
on o
p
erat
es
on t
h
e
reve
rse
osm
o
si
s phe
no
m
e
non
, w
h
i
c
h
con
s
i
s
t
s
of
t
h
e sp
o
n
t
a
ne
ou
s di
f
f
usi
on
o
f
a chem
i
cal
co
m
poun
d (
w
at
e
r) t
h
r
o
u
g
h
a se
m
i
perm
eabl
e
m
e
m
b
rane.
It
occu
rs
whe
n
a su
bst
a
nce i
s
prese
n
t
e
d at
di
ffe
rent
c
once
n
t
r
at
i
o
ns f
r
om
one si
de a
nd t
h
e
ot
he
r o
f
t
h
e
m
e
m
b
rane. Thi
s
di
ffe
re
nce caus
e
s excess p
r
ess
u
re cal
l
e
d osm
o
t
i
c
press
u
re
. The di
st
ri
but
i
o
n of t
h
i
s
c
o
m
pou
n
d
i
s
do
ne s
o
t
h
at
t
h
e t
w
o l
e
vel
s
on o
n
e si
de and t
h
e ot
her
of t
h
e m
e
m
b
rane t
e
nd t
o
eq
ual
i
ze. In co
nt
rast
, reve
rse o
s
m
o
si
s
enha
nces
conc
entration
of the m
o
re
concentrated s
o
lution by
sub
j
ecting th
e less con
c
en
trated so
l
u
tion
to a
press
u
re
greate
r
tha
n
t
h
e
os
mo
tic
pressu
re (Fig
ure 5
)
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
I
J
ECE Vo
l. 5
,
N
o
. 5
,
O
c
tob
e
r
20
15
:
905
–
9
17
91
2
Fi
gu
re
5.
The
r
e
verse
o
s
m
o
si
s st
at
i
o
n
To o
ffe
r a ge
n
e
ral
descri
pt
i
o
n o
f
t
h
e re
vers
e osm
o
si
s pro
cess, we
prese
n
t
kn
ot
A
-
0
S
ADT m
odel
(Figu
r
e
6).
Indeed
, am
o
n
g
the ele
m
en
ts o
f
th
e rev
e
rs
e
o
s
m
o
sis stat
io
n
,
we can m
e
n
tio
n
:
filter m
e
m
b
ran
e
,
pum
p. T
h
e S
A
DT m
e
t
hod i
s
essent
i
a
l
l
y
a st
ruct
ure
d
m
e
t
hod
of
re
prese
n
t
a
t
i
on
desi
g
n
e
d
f
r
om
sim
p
l
e
concept
s
.
It allows cons
idering the
process as
a structure com
pos
ed of sim
p
ler
syste
m
s inter
act. The hie
r
a
r
chical
structure
diagra
m
s
used t
o
re
prese
n
t
various
levels of
d
e
tail an
d
i
n
a
relativ
ely con
c
ise, t
o
th
e co
m
p
lex
si
m
p
le
sy
stem
s. W
e
u
s
ed
t
h
e
s
o
ft
war
e
Bp
W
i
n fo
r re
prese
n
tin
g diag
ram
s
S
A
D
T
m
o
d
e
l
.
Fi
gure
6. A
-
0 l
e
vel
of S
ADT
m
odel
of a r
e
ver
s
e o
s
m
o
sis
statio
n
5.
2
Val
i
d
ati
o
n of
the
Pr
op
ose
d
Mo
del
To
val
i
d
at
e t
h
e
fi
rst
pha
se o
f
ou
r
gene
ral
m
odel
o
f
a
n
al
y
s
i
s
and
su
pe
rvi
s
i
o
n
of a
n
i
n
d
u
st
r
i
al
pr
oces
s
,
we u
s
ed
th
e
SADT m
e
th
od
is a g
e
n
e
ral m
e
th
od
th
at
see
k
s
to foster com
m
unication bet
w
een a
p
plicant
s
and
users
on t
h
e
o
n
e ha
nd a
nd d
e
si
gne
rs an
d d
i
rect
ors
,
o
n
t
h
e ot
her
.
Aft
e
r
m
odel
i
ng t
h
e el
ect
ri
ci
t
y
pro
duct
i
o
n
pr
ocess TP
P t
h
e SAD
T
m
e
t
hod
, usi
ng act
i
g
ram
m
e
s of SA
DT,
we est
a
bl
i
s
he
d l
e
vel
s
A-
0, A
0
a
nd
A1
SA
D
T
m
odel
prese
n
t
e
d
by
t
h
e t
h
ree
f
i
gu
res
7 a
n
d
8
.
5.
2.
1
Application of
SADT Method
Th
e fi
g
u
re belo
w sh
ows
a SADT m
e
t
h
od
b
y
th
ermal p
o
w
er p
l
an
t. Th
e activ
ity d
i
ag
ram
(act
i
g
ram
m
e) devel
ope
d
hel
p
ed t
o
creat
e
or
gene
rat
e
a
gi
ve
n
out
put
(el
ect
ri
c ene
r
gy
,
l
o
s
s
es), t
o
t
r
a
n
s
f
o
r
m
,
or
m
odi
fy
or cha
nge st
at
e an i
n
put
d
a
t
a
(f
uel
,
ai
r, wat
e
r
)
an
d
seek t
h
e i
n
put
dat
a
fr
om
cont
rol
g
u
i
d
el
i
n
es
,
based
o
n
th
e
po
ten
tial
m
ech
an
ism
s
( bo
iler, t
u
rb
in
e and
g
e
n
e
rator).
US
ED
A
T
:
A
UT
H
O
R:
M
a
r
w
a
B
.
H
D
A
T
E
:
1
O
c
t
.
2
0
1
4
P
R
O
J
E
C
T
:
M
o
de
l 2
R
E
V
:
4
A
p
r
.
20
15
N
O
T
E
S
:
1
2
3
4 5
6
7
8
9
10
WO
R
K
I
N
G
DRA
F
T
RE
CO
M
M
E
NDE
D
PU
BL
I
C
A
T
I
O
N
R
EAD
ER
D
A
T
E
C
O
N
T
EXT
:
TO
P
NO
D
E
:
T
I
T
L
E
:
NUM
B
E
R
:
P
r
o
d
u
c
e
o
s
mo
si
s
w
a
te
r
A-
0
0
P
r
od
u
c
e
osm
o
s
i
s
w
a
ter
F
i
l
t
er
ed
w
a
t
e
r
Q
6
3
p
r
o
duc
t
P
r
od
u
c
t
i
on
p
r
ot
o
c
ol
Re
g
u
l
a
t
i
o
n
S
ecu
r
i
t
y
Os
m
o
s
e
d
w
a
t
e
r
L
o
sse
s
R
e
ve
r
s
e
os
m
o
si
s st
a
t
i
o
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
Dep
e
nd
ab
ility Eva
l
ua
tion
a
nd Su
p
e
rvisio
n in Th
erma
l Po
wer Plan
ts
(
M
arw
a
Be
n
H
a
m
o
ud
a)
91
3
Fi
gu
re
7.
A
-
0
l
e
vel
S
A
D
T
m
odel
o
f
a
t
h
erm
a
l
po
we
r
pl
ant
Th
e
g
o
a
l
o
f
th
e o
b
t
ai
n
e
d
m
o
del SADT is m
u
ltip
le: th
e
m
o
del SADT will allo
w to
tak
e
o
f
fen
ce at t
h
e
state of t
h
e st
udied
syste
m
, as shows it both
prese
n
te
d lev
e
l
s
, an
d to
clear t
h
e co
m
p
lex
ity o
f
th
e system
.
Ind
e
ed
, am
o
n
g
th
e co
m
p
lex
activ
ities in
th
e TPP,
we can q
u
o
t
e: th
e
p
r
ep
aration
o
f
t
h
e water; th
e
p
r
od
u
c
tion
of
th
e calorific en
erg
y
;
the
production of the
m
echanical en
erg
y
; th
e
prod
u
c
tion of electrical
energy; recycling steam
.
Fi
gu
re
8.
A
0
l
e
vel
o
f
SA
DT
m
odel
of a
t
h
e
r
m
a
l
powe
r
pl
a
n
t
5.
2.
2
Application of
Safe-SADT Method
In this secti
o
n, we
prese
n
t the
overview of t
h
e Sa
fe
-S
AD
T
m
odel
of t
h
e s
y
st
em
hy
dra
u
l
i
c
pa
rt
o
f
t
h
e
reve
rse osm
o
sis station. This
syste
m
shows that the re
presen
tatio
n
with
Safe- SADT
m
e
t
h
od
is m
o
re reliab
l
e
than t
h
e S
ADT m
e
thod
bec
a
use it takes
t
h
e
param
e
ters
of de
pe
nda
bility and t
h
e
various
operating m
odes
(fig
u
r
e 10
).
Th
e org
a
n
i
zatio
n
and
b
e
h
a
vio
r
of fun
c
tion
s
d
e
scrib
e
t
h
e activ
ities o
f
th
e au
to
m
a
ti
o
n
system
th
ro
ugh
ou
t its life cycle. In
our
e
x
am
pl
e, we
can
defi
ne t
h
e
fu
nct
i
o
ns
of
o
u
r
sy
st
em
as fol
l
o
ws:
“Ope
n
val
v
e
V
i
j”, “cl
ose
val
v
e Vi
j”
an
d “
c
o
n
t
r
ol
p
u
m
p
Pi
”
US
E
D
A
T
:
A
UT
H
O
R:
Be
n
Ha
m
o
u
d
a
M
.
DA
T
E
:
1
0
J
u
n
e
2
0
1
4
P
R
O
J
E
C
T
:
M
o
de
l
1
R
E
V
:
4
A
p
r
.
20
15
N
O
T
E
S
:
1
2
3
4
5
6
7
8
9
10
WO
R
K
I
N
G
DRA
F
T
RE
CO
M
M
E
NDE
D
PU
B
L
I
C
A
T
I
O
N
RE
A
D
E
R
DA
T
E
C
O
N
T
E
X
T
:
TO
P
EN
I
C
a
r
t
hag
e
NO
D
E
:
T
I
T
L
E
:
NUM
B
E
R
:
G
e
n
e
r
a
t
e
el
ec
t
r
i
c
i
t
y
A-
0
1
0
Ge
ne
r
a
te e
l
ec
tr
i
c
i
t
y
E3
:
W
a
t
er
E1
: F
u
e
l
E2
:
A
i
r
C
4
:
P
r
o
d
uct
i
o
n
pr
o
t
o
c
o
l
C
1
: R
e
g
u
l
a
ti
o
n
C2:
D
i
s
p
at
c
h
in
g
C
3
:
Secu
r
i
t
y
S2
:
L
o
s
s
e
s
S
1
:
El
ect
r
i
c po
w
e
r
M1
:
B
o
ile
r
M2
:
T
u
r
b
i
n
e
M3
:
A
l
t
e
r
n
a
t
o
r
M4
:
R
e
c
y
c
lin
g
S
t
a
t
io
n
M5
:
S
t
a
f
f
M6
:
A
u
x
ilia
r
y
M
7
:
Wa
te
r
tr
e
a
tm
e
n
t r
o
o
m
U
S
E
D
A
T
:
A
U
T
H
O
R
:
M
a
r
w
a
B
.
H
D
A
T
E
:
10
J
une
2
0
1
4
P
R
O
J
E
C
T
:
M
o
de
l
1
R
E
V
:
4
A
p
r
.
20
15
N
O
T
E
S
:
1
2
3
4 5
6
7
8
9
10
WO
R
K
I
N
G
DR
A
F
T
RE
C
O
M
M
E
N
DE
D
PU
BL
I
C
A
T
I
O
N
R
E
AD
ER
D
A
T
E
C
O
N
T
EX
T
:
A-
0
TO
P
NO
D
E
:
T
I
T
L
E
:
N
U
M
BER
:
Ge
n
e
r
a
t
e
e
l
e
c
t
r
i
c
i
t
y
A0
2
1
Pr
e
p
a
r
e
wa
t
e
r
2
Pr
o
d
u
c
e
ca
l
o
r
i
f
i
c
en
er
g
y
3
P
r
o
d
uce t
h
e
me
c
h
an
i
c
al
en
er
g
y
4
Pr
o
d
u
c
e
el
ec
tr
i
c
a
l
en
er
g
y
5
R
e
cy
cl
e
M
1
:
B
oile
r
C1
:
Re
gu
l
a
t
i
o
n
M2
:
T
u
r
b
i
n
e
C2
:
Di
s
p
a
t
c
h
i
n
g
M3
:
A
l
t
e
r
n
a
t
o
r
M4
:
R
e
c
y
c
l
i
n
g
Sta
t
ion
C3
:
S
ecu
r
i
t
y
M5
:
S
t
a
f
f
M6
:
A
u
x
i
l
i
a
r
y
C4
:
P
r
oduct
i
on
p
r
ot
o
c
o
l
M
7
: W
a
te
r
tr
e
a
tm
e
n
t r
o
o
m
E1
:
Fu
e
l
S1
:
El
e
c
t
r
i
c
po
w
e
r
E2
: A
i
r
S2
:
Lo
s
s
e
s
E3
:
W
a
t
e
r
E
x
pa
nd
ed
st
e
a
m
W.
C
a
lo
r
i
f
i
c
W
.
m
e
c
h
an
i
cal
Fe
e
d
i
n
g
pu
r
e
wa
t
e
r
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
I
J
ECE Vo
l. 5
,
N
o
. 5
,
O
c
tob
e
r
20
15
:
905
–
9
17
91
4
“Ope
n
val
v
e
V
i
j”, “cl
ose
val
v
e Vi
j”
an
d “
c
o
n
t
r
ol
p
u
m
p
Pi
”
The fi
r
s
t
archi
t
ect
ure t
o
be de
fi
ne
d i
n
t
h
e sy
s
t
em
desi
gn i
s
t
h
e f
unct
i
onal
a
r
chi
t
ect
u
r
e. T
h
e fu
nct
i
o
nal
d
e
sign
ph
ase i
s
to
d
e
co
m
p
ose th
e m
a
in
fu
n
c
tion
s
of
t
h
e system
in
to
sim
p
ler fun
c
t
i
o
n
s
(Tab
le1). Th
ese
fu
nct
i
o
ns a
r
e a
l
so
br
oke
n
d
o
w
n
i
n
t
o
eve
n
si
m
p
l
e
r sub
-
f
unc
t
i
ons.
Tabl
e1
. F
u
nct
i
onal
A
r
chi
t
ect
ure
Functional Ar
chitectur
e
Open valve V
11
,
Close valve V
11
Open valve V
21
,
Close valve V
21
Open valve V
12
,
Close valve V
12
Open valve V
22
,
Close valve V
22
Contr
o
l pu
m
p
P
1
Contr
o
l pu
m
p
P
2
Table
2. Material Arc
h
itecture
Materi
al A
r
chitect
ure
V
11
+PLC
1
V
21
+ PLC
2
V
12
+ PLC
1
V
22
+ PLC
2
P
1
+ PLC
1
P
2
+ PLC
2
C
a
rry
i
n
g
out
t
h
e
fu
nct
i
o
ns
o
f
t
h
e
f
u
nct
i
ona
l
archi
t
ect
u
r
e
r
e
qui
res a
ch
oi
ce o
f
e
q
ui
pm
ent
(
h
a
r
d
w
ar
e
and s
o
ft
ware
),
whi
c
h i
s
t
h
e
har
d
wa
re arc
h
i
t
ect
ure.
T
h
e
hardwa
re arc
h
itecture also specifies for eac
h
equi
pm
ent all of its global c
h
aracteristics
(Table
2).
T
h
e
har
d
ware a
r
c
h
i
t
ect
ure
of
o
u
r
sy
st
em
consi
s
t
s
o
f
:
val
v
es
,
pum
ps
and
i
n
du
st
ri
al
pr
o
g
ram
m
abl
e
l
ogi
c c
ont
rol
l
e
r
s
(PLC
).
The operati
o
nal architecture
is
th
e p
r
oj
ectio
n
of th
e fun
c
tion
a
l arch
itectu
r
e of th
e h
a
rdware
architecture. T
h
is projection
lead
s to
assign fun
c
tio
ns to
eq
u
i
p
m
en
t th
at
u
s
es th
e co
st fu
n
c
tion
to
ev
alu
a
te
d
i
fferen
t
so
l
u
tio
n
s
, and
u
s
es assessm
en
t to
o
l
s
su
ch
as
th
o
s
e used to
assess a lev
e
l o
f
d
e
p
e
n
d
a
b
i
lity o
r
per
f
o
r
m
a
nce l
e
vel
.
Tabl
e 3. O
p
era
t
i
onal
A
r
c
h
i
t
ect
ure
Operational Ar
chitecture
Open V
11
V
11
PLC
1
Close V
11
Open V
21
V
21
PLC
2
Close V
21
Open V
12
V
12
PLC
1
Close V
12
Open V
22
V
22
PLC
2
Close V
22
Contr
o
l P
2
P
2
PLC
2
Contr
o
l P
2
P
3
PLC
1
Fi
gu
re
9 sh
o
w
s t
h
e
de
vel
o
pm
ent
of Saf
e
-SA
D
T m
odel
of t
h
e hy
dra
u
l
i
c
sy
st
em
. Thi
s
m
e
t
hod
im
pl
em
ent
e
d t
h
e st
eps t
h
at
d
e
t
a
i
l
how t
h
e
sy
st
em
and i
t
s
fu
nct
i
o
nal
arc
h
i
t
ect
ure,
p
h
y
s
i
cal
and
ope
rat
i
onal
.
The a
n
al
y
s
i
s
b
y
t
h
e Sa
fe-
S
A
D
T m
e
t
hod
pr
ovi
des
g
o
o
d
d
a
t
a
carri
e
r
on
R
A
M
S
param
e
t
e
rs c
o
m
pone
nt
s i
n
or
der
t
o
det
e
r
m
i
n
e t
hose
o
f
t
h
e
gene
ral
sy
st
em
.
Evaluation Warning : The document was created with Spire.PDF for Python.