Inter
national
J
our
nal
of
Electrical
and
Computer
Engineering
(IJECE)
V
ol.
9,
No.
3,
June
2019,
pp.
1553
1560
ISSN:
2088-8708,
DOI:
10.11591/ijece.v9i3.pp1553-1560
r
1553
Impact
of
compr
essed
air
ener
gy
storage
system
into
diesel
po
wer
plant
with
wind
po
wer
penetration
Abdulla
Ahmed
1
,
T
ong
Jiang
2
1
State
K
e
y
Laboratory
of
Alternate
Electrical
Po
wer
System
with
Rene
w
able
Ener
gy
Sources,
Electrical
Po
wer
System
and
its
Automation
-
School
of
Electrical
and
Electronic
Engineering,
North
China
Electric
Po
wer
Uni
v
ersity
,
Beijing
-
Chi
na
1,2
Department
of
Electrical
and
Electronics
Engineering,
F
aculty
of
Engineering
Science,
Uni
v
ersity
of
Nyala,
Nyala
-
Sudan
Article
Inf
o
Article
history:
Recei
v
ed
Jun
7,
2018
Re
vised
Oct
11,
2018
Accepted
Dec
18,
2018
K
eyw
ords:
Compressed
air
ener
gy
storage
Diesel
po
wer
plant
Modeling
and
simulation
Unit
commitment
W
ind
po
wer
generation
ABSTRA
CT
The
wind
ener
gy
plays
an
important
role
in
po
wer
system
because
of
its
rene
w
able,
clean
and
free
ener
gy
.
Ho
we
v
er
,
the
penetration
of
wind
po
wer
(WP)
into
the
po
wer
grid
system
(PGS)
requires
an
ef
ficient
ener
gy
storage
systems
(ESS).
compressed
air
ener
gy
storage
(CAES)
system
is
one
of
the
most
ESS
technologies
which
can
alle
viate
the
int
ermittent
nature
of
the
rene
w
able
ener
gy
sources
(RES).
Nyala
city
po
wer
plant
in
Sudan
has
been
chosen
as
a
case
study
because
the
po
wer
supply
by
the
e
xisting
po
wer
plant
is
e
xpensi
v
e
due
to
high
costs
for
fuel
transport
and
the
reliability
of
po
wer
supply
is
lo
w
due
to
uncertain
fuel
pro
vision.
This
paper
presents
a
formulation
of
security-constrained
unit
commitment
(S
CUC)
of
diesel
po
wer
plant
(DPP)
with
the
inte
gration
of
CAES
and
PW
.
The
optimization
problem
is
modeled
and
coded
in
MA
TLAB
which
solv
ed
with
solv
er
GOR
UBI
8.0.
The
results
sho
w
that
the
proposed
model
is
suitable
for
inte
gration
of
rene
w
able
ener
gy
sources
(RES)
into
PGS
with
ESS
and
helpful
in
po
wer
system
operation
management.
Copyright
c
2019
Institute
of
Advanced
Engineering
and
Science
.
All
rights
r
eserved.
Corresponding
A
uthor:
Abdulla
Ahmed,
School
of
Electrical
and
Electronic
Engineering,
North
China
Electric
Po
wer
Uni
v
ersity
,
No.2
Beinong
road,
Changping
district
102206,
Beijing
-
China.
Phone:
008613241448448
Email:
elamin2018@gmail.com
1.
INTR
ODUCTION
The
RES
is
intermittent,
which
mak
es
it
more
dif
ficult
to
accurately
schedule
the
po
wer
generat
ion
from
these
sources.
Whilst
the
accurac
y
of
WP
forecasting
has
increased
substantially
i
n
recent
years,
there
are
still
frequently
significant
de
viations
between
forecast
and
actual
production
capability
.
T
ime
series
is
a
technique
which
used
for
fore
casting
the
wind
speed
and
feed-forw
ard
neural
netw
ork
technique
is
a
technique
which
used
for
forecasting
the
solar
radiations.
Therefore,
the
forecasting
results
are
used
to
determine
the
unit
commitment
(UC)
and
optimal
sizing
of
ESS
based
on
the
cost-benefit
analysis
in
a
microgrid
system
[1].
A
formulation
of
SCUC
with
the
inte
gration
of
CAES
and
WP
is
presented
to
obtain
a
commitm
ent
scheduling
at
a
maximum
po
wer
output
and
a
minimum
production
cost
and
the
optimization
model
is
for
-
mulated
as
mix
ed
inte
ger
programming
(MIP)for
solving
the
SCUC
[2],[3],[4].
The
SCUC
problem
with
the
impact
of
RES
and
CAES
subjected
to
se
v
eral
unit/system
constraints.
Unit
constraints
include
po
wer
gener
-
ation
limits,
on/of
f
time
indicator
limits,
ramping
limits,
fuel
consumption
and
emission
limits.
The
system
constraints
include
acti
v
e
and
reacti
v
e
po
wers
flo
w
limits
on
selected
transmission
lines
and
v
oltage
limits
on
b
uses.
The
proposed
problem
is
decomposed
into
a
master
problem
and
sub-problem
based
on
Benders
de-
composition
technique
to
optimizing
the
system
and
minimizing
the
netw
ork
violations
[5].The
w
ork
in
[6]
illustrated
the
thermodynamic
analysis
of
compressed
air
ener
gy
storage
with
inte
gration
of
wind
po
wer
and
J
ournal
homepage:
http://iaescor
e
.com/journals/inde
x.php/IJECE
Evaluation Warning : The document was created with Spire.PDF for Python.
1554
r
ISSN:
2088-8708
the
results
sho
w
good
performance.
The
central
unit
commitment
is
used
to
determine
the
cost
benefits
of
ESS
for
impacts
of
lar
ge-scale
wind
po
wer
in
the
Netherlands
electricity
po
wer
supply
.
The
proposed
model
illustrated
the
cost
benefits
analysis
and
sho
ws
that
the
ener
gy
storage
can
sa
v
e
the
operating
costs
with
the
increase
of
WP
installation
and
this
can
impro
v
e
the
future
of
the
po
wer
system
layout
of
the
Dutch
po
wer
generation
[7].
The
proposed
model
includes
RES
inte
grated
with
a
carbon
capture
po
wer
plant
is
presented.
The
rob
ust
optimization
technique
and
a
stochastic
unit
commitment
with
multiobjecti
v
e
optimization
models
and
a
linear
re-dispatch
strate
gy
are
emplo
yed
to
obtain
the
commitment
scheduling
of
units
and
decrease
the
emissions
[8],[9].
Inte
gration
of
po
wer
-to-h
ydrogen
is
one
of
t
he
solutions
for
balancing
between
the
supply
and
the
po
wer
demand
when
the
PGS
contains
RES.
Ref
[10]
proposed
a
model
that
includes
inte
gration
of
po
wer
-to-h
ydrogen
to
accommodate
a
high
penetration
of
wi
nd
generation
in
which
the
e
xcess
wind
generation
is
con
v
erting
into
h
ydrogen
and
stored
for
using
later
when
needed.
Ref
[12]
presented
the
modern
bio-inspired
algorithm
called
Gre
y
W
olf
Optimization
(GW
O)
algorithm
to
solv
e
the
proposed
problem
which
includes
ther
-
mal
generators
inte
grated
with
WP
and
the
optimization
problem
is
formulated
as
UC
model
and
the
results
sho
w
that
the
algorithm
has
an
ef
fecti
v
e
capability
to
obtain
the
economic
benefits
with
good
quality
.
The
uncertainty
nature
of
v
ariable
RES
mak
es
it
dif
ficult
to
schedule
the
po
wer
generator
units
ef
ficiently
because
the
system
operators
depend
on
v
ariable
outcomes.
Authors
in
[12]
modif
y
SCUC
to
capacity
constraints
by
defining
scenario
response
sets
for
predict
the
economic
cost
of
dispatching
backup
capacity
when
it
is
needed.
The
mathematical
models
and
se
v
eral
approaches
are
de
v
eloped
for
addressing
rene
w
able
po
wer
generations
ef
fects
and
uncertainties
[13].
This
paper
introduced
the
formulation
of
SCUC
problem
for
po
wer
grid
system
containing
DPP
,
WP
,
and
CAES
aim
s
to
find
the
best
scheduling
of
day-ahead
operation
planning.
The
rest
of
the
paper
is
or
g
anized
as
follo
ws:
section
2
presents
a
general
description
of
the
s
ystem.
The
mathematical
formulation
is
dealt
in
section
3.
The
case
studie
d
is
e
xplained
in
Section
4
and
the
simulation
results
with
discussions
are
presented
in
section
5,
the
conclusion
is
presented
in
section
6.
2.
SYSTEM
DESCRIPTION
The
proposed
system
is
composed
of
se
v
eral
parts
includes
DPP
,
WF
,
CAES
and
system
load
demand.
2.1.
Diesel
P
o
wer
Plant
The
total
e
xisting
po
wer
demands
of
Nyala
city
is
to
about
16
MW
and
the
electrical
ener
gy
is
entirely
pro
vided
by
14
Diesel
Generators
with
a
theoretical
maximum
capacity
of
30
MW
.
Peak
load
results
in
the
e
v
ening
between
19:00h
and
22:00h
while
lo
w
load
results
in
the
night
between
03:00
h
and
05:00
h.
The
fuel
cost
consumption
of
DG
can
be
calculated
by
the
quadratic
function
as:
C
f
T
X
t
=1
N
G
X
i
=1
(
aP
2
i;t
+
bP
i;t
+
c
)
(1)
Where
i
and
t
inde
x
es
for
the
time
period
and
diesel
unit
respecti
v
ely;
a,
b
and
c
are
the
cost
coef
ficients
related
to
DGs
fuel
consumption
carv
e;
C
f
is
the
price
of
diesel
fuel;
P
i;t
is
the
output
po
wer
of
DG
unit
(i)
at
the
time
interv
al
(t);
T
and
N
G
are
the
total
time
horizon
and
the
total
number
of
the
DGs
respecti
v
ely
.
2.2.
W
ind
P
o
wer
W
ind
ener
gy
is
the
source
of
po
wer
which
can
generate
electricity
by
pushed
the
wind
speed
ag
ainst
the
f
an
to
con
v
ert
it
to
mechanical
po
wer
then
generate
electricity
.
The
wind
turbine
captures
the
winds
kinetic
ener
gy
in
a
rotor
consisting
of
tw
o
or
more
blades
mechanically
coupled
to
an
electrical
generator
.
The
turbine
is
mounted
on
a
tall
to
wer
to
enhance
the
ener
gy
capture.
Numerous
wind
turbines
are
located
near
each
other
to
b
uild
a
wind
f
arm
of
the
desired
po
wer
production
capacity
.
The
electrical
po
wer
generated
from
the
wind
turbine
can
be
e
xpressed
as:
P
w
=
1
2
C
p
AV
3
(2)
where,
P
w
is
the
po
wer
in
the
wind
(kW),
A
is
the
swept
area
by
the
blades
(
m
2
)
,
is
the
air
density
and
it
can
be
tak
en
as
1.225
(
k
g
=m
3
)
,
V
is
the
wind
speed
(m/s)
and
C
p
is
the
po
wer
coef
ficient
of
the
turbine,
which
depends
on
the
blades
design
and
the
tip
speed
ratio.
The
coef
ficient
of
ef
ficienc
y
of
wind
ener
gy
con
v
ersion
to
turning
the
wind
ener
gy
into
ener
gy
which
can
be
used,
whether
electrical
or
mechanical
is
the
maximum
theoretical
v
alue
for
this
constant
is
about
0.593
and
kno
wn
(Betz
limit).
Thus,
the
maximum
po
wer
that
can
be
realized
from
a
wind
system
is
59.3
%
of
the
total
wind
po
wer
[10].
IJECE,
V
ol.
9,
No.
3,
June
2019
:
1553
–
1560
Evaluation Warning : The document was created with Spire.PDF for Python.
IJECE
ISSN:
2088-8708
r
1555
2.3.
Compr
essed
air
ener
gy
storage
The
function
of
CAES
is
stored
ener
gy
during
the
of
periods
and
reused
it
during
the
peak
periods.
There
are
man
y
applications
of
CAES
in
po
wer
grid
system
such
as
load
shifting,
mitig
ate
the
fluctuations
of
rene
w
able
ener
gies
and
mak
e
management
and
re
gulations
for
the
grid
system.
The
main
components
of
the
CAES
include
the
compressor
,
ca
v
ern,
and
the
e
xpander
.
There
are
tw
o
modes
of
operation,
the
first
mode
is
the
compression
mode
which
the
compressor
consumed
the
elect
ricity
from
wind
f
arm
or
from
the
grid
system
to
compress
air
and
stored
it
in
the
ca
v
ern
and
the
second
mode
is
the
generation
mode
which
the
air
stored
in
the
ca
v
ern
is
heated
up
by
g
as
and
then
entered
the
turbine
to
generate
electricity
.
Cost
of
producing
P
j
;t
MW
of
electricity
is
equal
to
g
as
price
multiply
by
heat
rate
v
alue
for
generating
P
j
;t
.
It
can
be
represented
as:
P
j
;t
=
r
j
v
r
j
;t
(3)
P
j
;t
=
inj
j
v
inj
j
;t
(4)
where,
r
j
and
inj
j
are
the
ef
ficienc
y
f
actor
for
producing
po
wer
and
the
ef
ficienc
y
f
actor
for
injecting
air
respecti
v
ely;
v
r
j
;t
and
v
inj
j
;t
are
the
a
mount
of
released
air
in
MW
at
hour
t
and
the
mount
of
injected
air
in
M
W
at
hour
t
respecti
v
ely
.
3.
MA
THEMA
TICAL
FORMULA
TION
The
mathematical
model
has
formulated
a
cording
t
o
the
system
parts.
As
mentioned,
the
main
parts
of
the
proposed
system
include
diesel
po
wer
generation,
wind
f
arm,
compressed
air
ener
gy
storage
and
loads.
3.1.
Objecti
v
e
function
The
objecti
v
e
function
is
to
minimize
the
total
operation
cost
consisting
of
tw
o
terms:
the
first
t
erm
is
diesel
operating
cost
including
fuel,
startup
and
shutdo
wn
costs
and
the
second
term
is
the
operating
cost
of
CAES
units
throughout
the
whole
operational
period.
The
operating
costs
of
wind
po
wer
generation
units
are
considered
to
be
zero
because
wind
is
free.
min
T
X
t
=1
8
<
:
N
G
X
i
=1
[
C
i
(
P
i;t
)
I
i;t
+
S
T
i;t
+
S
D
i;t
]
+
N
C
X
j
=1
C
j
(
P
j
;t
)
9
=
;
(5)
where,t,i
and
j
are
inde
x
es
of
time
inde
x,
number
of
hours
for
operating
period,
diesel
units
and
CAES
units
respecti
v
ely;
T
,
N
G
and
N
C
are
the
numbers
of
operating
hours,
diesel
units
and
CAES
units;
C
i
and
C
j
are
production
cost
functions
of
diesel
unit
i
and
CAES
unit
j;
S
T
i;t
and
S
D
i;t
are
startup
and
shutdo
wn
costs
of
unit
i
at
time
t
respecti
v
ely;and
P
i;t
and
P
J
;t
are
output
po
wers
from
diesel
units
and
CAES
units
respecti
v
ely
.
The
unit
status
indicator
I
i;t
is
an
inte
ger
term
can
be
1
or
zero.
3.2.
SCUC
constraints
The
objecti
v
e
function
is
subject
to
se
v
eral
constraints
including
the
unit
constraints
and
netw
ork
constraints.
3.2.1.
System
r
eal
po
wer
balance
constraints
The
total
real
po
wer
produced
by
the
DG,
CAES
in
addition
to
po
wer
generated
by
wind
turbines
is
must
be
equal
to
system
load
demand
plus
losses
as
e
xpressed
by:
N
G
X
i
=1
P
i;t
+
N
C
X
j
=1
P
j
;t
+
N
w
X
w
=1
P
w
t
=
X
n
2
i
P
D
t
+
P
L
t
;
8
t
2
T
(6)
where
P
w
t
and
P
D
t
are
the
wind
po
wer
generation
and
system
load
demand
at
time
(t)
respecti
v
ely
.
Impact
of
compr
essed
air
ener
gy
stor
a
g
e
system
into...(Abdulla
Ahmed)
Evaluation Warning : The document was created with Spire.PDF for Python.
1556
r
ISSN:
2088-8708
3.2.2.
Requir
ed
system
spinning
and
operating
r
eser
v
es
The
spinning
and
operating
reserv
es
of
the
DGs
should
be
lar
ge
enough
to
supply
electricity
to
the
system
during
the
WP
v
ariations
as
e
xpressed
in
(7)
and
(8)
respecti
v
ely
.
N
G
X
i
=1
r
s
i;t
+
N
C
X
j
=1
r
s
j
;t
R
S
(
t
)
;
8
t
(7)
where,
r
s
i;t
,
r
s
j
;t
and
R
S
(
t
)
are
the
spinning
reserv
e
of
diesel
unit
(i)
at
time
(t),
the
spinning
reserv
e
of
storage
system
unit
and
system
spinning
reserv
e
requirement
at
time
(t)
respecti
v
ely
.
N
G
X
i
=1
or
i;t
+
N
C
X
j
=1
or
j
;t
O
R
(
t
)
;
t
=
1
;
:::::T
(8)
where,
or
i;t
,
or
j
;t
and
O
R
(
t
)
are
the
operating
reserv
e
of
diesel
unit
(i)
at
time
(t),
the
operating
reserv
e
of
storage
system
unit
and
system
operating
reserv
e
requirement
at
time
(t)
respecti
v
ely
.
3.2.3.
Unit
ramping
limits
Unit
ramping
up
and
do
wn
limits
are
e
xpressed
as
in
(9)
and
(10)
respecti
v
ely
.
P
i;t
P
i;
(
t
1)
[1
I
i;t
(1
I
i;
(
t
1)
)]
P
R
U
i
+
I
i;t
(1
I
i;
(
t
1)
)
P
i;min
(9)
P
i;
(
t
1)
P
i;t
[1
I
i;
(
t
1)
(1
I
i;t
)]
P
R
D
i
+
I
i;
(
t
1)
(1
I
i;t
)
P
i;min
(10)
where
P
R
U
i
,
P
R
D
i
and
P
i;min
are
the
ramp-up,
ramp
do
wn
and
mini
mum
generation
limits
respec-
ti
v
ely
.
3.2.4.
Real
po
wer
generation
limits
The
real
po
wer
of
each
unit
are
restricted
by
the
lo
wer
and
upper
limits
as
e
xpressed
in
(11).
P
i;min
P
i;t
P
i;max
(11)
where
P
i;max
is
the
upper
limit
of
real
po
wer
generation
of
unit
i.
3.2.5.
Minimum
On/Off
time
limits
Unit
minimum
on
and
of
f
time
limits
are
e
xpressed
in
(12)
and
(13)
respecti
v
ely
.
8
>
>
>
>
>
>
>
>
>
>
>
>
>
<
>
>
>
>
>
>
>
>
>
>
>
>
>
:
T
U
i
=
max
(0
;
min
(
N
T
;
(
T
on
i
X
on
i
)
I
i
))
T
U
i
X
t
=1
(1
I
i;t
)
=
0
t
+
T
on
i
1
X
=
t
I
i
T
on
i
(
I
i;t
I
t
(
t
1)
)
8
t
=
T
U
i
+
1
;
:::;
N
T
T
on
i
1
N
T
X
=
t
[
I
i
(
I
i;t
I
t
(
t
1)
]
0
8
t
=
N
T
T
on
i
+
2
;
:::;
N
T
(12)
8
>
>
>
>
>
>
>
>
>
>
>
>
>
>
<
>
>
>
>
>
>
>
>
>
>
>
>
>
>
:
T
D
i
=
max
(0
;
min
(
N
T
;
(
T
of
f
i
X
of
f
i
)(1
I
i
)))
T
D
i
X
t
=1
(
I
i;t
)
=
0
t
+
T
of
f
i
1
X
=
t
(1
I
i
)
T
of
f
i
(
I
t
(
t
1)
I
i;t
)
;
8
t
=
T
D
i
+
1
;
:::;
N
T
T
of
f
i
1
N
T
X
=
t
[1
I
i
(
I
t
(
t
1)
I
i;t
)]
0
;
8
t
=
N
T
T
of
f
i
+
2
;
:::;
N
T
(13)
IJECE,
V
ol.
9,
No.
3,
June
2019
:
1553
–
1560
Evaluation Warning : The document was created with Spire.PDF for Python.
IJECE
ISSN:
2088-8708
r
1557
where
T
on
i
and
T
of
f
i
are
the
minimum
on/of
f
time
of
unit
i
respecti
v
ely;
T
U
i
and
T
D
i
are
the
hours
of
unit
i
must
be
initially
on/of
f
due
to
its
minimum
up/do
wn
time
limits
respecti
v
ely;
X
on
i
and
X
of
f
i
is
number
of
hours
of
unit
i
has
already
been
on/of
f
prior
to
the
first
hour
respecti
v
ely
.
3.3.
Constraints
f
or
CAES
As
mentioned,
CAES
has
t
w
o
m
od
e
s
of
w
orki
ng
(t
he
c
o
m
pression
and
the
e
xpansion
modes).
Three
w
orking
modes
of
CAES
are
considered,
the
compression
mode
which
CAES
acts
as
a
load,
the
e
xpansion
mode
which
acts
as
a
generator
and
an
idling
mode
which
CAES
is
not
operating
as
compression
or
e
xpansion
modes.
T
o
inte
grate
the
CAES
in
the
proposed
system
with
three
mentioned
modes,
it
should
consider
these
inte
ger
v
ariables
as
in
(14).
I
j
;t
+
I
c
j
;t
1
;
8
j
;
8
t
(14)
where,
I
j
;t
is
1
in
a
generation
mode
and
0
is
either
idle
or
compressor
mode;
I
c
j
;t
is
1
in
compressor
mode
and
0
is
idle
mode.
The
amount
of
injected
air
in
MW
at
hour
t
and
the
amount
of
released
air
in
MW
at
hour
t
are
li
mited
by
its
maximum
and
minimum
capacities
are
e
xpressed
in
(15)
and
(16)
respecti
v
ely
.
v
r
j
;min
v
r
j
;t
v
r
j
;max
(15)
v
inj
j
;min
v
inj
j
;t
v
inj
j
;max
(16)
Where,
v
r
j
;min
and
v
r
j
;max
are
minimum
and
amount
of
released
air
in
MW
while
v
inj
j
;min
and
v
inj
j
;max
are
the
minimum
and
maximum
amount
of
injected
air
in
MW
respecti
v
ely
.
In
the
compression
mode,
the
amount
of
compressed
air
is
limited
to
the
maximum
capacity
of
the
ca
v
ern
minus
the
current
in
v
entory
le
v
el
as
e
xpressed
in
(17)
and
(18)
respecti
v
ely
.
j
;t
+1
=
j
;t
+
v
inj
j
;t
v
r
j
;t
;
8
j
;
8
t
(17)
min
(
j
)
j
;t
max
(
j
)
;
8
j
;
8
t
(18)
where,
j
;t
and
j
;t
+1
are
the
in
v
entory
le
v
el
at
time
t
and
in
v
entory
le
v
el
at
time
t+1,
while
min
(
j
)
and
max
(
j
)
are
the
minimum
and
maximum
capacity
of
the
ca
v
ern
in
MWh
respecti
v
ely
.
The
mathematical
model
of
the
proposed
system
is
a
decision
problem
with
an
objecti
v
e
function
to
be
minimized
with
subjected
to
man
y
of
equality
and
i
nequality
constraints.
The
simulations
were
coded
in
MA
TLAB
with
GUR
OBI
8.0.0
solv
er
,
using
a
2.20
GHz
processor
with
4
GB
of
memory
and
64-bit
operating
system.
4.
CASE
STUD
Y
The
case
studies
are
performed
using
the
data
of
Nyala
DPP
and
n
yala
WF
with
the
inte
gration
of
CAES
to
obtain
the
operation
scheduling
of
the
system.
F
or
the
WF
,
the
t
o
t
al
generation
capacity
is
20
MW
produced
by
se
v
eral
turbines
unites
with
rated
po
wer
1.500MW
and
1800
MW
;
cut-in
and
cut-of
f
wind
speed
are
4
m/s
and
25
m/s
respecti
v
ely
.
The
parameters
of
LD
and
WP
forecasted
for
one
day
are
listed
in
T
able
1.
F
or
the
Nyala
DPP
,
the
electricity
is
generated
by
14
DGs
with
the
total
capacity
of
30
MW
as
listed
in
T
able
2.
The
CAES
has
a
maximum
and
minimum
generation
capacity
of
15
MW
and
3
MW
respecti
v
ely
.
There
are
three
cases
studied
i
s
performed.
The
first
one
is
only
Nyala
DPP
is
used
to
supply
the
LD;
the
second
one
is
the
inte
gration
of
Nyala
WF
into
Nyala
DPP
to
supply
the
LD
by
the
sharing
po
wer
and
the
third
case
is
the
inte
gration
of
CAES
instead
of
WF
where
there
is
no
wind
a
v
ailable.
T
able
1.
F
orecasted
Load
Lemand
and
W
ind
Po
wer
T
ime
(Hour)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
wind
(MW)
17
9
10
15
18
13
11
12
18
19
20
17
15
9
6
4
3
3
3
5
7
8
6
10
load
(MW)
22
14
15
20
23
19
21
24
29
28
29
27
29
22
18
16
14
14
9
22
23
21
17
16
Impact
of
compr
essed
air
ener
gy
stor
a
g
e
system
into...(Abdulla
Ahmed)
Evaluation Warning : The document was created with Spire.PDF for Python.
1558
r
ISSN:
2088-8708
T
able
2.
Diesel
Generators
P
arameters
of
Nyala
Po
wer
Plant
Units
Pmax
(MW)
Pmin
(MW)
c
($/
h)
b
($/kw)
a
($/
k
w
2
)
Min
On
Min
Off
ST
Ramp
Up
G1
1.600
0.480
37.2832
0.135318
0.00003056
3
-2
70
20
G2
1.600
0.480
44.0064
0.15005
0.000018336
1
-1
30
20
G3
1.876
0.562
23.5312
0.137826
0.00006112
1
-1
30
20
G4
1.876
0.562
11.9183
0.202613
0.000006112
4
-2
100
40
G5
1.876
0.562
31.1712
0.161051
0.000021392
3
-2
80
25
G6
1.200
0.480
58.9808
0.166858
0.000006112
3
-2
80
25
G7
1.200
0.480
16.1968
0.187944
0.000009168
5
-2
200
50
G8
3.520
1.408
37.2832
0.135318
0.00003056
4
-3
95
35
G9
3.520
1.408
44.0064
0.15005
0.000018336
4
-2
95
30
G10
1.840
0.552
23.5312
0.137826
0.00006112
3
-2
70
20
G11
1.840
0.552
11.91834
0.202613
0.000006112
1
-2
30
20
G12
2.640
0.792
31.1712
0.161051
0.000021392
1
-1
30
20
G13
2.640
0.792
58.9808
0.166858
0.000006112
4
-1
100
40
G14
2.640
0.792
16.1968
0.187944
0.000009168
3
-3
80
25
5.
RESUL
TS
AND
DISCUSSION
There
are
three
cases
are
discussed
in
this
section:
the
first
one
is
the
basic
case
which
the
s
y
s
tem
load
is
supplied
by
the
DPP
only;
the
second
one
is
the
inte
gration
of
the
wind
po
wer
into
DPP
without
CAES
system,
and
the
third
one
is
the
inte
gration
of
CAES
into
DPP
when
the
wind
po
wer
is
not
a
v
ailable.
5.1.
Case
I:
The
normal
operation
of
diesel
po
wer
plant
without
integration
of
the
wind
po
wer
and
CAES
In
this
case,
the
operation
schedule
is
determined
for
the
DPP
as
a
basic
supply
the
load
without
inte
gration
of
wind
po
wer
and
CAES
during
the
day-ahead
period
and
the
simulation
results
are
sho
wn
in
T
able
3.
The
re
sults
sho
w
that,
during
the
dispatched
period
only
the
cheaper
generator
units
are
committed
to
supplying
the
load
while
the
e
xpensi
v
e
generator
units
G1,
G4,
G7
and
G10
can
not
generate
an
y
po
wer
during
the
first
se
v
en
hou
r
s
and
then
started
to
produce
electricity
during
the
period
9:00-13:00
and
then
turn
of
f
ag
ain
during
the
period
14:00-24:00.
The
operation
cost
is
v
ery
high
because
the
operation
cost
only
depending
on
the
diesel
and
the
total
operating
costs
are
equal
to
$114740.
T
able
3.
Nyala
Diesel
Po
wer
Plant
only
Hours
Units
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
G1
0
0
0
0
0
0
0
0
1
1
1
1
1
0
0
0
0
0
0
0
0
0
0
0
G2
1
0
1
1
1
0
1
1
1
1
1
1
1
1
0
0
0
0
0
1
1
1
0
0
G3
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0
1
1
1
0
1
1
1
1
1
G4
0
0
0
0
0
0
0
1
1
1
1
1
1
0
0
0
0
0
0
0
0
0
0
0
G5
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
G6
0
1
0
1
1
1
1
0
1
1
1
1
1
0
1
0
1
1
1
0
1
1
1
0
G7
0
0
0
1
0
0
0
0
1
1
1
1
1
0
0
0
0
0
0
0
0
0
0
0
G8
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0
1
1
1
1
1
G9
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
G10
0
0
0
0
0
0
0
0
1
1
1
1
1
0
0
0
0
0
0
0
0
0
0
0
G11
1
0
0
0
1
1
0
1
1
1
1
1
1
1
0
0
0
0
0
1
1
0
0
0
G12
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
G13
1
0
0
0
1
0
1
1
1
1
1
1
1
1
1
0
0
0
0
1
1
1
0
0
G14
1
0
0
1
1
1
1
1
1
1
1
1
1
1
1
1
0
0
0
1
1
1
1
1
5.2.
Case
II:
The
operation
of
the
system
with
diesel
po
wer
plant
and
wind
po
wer
integration
This
case
e
xplained
the
inte
gration
of
wind
po
wer
into
the
diesel
po
wer
plant
to
serv
e
the
load
by
the
shared
po
wer
and
the
results
are
sho
wn
in
T
able
4.
From
the
results,
all
the
e
xpensi
v
e
generator
units
G1,
G2
G4,
G7,
G10,
G11,
G13
are
turned
of
f
during
the
whole
operation
period
and
the
load
demand
can
be
supplied
by
the
other
cheapest
generators
only
.
The
total
operating
costs
is
equal
to
$45576
and
this
cost
is
less
than
the
first
case
because
the
e
xpensi
v
e
generator
units
are
substituted
by
the
wind
po
wer
to
serv
e
the
load.
IJECE,
V
ol.
9,
No.
3,
June
2019
:
1553
–
1560
Evaluation Warning : The document was created with Spire.PDF for Python.
IJECE
ISSN:
2088-8708
r
1559
T
able
4.
Nyala
Diesel
Po
wer
Plant
with
W
ind
Po
wer
Hours
Units
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
G1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
G2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
G3
0
0
0
0
0
0
0
0
1
0
1
0
1
1
0
0
1
1
0
1
1
1
1
0
G4
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
G5
1
0
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
G6
0
0
0
0
0
1
1
1
1
1
0
1
1
0
1
1
1
1
0
1
0
0
1
1
G7
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
G8
0
0
0
0
0
0
1
1
0
1
0
1
1
1
0
1
0
0
0
1
1
1
0
0
G9
1
0
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
G10
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
G11
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
G12
0
0
0
0
0
0
0
1
1
0
1
1
1
1
1
1
1
1
0
1
1
1
1
0
G13
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
G14
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
0
0
0
5.3.
Case
III:
System
operation
with
diesel
po
wer
plant
and
impact
of
compr
essed
air
ener
gy
storage
system
This
case
e
xplains
the
performance
of
the
proposed
system
when
CAES
is
char
ged
by
using
the
e
xcessi
v
e
po
wer
from
the
wind
during
the
of
f-peak
period
and
then
dischar
ge
it
when
there
is
no
wind
po
wer
during
the
peak
period
and
the
simulation
results
are
sho
wn
in
T
able
5.
As
sho
wn
in
the
results,
the
generator
units
G1,
G4
and
G10
are
turned
of
f
during
the
whole
operation
periods
because
the
y
are
e
xpensi
v
e
while
the
other
generator
units
are
committed
e
xcept
units
G7
and
G11
are
committed
only
at
9:00
and
8:00
respecti
v
ely
.
The
total
operating
costs
in
this
situation
are
$45576
and
this
cost
is
relati
v
ely
high
compared
with
the
second
case
because
the
capacity
of
ener
gy
storage
system
is
less
than
the
capacity
of
the
wind
f
arm.
T
able
5.
Nyala
Diesel
Po
wer
Plant
with
Battery
Ener
gy
Storage
Hours
Units
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
G1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
G2
1
0
0
0
0
0
0
0
1
0
0
1
1
1
1
0
0
0
0
0
0
0
0
0
G3
1
0
0
0
1
0
1
0
1
1
1
1
1
1
1
1
0
0
1
1
1
1
1
1
G4
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
G5
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
G6
1
0
1
1
1
1
0
1
1
1
1
1
1
0
1
1
1
1
0
1
1
1
1
1
G7
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
G8
1
0
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0
1
1
1
1
0
G9
1
0
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
G10
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
G11
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
G12
0
1
0
1
1
1
1
1
1
1
1
1
1
1
0
1
1
1
0
1
1
1
1
1
G13
0
0
0
0
0
0
0
1
0
1
1
0
1
0
0
0
0
0
0
1
1
0
0
0
G14
0
0
0
0
0
0
1
1
1
1
1
1
1
1
0
0
0
0
0
1
1
1
1
1
6.
CONCLUSION
The
h
ybrid
system
includes
diesel-wind-CAES
has
been
Modeled
and
simulated
to
obtain
the
opera-
tional
pl
anning
of
the
proposed
system.
From
the
simulation,
its
observ
ed
that
the
inte
gration
of
wind
f
arm
into
diesel
po
wer
plant
i
s
the
best
choice
for
minimizing
the
use
of
fossil
fuels
as
well
as
decreasing
the
greenhouse
emissions
such
as
C
O
2
.
The
CAES
plays
an
essential
role
in
the
h
ybrid
system
to
mitig
ate
the
v
ariabilit
y
nature
of
the
wind
ener
gy
by
storing
the
e
xcessi
v
e
wind
ener
gy
during
the
of
f-peak
period
and
reused
it
during
the
peak
period.
Thus
the
proposed
system
is
useful
for
helping
in
inte
grating
rene
w
able,
backup
system
and
mak
e
management
for
the
grid
system.
A
CKNO
WLEDGEMENT
This
research
w
as
financially
supported
by
t
he
Science
and
T
echnology
Proj
ect
of
State
Grid
Corpo-
ration
of
China
No.SGHE0000KXJS1700086.
Impact
of
compr
essed
air
ener
gy
stor
a
g
e
system
into...(Abdulla
Ahmed)
Evaluation Warning : The document was created with Spire.PDF for Python.
1560
r
ISSN:
2088-8708
REFERENCES
[1]
Chen
SX,
H.
B.
Gooi
and
M.
Q.
W
ang.,
”Sizing
of
ener
gy
storage
for
microgrids,
”
IEEE
T
ransactions
on
Smart
Grid
,
v
ol.
3,
pp.
142-51,
Mar
2012.
[2]
Daneshi
A,
et
al.,
”Inte
gration
of
wind
po
wer
and
ener
gy
storage
in
SCUC
problem,
”
in
W
orld
Non-Grid-
Connected
W
ind
Po
wer
and
Ener
gy
Conference,
2010.
WNWEC
2010.IEEE
,
2010,
pp.
1-8.
[3]
Daneshi
H,
A.K.
Sri
v
asta
v
a,
et
al.,
”Generation
scheduling
with
inte
gration
of
wind
po
wer
and
compressed
air
ener
gy
storage.,
”
in
T
ransmission
and
Distrib
ution
Conference
and
Exposition,2010.
IEEE
PES
2010.
,
pp.
1-6.
[4]
Daneshi
H,
et
al.,
”Security-constrained
unit
commitment
in
a
syste
m
with
wind
generation
and
compressed
air
ener
gy
storage,
”
in
Ener
gy
Mark
et,
2009.
6th
International
Conference
on
the
European
,
2009,
pp.
1-6.
[5]
Daneshi
H
and
Sri
v
asta
v
a
AK.,
”Security-constrained
unit
commitment
with
wind
generati
on
and
com-
pressed
air
ener
gy
storage,
”
IET
Generation,
T
ransmission
and
Distrib
ution
,
v
ol.
6,
pp.
167-75,
Feb
2012.
[6]
ABDULLA
AHMED
and
T
ONG
JIANG.
”Thermodynamic
Analysis
of
Compressed
Air
Ener
gy
Storage
with
Inte
gration
of
W
ind
Po
wer
.
”
DEStech
T
ransactions
on
En
vironment,
Ener
gy
and
Earth
Sciences,
2018,
pp.
37-43.
[7]
B.C.Ummels,
E.
Pelgrum
and
W
.L.
Kling.,
”Inte
gration
of
lar
ge-scale
wind
po
wer
and
use
of
ener
gy
storage
in
the
Netherlands
electricity
supply
,
”
IET
Rene
w
able
Po
wer
Generation
,
v
ol.
2,
pp.
3446,
Feb
2008.
[8]
Jiaming
LI,
Jin
yu
WEN
and
Xingning
HAN.,
”Lo
w-carbon
unit
commitment
with
intensi
v
e
wind
po
wer
generation
and
carbon
capture
po
wer
plant,
”
J.
Mod.
Po
wer
Syst.
Clean
Ener
gy
,
v
ol.
3,
pp.
6371,
Mar
2015.
[9]
K
en
KUR
OD
A.,
et
al.,
”A
h
ybrid
multi-objecti
v
e
optimization
method
considering
optimization
problems
in
po
wer
distrib
ution
systems,
”
J.
Mod.
Po
wer
Syst.
Clean
Ener
gy
,
v
ol.
3,
pp.
4150,
Feb
2015.
[10]
Ahmed,
Abdulla,
and
T
ong
Jiang.
”Modeling
and
Performance
Analysis
of
V
ariable
Speed
W
ind
T
ur
-
bines.
”
In
2018
International
Conference
on
Computer
,
Control
,
Electrical,
and
Electronics
Engineering
(ICCCEEE),
pp.
1-5.
IEEE,
2018.
[11]
X.
S.
Li.,
et
al.,
”W
ind
Inte
grated
Thermal
Unit
Commitment
Solution
using
Gre
y
W
olf
Optimizer
,
”
International
Journal
of
Electrical
and
Computer
Engineering
,
v
ol.
7,
pp.
2309
2320,
Oct
2017.
[12]
L
yon
JD,
Zhang
M
and
Hedman
KW
.,
”Capacity
response
sets
for
security-constrained
unit
commitment
with
wind
uncertainty
,
”
Electric
Po
wer
Systems
Research
,
v
ol.
136,
pp.
21-30.,
Jul
2016.
[13]
Ahmed,
Abdulla,
and
T
ong
Jiang.
”Inte
gration
of
Hybrid
W
ind/Battery
System
into
Diesel
Po
wer
Plant.
”
In
2018
International
Conference
on
Computer
,
Control,
Electrical,
and
Electronics
Engineering
(ICC-
CEEE),
pp.
1-6.
IEEE,
2018.
BIOGRAPHY
OF
A
UTHORS
Abdulla
Ahmed
is
a
lecturer
at
Department
of
Electrical
and
Electronics
Engineering,
F
aculty
of
Engineering
Science,
Uni
v
ersity
of
Nyala,
Nyala
-
Sudan
with
Mas
ter
of
science
in
electrical
po
wer
from
Sudan
Uni
v
ersity
of
Science
and
T
echnology
,
Khartoum
-
Sudan
(2012).
He
obtained
his
Bachelor
De
gree
in
Electrical
Engineering
from
Omdurman
Islamic
Uni
v
ersity
,
Omdurman
-
Sudan
(2008).
Currently
,
he
is
w
orking
to
w
ard
the
Ph.D.
de
gree
at
t
he
school
of
electrical
and
electronic
engineering,
North
China
Electric
Po
wer
Uni
v
ersity
,
Beijing-China.
His
researches
are
in
fields
of
po
wer
system
operation
and
control,
inte
gration
of
rene
w
able
ener
gy
into
the
po
wer
grid
system
and
smart
po
wer
grids
systems.
He
is
af
filiated
with
IEEE
as
a
student
member
.
T
ong
Jiang
is
a
professor
at
the
School
of
Electrical
and
Electronic
Engineering,
North
China
Electric
Po
wer
Uni
v
ersity
,
Beijing-China.
He
obtained
Bachelor
De
gree,
Master
de
gree
and
a
doctorate
in
Electrical
Engineering
from
Harbin
Institute
of
T
echnology
(China)
in
1992,
1995
and
2002
respecti
v
ely
.
From
2003
to
2005,
he
conducted
postdoctoral
research
in
the
Department
of
Electrical
Engineering,
Tsinghua
Uni
v
ersity
,
Beijing
-
China.
His
researches
are
in
fields
of
electric
po
wer
system
analysis
and
control,
lar
ge-scale
electric
po
wer
ener
gy
storage
technology
,
smart
distrib
ution
grid
and
f
ault
location.
He
has
made
bril-
liant
achie
v
ements
in
electric
po
wer
system
f
ault
calculations,
load
flo
w
calculations,
and
ne
w
compressed
air
ener
gy
storage.
IJECE,
V
ol.
9,
No.
3,
June
2019
:
1553
–
1560
Evaluation Warning : The document was created with Spire.PDF for Python.