14 trang 11 lượt tải

An outlook towards hydrogen supply chainnetworks in 2050 — Design of novel fuelinfrastructures in Germany

22

An outlook towards hydrogen supply chainnetworks in 2050 — Design of novel fuelinfrastructures in Germany. Tài liệu được sưu tầm giúp bạn tham khảo, ôn tập và đạt kết quả cao. Mời bạn đọc đón xem.

Môn: Tài liệu Tổng hợp 3.5 K tài liệu

Trường: Tài liệu khác 3.8 K tài liệu

Tác giả:

Profile

HUU TUAN NGUYEN BUI

2 ngày trước
Tải xuống
Báo cáo
Danh sách Quiz
Tải xuống
/14
Chemical
Engineering
Research
and
Design
1
3
4
(
2
0
1
8
)
90–103
Contents
lists
available
at
ScienceDirect
Chemical
Engineering
Research
and
Design
journal
h
om
epage:
www.elsevier.com/locate/cherd
An
outlook
towards
hydrogen
supply
chain
networks
in
2050
—
Design
of
novel
fuel
infrastructures
in
Germany
Anton
Ochoa
Bique
∗
,
Edwin
Zondervan
Laboratory
of
Process
Systems
Engineering,
Department
of
Production
Engineering,
Universität
Bremen,
Leobener
Str.
6,
28359
Bremen,
Germany
a
r
t
i
c
l
e
i
n
f
o
Article
history:
Received
30
August
2017
Received
in
revised
form
20
March
2018
Accepted
23
March
2018
Available
online
3
April
2018
Keywords:
Hydrogen
supply
chain
design
Fuel
infrastructures
Mixed
integer
linear
programming
Germany
AIMMS
a
b
s
t
r
a
c
t
This
work
provides
a
comprehensive
investigation
of
the
feasibility
of
hydrogen
as
trans-
portation
fuel
from
a
supply
chain
point
of
view.
It
introduces
an
approach
for
the
identification
the
best
hydrogen
infrastructure
pathways
making
decision
of
primary
energy
source,
production,
storage
and
distribution
networks
to
aid
the
target
of
greenhouse
gas
emissions
reduction
in
Germany.
The
minimization
of
the
total
hydrogen
supply
chain
(HSC)
network
cost
for
Germany
in
2030
and
2050
years
is
the
objective
of
this
study.
The
model
presented
in
this
paper
is
expanded
to
take
into
account
water
electrolysis
technology
driven
by
solar
and
wind
energy.
Two
scenarios
are
evaluated,
including
a
full
range
of
technologies
and
“green”
technologies
using
only
renewable
resources.
The
resulting
model
is
a
mixed
integer
linear
program
(MILP)
that
is
solved
with
the
Advanced
Integrated
Multidimensional
Modeling
System
(AIMMS).
The
results
show
that
renewable
energy
as
a
power
source
has
the
potential
to
replace
common
used
fossil
fuel
in
the
near
future
even
though
currently
coal
gasification
technology
is
the
still
the
dominant
technology.
©
2018
Institution
of
Chemical
Engineers.
Published
by
Elsevier
B.V.
All
rights
reserved.
1.
Introduction
Up
till
now,
fossil
fuels,
which
include
natural
gas,
oil,
and
coal
are
the
primary
energy
sources
for
transportation,
electricity,
and
residen-
tial
services.
Based
on
a
report
by
the
International
Energy
Agency
(IEA)
and
the
U.S.
EIA,
the
global
energy
demand
will
grow
with
30%
in
2040
(International
Energy
Agency
(IEA),
2016;
U.S.
Energy
Information
Administration,
2017).
This
means
a
progressively
growing
fuel
con-
sumption
in
the
near
future
i.e.
greenhouse
gas
emissions
such
as
carbon
dioxide
also
increase.
Fossil
fuel
is
a
nonrenewable
energy
source.
The
depletion
time
for
fossil
fuel
is
estimated
to
be
around
100
years,
where
oil
and
gas
will
be
exhausted
earlier
than
coal
(Shafiee
and
Topal,
2009).
Moreover,
due
to
increasing
fuel
consumption,
cause
of
concern
is
the
fast
rise
of
CO
2
level,
now
already
exceeding
400
ppm
level
and,
left
unmitigated,
can
possibly
double
in
100
years
to
800
ppm
(
CO2.earth,
n.d.).
Due
to
the
increasing
demand
of
electric
energy
and
a
decreas-
ing
amount
of
fossil
fuel
sources,
the
development
of
solar-,
wind-
∗
Corresponding
author.
E-mail
address:
antochoa@uni-bremen.de
(A.
Ochoa
Bique).
and
biomass
based
production
of
energy
and
chemicals
is
strongly
supported
by
governments
(Schill,
2014).
For
example,
the
German
government
decided
to
completely
phase
out
nuclear
energy
by
2022
(
Pregger
et
al.,
2013)
and
replace
it
with
renewable
energy
production.
The
largest
part
of
renewable
power
will
come
from
solar
and
wind
as
shown
in
Fig.
1.
Electric
power
from
wind
mills
increases
its
contribu-
tion
by
225
TWh
in
2050,
which
is
39%
of
the
final
produced
energy;
solar
contributes
17%,
at
100
TWh
per
year,
while
biomass
reaches
60
TWh
per
year.
While
biomass
as
a
raw
material
might
be
stored
for
a
long
period
of
time,
wind
and
solar
are
more
difficult
to
handle.
As
battery
sys-
tems
do
currently
not
have
enough
capacity
and
storage
of
electricity
is
very
expensive,
the
developments
in
new
long-term
storage
technol-
ogy
is
one
of
the
main
challenges.
Industrial
key
players,
like
Siemens
currently
work
on
a
new
type
of
energy
storage
system
based
on
hydro-
gen
production
(Siemens,
n.d.).
The
main
idea
is
that
excess
energy
from
renewable
energy
sources
can
be
converted
into
hydrogen
from
water
by
electrolysis,
which
is
a
non-toxic
source
of
energy
to
con-
https://doi.org/10.1016/j.cherd.2018.03.037
0263-8762/©
2018
Institution
of
Chemical
Engineers.
Published
by
Elsevier
B.V.
All
rights
reserved.
Chemical
Engineering
Research
and
Design
1
3
4
(
2
0
1
8
)
90–103
91
Nomenclature
Indices
e
Type
of
energy
source
f
Type
of
hydrogen
physical
form
g
Grid
points,
each
grid
point
represents
German
state
p
Type
of
hydrogen
production
facility
s
Type
of
storage
facility
t
Type
of
transportation
mode
Abbreviation
CH
Compressed-gaseous
hydrogen
FCEV
Fuel
cell
electric
vehicle
HSC
Hydrogen
supply
chain
IEA
International
Energy
Agency
LH
Liquid
hydrogen
USEIA
U.S.
Energy
Information
Administration
Continuous
variable
EESAv
e,g
Amount
of
available
energy
source
e
in
the
grid
point
g,
which
is
used
to
satisfy
energy
demand
in
grid
point
g
[kWh
d
−1
]
EESN
e,g

,
g
The
flowrate
of
the
supplied
energy
source
e
from
neighboring
grid
point
g

to
grid
point
g,
which
is
used
to
satisfy
energy
demand
in
grid
point
g
[kWh
d
−1
]
ESAv
e,g
Amount
of
available
energy
source
e
in
the
grid
point
g
[unit
e
d
−1
]
ESC
Total
cost
for
the
energy
source
consumed
for
hydrogen
production
[$
d
−1
]
ESD
g
Daily
energy
source
demand
[kWh
d
−1
]
HD
g
Hydrogen
demand
by
grid
point
[kg
d
−1
]
HF
g,g
,
t,
f
Hydrogen
flowrate
in
the
form
f
from
grid
point
g
to
g
via
transportation
mode
t
[kg
d
−1
]
HHED
g
Total
energy
demand
in
the
grid
point
g
[kWh
d
−1
]
HP
g,
f
Hydrogen
generation
in
the
form
f
at
grid
point
g
[kg
d
−1
]
HP
p,g,
f
Amount
of
produced
hydrogen
in
the
produc-
tion
facility
p
in
the
form
f
at
grid
point
g
[kg
d
−1
]
PC
Daily
production
costs
[$
d
−1
]
PCC
Production
capital
cost
[$
d
−1
]
PESAv
e,g
Amount
of
available
energy
source
e
in
the
grid
point
g,
which
is
used
to
satisfy
energy
source
demand
for
hydrogen
production
[unit
e
d
−1
]
PESIm
e,g

,
g
Flowrate
of
imported
energy
source
e
from
neighboring
grid
point
g

to
grid
point
g,
which
is
used
to
satisfy
energy
source
demand
for
hydrogen
production
[unit
e
d
−1
]
PESN
e,g

,
g
Flowrate
supplying
energy
source
e
from
neighboring
grid
point
g

to
grid
point
g
[unit
e
d
−1
]
POC
Production
operating
cost
[$
d
−1
]
PCC
p,
f
Capital
cost
of
facility
p,
producing
hydrogen
in
form
f
[$]
POC
p,
f
Hydrogen
production
operating
cost
in
form
f
at
facility
p
[$
kg
−1
]
SC
The
total
hydrogen
storage
cost
[$]
SCC
p,
f
Capital
cost
for
storage
facility
s
holding
hydro-
gen
in
the
form
f
[$]
SOC
p,
f
Operating
cost
to
store
1
kg
of
hydrogen
in
the
from
f
inside
of
storage
facility
s
[$
kg
−1
d
−1
]
TC
Daily
distribution
cost
[$
d
−1
]
TCC
f
,t
Capital
cost
of
transport
mode
t
for
distribution
hydrogen
in
the
form
f
[$]
Total
Total
cost
of
HSC
network
[$]
Integer
variables
NPF
p,
f
,g
Number
of
production
facility
p
generating
hydrogen
in
from
f
at
grid
point
g
NSF
s,
f
,g
Number
of
storage
facility
s
holding
hydrogen
in
the
form
f
at
grid
point
g
NTF
g,g
,
f
,t
The
number
of
transport
mode
t
used
for
hydrogen
distribution
in
the
form
f
from
g
to
g
PN
g
Population
at
the
grid
point
g
Parameters
AvCon Average
of
household
energy
consumption
[kWh
d
1
]
AvD
The
average
distance
travelled
by
personal
car
[km
y
1
]
AvT
The
average
amount
of
personal
car
per
1000
people
AF
p
Annual
factor
for
the
facility
p
[%]
AF
s
Annual
factor
for
the
s
storage
facility
s
[%]
AF
t
Annual
factor
for
the
transport
mode
t
[%]
Dis
g

,
g
Distance
between
grid
points
[km]
Dis
g,g
,
t
Distance
between
grid
points
depending
of
type
of
transport
[km]
ESCost
e
Energy
source
e
price
in
year
y,
generated
locally
[$
unit
−1
e]
ESDis
e
Delivery
price
for
energy
source
e
[$
unit
−1
km
−1
]
ESICost
e
Energy
source
e
import
price
[$
unit
−1
]
FE
The
fuel
economy
[kg
H
2
km
−1
]
FP
t
Fuel
price
for
transport
mode
t
[$
l
−1
]
MaxPCap
p
/MinPCap
p
Max/min
production
capacity
for
hydrogen
production
facility
p
[kg
d
−1
]
OP
Operating
period
[d
y
−1
]
SCap
s,f
Capacity
of
storage
facility
s
for
holding
hydro-
gen
in
the
from
f
[kg]
Greek
letters
˛
e,p
The
ratio
between
energy
sources
e
consump-
tion
to
produce
1
kg
[unit
e
kg
1
H
2
]
FCEVs
penetration
rate
[%]
Is
total
product
storage
period
[d]
sumers
allowing
a
greater
energy
security
and
flexibility.
As
soon
as
there
is
energy
shortage,
hydrogen
might
be
used
in
different
appli-
cations
such
as
power
generation,
domestic
and
industrial
services,
navigation
and
space
(Hake
et
al.,
2006).
However,
hydrogen
is
not
a
naturally
occurring
fuel
of
mineral
origin;
it
can
be
produced
from
both
renewable
and
non-renewable
resources:
from
coal
and
biomass
gasification,
the
reforming
of
natural
gas,
from
water
electrolysis,
photo-electrolysis,
water-splitting
thermochemical
cycle,
photobio-
logical
production,
and
high
temperature
decomposition.
Moreover,
hydrogen
generation
is
only
a
part
of
the
hydrogen
production
net-
work,
which
can
be
defined
as
a
supply
chain
consisting
of
several
components
(such
as
production,
storage
and
distribution).
For
each
of
these
stages
a
wide
range
of
potential
technological
options
exist.
Due
to
increasing
demand
for
energy,
the
development
of
sustainable
and
environmental
friendly
concepts
such
as
the
HSC
should
be
devel-
oped
to
replace
non-sustainable
alternatives
to
meet
the
global
need
for
92
Chemical
Engineering
Research
and
Design
1
3
4
(
2
0
1
8
)
90–103
0
50000
100000
150000
200000
250000
1990
1993
1996
1999
2002
2005
2008
2011
2014
2017
2020
2023
2026
2029
2032
2035
2038
2041
2044
2047
Year
Wind electricity production Solar electricity prodution
Biomass Projection for Wind electricity production
Fig.
1
–
Projection
of
energy
generation.
energy
(Ball
et
al.,
2007).
The
work
of
Hugo
et
al.
(2005)
takes
all
possible
hydrogen
alternatives
for
the
design
of
an
optimal
hydrogen
infras-
tructure
in
Germany
in
to
consideration.
However,
their
model
does
not
include
energy
sources
distribution
and
the
ability
of
centralized
hydrogen
storage
to
satisfy
the
local
demand.
A
study
of
Almansoori
investigated
a
number
of
strategic
decisions
for
hydrogen
fuel
pro-
duction
and
hydrogen
delivery
networks
in
Germany
at
large-scale
considering
emission
targets
and
carbon
tax
as
a
part
of
the
model
for-
mulation
for
2030
(Almansoori
and
Betancourt-Torcat,
2016).
The
main
objective
in
that
study
was
to
satisfy
the
hydrogen
demand
which
was
determined
by
a
fuel
cell
electric
vehicles
(FCEVs)
penetration
of
10%
of
the
overal
passenger
transport.
The
results
showed
that
liquefied
hydrogen
production
by
coal
gasification
facilities
at
large-scale
and
delivery
via
railway
tank
cars
results
in
the
best
HSC
network
struc-
ture.
Large-size
facilities
showed
benefit
compared
to
a
small-scale
facility
since
large
facilities
have
a
high
energy
efficiency.
Renewable
energy
such
as
wind
and
solar
were
not
included
in
that
study
due
to
technical-
and
economical
hurdles
such
as
expenses
of
electricity
price
for
water
electrolysis
technology
and
size-independent
electrolyzer
efficiency.
The
rate
of
renewable
energy
consumption
to
generate
a
unit
of
hydrogen
for
both
sizes
of
electrolysis
facility
is
identical
as
the
elec-
trolyzer
efficiency
is
independent
of
the
facility
size.
A
similar
model
was
developed
for
the
United
Kingdom
(Almansoori
and
Shah,
2012).
The
objective
was
the
minimization
of
the
cost
of
the
network
consider-
ing
capital-
and
operating
costs.
The
results
showed
the
dominance
of
steam
methane
reforming
technology.
Large-scale
electrolysis
facilities
were
not
considered
due
to
a
size-independent
electrolyzer
efficiency
that
was
mentioned
before.
The
aim
of
this
paper
is
to
develop
and
evaluate
an
optimization
model
that
can
be
used
to
solve
a
HSC
network
design
problem
forecast-
ing
in
2030
and
2050
years
while
considering
a
full
range
of
local
factors
such
as
(i)
energy
sources
distribution
for
hydrogen
production,
(ii)
local
hydrogen
demand
and
(iii)
distribution
between
the
place
of
hydrogen
production
and
hydrogen
demand.
The
model
is
used
to
define
the
procurement
of
energy
sources
from
the
supplier,
the
type,
the
num-
ber
and
the
location
of
a
production
facility,
the
hydrogen
production
form
and
the
delivery
of
hydrogen
to
consumers.
The
logistics
of
renew-
able
sources
is
also
included
into
the
model
by
accounting
for
personal
needs
such
as
household
energy
and
hydrogen
based
fuel
consump-
tion.
In
addition,
this
work
also
considers
environmental
influences.
Moreover,
all
techno-economic
parameters
were
collected
for
2015
and
were
assumed
stay
same
as
for
the
reference
years.
The
German
land-
scape
provides
an
important
case
study
as
Germany
has
an
immense
potential
to
develop
a
sustainable
hydrogen
infrastructure
(Hake
et
al.,
2006
).
In
the
following
sections
we
will
define
the
problem
statement,
develop
and
describe
the
model,
setup
a
case
study
and
interpret
and
discuss
the
results
obtained.
2.
Network
description
and
problem
statement
2.1.
Problem
statement
Given
are
the
location
and
capacity
of
energy
source
suppliers,
capital
and
operating
costs
for
a
large-scale
hydrogen
pro-
duction,
transportation
and
storage
facilities
of
the
network,
under
the
conditions
that:
1.
location
of
storage
facilities
is
fixed;
2.
all
natural
gas
is
imported
(despite
a
national
12%
produc-
tion
of
natural
gas);
3.
weighted
average
cost
of
capital
for
production,
transporta-
tion
and
storage
is
10%;
4.
electricity
is
the
main
energy
source
to
power
rail
freight
transport
(International
Union
of
Railways,
2012);
5.
the
way
of
handling
of
residual
waste
is
not
considered;
6.
secondary
energy
carriers
have
no
economic
value
in
this
network
model;
7.
electricity
price
based
on
industrial
electricity
price
for
Germany.
The
HSC
consists
of
energy
sources
from
different
origins,
large-scale
hydrogen
production
technologies,
hydrogen
prod-
uct
form
and
the
hydrogen
distribution
and
storage
options
(
Fig.
2).
Five
types
of
energy
sources
are
considered:
wind-
and
solar
energy,
biomass,
natural
gas
and
coal.
In
addition,
four
types
of
hydrogen
production
technologies
are
included
into
the
model:
steam
methane
reforming,
coal
gasifica-
tion,
biomass
gasification
and
water
electrolysis.
As
hydrogen
might
be
generated
by
different
production
technologies,
it
may
be
transported
into
two
forms
(i.e.
liquid
or
gaseous),
which
determines
the
transportation
mode
that
will
be
used.
The
liquid
form
(LH)
could
be
stored
in
super-insulated
spher-
ical
tanks
and
be
distributed
via
two
types
of
transportation
modes:
by
railway
tank
car
or
via
tanker
truck.
Gaseous
hydro-
gen
(CH)
could
be
stored
into
pressurized
cylindrical
vessels
Chemical
Engineering
Research
and
Design
1
3
4
(
2
0
1
8
)
90–103
93
Biomass
Wind
H
2
Electrolysis
Gasification
ENERGY
SOURCE
TYPE OF PLANT PRODUCT
PRODUCT FORM
DISTRIBUTION
MODE
STORAGE
OPTION
Liquefaction
Compression
Tube trailer
Railway tube
car
Railway tank
car
Tanker truck
Pressurized
Cylindrical
vessels
Super-insulated
Spherical tanks
PRODUCTION DISTRIBUTION
Solar
Steam
reforming
Coal
Natural gas
Fig.
2
–
Structure
of
the
hydrogen
supply
and
delivery
chain.
Table
1
–
Parameters
used
for
total
hydrogen
demand
calculation
in
Germany.
Parameter
Passenger
transport
system
in
Germany
2015
Number
45,046,564
Average
distance
travelled
(km
y
−1
)
13,000
Fuel
economy
(kg
H
2
km
−1
)
0.018
and
be
distributed
via
railway
tube
car
or
tube
trailer.
Each
facility
of
the
HSC
includes:
a
technological
option,
a
capacity,
a
location.
The
problem
is
concerned
with
finding
the
number
and
locations
of
the
production
facilities
for
a
given
demand,
while
minimizing
the
total
operating
HSC
network
cost.
2.2.
Model
description
In
Fig.
3
the
superstructure
of
HSC
model
is
shown
(Hugo
et
al.,
2005).
The
superstructure
includes
all
the
possible
connections
between
the
model
components.
Ultimately,
an
optimization
algorithm
is
used
to
search
for
the
best
strategy
to
minimize
the
costs
of
the
HSC
network.
The
superstructure
consists
a
set
of
grid
points
(g,
each
grid
point
represents
a
German
state),
energy
sources
(e),
different
transportation
(t)
modes,
different
hydrogen
production-(p)
and
storage
(s)
facil-
ities.
The
transportation
modes
are
used
to
distribute
different
types
of
hydrogen
(f)
from
production
facility
to
storage
facil-
ity.
In
the
following
subsections,
each
component
of
the
HSC
model
will
be
described
in
more
detail.
2.2.1.
Grid
In
this
study,
the
landscape
of
Germany
is
divided
into
16
grid
points,
each
of
this
grid
points
represents
a
German
region.
Moreover,
the
region’s
largest
city
is
taken
as
the
potential
cen-
ter
for
a
hydrogen
production
facility
and
for
a
storage
facility
to
satisfy
the
local
demand
and
further
product
distribution
to
another
region
(Almansoori
and
Betancourt-Torcat,
2016).
The
total
hydrogen
demand
was
estimated
based
on
FCEVs
pene-
tration
rate
of
10%
in
the
total
number
of
passenger
transports
(public
buses,
light
motor
vehicle)
available
by
2030,
and
30%
by
2050,
average
distance
travelled
and
transport
fuel
economy
(see
Table
2)
(BM
Verkehr
Bau
und
Stadtentwicklung,
2013).
2015
was
used
as
the
reference
year
for
the
calculations.
All
relevant
parameters
are
listed
in
Table
1.
Based
on
the
pro-
jections
of
energy
consumption
from
1960
to
2050
(as
shown
in
Fig.
4),
the
household
energy
demand
was
estimated
over
time.
2.2.2.
Primary
energy
sources
Hydrogen
can
be
produced
from
different
sources
such
as
water,
natural
gas,
biomass
and
coal.
This
resource
availability
at
each
grid
point
plays
an
important
role
in
defining
the
type
and
location
of
production
technologies
(see
Appendix
F).
In
addition,
the
main
problem
of
a
domestic
production
facility
is
concerned
with
finding
an
appropriate
energy
source
supplier.
There
are
three
opportunities
related
with
the
energy
source
consumption
from
(i)
a
domestic
grid
point
or
(ii)
supply
from
neighboring
grid
points
or
(iii)
import
from
abroad.
2.2.3.
Hydrogen
production
Considering
that
hydrogen
is
not
a
naturally
occurring
fuel
of
mineral
origin,
different
production
technologies,
including
steam
methane
reforming,
coal
gasification,
biomass
gasifica-
tion
and
water
electrolysis,
are
generally
used
to
generate
it.
Each
alternative
has
fixed
capital
and
operational
costs
(see
Appendix
A).
The
main
decision
is
to
determine
the
type,
location
and
number
of
production
facilities.
Each
facility
car-
ries
out
large-scale
hydrogen
production
(960
t
per
day)
(see
Table
3).
2.2.4.
Hydrogen
physical
form
Hydrogen
can
be
carried
in
two
physical
forms:
liquid
and
gaseous.
Each
form
is
distributed
by
different
transportation
modes
and
might
be
stored
in
special
storage
facilities.
The
hydrogen
form
plays
an
important
role
in
defining
which
transportation
mode
and
storage
facilities
should
be
used.
These
decisions
affect
the
final
costs
of
the
HSC
network.
2.2.5.
Transportation
mode
The
transportation
mode
is
related
to
the
hydrogen
form
(gas
or
liquid).
The
main
decision
is
to
define
the
type
transporta-
tion
mode
and
its
number
of
vehicles
used
to
deliver
the
final
product
from
production
point
to
storage
point.
Each
trans-
portation
mode
has
a
specific
capacity,
capital
cost,
operating
cost
and
delivery
distance
(see
Table
4).
It
is
noted
that
the
operating
cost
is
associated
with
the
delivery
distance.
2.2.6.
Storage
facility
The
storage
facility,
just
like
the
transportation
mode,
is
linked
to
the
hydrogen
form.
Each
type
has
a
specific
capacity,
capital
94
Chemical
Engineering
Research
and
Design
1
3
4
(
2
0
1
8
)
90–103
Region g
Production technology type p
Distributi
on mode t
Storage fac
ility s
Energy source e
Hydrogen form f
Combiner
Energy/Material flow
Fig.
3
–
Model
superstructure.
Table
2
–
Local
energy
and
hydrogen
demand
for
the
2030
and
2050.
Grid
points,
g
German
region
Population
(MM)
Household
energy
consumption
(GWh
d
−1
)
Hydrogen
demand
(t
d
−1
)
2030
2050
2030
2050
2030
2050
1.
Baden-Wurttemberg
10.80
10.10
34.20
32.74
380.81
1068.39
2.
Bavaria
12.90
12.10
40.85
39.23
454.86
1279.95
3.
Berlin
3.70
3.60
11.72
11.67
130.46
380.81
4.
Brandenburg 2.30
1.90
7.28
6.16
81.10
200.98
5.
Bremen
0.60
0.60
1.90
1.95
21.16
63.47
6.
Hamburg
1.80
1.80
5.70
5.84
63.47
190.41
7.
Hesse
6.00
5.60
19.00
18.16
211.56
592.37
8.
Mecklenburg-Vorpommern
1.40
1.20
4.43
3.89
49.36
126.94
9.
Lower
Saxony
7.50
6.70
23.75
21.72
264.45
708.73
10.
North
Rhine-Westphalia
16.90
15.30
53.52
49.60
595.90
1618.45
11.
Rhineland-Palatinate
3.80
3.40
12.03
11.02
133.99
359.65
12.
Saarland
0.90
0.80
2.85
2.59
31.73
84.62
13.
Saxony
3.80
3.30
12.03
10.70
133.99
349.08
14.
Saxony-Anhalt
1.90
1.60
6.02
5.19
66.99
169.25
15.
Schleswig-Holstein
2.80
2.40
8.87
7.78
98.73
253.87
16.
Thuringia
1.90
1.60
6.02
5.19
66.99
169.25
Total
79.00
72.00
250.18
233.42
2785.56
7616.22
1000
2000
3000
4000
5000
6000
7000
8000
9000
1960
1968
1976
1984
1992
2000
2008
2016
2024
2032
2040
2048
Year
3100
3
150
3
200
3
250
3
300
3
350
3
400
3
450
3
500
3
550
3
600
1990
1995
2000
2005
2010
2015
2020
2025
2030
2035
2040
2045
Energy consumption, kWh
Year
ousehold
Fig.
4
–
Projection
of
energy
consumption.
Table
3
–
Capital
and
unit
production
costs
of
hydrogen
production
technologies
(Simbeck
and
Chang,
2002).
Parameters
Facility
type
Steam
reforming
Coal
gasification
Electrolysis
Biomass
gasification
Capacity
(kg
d
−1
)
960,000
960,000
960,000
960,000
Product
form
LH
CH
LH
CH
LH
CH
LH
CH
Facility
capital
cost
(Mio
$)
1082
775
1668
1123
1910
1663
1895
1518
Unit
production
cost
($
kg
−1
)
1.55
0.97
1.66
0.95
6.40
5.86
2,19
1,58
Chemical
Engineering
Research
and
Design
1
3
4
(
2
0
1
8
)
90–103
95
Table
4
–
Parameters
used
to
estimate
the
capital
and
operating
costs
of
transportation
modes
(Amos,
1999).
Transpiration
mode
Tanker
truck
Tube
trailer
Railway
tank
car
Railway
tube
car
Capacity
(kg
trip
−1
) 4082
181
9072
454
Total
cost
($)
500.000
250,000
500,000
300,000
Fuel
economy
(km
unit
−1a
)
2,550
1.133
22.637
Fuel
price
($
unit
−1a
)
1,22
0,05
a
Unit
for
truck
and
trailer
in
l,
for
railway
car
in
kWh.
Table
5
–
Capital
and
unit
storage
costs
of
hydrogen
storage
facilities
(Almansoori
and
Shah,
2012).
Storage
type
Super-insulated
spherical
tanks
Pressurized
cylindrical
vessel
Product
form
LH
CH
Capacity
(kg)
540,000
540,000
Storage
capital
cost
(Mio
$)
122
1894
Unit
storage
cost
($
kg
−1
d
−1
)
0,005
0,076
and
operating
cost
(see
Table
5).
Storage
facilities
are
installed
at
each
grid
point
to
satisfy
the
local
hydrogen
demand.
Stor-
age
facilities
could
be
located
next
to
production
plant
or
away
from
it.
3.
Model
formulation
This
section
represents
the
model
constraints,
the
compo-
nents
and
objective
function,
resulting
in
a
MILP.
3.1.
Household
energy
demand
As
mentioned
earlier,
the
household’s
energy
demand
by
grid
point
was
estimated
by
projections
of
the
German
popula-
tion
(
Statistisches
Bundesamt,
n.d.)
and
energy
consumption
till
2050.
The
household
energy
demand
can
be
calculated
as
follows:
HHED
g
=
PN
g
AvCon,
∀g
(1)
where
HHED
g
is
the
total
energy
demand
at
grid
point
g,
PN
g
represents
the
population
at
grid
point
g,
AvCon
denotes
aver-
age
of
household
energy
consumption.
The
demand
must
be
covered
by
local
energy
sources
generation
and/or
imports
from
neighboring
grid
points
as
follows:
HHED
g
≤
e
EESAv
e,g
+
g

EESN
e,g

,g
,
∀g
(2)
where
EESAv
e,g
is
amount
of
available
energy
source
e
in
grid
point
g,
which
is
used
to
satisfy
the
energy
demand
in
grid
point
g,
and
EESN
e,g

,g
is
the
flowrate
of
the
supplied
energy
source
e
from
neighboring
grid
point
g

to
grid
point
g.
Prefer-
ably,
the
renewable
energy
source
e
will
be
used
to
satisfy
the
household
energy
demand.
3.2.
Demand
for
a
certain
energy
source
The
demand
for
a
certain
energy
source
must
be
satisfied
to
ensure
production.
The
demand
for
a
certain
energy
source
is
calculated
as
follows:
ESD
g,e
=
f,p
HP
p,g,f
˛
e,p,
∀e,
g
(3)
where
HP
p,g,
f
denotes
the
amount
of
produced
hydrogen
in
the
production
facility
p
in
the
form
f
at
grid
point
g
and
˛
e,p
denotes
the
ratio
between
the
energy
sources
e
consumption
to
produce
1
kg
of
hydrogen
in
production
facility
p.
The
demand
must
be
covered
by
local
power
generation
and/or
imports
from
neighboring
grid
points
as
follows:
ESD
e,g
≤
PESAv
e,g
+
g

PESN
e,g

,g
+PESIm
e,g
, ∀g
(4)
where
PESAv
e,g
is
the
amount
of
available
energy
source
e
at
grid
point
g,
which
is
used
to
satisfy
the
demand
for
energy
source
e
at
grid
point
g,
PESN
e,g

,
g
is
the
flowrate
supplying
energy
source
e
from
neighboring
grid
point
g

to
grid
point
g,
and
PESIm
e,g
is
the
flowrate
importing
energy
source
e
to
grid
point
g.
The
price
for
the
energy
source
consumed
in
year
y
is
cal-
culated
as
follows:
ESC
=
g

,g,e
(PESN
e,g

,g
ESDis
e
Dis
g

,g
+PESAv
e,g
ESCost
e
+PESIm
e,g
ESICost
e
)
(5)
where
ESICost
e
represents
the
energy
source
e
import
price,
ESCost
e
denotes
the
energy
source
e
price,
generated
locally,
ESDis
e
is
the
delivery
price
for
energy
source
e,
and
Dis
g

,
g
is
the
distance
between
grid
points.
3.3.
Hydrogen
demand
The
hydrogen
demand
by
grid
point
can
be
calculated
as
fol-
lows:
HD
g
=
␥PN
g
AvT
AvD
FE,
∀g
(6)
where
represents
the
FCEVs
penetration
rate,
AvT
repre-
sents
the
average
number
of
personal
cars
per
1000
people,
AvD
is
the
average
distance
travelled
by
a
personal
car,
and
FE
denotes
the
fuel
economy.
The
demand
must
be
satisfied
by
local
production
and/or
import
from
neighboring
grid
points
as
follows:
HD
g
≤
f
HP
g,f
+
t,g
HF
g
,g,t,f
,
∀g
(7)
where
HP
g,f
represents
the
hydrogen
generation
in
the
form
f
at
grid
point
g,
and
HF
g
,g
,t ,
f
is
hydrogen
flowrate
in
the
form
f
from
a
neighboring
grid
point
g
to
g
via
transportation
mode
t.
3.4.
Hydrogen
generation
The
hydrogen
production
is
described
as
follows:
HP
g,f
=
p
HP
p,g,f
,
∀g,
f
(8)
where
HP
p,g,
f
,
denotes
the
amount
of
produced
hydrogen
in
the
production
facility
p
in
the
form
f
at
grid
point
g.
The
96
Chemical
Engineering
Research
and
Design
1
3
4
(
2
0
1
8
)
90–103
hydrogen
production
rate
is
constrained
by
a
maximum
and
minimum
capacities
as
follows:
MinPCap
p
NPF
p,f,g
≤
HP
p,g,f
≤
MaxPCap
p
NPF
p,f,g
, ∀p,
g,
f
(9)
where
MaxPCap
p
,
MinPCap
p
is
the
max/min
production
capac-
ity
for
hydrogen
production
facility
p,
NPF
p,
f
,g
represents
number
of
installed
production
technologies
p
at
grid
point
g,
Each
production
plant
has
an
associated
capital-
and
oper-
ating
cost,
the
total
daily
production
cost
is
given
by:
PC
=
PCC+POC
(10)
where
PCC
represents
the
production
capital
cost,
and
POC
is
production
operating
cost.
Each
cost
factor
can
be
calculated
as
follows:
PC
=
p,f,g
PCC
p,f
NPF
p,f,g
AF
p
OP
+HP
p,g,f
POC
p,f
(11)
where
PCC
p,
f
represents
the
capital
cost
of
facility
p,
producing
hydrogen
in
form
f,
NPF
p,
f
,g
denotes
the
number
of
production
facilities
p
generating
hydrogen
in
form
f
at
grid
point
g,
AF
p
is
an
annuity
factor
for
facility
p,
OP
represents
the
operating
period,
and
POC
p,
f
denotes
the
hydrogen
production
cost
in
form
f
at
facility
p.
3.5.
Hydrogen
distribution
The
product
flowrate
by
transportation
mode
t
from
g
to
g
is
given
as
follows:
HP
g,f
≥
t,g
HF
g,g
,t,f
,
∀g
(12)
It
is
noted
that
the
product
can
only
move
in
one
direction
between
grid
points.
The
total
distribution
cost,
calculated
as
the
sum
of
the
operating
and
capital
costs,
is
represented
as:
TC
=
f,t,g,g
TCC
f,t
NTF
g,g
f,t
AF
t
OP
+
NTF
g,g
,f,t
Dis
g,g
,t
FP
t
(13)
where
TCC
f
,t
denotes
the
capital
cost
of
transport
mode
t
for
the
distribution
of
hydrogen
in
form
f,
NTF
g,g
,
f
,t
is
the
number
of
transport
mode
t
used
for
the
hydrogen
distribution
in
the
form
f
from
g
to
g
,
AF
t
is
an
annuity
factor
for
transport
mode
t,
Dis
g,g
,
t
denotes
the
distance
between
grid
points
depending
on
the
type
of
transport,
and
FP
t
is
the
fuel
price
for
trans-
port
mode
t.
The
driver’s
wage
and
maintenance
costs
are
not
included.
3.6.
Hydrogen
storage
The
required
hydrogen
storage
is
constrained
by
a
maximum
and
minimum
capacities
as
follows:
s,f
NSF
s,f,g
MinSCap
s,f
≤
f
HP
g,f
+
t,g
HF
g
,g,t,f
≤
s,f
NSF
s,f,g
MaxSCap
s,f
,
∀g
(14)
where
NSF
s,
f
,g
denotes
the
number
of
storage
facility
s
holding
hydrogen
in
form
f
at
grid
point
g,
and
MaxSCap
s,
f
,
MinSCap
s,
f
represents
maximum
and
minimum
capacity
of
storage
facil-
ity
s
for
holding
hydrogen
in
the
from
f,
␶
is
total
product
storage
period.
The
total
hydrogen
storage
cost
is
calculated
as
follows:
SC
=
f,s,g
SCC
s,f
NSF
s,f,g
AF
s
OP
+
SOC
s,f
f
(HF
g,f
+
t,g
HF
g
,g,t,f
)
(15)
where
SCC
p,
f
denotes
the
capital
cost
for
storage
facility
s
holding
hydrogen
in
the
form
f,
AF
s
is
annuity
factor
for
the
s
storage
facility,
SOC
p,
f
is
the
operating
cost
to
store
1
kg
of
hydrogen
in
the
form
f
at
storage
facility
s.
3.7.
Objective
function
The
total
cost
of
HSC
network
is
given
as
follows:
min
Total
=
PC
+
TC
+
SC
+
ESC (16)
The
right-hand
side
of
Eq.
(16)
contains
four
parts:
the
costs
of
hydrogen
production
(PC),
transport
(TC),
storage
(SC),
and
energy
sources
(ESC).
The
objective
is
to
minimize
the
total
annualized
cost
finding
the
combination
of
network
compo-
nents
to
satisfy
the
local
hydrogen
demand
under
the
given
constraints.
The
model
is
coded
in
AIMMS
and
is
solved
with
CPLEX
12.6.3.
The
model
consists
8922
equations,
3694
contin-
uous
variables,
and
2241
integer
variables.
4.
Case
study
Almansoori
and
Betancourt-Torcat
(2016)
concluded
that
the
development
of
a
HSC
in
Germany
is
economically
feasible
for
the
following
reasons:
there
are
already
20
hydrogen
fueling
stations
across
country
and
the
government
is
reaching
the
decarbonization
target
for
private
transport
and
reduction
of
greenhouse
gases
of
at
least
85%
by
2050.
To
validate
the
model,
a
future
HSC
scenario
analysis
for
Germany
was
performed.
The
data
was
collected
from
the
Fed-
eral
Statistical
Office
of
Germany
(Statistisches
Bundesamt,
n.d.
),
the
Fraunhofer
Institute
(ISE,
n.d.)
for
Solar
Energy
Sys-
tems
ISE,
and
Almansoori
and
Betancourt-Torcat
(2016).
This
work
considers
two
case
studies.
Each
case
represents
a
design
of
an
HSC
network
for
Germany
for
2030
and
2050.
The
first
case
study
considers
a
scenario
to
satisfy
local
hydro-
gen
demand
on
the
HSC
by
using
the
whole
range
of
available
technologies.
The
second
case
considers
a
“green”
scenario,
which
represents
the
ability
to
satisfy
local
personal
needs
(local
household
energy
demand
first
and
hydrogen
based
fuel
demand
after
using
rest
of
energy
sources)
by
using
only
renewable
sources
(see
Appendix
E).
5.
Results
and
discussion
5.1.
Base
case
scenario
The
optimization
results
show
that
3
or
8
large
coal
gasifica-
tion
hydrogen
facilities
are
selected
as
most
economic
option
to
satisfy
hydrogen
demand
in
2030
and
2050
respectively
for
the
first
scenario
(see
Appendix
B).
Coal
is
one
of
the
main
power
sources
in
Germany.
Moreover,
capital
and
operating
costs
for
a
coal
gasification
facility
are
very
low.
In
energy
use
the
coal
gasification
facility
only
costs
0.03
$
kg
−1
which
is
around
5
times
less
then
natural
gas
(0.14
$
kg
−1
)
and
1.7
times
Chemical
Engineering
Research
and
Design
1
3
4
(
2
0
1
8
)
90–103
97
21
1
1
1
1
1
1 1
1
1 1
1
1
1
1
1
1
1
2030
Storage system
Hydrogen production plant
Number of fac
ility
N
Hydrogen flowrate
Fig.
5
–
Hydrogen
supply
chain
network
for
2030
(Base
scenario).
less
than
biomass
or
wind
and
solar
energy
(0.05
$
kg
−1
)
(see
Appendix
A).
Facilities
and
their
interconnections
are
shown
in
Fig.
5.
The
result
is
comparable
with
the
outcomes
of
the
by
work
of
Almansoori
and
Betancourt-Torcat
(2016)
for
2030.
In
both
studies,
coal
gasification
technology
is
selected
as
the
most
economic
option.
One
of
the
production
facility
is
installed
in
Hannover,
another
in
Cologne,
and
the
last
one
in
Munich
for
both
studies.
The
production
facilities
locations
promotes
the
product
distribution
to
regional
storage
facil-
ities.
Additionally,
each
production
facility
includes
nearby
storage
facilities
to
satisfy
the
local
hydrogen
demand.
Furthermore,
hydrogen
is
generated
in
liquid
form.
Germany
has
a
well-developed
railway
infrastructure,
i.e.
the
railway
tank
car
is
selected
as
preferred
transportation
mode
in
both
studies.
It
is
noted
that
a
large
part
of
the
German
rail
freight
transport
is
electrified,
which
means
that
the
rail
transport
is
a
clean
type
of
distribution.
As
hydrogen
is
gener-
ated
in
liquid
form,
super-insulated
spherical
tanks
are
used
to
minimize
heat
loss.
The
total
cost
of
HSC
is
approximately
7.8
and
19.3
Mio
$
d
−1
for
2030
and
2050
respectively,
which
means
2.63
and
2.51
$
kg
−1
of
H
2
(the
hydrogen
price
is
4.5%
less
in
2050
than
in
2030).
In
case
of
hydrogen
price
decreas-
ing
up
to
2050,
it
might
motivate
to
replace
gasoline
cars
by
FCEVs
in
future.
However,
the
hydrogen
price
(2.63
$
kg
−1
)
is
lower
than
from
Almansoori’s
work
(3.03
$
kg
−1
)
as
the
cur-
rent
work
is
time-invariant:
load/unload
time,
driver
wage
expenses
of
transportation
modes
were
not
included,
and
it
is
close
to
the
average
unit
cost
expected
in
Europe
in
2030
(around
3.2
$
kg
−1
).
In
2050,
one
extra
facility
is
installed
in
Cologne.
Additionally,
plants
are
installed
in
Stuttgart,
Berlin,
Frankfurt
and
Dresden
(see
Fig.
6).
5.2.
“Green”
scenario
Despite
the
costs
for
water
electrolysis
technology,
the
“green”
scenario
considers
the
opportunity
to
satisfy
the
local
hydro-
gen
demand
and
the
household
energy
demand
by
wind-
and
solar
energy
(see
Appendix
C).
It
was
founded
that
3
or
8
large
hydrogen
facilities
are
required.
The
“green”
scenario
shows
that
hydrogen
facilities
need
to
be
built
in
Potsdam
and
Hannover
by
2030,
and
in
2050
they
need
to
be
installed
in
Hannover,
Potsdam,
Munich,
Mainz,
Rostock,
Halle,
and
Kiel.
The
total
costs
of
the
HSC
are
approximately
27.9
and
73.5
Mio
$
d
−1
for
2030
and
2050
respectively,
which
means
9.69
and
9.57
$
kg
−1
of
H
2
.
The
hydrogen
price
decreased
by
4%
from
2030
to
2050.
However,
it
is
not
a
reasonable
price
for
industry.
In
addition,
expenses
related
with
household
energy
consumption
account
for
25.3
and
23.4
Mio
$
d
−1
for
2030
and
2050
respectively,
which
shows
electricity
price
reduction
for
this
sector.
98
Chemical
Engineering
Research
and
Design
1
3
4
(
2
0
1
8
)
90–103
32
1
1
2
1
1
1 2
2
1 3
1
1
1
1
1
2050
Storage system
Hydrogen production plant
Number of f
acility
N
Hydrogen flowrate
1
1
1
1
1
1
Fig.
6
–
Hydrogen
supply
chain
network
for
2050
(Base
scenario).
Table
6
–
HSC
cost
for
two
case
studies
in
2030.
HSC
costs
(k$
d
−1
)
Base
case
“Green”
scenario
Operating
costs
Production
and
storage
facilities
4795.0
18,442.5
Production
feedstock
fuel
471.0
6912.0
Transportation
2.2
5.3
Capital
costs
Production
and
storage
facilities
2277.0
2512.0
Transport
37.2
61.3
Total
network
cost
7582.4
27,933.1
Hydrogen
unit
cost
($
kg
−1
)
2.6
9.7
5.3.
Case
comparison
As
shown
in
Table
6,
the
cost
analysis
has
been
done
for
the
two
cases
in
2030.
For
the
hydrogen
pathways,
water
electrolysis
consumes
more
electricity
than
other
technolo-
gies
(see
Appendix
A).
This
energy
consumption
is
about
50
kWh
kg
−1
H
2
,
with
a
specific
power
cost
to
supply
the
elec-
trolysers
of
0.05
$
kWh
−1
.
The
capital-
and
operating
costs
for
a
hydrogen
production
facility
are
only
related
to
the
required
power
to
achieve
the
targeted
hydrogen
production
rate.
The
actual
hydrogen
production
facility’s
specific
energy
consumption
is
the
sum
of
the
specific
power
demand
for
hydrogen
production
and
the
general
power
demand
for
any
electrical
facility
(excluding
compression).
Due
to
the
high
energy
consumption
an
electricity
price
reduction
can
make
water
electrolysis
technology
feasible
(see
Appendix
E).
The
electricity
price
might
be
decreased
by
several
strategies
such
as
integration
of
more
renewable
electricity
generation
facilities
such
as
wind
mills,
solar
panels,
wave
pumps,
or
using
off-peak
electricity.
However,
the
installa-
tion
of
new
energy
facilities
is
only
realistic
in
places
with
suitable
geographical
conditions
(location
with
high
avail-
ability
of
renewable
energy
sources)
providing
continuous
energy
generation.
Moreover,
the
modification
of
existing
con-
ventional
methods
or
development
of
innovative
methods
is
necessary
for
reduction
of
conversion
losses
and
capital
costs
investment.
There
are
a
number
of
problems
related
with
conventional
electrolysers
such
as
safety
risks
due
to
leaks,
stack
degradation,
membrane
deterioration,
difficulties
with
starting
the
system
after
shutdown,
and
freezing
of
mem-
branes,
especially
during
cold
weather.
All
these
problems
require
technological
improvements.
In
addition,
analysis
of
the
impact
of
intermittency
of
renewable
energy
sources
on
the
electrolysis
system
performances
and
reliability,
analysis
of
possible
cost
parameters
changing
and
demand
shifting
are
required
to
map
the
uncertainty.
Thus,
there
is
a
necessity
of
further
R&D
efforts
on
system
design,
power
electronics
Chemical
Engineering
Research
and
Design
1
3
4
(
2
0
1
8
)
90–103
99
and
process
control
to
make
hydrogen
production
by
water
electrolysis
competitive.
6.
Conclusion
In
this
work,
general
optimization
model
for
a
HSC
network
is
proposed.
The
results
shows
that
H
2
has
the
potential
to
replace
fossil
fuel
as
a
source
of
energy,
especially
for
transportation.
The
model
was
applied
to
design
strategies
of
developing
the
future
HSC
network
in
Germany,
considering
a
full
range
of
local
factors
and
geographical
conditions.
More-
over,
the
use
of
the
water
electrolysis
technology
is
suitable
option
to
satisfy
the
local
hydrogen
demand
and
the
house-
hold
energy
demand
by
existing
installed
wind-
and
solar
power
plants
without
environmental
impact.
Due
to
the
high
energy
consumption
an
electricity
price
reduction
can
make
water
electrolysis
technology
competitive.
However,
currently
coal
gasification
technology
is
still
the
dominant
technology,
in
terms
of
economy
as
well
as
implementation.
Hydrogen
transport
in
liquid
form
is
preferable,
distributed
via
rail-
way
tank
car
and
stored
into
super-insulated
spherical
tanks.
FCEVs
penetration
rate
of
10%
in
2030
and
30%
in
2050
into
pas-
senger
transport
shows
a
4%
reduction
of
the
hydrogen
price,
it
can
motivate
to
replace
gasoline
cars
by
FCEVs
in
future
to
diminish
environmental
impact.
Based
on
the
obtained
results,
the
building
of
HSC
network
would
be
expensive
at
the
current
stage
of
technological
devel-
opment.
However,
the
potential
impact
of
renewable
energy
sources
on
the
development
of
the
cost
effective
and
sustain-
able
HSC
network
takes
place
in
future.
Moreover,
that
can
be
complemented
by
CO
2
network,
captured
in
industry,
to
produce
chemicals
like
methanol
or
heavy
hydrocarbons
can
further
motivate
a
hydrogen
based
economy.
Appendix
A.
Capital
and
unit
production
costs
of
hydrogen
production
technologies
(Simbeck
and
Chang,
2002)
Production
technology
Steam
reforming
Coal
gasification
Water
electrolysis
Biomass
gasification
Product
form
LH
LH
LH
LH
Design
production
capacity
t
d
−1
960.00
960.00
960.00
960.00
Plant
availability
d
329.00
329.00
329.00
329.00
Annual
production
10
3
t
315.84
315.84
315.84
315.84
Fuel
required
per
H
2
generated
unit
kg
−1
H
2
3.16
5.33
47.60
11.26
Fuel
consumed
unit
d
−1
3033.60
5116.80
45,696.00
10,809.60
Fuel
price
$
unit
−1
0.14
0.03
0.05
0.05
CO
2
produced
kg
kg
−1
H
2
17.40
30.30
0.00
32.10
SMR/gasifier/electrolyzer
$
unit
−1
317.25
239.85
1110.67
506.70
CO
2
cost
$
kg
−1
0.06
0.06
0.06
0.06
Energy
cost
$
kW
−1
560.00
CO
shift,
cool
and
cleanup
$
kg
−1
d
−1
CO
2
20.00
15.00
Air
separation
unit
$
kg
−1
d
−1
O
2
28.00
27.00
O
2
consumed
per
H
2
generated
1.08
1.41
Dispenser
rate
kg
h
−1
4000.00
4000.00
4000.00
4000.00
Number
of
dispenser
10.00
10.00
10.00
10.00
H
2
dispenser
unit
cost
$
100,000.00
100,000.00
100,000.00
100,000.00
Power
consumption
kWh
kg
−1
H
2
11.00
11.00
58.60
11.00
Electricity
cost
$
kWh
−1
0.05
0.05
0.05
0.05
Unit
cost
$
kg
−1
d
−1
H
2
318.22
877.74
1110.67
1027.94
Size
factor
process
0.75
0.75
0.75
0.80
Unit
liq/gas
cost
$
kg
−1
d
−1
H
2
700.00
700.00
700.00
700.00
Size
factor
liq/gas
0.75
0.75
0.75
0.75
Unit
storage
cost
$
kg
−1
H
2
19.00
19.00
19.00
19.00
Size
factor
of
storage
0.70
0.70
0.70
0.70
Total
process
unit
cost
(UC)
M$
746.89
1149.74
1317.45
1307.23
General
facilities
cost
20%
of
UC
M$
149.38
229.95
263.49
261.44
Engineering
Permitting
10%
of
UC
M$
74.69
114.97
131.74
130.72
Contingencies
10%
of
UC
M$
74.69
114.97
131.74
130.72
Working
Capital,
Land
5%
of
UC
M$
37.34
57.49
65.87
65.36
Total
capital
cost
(CC)
M$
1082.99
1667.13
1910.30
1895.48
O&M
1%
of
CC
M$
y
−1
10.83
16.67
19.10
18.95
100
Chemical
Engineering
Research
and
Design
1
3
4
(
2
0
1
8
)
90–103
Fuel
price
M$
y
−1
139.73
50.50
751.70
177.82
Electricity
cost
M$
y
−1
173.71
173.71
925.41
173.71
Fixed
operating
cost
5%
M$
y
−1
54.15
83.36
95.51
94.78
Capital
charges
12%
of
capital
M$
y
−1
129.96
200.06
22.92
227.46
Total
Operating
Cost
M$
y
−1
508.38
524.30
2020.96
692.71
Unit
Production
cost
$
kg
−1
1.61
1.66
6.40
2.19
Appendix
B.
Results
of
the
hydrogen
supply
by
grid
point
(Base
scenario)
Grid
point,
g Year
2030 Year
2050
H
2
produced
by
grid
point
(t
d
−1
)
H
2
imports
by
grid
point
(t
d
−1
)
Local
satisfaction
of
H
2
demand
(t
d
−1
)
H
2
produced
by
grid
point
(t
d
−1
)
H
2
imports
by
grid
point
(t
d
−1
)
Local
satisfaction
of
H
2
demand
(t
d
−1
)
1
381.02
959.52
108.86
959.52
2
917.53
454.86
953.36
326.59
953.36
3
136.08
952.35
380.81
4
81.658
208.66
5
27.22
63.50
6
63.50
190.51
7
217.73
955.25
592.37
8
54.43
127.01
9
953.92
264.45
953.68
708.73
10
958.78
595.90
1908.75
1618.45
11
136.08
362.88
12
36.29
90.72
13
136.08
956.90
349.08
14
72.586
172.37
15
99.79
254.02
16
72.58
172.37
Total
2830.23
1515.02
1315.21
7639.81
2077.49
5562.32
Appendix
C.
Results
of
the
hydrogen
distribution
by
grid
point
(“Green”
scenario)
Grid
point,
g
Year
2030
Year
2050
H
2
produced
by
grid
point
(t
d
−1
)
H
2
imports
by
grid
point
(t
d
−1
)
Local
satisfaction
of
H
2
demand
(t
d
−1
)
H
2
produced
by
grid
point
(t
d
−1
)
H
2
imports
by
grid
point
(t
d
−1
)
Local
satisfaction
of
H
2
demand
(t
kg
d
−1
)
1
381.02
1070.50
2
462.67
953.36
326.59
953.36
3
136.08
381.02
4
942.94
81.10
953.96
200.98
5
27.22
63.50
6
63.50
190.51
7
217.73
598.75
8
54.43
952.49
126.94
Chemical
Engineering
Research
and
Design
1
3
4
(
2
0
1
8
)
90–103
101
9
1897.41
264.45
1915.31
708.73
10
598.75
1623.89
11
136.08
958.41
359.66
12
36.29
90.72
13
136.08
353.81
14
72.58
958.51
169.25
15
99.79
952.42
253.87
16
72.58
172.37
Total 2840.35
2494.80
345.55
7644.45
4871.66
2772.79
Appendix
D.
Hydrogen
flowrate
for
different
scenario
Base
scenario
“Green”
scenario
Year
2030
Year
2050
Year
2030
Year
2050
From
grid
To
grid
H
2
flowrate
(t
d
−1
)
From
grid
To
grid
H
2
flowrate
(t
d
−1
)
From
grid
To
grid
H
2
flowrate
(t
d
−1
)
From
grid
To
grid
H
2
flowrate
(t
d
−1
)
2
1
381.02
1
1
959.52
4
2
462.67
2
2
953.36
2
7
27.22
2
2
953.36
4
3
136.08
4
2
18.14
2
13
54.43
3
3
380.81
4
8
54.43
4
3
381.02
9
3
136.08
3
4
208.66
4
13
136.08
4
13
353.81
9
4
81.65
3
8
127.01
4
14
72.58
8
1
254.02
9
5
27.22
3
15
235.87
9
1
381.02
8
7
571.54
9
6
63.50
7
1
108.86
9
5
27.22
9
7
27.22
9
8
54.43
7
2
63.50
9
6
63.50
9
10
1179.36
9
13
81.65
7
7
592.37
9
7
217.73
11
1
508.03
9
14
72.58
7
11
190.51
9
10
598.75
11
12
90.72
9
15
99.80
9
5
36.29
9
11
136.08
14
1
308.45
9
16
72.58
9
6
190.51
9
12
36.29
14
2
308.45
10
7
190.51
9
9
708.73
9
15
99.79
14
16
172.37
10
11
136.08
9
15
18.14
9
16
72.58
15
5
63.50
10
12
36.29
10
5
27.22
15
6
190.51
10
10
1618.45
15
10
444.53
10
11
172.37
10
12
90.72
13
2
263.09
13
13
349.08
13
14
172.37
13
16
172.37
Appendix
E.
HSC
network
cost
depending
of
the
energy
price
HSC
costs
“Green”
scenario
Electricity
price
reduction
10%
50%
100%
Operating
costs
Production
and
storage
facilities
(M$
d
−1
)
16.91
10.80
3.15
102
Chemical
Engineering
Research
and
Design
1
3
4
(
2
0
1
8
)
90–103
Production
feedstock
fuel
(M$
d
−1
)
6.64
3.65
0.01
Transportation
($
d
−1
)
3475.30
2441.66
2213.80
Capital
costs
Production
and
storage
facilities
(M$
d
−1
)
2.51
2.51
2.51
Transport
($
d
−1
)
51,498.89
41,912.51
37,230.80
Total
network
cost
(M$
d
−1
)
26.12
17.01
5.71
Hydrogen
unit
cost
($
d
−1
)
9.38
6.10
2.05
Appendix
F.
Initial
availability
of
energy
sources
Grid
point,
g
Primary
energy
source,
e
Biomass
(t
d
−1
)
Coal
(t
d
−1
)
Natural
gas
(t
d
−1
)
Renewable
energy
source
(GWh
d
−1
)
Base
scenario
“Green”
scenario
2030
2050
2030
2050
1
1.99
0.00
0.00
25.85
43.57
0.00
10.83
2
4.62
0.00
0.00
61.50
104.09
20.64
64.87
3
0.00
0.00
0.00
0.34
0.57
0.00
0.00
4
1.92
95,890.41
0.00
55.73
98.52
37.07
81.26
5
0.00
0.00
0.00
1.70
3.04
0.00
1.09
6
0.00
0.00
0.00
0.59
1.04
0.00
0.00
7
1.13
0.00
0.00
16.43
28.60
0.00
10.45
8
4.39
0.00
0.00
32.51
57.78
28.08
53.89
9
5.34
0.00
0.00
112.46
200.64
88.51
178.92
10
2.19
293,041.10
0.00
46.47
81.43
0.00
31.83
11
0.63
0.00
0.00
29.67
52.37
0.00
41.34
12
0.06
0.00
0.00
3.81
6.63
0.63
4.04
13
2.46
95,890.41
0.00
15.58
27.09
3.55
16.39
14
3.07
26,027.40
0.00
42.32
75.07
36.31
69.88
15
2.91
0.00
0.00
47.50
84.74
33.52
72.16
16
2.26
0.00
0.00
14.38
25.22
8.36
20.04
Total
32.97
510,849.32
0.00
506.85
890.41
256.67
656.99
References
Almansoori,
A.,
Betancourt-Torcat,
A.,
2016.
Design
of
optimization
model
for
a
hydrogen
supply
chain
under
emission
constraints
—
a
case
study
of
Germany.
Energy
111,
414–429,
http://dx.doi.org/10.1016/j.energy.2016.05.123.
Almansoori,
A.,
Shah,
N.,
2012.
Design
and
operation
of
a
stochastic
hydrogen
supply
chain
network
under
demand
uncertainty.
Int.
J.
Hydrogen
Energy
37,
3965–3977,
http://dx.doi.org/10.1016/j.ijhydene.2011.11.091.
Amos,
W.A.,
1999.
Costs
of
Storing
and
Transporting
Hydrogen.
Other
Inf.
PBD
27
Jan
1999;
PBD
27
Jan
1999;
PBD
27
Jan
1999.
Ball,
M.,
Wietschel,
M.,
Rentz,
O.,
2007.
Integration
of
a
hydrogen
economy
into
the
German
energy
system:
an
optimising
modelling
approach.
Int.
J.
Hydrogen
Energy
32,
1355–1368,
http://dx.doi.org/10.1016/j.ijhydene.2006.10.016
.
BM
Verkehr
Bau
und
Stadtentwicklung,
2013.
The
Mobility
and
Fuels
Strategy
of
the
German
Government
(MFS).
CO2.earth,
n.d.
2100
Projections
[WWW
Document].
https://www.co2.earth
(accessed
5.25.17).
Hake,
J.F.,
Linssen,
J.,
Walbeck,
M.,
2006.
Prospects
for
hydrogen
in
the
German
energy
system.
Energy
Policy
34,
1271–1283,
http://dx.doi.org/10.1016/j.enpol.2005.12.015.
Hugo,
A.,
Rutter,
P.,
Pistikopoulos,
S.,
Amorelli,
A.,
Zoia,
G.,
2005.
Hydrogen
infrastructure
strategic
planning
using
multi-objective
optimization.
Int.
J.
Hydrogen
Energy
30,
1523–1534,
http://dx.doi.org/10.1016/j.ijhydene.2005.04.017.
International
Energy
Agency
(IEA),
2016.
World
Energy
Outlook
2016.
http://www.iea.org/publications/freepublications/publication/
WEB
WorldEnergyOutlook2015ExecutiveSummaryEnglishFinal.pdf.
International
Union
of
Railways,
2012.
Railway
Handbook
2012
—
Energy
Consumption
and
CO
2
Emissions.
ISE,
F.I.
for
S.E.S.,
n.d.
Energy
Charts
[WWW
Document].
https://www.energy-charts.de/index.htm
(accessed
5.25.17).
Pregger,
T.,
Nitsch,
J.,
Naegler,
T.,
2013.
Long-term
scenarios
and
strategies
for
the
deployment
of
renewable
energies
in
Chemical
Engineering
Research
and
Design
1
3
4
(
2
0
1
8
)
90–103
103
Germany.
Energy
Policy
59,
350–360,
http://dx.doi.org/10.1016/j.enpol.2013.03.049.
Schill,
W.P.,
2014.
Residual
load,
renewable
surplus
generation
and
storage
requirements
in
Germany.
Energy
Policy
73,
65–79,
http://dx.doi.org/10.1016/j.enpol.2014.05.032.
Shafiee,
S.,
Topal,
E.,
2009.
When
will
fossil
fuel
reserves
be
diminished?
Energy
Policy
37,
181–189,
http://dx.doi.org/10.1016/j.enpol.2008.08.016.
Siemens,
L.,
n.d.
Energiepark
Mainz
[WWW
Document].
http://www.energiepark-mainz.de/en/.
(Accessed
20
March
2017).
Simbeck,
D.R.,
Chang,
E.,
2002.
Hydrogen
Supply:
Cost
Estimate
for
Hydrogen
Pathways–Scoping
Analysis.
National
Renewable
Energy
Laboratory,
doi:NREL/SR-540-32525.
Statistisches
Bundesamt,
n.d.
Koordinierte
Bevölkerungsvorausberechnung
nach
Bundesländern
[WWW
Document].
https://www.destatis.de/EN/Homepage.html.
(Accessed
25
May
2017).
U.S.
Energy
Information
Administration,
2017.
Annual
Energy
Outlook,
doi:DOE/EIA-0383(2017).

Tài liệu khác của Tài liệu khác

Preview text:

Chemical Engineering Research and Design 1 3 4 ( 2 0 1 8 ) 90–103
Contents lists available at ScienceDirect
Chemical Engineering Research and Design
j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / c h e r d
An outlook towards hydrogen supply chain
networks in 2050 — Design of novel fuel
infrastructures in Germany
Anton Ochoa Bique ∗, Edwin Zondervan
Laboratory of Process Systems Engineering, Department of Production Engineering, Universität Bremen, Leobener
Str. 6, 28359 Bremen, Germany a r t i c l e i n f o a b s t r a c t Article history:
This work provides a comprehensive investigation of the feasibility of hydrogen as trans- Received 30 August 2017
portation fuel from a supply chain point of view. It introduces an approach for the
Received in revised form 20 March
identification the best hydrogen infrastructure pathways making decision of primary energy 2018
source, production, storage and distribution networks to aid the target of greenhouse gas Accepted 23 March 2018
emissions reduction in Germany. The minimization of the total hydrogen supply chain (HSC) Available online 3 April 2018
network cost for Germany in 2030 and 2050 years is the objective of this study. The model
presented in this paper is expanded to take into account water electrolysis technology driven Keywords:
by solar and wind energy. Two scenarios are evaluated, including a full range of technologies Hydrogen supply chain design
and “green” technologies using only renewable resources. The resulting model is a mixed Fuel infrastructures
integer linear program (MILP) that is solved with the Advanced Integrated Multidimensional
Mixed integer linear programming
Modeling System (AIMMS). The results show that renewable energy as a power source has Germany
the potential to replace common used fossil fuel in the near future even though currently AIMMS
coal gasification technology is the still the dominant technology.
© 2018 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved. 1. Introduction
and biomass based production of energy and chemicals is strongly
supported by governments (Schill, 2014). For example, the German
Up till now, fossil fuels, which include natural gas, oil, and coal are
government decided to completely phase out nuclear energy by 2022
the primary energy sources for transportation, electricity, and residen-
(Pregger et al., 2013) and replace it with renewable energy production.
tial services. Based on a report by the International Energy Agency
The largest part of renewable power will come from solar and wind as
(IEA) and the U.S. EIA, the global energy demand will grow with 30% in
shown in Fig. 1. Electric power from wind mills increases its contribu-
2040 (International Energy Agency (IEA), 2016; U.S. Energy Information
tion by 225 TWh in 2050, which is 39% of the final produced energy; solar
Administration, 2017). This means a progressively growing fuel con-
contributes 17%, at 100 TWh per year, while biomass reaches 60 TWh
sumption in the near future i.e. greenhouse gas emissions such as per year.
carbon dioxide also increase. Fossil fuel is a nonrenewable energy
While biomass as a raw material might be stored for a long period
source. The depletion time for fossil fuel is estimated to be around
of time, wind and solar are more difficult to handle. As battery sys-
100 years, where oil and gas will be exhausted earlier than coal (Shafiee
tems do currently not have enough capacity and storage of electricity
and Topal, 2009). Moreover, due to increasing fuel consumption, cause
is very expensive, the developments in new long-term storage technol-
of concern is the fast rise of CO
ogy is one of the main challenges. Industrial key players, like Siemens
2 level, now already exceeding 400 ppm
level and, left unmitigated, can possibly double in 100 years to 800 ppm
currently work on a new type of energy storage system based on hydro- (CO2.earth, n.d.).
gen production (Siemens, n.d.). The main idea is that excess energy
Due to the increasing demand of electric energy and a decreas-
from renewable energy sources can be converted into hydrogen from
ing amount of fossil fuel sources, the development of solar-, wind-
water by electrolysis, which is a non-toxic source of energy to con-
∗ Corresponding author.
E-mail address: antochoa@uni-bremen.de (A. Ochoa Bique).
https://doi.org/10.1016/j.cherd.2018.03.037
0263-8762/© 2018 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
Chemical Engineering Research and Design 1 3 4 ( 2 0 1 8 ) 90–103 91 Nomenclature TC
Daily distribution cost [$ d−1] TCCf ,t
Capital cost of transport mode t for distribution Indices
hydrogen in the form f [$] e Type of energy source Total Total cost of HSC network [$] f Type of hydrogen physical form g
Grid points, each grid point represents German Integer variables state
NPFp,f ,g Number of production facility p generating p
Type of hydrogen production facility
hydrogen in from f at grid point g s Type of storage facility NSF
of storage facility s holding hydrogen s,f ,g Number t Type of transportation mode
in the form f at grid point g
NTFg,g,f ,t The number of transport mode t used for Abbreviation
hydrogen distribution in the form f from g to CH Compressed-gaseous hydrogen g FCEV Fuel cell electric vehicle PNg
Population at the grid point g HSC Hydrogen supply chain IEA International Energy Agency Parameters LH Liquid hydrogen AvCon
Average of household energy consumption USEIA
U.S. Energy Information Administration [kWh d1] AvD
The average distance travelled by personal car Continuous variable [km y1]
EESAve,g Amount of available energy source e in the grid AvT
The average amount of personal car per 1000
point g, which is used to satisfy energy demand people
in grid point g [kWh d−1] AFp
Annual factor for the facility p [%]
EESNe,g,g The flowrate of the supplied energy source e AFs
Annual factor for the s storage facility s [%]
from neighboring grid point g to grid point g, AFt
Annual factor for the transport mode t [%]
which is used to satisfy energy demand in grid Disg,g
Distance between grid points [km] point g [kWh d−1]
Disg,g,t Distance between grid points depending of type
ESAve,g Amount of available energy source e in the grid of transport [km] point g [unit e d−1]
ESCoste Energy source e price in year y, generated locally ESC
Total cost for the energy source consumed for [$ unit−1 e] hydrogen production [$ d−1] ESDise Delivery price for energy source e ESDg
Daily energy source demand [kWh d−1] [$ unit−1 km−1] HDg
Hydrogen demand by grid point [kg d−1]
ESICoste Energy source e import price [$ unit−1]
HFg,g,t,f Hydrogen flowrate in the form f from grid point FE
The fuel economy [kg H2 km−1]
g to g via transportation mode t [kg d−1] FPt
Fuel price for transport mode t [$ l−1] HHEDg
Total energy demand in the grid point g
MaxPCapp/MinPCapp Max/min production capacity for [kWh d−1]
hydrogen production facility p [kg d−1] HPg,f
Hydrogen generation in the form f at grid point OP Operating period [d y−1] g [kg d−1] SCaps,f
Capacity of storage facility s for holding hydro- HPp,g,f
Amount of produced hydrogen in the produc- gen in the from f [kg]
tion facility p in the form f at grid point g [kg d−1] PC
Daily production costs [$ d−1] Greek letters PCC
Production capital cost [$ d−1] ˛
ratio between energy sources e consump- e,p The
PESAve,g Amount of available energy source e in the grid
tion to produce 1 kg [unit e kg1 H 2]
point g, which is used to satisfy energy source FCEVs penetration rate [%]
demand for hydrogen production [unit e d−1]
Is total product storage period [d]
PESIme,g,g Flowrate of imported energy source e from
neighboring grid point g to grid point g, which
is used to satisfy energy source demand for
hydrogen production [unit e d−1]
sumers allowing a greater energy security and flexibility. As soon as PESNe
there is energy shortage, hydrogen might be used in different appli-
,g,g Flowrate supplying energy source e from
neighboring grid point g to grid point g
cations such as power generation, domestic and industrial services, [unit e d−1]
navigation and space (Hake et al., 2006). However, hydrogen is not a
naturally occurring fuel of mineral origin; it can be produced from POC
Production operating cost [$ d−1]
both renewable and non-renewable resources: from coal and biomass PCC p,f
Capital cost of facility p, producing hydrogen in
gasification, the reforming of natural gas, from water electrolysis, form f [$]
photo-electrolysis, water-splitting thermochemical cycle, photobio- POCp,f
Hydrogen production operating cost in form f
logical production, and high temperature decomposition. Moreover,
at facility p [$ kg−1]
hydrogen generation is only a part of the hydrogen production net- SC
The total hydrogen storage cost [$]
work, which can be defined as a supply chain consisting of several SCCp,f
Capital cost for storage facility s holding hydro-
components (such as production, storage and distribution). For each gen in the form f [$]
of these stages a wide range of potential technological options exist. SOCp
Due to increasing demand for energy, the development of sustainable ,f
Operating cost to store 1 kg of hydrogen in the
from f inside of storage facility s [$ kg−1 d−1]
and environmental friendly concepts such as the HSC should be devel-
oped to replace non-sustainable alternatives to meet the global need for 92
Chemical Engineering Research and Design 1 3 4 ( 2 0 1 8 ) 90–103 250000 200000 150000 100000 50000 0 1990 1993 1996 1999 2002 2005 2008 2011 2014 2017 2020 2023 2026 2029 2032 2035 2038 2041 2044 2047 Year Wind electricity production Solar electricity prodution Biomass
Projection for Wind electricity production
Fig. 1 – Projection of energy generation.
energy (Ball et al., 2007). The work of Hugo et al. (2005) takes all possible
develop and describe the model, setup a case study and interpret and
hydrogen alternatives for the design of an optimal hydrogen infras- discuss the results obtained.
tructure in Germany in to consideration. However, their model does
not include energy sources distribution and the ability of centralized 2.
Network description and problem
hydrogen storage to satisfy the local demand. A study of Almansoori statement
investigated a number of strategic decisions for hydrogen fuel pro-
duction and hydrogen delivery networks in Germany at large-scale 2.1.
Problem statement
considering emission targets and carbon tax as a part of the model for-
mulation for 2030 (Almansoori and Betancourt-Torcat, 2016). The main
Given are the location and capacity of energy source suppliers,
objective in that study was to satisfy the hydrogen demand which was
determined by a fuel cell electric vehicles (FCEVs) penetration of 10%
capital and operating costs for a large-scale hydrogen pro-
of the overal passenger transport. The results showed that liquefied
duction, transportation and storage facilities of the network,
hydrogen production by coal gasification facilities at large-scale and under the conditions that:
delivery via railway tank cars results in the best HSC network struc-
ture. Large-size facilities showed benefit compared to a small-scale
1. location of storage facilities is fixed;
facility since large facilities have a high energy efficiency. Renewable
2. all natural gas is imported (despite a national 12% produc-
energy such as wind and solar were not included in that study due to tion of natural gas);
technical- and economical hurdles such as expenses of electricity price
3. weighted average cost of capital for production, transporta-
for water electrolysis technology and size-independent electrolyzer tion and storage is 10%;
efficiency. The rate of renewable energy consumption to generate a unit
of hydrogen for both sizes of electrolysis facility is identical as the elec-
4. electricity is the main energy source to power rail freight
trolyzer efficiency is independent of the facility size. A similar model
transport (International Union of Railways, 2012);
was developed for the United Kingdom (Almansoori and Shah, 2012).
5. the way of handling of residual waste is not considered;
The objective was the minimization of the cost of the network consider-
6. secondary energy carriers have no economic value in this
ing capital- and operating costs. The results showed the dominance of network model;
steam methane reforming technology. Large-scale electrolysis facilities
7. electricity price based on industrial electricity price for
were not considered due to a size-independent electrolyzer efficiency Germany. that was mentioned before.
The aim of this paper is to develop and evaluate an optimization
The HSC consists of energy sources from different origins,
model that can be used to solve a HSC network design problem forecast-
ing in 2030 and 2050 years while considering a full range of local factors
large-scale hydrogen production technologies, hydrogen prod-
such as (i) energy sources distribution for hydrogen production, (ii) local
uct form and the hydrogen distribution and storage options
hydrogen demand and (iii) distribution between the place of hydrogen
(Fig. 2). Five types of energy sources are considered: wind-
production and hydrogen demand. The model is used to define the
and solar energy, biomass, natural gas and coal. In addition,
procurement of energy sources from the supplier, the type, the num-
four types of hydrogen production technologies are included
ber and the location of a production facility, the hydrogen production
into the model: steam methane reforming, coal gasifica-
form and the delivery of hydrogen to consumers. The logistics of renew-
tion, biomass gasification and water electrolysis. As hydrogen
able sources is also included into the model by accounting for personal
might be generated by different production technologies, it
needs such as household energy and hydrogen based fuel consump-
may be transported into two forms (i.e. liquid or gaseous),
tion. In addition, this work also considers environmental influences.
which determines the transportation mode that will be used.
Moreover, all techno-economic parameters were collected for 2015 and
were assumed stay same as for the reference years. The German land-
The liquid form (LH) could be stored in super-insulated spher-
scape provides an important case study as Germany has an immense
ical tanks and be distributed via two types of transportation
potential to develop a sustainable hydrogen infrastructure (Hake et al.,
modes: by railway tank car or via tanker truck. Gaseous hydro-
2006). In the following sections we will define the problem statement,
gen (CH) could be stored into pressurized cylindrical vessels
Chemical Engineering Research and Design 1 3 4 ( 2 0 1 8 ) 90–103 93 PRODUCTION DISTRIBUTION STORAGE ENERGY OPTION SOURCE TYPE OF PLANT PRODUCT PRODUCT FORM DISTRIBUTION MODE Wind Tube trailer Electrolysis Pressurized Solar Compression Railway tube Cylindrical car vessels Biomass H2 Gasification Railway tank Coal car Liquefaction Super-insulated Spherical tanks Tanker truck Steam Natural gas reforming
Fig. 2 – Structure of the hydrogen supply and delivery chain.
in Fig. 4), the household energy demand was estimated over
Table 1 – Parameters used for total hydrogen demand calculation in Germany. time. Parameter Passenger transport system in Germany 2015 2.2.2. Primary energy sources
Hydrogen can be produced from different sources such as Number 45,046,564
water, natural gas, biomass and coal. This resource availability
Average distance travelled (km y−1) 13,000
at each grid point plays an important role in defining the type Fuel economy (kg H 2 km−1) 0.018
and location of production technologies (see Appendix F). In
addition, the main problem of a domestic production facility is
and be distributed via railway tube car or tube trailer. Each
concerned with finding an appropriate energy source supplier.
facility of the HSC includes: a technological option, a capacity,
There are three opportunities related with the energy source
a location. The problem is concerned with finding the number
consumption from (i) a domestic grid point or (ii) supply from
and locations of the production facilities for a given demand,
neighboring grid points or (iii) import from abroad.
while minimizing the total operating HSC network cost. 2.2.3. Hydrogen production 2.2.
Model description
Considering that hydrogen is not a naturally occurring fuel
of mineral origin, different production technologies, including
In Fig. 3 the superstructure of HSC model is shown (Hugo
steam methane reforming, coal gasification, biomass gasifica-
et al., 2005). The superstructure includes all the possible
tion and water electrolysis, are generally used to generate it.
connections between the model components. Ultimately, an
Each alternative has fixed capital and operational costs (see
optimization algorithm is used to search for the best strategy
Appendix A). The main decision is to determine the type,
to minimize the costs of the HSC network. The superstructure
location and number of production facilities. Each facility car-
consists a set of grid points (g, each grid point represents a
ries out large-scale hydrogen production (960 t per day) (see
German state), energy sources (e), different transportation (t) Table 3).
modes, different hydrogen production-(p) and storage (s) facil-
ities. The transportation modes are used to distribute different 2.2.4. Hydrogen physical form
types of hydrogen (f) from production facility to storage facil-
Hydrogen can be carried in two physical forms: liquid and
ity. In the following subsections, each component of the HSC
gaseous. Each form is distributed by different transportation
model will be described in more detail.
modes and might be stored in special storage facilities. The
hydrogen form plays an important role in defining which 2.2.1. Grid
transportation mode and storage facilities should be used.
In this study, the landscape of Germany is divided into 16 grid
These decisions affect the final costs of the HSC network.
points, each of this grid points represents a German region.
Moreover, the region’s largest city is taken as the potential cen- 2.2.5. Transportation mode
ter for a hydrogen production facility and for a storage facility
The transportation mode is related to the hydrogen form (gas
to satisfy the local demand and further product distribution to
or liquid). The main decision is to define the type transporta-
another region (Almansoori and Betancourt-Torcat, 2016). The
tion mode and its number of vehicles used to deliver the final
total hydrogen demand was estimated based on FCEVs pene-
product from production point to storage point. Each trans-
tration rate of 10% in the total number of passenger transports
portation mode has a specific capacity, capital cost, operating
(public buses, light motor vehicle) available by 2030, and 30%
cost and delivery distance (see Table 4). It is noted that the
by 2050, average distance travelled and transport fuel economy
operating cost is associated with the delivery distance.
(see Table 2) (BM Verkehr Bau und Stadtentwicklung, 2013).
2015 was used as the reference year for the calculations. All 2.2.6. Storage facility
relevant parameters are listed in Table 1. Based on the pro-
The storage facility, just like the transportation mode, is linked
jections of energy consumption from 1960 to 2050 (as shown
to the hydrogen form. Each type has a specific capacity, capital 94
Chemical Engineering Research and Design 1 3 4 ( 2 0 1 8 ) 90–103 Region g Energy source e Production technology type p Hydrogen form f Distribution mode t Combiner Storage facility s Energy/Material flow
Fig. 3 – Model superstructure.
Table 2 – Local energy and hydrogen demand for the 2030 and 2050. Grid points, g German region Population (MM)
Household energy consumption (GWh d−1) Hydrogen demand (t d−1) 2030 2050 2030 2050 2030 2050 1. Baden-Wurttemberg 10.80 10.10 34.20 32.74 380.81 1068.39 2. Bavaria 12.90 12.10 40.85 39.23 454.86 1279.95 3. Berlin 3.70 3.60 11.72 11.67 130.46 380.81 4. Brandenburg 2.30 1.90 7.28 6.16 81.10 200.98 5. Bremen 0.60 0.60 1.90 1.95 21.16 63.47 6. Hamburg 1.80 1.80 5.70 5.84 63.47 190.41 7. Hesse 6.00 5.60 19.00 18.16 211.56 592.37 8. Mecklenburg-Vorpommern 1.40 1.20 4.43 3.89 49.36 126.94 9. Lower Saxony 7.50 6.70 23.75 21.72 264.45 708.73 10. North Rhine-Westphalia 16.90 15.30 53.52 49.60 595.90 1618.45 11. Rhineland-Palatinate 3.80 3.40 12.03 11.02 133.99 359.65 12. Saarland 0.90 0.80 2.85 2.59 31.73 84.62 13. Saxony 3.80 3.30 12.03 10.70 133.99 349.08 14. Saxony-Anhalt 1.90 1.60 6.02 5.19 66.99 169.25 15. Schleswig-Holstein 2.80 2.40 8.87 7.78 98.73 253.87 16. Thuringia 1.90 1.60 6.02 5.19 66.99 169.25 Total 79.00 72.00 250.18 233.42 2785.56 7616.22 ousehold 9000 3600 h 8000 3550 W 7000 3500 , k 3450 on 6000 3400 pti 5000 3350 4000 3300 nsum 3250 3000 3200 gy co 2000 3150 ner 1000 E 3100 0 8 6 4 2 0 8 6 4 2 0 8 0 5 0 5 0 5 0 5 0 5 0 5 196 196 197 198 199 200 200 201 202 203 204 204 199 199 200 200 201 201 202 202 203 203 204 204 Year Year
Fig. 4 – Projection of energy consumption.
Table 3 – Capital and unit production costs of hydrogen production technologies (Simbeck and Chang, 2002). Parameters Facility type Steam reforming Coal gasification Electrolysis Biomass gasification Capacity (kg d−1) 960,000 960,000 960,000 960,000 Product form LH CH LH CH LH CH LH CH Facility capital cost (Mio $) 1082 775 1668 1123 1910 1663 1895 1518
Unit production cost ($ kg−1) 1.55 0.97 1.66 0.95 6.40 5.86 2,19 1,58
Chemical Engineering Research and Design 1 3 4 ( 2 0 1 8 ) 90–103 95
Table 4 – Parameters used to estimate the capital and operating costs of transportation modes (Amos, 1999). Transpiration mode Tanker truck Tube trailer Railway tank car Railway tube car Capacity (kg trip−1) 4082 181 9072 454 Total cost ($) 500.000 250,000 500,000 300,000 Fuel economy (km unit−1a) 2,550 1.133 22.637 Fuel price ($ unit−1a) 1,22 0,05
a Unit for truck and trailer in l, for railway car in kWh.
denotes the ratio between the energy sources e consumption to
Table 5 – Capital and unit storage costs of hydrogen
storage facilities (Almansoori and Shah, 2012).
produce 1 kg of hydrogen in production facility p. The demand
must be covered by local power generation and/or imports Storage type Super-insulated Pressurized
from neighboring grid points as follows: spherical tanks cylindrical vessel Product form LH CH ESD PESAv +PESIm e,g ≤ e,g+ PESN e,g, ∀g (4) g e,g ,g Capacity (kg) 540,000 540,000 Storage capital cost (Mio $) 122 1894 where PESAve
the amount of available energy source e at
Unit storage cost ($ kg−1 d−1) 0,005 0,076 ,g is
grid point g, which is used to satisfy the demand for energy
source e at grid point g, PESNe,g the flowrate supplying
and operating cost (see Table 5). Storage facilities are installed ,g is
energy source e from neighboring grid point g to grid point g,
at each grid point to satisfy the local hydrogen demand. Stor-
and PESIme,g is the flowrate importing energy source e to grid
age facilities could be located next to production plant or away point g. from it.
The price for the energy source consumed in year y is cal- culated as follows: 3. Model formulation ESC = (PESN ESDise Dis +PESAve,g ESCoste
This section represents the model constraints, the compo- e,g ,g g ,g g ,g,e
nents and objective function, resulting in a MILP. +PESIme,g ESICoste) (5) 3.1.
Household energy demand
where ESICoste represents the energy source e import price,
ESCoste denotes the energy source e price, generated locally,
As mentioned earlier, the household’s energy demand by grid
ESDise is the delivery price for energy source e, and Disg ,g is
point was estimated by projections of the German popula-
the distance between grid points.
tion (Statistisches Bundesamt, n.d.) and energy consumption
till 2050. The household energy demand can be calculated as 3.3. Hydrogen demand follows:
The hydrogen demand by grid point can be calculated as fol- HHEDg = PNgAvCon, ∀g (1) lows:
where HHEDg is the total energy demand at grid point g, PNg HD ␥PN AvD FE, ∀g (6)
represents the population at grid point g, AvCon denotes aver- g = g AvT
age of household energy consumption. The demand must be
where represents the FCEVs penetration rate, AvT repre-
covered by local energy sources generation and/or imports
sents the average number of personal cars per 1000 people,
from neighboring grid points as follows:
AvD is the average distance travelled by a personal car, and FE
denotes the fuel economy. The demand must be satisfied by HHED g ≤ e EESAve,g+ EESN , ∀g (2) g e,g ,g
local production and/or import from neighboring grid points as follows: where EESAv
e,g is amount of available energy source e in grid
point g, which is used to satisfy the energy demand in grid HD HP , ∀g (7) point g, and EESN
the flowrate of the supplied energy g ≤ f g,f + t,g HFg,g,t,f e,g,g is
source e from neighboring grid point g to grid point g. Prefer-
ably, the renewable energy source e will be used to satisfy the
where HPg,f represents the hydrogen generation in the form f household energy demand.
at grid point g, and HFg,g,t,f is hydrogen flowrate in the form f
from a neighboring grid point g to g via transportation mode 3.2.
Demand for a certain energy source t.
The demand for a certain energy source must be satisfied to 3.4.
Hydrogen generation
ensure production. The demand for a certain energy source is calculated as follows:
The hydrogen production is described as follows: ESD ∀ g,e = g (3) HPg,f = HPp,g,f, g, f (8) f,p HPp,g,f ˛e,p, ∀e, p
where HPp,g,f denotes the amount of produced hydrogen in
where HPp,g,f , denotes the amount of produced hydrogen in
the production facility p in the form f at grid point g and ˛e,p
the production facility p in the form f at grid point g. The 96
Chemical Engineering Research and Design 1 3 4 ( 2 0 1 8 ) 90–103
hydrogen production rate is constrained by a maximum and
ity s for holding hydrogen in the from f, ␶ is total product minimum capacities as follows: storage period.
The total hydrogen storage cost is calculated as follows: MinPCapp NPFp,f,g ≤ HP MaxPCap g, f (9) p,g,f ≤ p NPFp,f,g, ∀p, SCCs,f NSFs,f,g AFs SC = + SOC (HF s,f g,f+ where MaxPCap t,g HFg,g,t,f)
p, MinPCapp is the max/min production capac- f,s,g OP f
ity for hydrogen production facility p, NPF (15) p,f ,g represents
number of installed production technologies p at grid point
g, Each production plant has an associated capital- and oper- where SCCp
the capital cost for storage facility s
ating cost, the total daily production cost is given by: ,f denotes
holding hydrogen in the form f, AFs is annuity factor for the
s storage facility, SOCp,f is the operating cost to store 1 kg of PC = PCC+POC (10)
hydrogen in the form f at storage facility s.
where PCC represents the production capital cost, and POC is 3.7.
Objective function
production operating cost. Each cost factor can be calculated as follows:
The total cost of HSC network is given as follows: PCCp,f NPFp,f,g AFp PC = +HP (11) min Total = PC + TC + SC + ESC (16) p,g,f POCp,f p,f,g OP
The right-hand side of Eq. (16) contains four parts: the costs where PCC
p,f represents the capital cost of facility p, producing
of hydrogen production (PC), transport (TC), storage (SC), and
hydrogen in form f, NPF
p,f ,g denotes the number of production
energy sources (ESC). The objective is to minimize the total
facilities p generating hydrogen in form f at grid point g, AF p
annualized cost finding the combination of network compo-
is an annuity factor for facility p, OP represents the operating
nents to satisfy the local hydrogen demand under the given period, and POC
p,f denotes the hydrogen production cost in
constraints. The model is coded in AIMMS and is solved with
form f at facility p.
CPLEX 12.6.3. The model consists 8922 equations, 3694 contin-
uous variables, and 2241 integer variables. 3.5.
Hydrogen distribution 4. Case study
The product flowrate by transportation mode t from g to g is given as follows:
Almansoori and Betancourt-Torcat (2016) concluded that the
development of a HSC in Germany is economically feasible for HP g,f ≥ ∀g (12) t,g HFg,g,t,f,
the following reasons: there are already 20 hydrogen fueling
stations across country and the government is reaching the
It is noted that the product can only move in one direction
decarbonization target for private transport and reduction of
between grid points. The total distribution cost, calculated as
greenhouse gases of at least 85% by 2050.
the sum of the operating and capital costs, is represented as:
To validate the model, a future HSC scenario analysis for
Germany was performed. The data was collected from the Fed- TCC f,t NTFg,gf,t AFt TC = + NTF
eral Statistical Office of Germany (Statistisches Bundesamt, f,t,g,g OP g,g,f,t Disg,g,t FPt
n.d.), the Fraunhofer Institute (ISE, n.d.) for Solar Energy Sys- (13)
tems ISE, and Almansoori and Betancourt-Torcat (2016).
This work considers two case studies. Each case represents
where TCCf ,t denotes the capital cost of transport mode t for
a design of an HSC network for Germany for 2030 and 2050.
the distribution of hydrogen in form f, NTFg,g the number ,f ,t is
The first case study considers a scenario to satisfy local hydro-
of transport mode t used for the hydrogen distribution in the
gen demand on the HSC by using the whole range of available
form f from g to g, AFt is an annuity factor for transport mode
technologies. The second case considers a “green” scenario, t, Disg,g
the distance between grid points depending ,t denotes
which represents the ability to satisfy local personal needs
on the type of transport, and FPt is the fuel price for trans-
(local household energy demand first and hydrogen based
port mode t. The driver’s wage and maintenance costs are not
fuel demand after using rest of energy sources) by using only included.
renewable sources (see Appendix E). 3.6. Hydrogen storage 5. Results and discussion
The required hydrogen storage is constrained by a maximum 5.1.
Base case scenario
and minimum capacities as follows:
The optimization results show that 3 or 8 large coal gasifica- NSF ≤ HP s,f s,f,g MinSCaps,f f g,f + t,g HFg,g,t,f
tion hydrogen facilities are selected as most economic option (14)
to satisfy hydrogen demand in 2030 and 2050 respectively for ≤ NSF , ∀g
the first scenario (see Appendix B). Coal is one of the main s,f s,f,g MaxSCaps,f
power sources in Germany. Moreover, capital and operating
where NSFs,f ,g denotes the number of storage facility s holding
costs for a coal gasification facility are very low. In energy
hydrogen in form f at grid point g, and MaxSCaps,f , MinSCaps,f
use the coal gasification facility only costs 0.03 $ kg−1 which is
represents maximum and minimum capacity of storage facil-
around 5 times less then natural gas (0.14 $ kg−1) and 1.7 times
Chemical Engineering Research and Design 1 3 4 ( 2 0 1 8 ) 90–103 97 2030 Storage system Hydrogen production plant N Number of facility 1 Hydrogen flowrate 1 1 1 1 1 1 1 1 1 2 1 1 1 1 1 1 1 1
Fig. 5 – Hydrogen supply chain network for 2030 (Base scenario).
less than biomass or wind and solar energy (0.05 $ kg−1) (see
rent work is time-invariant: load/unload time, driver wage
Appendix A). Facilities and their interconnections are shown
expenses of transportation modes were not included, and it
in Fig. 5. The result is comparable with the outcomes of the
is close to the average unit cost expected in Europe in 2030
by work of Almansoori and Betancourt-Torcat (2016) for 2030.
(around 3.2 $ kg−1). In 2050, one extra facility is installed in
In both studies, coal gasification technology is selected as
Cologne. Additionally, plants are installed in Stuttgart, Berlin,
the most economic option. One of the production facility is
Frankfurt and Dresden (see Fig. 6).
installed in Hannover, another in Cologne, and the last one in
Munich for both studies. The production facilities locations 5.2.
“Green” scenario
promotes the product distribution to regional storage facil-
ities. Additionally, each production facility includes nearby
Despite the costs for water electrolysis technology, the “green”
storage facilities to satisfy the local hydrogen demand.
scenario considers the opportunity to satisfy the local hydro-
Furthermore, hydrogen is generated in liquid form.
gen demand and the household energy demand by wind- and
Germany has a well-developed railway infrastructure, i.e. the
solar energy (see Appendix C). It was founded that 3 or 8
railway tank car is selected as preferred transportation mode
large hydrogen facilities are required. The “green” scenario
in both studies. It is noted that a large part of the German
shows that hydrogen facilities need to be built in Potsdam
rail freight transport is electrified, which means that the rail
and Hannover by 2030, and in 2050 they need to be installed
transport is a clean type of distribution. As hydrogen is gener-
in Hannover, Potsdam, Munich, Mainz, Rostock, Halle, and
ated in liquid form, super-insulated spherical tanks are used
Kiel. The total costs of the HSC are approximately 27.9 and
to minimize heat loss. The total cost of HSC is approximately
73.5 Mio $ d−1 for 2030 and 2050 respectively, which means
7.8 and 19.3 Mio $ d−1 for 2030 and 2050 respectively, which
9.69 and 9.57 $ kg−1 of H2. The hydrogen price decreased by
means 2.63 and 2.51 $ kg−1 of H2 (the hydrogen price is 4.5%
4% from 2030 to 2050. However, it is not a reasonable price for
less in 2050 than in 2030). In case of hydrogen price decreas-
industry. In addition, expenses related with household energy
ing up to 2050, it might motivate to replace gasoline cars by
consumption account for 25.3 and 23.4 Mio $ d−1 for 2030 and
FCEVs in future. However, the hydrogen price (2.63 $ kg−1) is
2050 respectively, which shows electricity price reduction for
lower than from Almansoori’s work (3.03 $ kg−1) as the cur- this sector. 98
Chemical Engineering Research and Design 1 3 4 ( 2 0 1 8 ) 90–103 2050 Storage system Hydrogen production plant N Number of facility 1 Hydrogen flowrate 1 1 1 1 1 1 2 1 1 2 3 1 1 1 1 1 2 1 1 2 1 3
Fig. 6 – Hydrogen supply chain network for 2050 (Base scenario).
hydrogen production and the general power demand for any
Table 6 – HSC cost for two case studies in 2030.
electrical facility (excluding compression). Due to the high HSC costs (k$ d−1) Base case “Green” scenario
energy consumption an electricity price reduction can make Operating costs
water electrolysis technology feasible (see Appendix E).
Production and storage facilities 4795.0 18,442.5
The electricity price might be decreased by several Production feedstock fuel 471.0 6912.0
strategies such as integration of more renewable electricity Transportation 2.2 5.3
generation facilities such as wind mills, solar panels, wave Capital costs
pumps, or using off-peak electricity. However, the installa-
Production and storage facilities 2277.0 2512.0
tion of new energy facilities is only realistic in places with Transport 37.2 61.3
suitable geographical conditions (location with high avail- Total network cost 7582.4 27,933.1
ability of renewable energy sources) providing continuous
Hydrogen unit cost ($ kg−1) 2.6 9.7
energy generation. Moreover, the modification of existing con-
ventional methods or development of innovative methods is
necessary for reduction of conversion losses and capital costs 5.3. Case comparison
investment. There are a number of problems related with
conventional electrolysers such as safety risks due to leaks,
As shown in Table 6, the cost analysis has been done for
stack degradation, membrane deterioration, difficulties with
the two cases in 2030. For the hydrogen pathways, water
starting the system after shutdown, and freezing of mem-
electrolysis consumes more electricity than other technolo-
branes, especially during cold weather. All these problems
gies (see Appendix A). This energy consumption is about
require technological improvements. In addition, analysis of 50 kWh kg−1 H
2, with a specific power cost to supply the elec-
the impact of intermittency of renewable energy sources on
trolysers of 0.05 $ kWh−1. The capital- and operating costs
the electrolysis system performances and reliability, analysis
for a hydrogen production facility are only related to the
of possible cost parameters changing and demand shifting are
required power to achieve the targeted hydrogen production
required to map the uncertainty. Thus, there is a necessity
rate. The actual hydrogen production facility’s specific energy
of further R&D efforts on system design, power electronics
consumption is the sum of the specific power demand for
Chemical Engineering Research and Design 1 3 4 ( 2 0 1 8 ) 90–103 99
and process control to make hydrogen production by water
in terms of economy as well as implementation. Hydrogen electrolysis competitive.
transport in liquid form is preferable, distributed via rail-
way tank car and stored into super-insulated spherical tanks. 6. Conclusion
FCEVs penetration rate of 10% in 2030 and 30% in 2050 into pas-
senger transport shows a 4% reduction of the hydrogen price,
In this work, general optimization model for a HSC network
it can motivate to replace gasoline cars by FCEVs in future to
is proposed. The results shows that H2 has the potential diminish environmental impact.
to replace fossil fuel as a source of energy, especially for
Based on the obtained results, the building of HSC network
transportation. The model was applied to design strategies of
would be expensive at the current stage of technological devel-
developing the future HSC network in Germany, considering a
opment. However, the potential impact of renewable energy
full range of local factors and geographical conditions. More-
sources on the development of the cost effective and sustain-
over, the use of the water electrolysis technology is suitable
able HSC network takes place in future. Moreover, that can
option to satisfy the local hydrogen demand and the house-
be complemented by CO2 network, captured in industry, to
hold energy demand by existing installed wind- and solar
produce chemicals like methanol or heavy hydrocarbons can
power plants without environmental impact. Due to the high
further motivate a hydrogen based economy.
energy consumption an electricity price reduction can make
water electrolysis technology competitive. However, currently
Appendix A. Capital and unit production costs
coal gasification technology is still the dominant technology,
of hydrogen production technologies (Simbeck and Chang, 2002) Production technology Steam reforming Coal gasification Water electrolysis Biomass gasification Product form LH LH LH LH
Design production capacity t d−1 960.00 960.00 960.00 960.00 Plant availability d 329.00 329.00 329.00 329.00 Annual production 103 t 315.84 315.84 315.84 315.84
Fuel required per H2 generated unit kg−1 H2 3.16 5.33 47.60 11.26 Fuel consumed unit d−1 3033.60 5116.80 45,696.00 10,809.60 Fuel price $ unit−1 0.14 0.03 0.05 0.05 CO2 produced kg kg−1 H2 17.40 30.30 0.00 32.10
SMR/gasifier/electrolyzer $ unit−1 317.25 239.85 1110.67 506.70 CO2 cost $ kg−1 0.06 0.06 0.06 0.06 Energy cost $ kW−1 560.00
CO shift, cool and cleanup $ kg−1 d−1 CO2 20.00 15.00
Air separation unit $ kg−1 d−1 O2 28.00 27.00 O2 consumed per H2 generated 1.08 1.41 Dispenser rate kg h−1 4000.00 4000.00 4000.00 4000.00 Number of dispenser 10.00 10.00 10.00 10.00 H2 dispenser unit cost $ 100,000.00 100,000.00 100,000.00 100,000.00
Power consumption kWh kg−1 H2 11.00 11.00 58.60 11.00 Electricity cost $ kWh−1 0.05 0.05 0.05 0.05 Unit cost $ kg−1 d−1 H2 318.22 877.74 1110.67 1027.94 Size factor process 0.75 0.75 0.75 0.80
Unit liq/gas cost $ kg−1 d−1 H2 700.00 700.00 700.00 700.00 Size factor liq/gas 0.75 0.75 0.75 0.75 Unit storage cost $ kg−1 H2 19.00 19.00 19.00 19.00 Size factor of storage 0.70 0.70 0.70 0.70
Total process unit cost (UC) M$ 746.89 1149.74 1317.45 1307.23
General facilities cost 20% of UC M$ 149.38 229.95 263.49 261.44
Engineering Permitting 10% of UC M$ 74.69 114.97 131.74 130.72 Contingencies 10% of UC M$ 74.69 114.97 131.74 130.72
Working Capital, Land 5% of UC M$ 37.34 57.49 65.87 65.36
Total capital cost (CC) M$ 1082.99 1667.13 1910.30 1895.48 O&M 1% of CC M$ y−1 10.83 16.67 19.10 18.95 100
Chemical Engineering Research and Design 1 3 4 ( 2 0 1 8 ) 90–103 Fuel price M$ y−1 139.73 50.50 751.70 177.82 Electricity cost M$ y−1 173.71 173.71 925.41 173.71
Fixed operating cost 5% M$ y−1 54.15 83.36 95.51 94.78
Capital charges 12% of capital M$ y−1 129.96 200.06 22.92 227.46
Total Operating Cost M$ y−1 508.38 524.30 2020.96 692.71
Unit Production cost $ kg−1 1.61 1.66 6.40 2.19
Appendix B. Results of the hydrogen supply by grid point (Base scenario) Grid point, g Year 2030 Year 2050 H2 produced H2 imports by Local H2 produced H2 imports by Local by grid point grid point satisfaction of by grid point grid point satisfaction of (t d−1) (t d−1) H2 demand (t d−1) (t d−1) H2 demand (t d−1) (t d−1) 1 381.02 959.52 108.86 959.52 2 917.53 454.86 953.36 326.59 953.36 3 136.08 952.35 380.81 4 81.658 208.66 5 27.22 63.50 6 63.50 190.51 7 217.73 955.25 592.37 8 54.43 127.01 9 953.92 264.45 953.68 708.73 10 958.78 595.90 1908.75 1618.45 11 136.08 362.88 12 36.29 90.72 13 136.08 956.90 349.08 14 72.586 172.37 15 99.79 254.02 16 72.58 172.37 Total 2830.23 1515.02 1315.21 7639.81 2077.49 5562.32
Appendix C. Results of the hydrogen distribution by grid point (“Green” scenario) Grid point, g Year 2030 Year 2050 H2 produced H2 imports by Local H2 produced H2 imports by Local by grid point grid point satisfaction of by grid point grid point satisfaction of (t d−1) (t d−1) H2 demand (t d−1) (t d−1) H2 demand (t d−1) (t kg d−1) 1 381.02 1070.50 2 462.67 953.36 326.59 953.36 3 136.08 381.02 4 942.94 81.10 953.96 200.98 5 27.22 63.50 6 63.50 190.51 7 217.73 598.75 8 54.43 952.49 126.94
Chemical Engineering Research and Design 1 3 4 ( 2 0 1 8 ) 90–103 101 9 1897.41 264.45 1915.31 708.73 10 598.75 1623.89 11 136.08 958.41 359.66 12 36.29 90.72 13 136.08 353.81 14 72.58 958.51 169.25 15 99.79 952.42 253.87 16 72.58 172.37 Total 2840.35 2494.80 345.55 7644.45 4871.66 2772.79
Appendix D. Hydrogen flowrate for different scenario Base scenario “Green” scenario Year 2030 Year 2050 Year 2030 Year 2050 From To H2 From To H2 From To H2 From To H2 grid grid flowrate grid grid flowrate grid grid flowrate grid grid flowrate (t d−1) (t d−1) (t d−1) (t d−1) 2 1 381.02 1 1 959.52 4 2 462.67 2 2 953.36 2 7 27.22 2 2 953.36 4 3 136.08 4 2 18.14 2 13 54.43 3 3 380.81 4 8 54.43 4 3 381.02 9 3 136.08 3 4 208.66 4 13 136.08 4 13 353.81 9 4 81.65 3 8 127.01 4 14 72.58 8 1 254.02 9 5 27.22 3 15 235.87 9 1 381.02 8 7 571.54 9 6 63.50 7 1 108.86 9 5 27.22 9 7 27.22 9 8 54.43 7 2 63.50 9 6 63.50 9 10 1179.36 9 13 81.65 7 7 592.37 9 7 217.73 11 1 508.03 9 14 72.58 7 11 190.51 9 10 598.75 11 12 90.72 9 15 99.80 9 5 36.29 9 11 136.08 14 1 308.45 9 16 72.58 9 6 190.51 9 12 36.29 14 2 308.45 10 7 190.51 9 9 708.73 9 15 99.79 14 16 172.37 10 11 136.08 9 15 18.14 9 16 72.58 15 5 63.50 10 12 36.29 10 5 27.22 15 6 190.51 10 10 1618.45 15 10 444.53 10 11 172.37 10 12 90.72 13 2 263.09 13 13 349.08 13 14 172.37 13 16 172.37
Appendix E. HSC network cost depending of the energy price HSC costs “Green” scenario Electricity price reduction 10% 50% 100% Operating costs
Production and storage facilities (M$ d−1) 16.91 10.80 3.15 102
Chemical Engineering Research and Design 1 3 4 ( 2 0 1 8 ) 90–103
Production feedstock fuel (M$ d−1) 6.64 3.65 0.01 Transportation ($ d−1) 3475.30 2441.66 2213.80 Capital costs
Production and storage facilities (M$ d−1) 2.51 2.51 2.51 Transport ($ d−1) 51,498.89 41,912.51 37,230.80
Total network cost (M$ d−1) 26.12 17.01 5.71
Hydrogen unit cost ($ d−1) 9.38 6.10 2.05
Appendix F. Initial availability of energy sources Grid point, g Primary energy source, e Biomass (t d−1) Coal (t d−1) Natural gas (t d−1)
Renewable energy source (GWh d−1) Base scenario “Green” scenario 2030 2050 2030 2050 1 1.99 0.00 0.00 25.85 43.57 0.00 10.83 2 4.62 0.00 0.00 61.50 104.09 20.64 64.87 3 0.00 0.00 0.00 0.34 0.57 0.00 0.00 4 1.92 95,890.41 0.00 55.73 98.52 37.07 81.26 5 0.00 0.00 0.00 1.70 3.04 0.00 1.09 6 0.00 0.00 0.00 0.59 1.04 0.00 0.00 7 1.13 0.00 0.00 16.43 28.60 0.00 10.45 8 4.39 0.00 0.00 32.51 57.78 28.08 53.89 9 5.34 0.00 0.00 112.46 200.64 88.51 178.92 10 2.19 293,041.10 0.00 46.47 81.43 0.00 31.83 11 0.63 0.00 0.00 29.67 52.37 0.00 41.34 12 0.06 0.00 0.00 3.81 6.63 0.63 4.04 13 2.46 95,890.41 0.00 15.58 27.09 3.55 16.39 14 3.07 26,027.40 0.00 42.32 75.07 36.31 69.88 15 2.91 0.00 0.00 47.50 84.74 33.52 72.16 16 2.26 0.00 0.00 14.38 25.22 8.36 20.04 Total 32.97 510,849.32 0.00 506.85 890.41 256.67 656.99 References
Almansoori, A., Betancourt-Torcat, A., 2016. Design of
optimization model for a hydrogen supply chain under
Hake, J.F., Linssen, J., Walbeck, M., 2006. Prospects for hydrogen in
emission constraints — a case study of Germany. Energy 111,
the German energy system. Energy Policy 34, 1271–1283,
414–429, http://dx.doi.org/10.1016/j.energy.2016.05.123.
http://dx.doi.org/10.1016/j.enpol.2005.12.015.
Almansoori, A., Shah, N., 2012. Design and operation of a
Hugo, A., Rutter, P., Pistikopoulos, S., Amorelli, A., Zoia, G., 2005.
stochastic hydrogen supply chain network under demand
Hydrogen infrastructure strategic planning using
uncertainty. Int. J. Hydrogen Energy 37, 3965–3977,
multi-objective optimization. Int. J. Hydrogen Energy 30,
http://dx.doi.org/10.1016/j.ijhydene.2011.11.091.
1523–1534, http://dx.doi.org/10.1016/j.ijhydene.2005.04.017.
Amos, W.A., 1999. Costs of Storing and Transporting Hydrogen.
International Energy Agency (IEA), 2016. World Energy Outlook
Other Inf. PBD 27 Jan 1999; PBD 27 Jan 1999; PBD 27 Jan 1999. 2016.
Ball, M., Wietschel, M., Rentz, O., 2007. Integration of a hydrogen
http://www.iea.org/publications/freepublications/publication/
economy into the German energy system: an optimising
WEB WorldEnergyOutlook2015ExecutiveSummaryEnglishFinal.pdf.
modelling approach. Int. J. Hydrogen Energy 32, 1355–1368,
International Union of Railways, 2012. Railway Handbook 2012 —
http://dx.doi.org/10.1016/j.ijhydene.2006.10.016.
Energy Consumption and CO2 Emissions.
BM Verkehr Bau und Stadtentwicklung, 2013. The Mobility and
ISE, F.I. for S.E.S., n.d. Energy Charts [WWW Document].
Fuels Strategy of the German Government (MFS).
https://www.energy-charts.de/index.htm (accessed 5.25.17).
CO2.earth, n.d. 2100 Projections [WWW Document].
Pregger, T., Nitsch, J., Naegler, T., 2013. Long-term scenarios and
https://www.co2.earth (accessed 5.25.17).
strategies for the deployment of renewable energies in
Chemical Engineering Research and Design 1 3 4 ( 2 0 1 8 ) 90–103 103
Germany. Energy Policy 59, 350–360,
Simbeck, D.R., Chang, E., 2002. Hydrogen Supply: Cost Estimate
http://dx.doi.org/10.1016/j.enpol.2013.03.049.
for Hydrogen Pathways–Scoping Analysis. National
Schill, W.P., 2014. Residual load, renewable surplus generation
Renewable Energy Laboratory, doi:NREL/SR-540-32525.
and storage requirements in Germany. Energy Policy 73, 65–79,
Statistisches Bundesamt, n.d. Koordinierte
http://dx.doi.org/10.1016/j.enpol.2014.05.032.
Bevölkerungsvorausberechnung nach Bundesländern [WWW
Shafiee, S., Topal, E., 2009. When will fossil fuel reserves be
Document]. https://www.destatis.de/EN/Homepage.html.
diminished? Energy Policy 37, 181–189, (Accessed 25 May 2017).
http://dx.doi.org/10.1016/j.enpol.2008.08.016.
U.S. Energy Information Administration, 2017. Annual Energy
Siemens, L., n.d. Energiepark Mainz [WWW Document].
Outlook, doi:DOE/EIA-0383(2017).
http://www.energiepark-mainz.de/en/. (Accessed 20 March 2017).
Document Outline

  • An outlook towards hydrogen supply chain networks in 2050 — Design of novel fuel infrastructures in Germany
    • 1 Introduction
    • 2 Network description and problem statement
      • 2.1 Problem statement
      • 2.2 Model description
        • 2.2.1 Grid
        • 2.2.2 Primary energy sources
        • 2.2.3 Hydrogen production
        • 2.2.4 Hydrogen physical form
        • 2.2.5 Transportation mode
        • 2.2.6 Storage facility
    • 3 Model formulation
      • 3.1 Household energy demand
      • 3.2 Demand for a certain energy source
      • 3.3 Hydrogen demand
      • 3.4 Hydrogen generation
      • 3.5 Hydrogen distribution
      • 3.6 Hydrogen storage
      • 3.7 Objective function
    • 4 Case study
    • 5 Results and discussion
      • 5.1 Base case scenario
      • 5.2 “Green” scenario
      • 5.3 Case comparison
    • 6 Conclusion
    • Appendix A Capital and unit production costs of hydrogen production technologies (Simbeck and Chang, 2002)
    • Appendix B Results of the hydrogen supply by grid point (Base scenario)
    • Appendix C Results of the hydrogen distribution by grid point (“Green” scenario)
    • Appendix D Hydrogen flowrate for different scenario
    • Appendix E HSC network cost depending of the energy price
    • Appendix F Initial availability of energy sources
    • References

Tài liệu liên quan: