Preprints and early-stage research may not have been peer reviewed yet.
Preprint

Integrating Hydrogen in Single-Price Electricity Systems: The Effects of Spatial Economic Signals (Working Paper)

Abstract

For the final, published version of this article see: https://www.sciencedirect.com/science/article/pii/S0301421521005930 ------- Hydrogen can contribute substantially to the reduction of carbon emissions in industry and transportation. However, the production of hydrogen through electrolysis creates interdependencies between hydrogen supply chains and electricity systems. Therefore, as governments worldwide are planning considerable financial subsidies and new regulation to promote hydrogen infrastructure investments in the next years, energy policy research is needed to guide such policies with holistic analyses. In this study, we link a electrolytic hydrogen supply chain model with an electricity system dispatch model, for a cross-sectoral case study of Germany in 2030. We find that hydrogen infrastructure investments and their effects on the electricity system are strongly influenced by electricity prices. Given current uniform prices, hydrogen production increases congestion costs in the electricity grid by 17%. In contrast, passing spatially resolved electricity price signals leads to electrolyzers being placed at low-cost grid nodes and further away from consumption centers. This causes lower end-use costs for hydrogen. Moreover, congestion management costs decrease substantially, by up to 20% compared to the benchmark case without hydrogen. These savings could be transferred into according subsidies for hydrogen production. Thus, our study demonstrates the benefits of differentiating economic signals for hydrogen production based on spatial criteria.
Integrating Hydrogen in Single-Price Electricity Systems: The Effects
of Spatial Economic Signals
Frederik vom Scheidta,,Jingyi Qub,Philipp Staudta,Dharik S. Mallapragadacand
Christof Weinhardta
aKarlsruhe Institute of Technology, Karlsruhe, Germany
bFZI Research Center for Information Technology, Karlsruhe, Germany
cMassachusetts Institute of Technology, Cambridge, USA
ARTICLE INFO
Keywords:
Hydrogen
Electricity markets
Nodal pricing
Congestion management
Sector coupling
Power-to-Gas
Electrolysis
ABSTRACT
Hydrogen can contribute substantially to the reduction of carbon emissions in industry and transporta-
tion. However, the production of hydrogen through electrolysis creates interdependencies between
hydrogen supply chains and electricity systems. Therefore, as governments worldwide are planning
considerable financial subsidies and new regulation to promote hydrogen infrastructure investments
in the next years, energy policy research is needed to guide such policies with holistic analyses. In this
study, we link a electrolytic hydrogen supply chain model with an electricity system dispatch model.
We use this methodology for a cross-sectoral case study of Germany in 2030. We find that hydrogen
infrastructure investments and their effects on the electricity system are strongly influenced by elec-
tricity prices. Given current uniform zonal prices, hydrogen production increases congestion costs in
the electricity grid by 11%. In contrast, passing spatially resolved electricity price signals leads to
electrolyzers being placed at low-cost grid nodes and further away from consumption centers. This
causes lower end-use costs for hydrogen. Moreover, congestion management costs decrease substan-
tially, by 24% compared to the benchmark case without hydrogen. These savings could be transferred
into according subsidies for hydrogen production. Thus, our study demonstrates the benefits of dif-
ferentiating subsidies for hydrogen production based on spatial criteria.
1. Introduction
Hydrogen produced from low-carbon sources can con-
tribute substantially to mitigating emissions in sectors that
are difficult or impossible to electrify directly. Governments
worldwide, and in particular in Europe, have announced strate-
gies and billions of public funding to develop large-scale hy-
drogen infrastructure, that is centered on electrolytic hydro-
gen supply (Hydrogen Council and McKinsey & Company,
2021). Since hydrogen production from electrolysis uses
large amounts of electricity, a future hydrogen sector will
introduce new interdependencies with the electricity sector.
While electricity prices influence the cost-minimal installa-
tion (vom Scheidt et al.,2021) and operation (Guerra et al.,
2019) of electrolyzers, these electrolyzers in turn introduce
new electricity demand into the bulk power system, influenc-
ing in the short term the usage of renewable energy (Ruhnau,
2020;Bødal et al.,2020), as well as congestion of power net-
works (vom Scheidt et al.,2021;Xiong et al.,2021), and in
the long term the need for electricity generation and trans-
mission capacity (Bødal et al.,2020). Most importantly, the
effects of these interdependencies will be strong and will pre-
vail for a long time, because electrolyzers are large-scale,
stationary consumers with typical lifetimes of ten years and
more (Schmidt et al.,2017).
The integration of electrolyzers in European grids raises
Corresponding author
frederik.scheidt@kit.edu (F.v. Scheidt)
frederik.scheidt@kit.edu (F.v. Scheidt)
ORCID(s): 0000-0001-6493-5073
some unique questions as European wholesale power mar-
kets are designed as single-price zonal markets that over-
look intra-zonal transmission capacities and price variations.
Such single-price zonal market designs are already leading
to rising congestion management costs in many electricity
systems (Staudt et al.,2017). In Germany, the costs for con-
gestion management have risen to almost a billion Euro an-
nually, and especially the curtailment of renewable energy
plants is increasing (Xiong et al.,2021). Without appropri-
ate policy to guide system-beneficial integration, hydrogen
production might strongly aggravate these effects. While
the importance of market cost-reflective price-regulation and
subsidization of electrolyzers has been voiced in the political
sphere (European Commission,2020), there is a prevailing
lack of energy policy research to guide efficient integration
of hydrogen infrastructure into the electricity sector.
Therefore, in this study, we link an electrolytic hydro-
gen supply chain model with an electricity system dispatch
model to analyze the cost-minimal hydrogen infrastructure
setup in electricity markets using zonal vs. nodal pricing
structure, using Germany 2030 as a case study. We find that
under current regulation with uniform electricity prices, the
cost-minimal solution is to produce hydrogen close to lo-
cations of consumption as one would expect. These loca-
tions partly coincide with high locational marginal electric-
ity costs. Consequently, hydrogen production aggravates the
inefficiencies of single-price markets and increases conges-
tion management costs substantially.
We compare this benchmark scenario to a case in which
vom Scheidt et al.: Preprint submitted to Elsevier Page 1 of 19
arXiv:2105.00130v1 [econ.GN] 1 May 2021
Integrating Hydrogen in Single-Price Electricity Systems
electrolyzers are offered spatially differentiated (nodal) price
signals based on the locational marginal prices that would
form in a nodal pricing system. We find that such nodal
signals lead to lower costs for hydrogen, higher shares of
hydrogen production at low-price nodes, and longer trans-
port distances. This demonstrates the sensitivity of hydro-
gen supply chains to spatial prices or subsidies. Moreover,
in this scenario, the integration of hydrogen leads to con-
gestion management costs that are substantially lower than
in the benchmark scenario and even below a scenario with-
out hydrogen. Interestingly, these avoided redispatch costs
could already cover most of the subsidies a regulator would
have to pay to mimic nodal prices for hydrogen within the
existing single-price market design.
Thus, in a time in which many policy makers and regu-
lators in single-price markets are planning future hydrogen
supply systems and corresponding subsidies for hydrogen in-
frastructure, our study highlights the considerable benefits of
differentiating those subsidies with respect to spatial criteria.
2. Background
Several past studies have addressed the spatial aspects of
hydrogen supply chains that use grid electricity for hydro-
gen production. Robinius et al. (2017); Reuß et al. (2019);
Emonts et al. (2019) present models that link a hydrogen
supply chain with a national electricity grid. Their studies
provide a high spatial granularity, i.e. the third level of the
Nomenclature of Territorial Units for Statistics (NUTS-3)
(European parliament and European council,2003). The
authors apply their model to the case of hydrogen fueled
passenger cars in Germany in 2050 and identify favorable
regions for hydrogen production in Germany. Runge et al.
(2019) optimize supply chains for synthetic fuels, including
hydrogen stored in liquid organic hydrogen carrier (LOHC)
material. Besides considering uniform single-prices, the au-
thors also present a case in which they calculate state-level
representative nodal prices for two exemplary states in Ger-
many (NUTS-2 level) and allow transportation of hydrogen
between the two states. This causes increased hydrogen pro-
duction in the state with lower prices. The authors acknowl-
edge the importance of future work analyzing feedback ef-
fects on the electricity system. We fill their identified gap
with this study.
Zhang et al. (2020) analyze the flexible operation of elec-
trolyzers that produce hydrogen for light, medium- and heavy-
duty fuel cell electric vehicles (FCEVs) in the Western United
States of America. They find evidence that increasing elec-
trolyzer flexibility lowers hydrogen and electricity genera-
tion cost and 𝐶𝑂2emissions. With a similar focus on tempo-
ral aspects, it has been demonstrated that flexibility of elec-
trolytic hydrogen production enables more renewable inte-
gration for case studies in Texas, USA (Bødal et al.,2020)
and Germany (Ruhnau,2020).
Rose and Neumann (2020a) focus on hydrogen supply
for heavy-duty trucks from on-site electrolysis at highway
fuel stations. They jointly optimize the infrastructure of fuel
stations and the electricity system. They find that using 100%
hydrogen fueled heavy-duty trucks in Germany in 2050 would
increase the total electricity demand by about 60 TWh and
cause additional infrastructure costs of about 12 billion euro
per year. They note that nodal prices contain important infor-
mation about ”cost-effective energy consumption from a sys-
tem perspective” and that investors in hydrogen infrastruc-
ture should consider the system perspective. This idea is ex-
panded and implemented by vom Scheidt et al. (2021). The
authors link a hydrogen supply chain optimization model
and a nodal electricity system dispatch model and observe
their interdependence in an initial case study of hydrogen-
fueled trucks and passenger cars. They find that compared
to current uniform single-price prices, nodal prices would
lead to more hydrogen generation at low-price nodes. This
in turn causes substantially lower congestion management
costs. However, their analysis, like all previous ones, fo-
cuses on a small subset of hydrogen demand, i.e. demand
from the transport sector.
Xiong et al. (2021) provide another perspective on the
topic of hydrogen integration in single-price markets. They
do not consider the effects of hydrogen production on day-
ahead energy wholesale markets, but analyze how Power-
to-Gas plants (i.e. electrolyzers) can serve as a redispatch
option. They find that curtailment of renewable generation
can be reduced by 12% when electrolyzers are installed for
performing redispatch at a few frequently curtailed nodes in
the German grid of 2015. The study thus showcases the im-
portance of spatial consideration in hydrogen infrastructure
planning. However, the study disregards spatial aspects of
the hydrogen supply chain and disregards how policy mak-
ers could incentivize investors to build electrolyzers at the
identified nodes. Moreover, the direct political applicability
of the study is restricted, because future hydrogen volumes,
shares of renewable and conventional generation, and spatial
distribution of generation will be much different than in the
used scenario from 2015.
In summary, past research indicates that the spatial di-
mension of hydrogen integration matters. Within the limita-
tions of single sector analyses and/or a reduced network con-
sideration, studies have demonstrated that electrolyzer loca-
tions influence grid congestion. However, to the best of our
knowledge, no past study has evaluated the cost-optimal hy-
drogen supply chain for electrolytic hydrogen use for a broad
range of hydrogen demand sectors, namely steel, ammonia,
methanol, refineries, and transportation, considered the ef-
fect of alternative electricity price signals, and assessed the
feedback effects of the resulting supply chains. Such anal-
ysis is timely from a policy perspective given the prospect
of significant electrolyzer capacity integration over the next
decade in the German power grid. Therefore, we pose the
following research questions to guide future policy decision
on hydrogen subsidization:
1. What is the cost-minimal supply chain design using
electrolytic hydrogen production for the combined hy-
drogen demand from all major relevant sectors in 2030
vom Scheidt et al.: Preprint submitted to Elsevier Page 2 of 19
Integrating Hydrogen in Single-Price Electricity Systems
Spatially resolved
hydrogen demand
Techno-economic
parameters:
Production
Conversion
Transport
Fueling stations
Grid topology
Transmission line
capacity
Spatially resolved:
Consumption
time series
Conventional
generation
capacity
Renewable
generation time
series
Generation cost
assumptions
Electricity
input data
(Chapter 4.2)
Hydrogen
input data
(Chapter 4.1)
Electrolyzer locations
Electricity consumption
Hydrogen costs
Hydrogen output data
Electricity output data
Uniform electricity prices
Nodal electricity prices
Redispatch costs
Data flow
Feedback data flow
Electricity system
dispatch model
(Chapter 3.2)
Hydrogen supply
chain model
(Chapter 3.1)
Figure 1: Methodology: Modelling the hydrogen and electricity system and linking them through inputs and outputs
in Germany?
2. How do electricity price signals (single-price versus
nodal) influence the cost-minimal locations of elec-
trolyzers and the costs of hydrogen?
3. How does hydrogen production change electricity whole-
sale prices and congestion management costs under
single-price and nodal price signals?
3. Methodology
To address the research questions, we model the hydro-
gen supply chain and the electricity system and link both
models through their respective inputs and outputs. As shown
in 1, we proceed in three steps. First, we parametrize an elec-
tricity system dispatch model without hydrogen and com-
pute uniform prices, nodal prices, and redispatch costs. Sec-
ond, utilizing the computed electricity prices, we run the hy-
drogen model to identify the cost-minimal spatial siting of
electrolyzers, their capacities, and the hydrogen transporta-
tion. We calculate one scenario for uniform prices to reflect
current regulation and one scenario for nodal prices, to re-
flect a more efficient solution. Third, we feed back the result-
ing locations and capacities of the electrolyzers as additional
regional loads into the electricity model. We calculate con-
sequential changes in wholesale electricity prices and con-
gestion management costs. Both models are implemented in
Python 3.7.3, and solved using the Gurobi solver 8.1.1.
3.1. Hydrogen supply chain model
In the following, we describe the details of the hydrogen
supply chain model. It represents an enhanced version of the
model in (vom Scheidt et al.,2021). Due to the temporal un-
certainty in demand from end-use sectors the model assumes
time-invariant hydrogen consumption.
3.1.1. Objective function
The model minimizes the total end-use costs of hydro-
gen, which consist of capital costs and operating costs for
electrolytic production (𝑃 𝐶𝐶 ,𝑃 𝑂𝐶 ), conversion (𝐶𝐶𝐶,
𝐶𝑂𝐶) and transportation (𝑇 𝐶 𝐶,𝑇 𝑂𝐶 ) of hydrogen, and the
fueling stations (𝑆𝐶 𝐶 ,𝑆 𝑂𝐶) (1). There are four decision
variables. (i) 𝑋𝑝is a binary variable that indicates whether
an electrolyzer is installed at a location 𝑝(1) or not (0). (ii)
𝐻𝑃𝑝∈ [0,∞) denotes the amount of hydrogen produced at
𝑝in 𝑘𝑔𝐻2per day. (iii) 𝑌𝑝,𝑐 is binary and indicates whether
hydrogen is transported from a production location 𝑝to a
consumption location 𝑐(1) or not (0). (iv) 𝐻 𝑇𝑝,𝑐 ∈ [0,∞)
denotes the amount of hydrogen transported from 𝑝to 𝑐in
𝑘𝑔𝐻2per day. 𝑃and 𝐶represent the set of all potential elec-
trolyzer plant locations 𝑝, and consumption locations 𝑐, re-
spectively. Thus, the model outputs the cost-minimal loca-
tions of electrolyzers, their individual daily production, the
transportation volume between each electrolyzer and point
of consumption, and the resulting end-use costs of hydro-
gen.
The model can be parametrized for three possible hydro-
gen states 𝑠for transportation via delivery trailers: lique-
fied (LH2), and bound in LOHC. These three states require
different technologies for conversion, transportation and fu-
eling stations and thus cause different costs. The annotation
of decision variables, indices and input variables is provided
in Table 1.
vom Scheidt et al.: Preprint submitted to Elsevier Page 3 of 19
Integrating Hydrogen in Single-Price Electricity Systems
Table 1
Notation for hydrogen model – Decision variables, indices and
variables
Decision variables
𝑋𝑝∈ {0,1} Hydrogen production/import at location
p (1), or not (0)
𝐻𝑃𝑝∈ [0,∞) Daily amount of hydrogen production at
p [𝑘𝑔𝐻2/day]
𝑌𝑝,𝑐 ∈ {0,1} Hydrogen transport from p to c (1), or
not (0)
𝐻𝑇𝑝,𝑐 ∈ [0,∞) Daily amount of hydrogen transportation
from p to c [𝑘𝑔𝐻2/day]
Indices
𝐶Set of consumption locations
𝑃𝑃 𝑟𝑜𝑑𝑢𝑐𝑡𝑖𝑜𝑛 Set of domestic hydrogen production
locations
𝑃𝐼𝑚𝑝𝑜𝑟𝑡 Set of hydrogen import locations
𝑆Set of hydrogen transportation states
{GH2, LH2, LOHC}
Input variables
𝑃 𝐶𝐶𝑝Production capital cost at 𝑝[EUR]
𝐶𝐶𝐶𝑝,𝑠 Conversion capital cost of 𝑠at 𝑝[EUR]
𝑇 𝐶𝐶𝑠Transportation capital cost of 𝑠[EUR]
𝑆𝐶 𝐶𝑠Total fueling station capital cost of 𝑠
[EUR]
𝑃 𝑂𝐶𝑝Production operating cost at 𝑝[EUR]
𝐶𝑂𝐶𝑝,𝑠 Conversion operating cost of 𝑠at 𝑝
[EUR]
𝑇 𝑂𝐶𝑝,𝑐,𝑠 Transportation operating cost of 𝑠from
𝑝to 𝑐[EUR]
𝑆𝑂𝐶𝑠Total fueling station operating cost of 𝑠
[EUR]
The four components of capital costs include specific an-
nual operation and management costs (𝑂&𝑀) and annuity
factors (𝐴𝐹 ). The annuity factors account for the deprecia-
tion of one-time investments over multiple years and depend
on the weighted average cost of capital (𝑊 𝐴𝐶𝐶 [%]) and
depreciation periods (𝑎[years] (2). Equations 3to 7and the
notations in Table 2provide details on the capital cost com-
ponents. Notably, the model includes overseas imports of
hydrogen. Imports do not incur any capital costs for produc-
tion and conversion (4), but specific production operating
costs (9). Moreover, conversion capital costs are assumed to
occur in bulk and independently of the location (5). Details
for operating costs are provided in Equations 8to 17. They
include operating costs for domestic production (8) and im-
port (9) of hydrogen. Furthermore, they include the costs
of converting hydrogen. In the case of gaseous delivery, hy-
drogen is compressed (10). For liquid delivery, hydrogen
needs to be liquefied and later evaporated at the location of
consumption (11). For LOHC delivery, the carrier material
needs to be hydrogenated and later dehydrogenated at the lo-
cation of consumption (12). For LH2 and LOHC, we assume
that imports already arrive in the respective form and thus do
not require the first conversion step for domestic delivery.
Table 2
Notation for hydrogen model - input parameters
𝑎Depreciation period [years]
𝐴𝐹 Annuity factor [%]
𝐶𝐴𝑃𝐼 𝑚𝑝𝑜𝑟𝑡 Import capacity [𝑘𝑔𝐻2/day]
𝐶𝐴𝑃𝑃 𝑟𝑜𝑑 𝑢𝑐𝑡𝑖𝑜𝑛,𝑚𝑎𝑥 Maximum production capacity [𝑘𝑔𝐻2/day]
𝐶𝐴𝑃𝑃 𝑟𝑜𝑑 𝑢𝑐𝑡𝑖𝑜𝑛,𝑚𝑖𝑛 Minimum production capacity [𝑘𝑔𝐻2/day]
𝐶𝐴𝑃𝑇 𝑟𝑎𝑖𝑙 𝑒𝑟𝑠Capacity of delivery trailer for state 𝑠
[𝑘𝑔𝐻2]
𝐷𝐼𝑆𝑇𝑝,𝑐 Air-line distance between p and c [km]
𝐷𝐹 Detour factor [-]
𝐷𝑆 Driving speed [𝑘𝑚]
𝐷𝑈 𝑅𝐷𝑟𝑖𝑣𝑖𝑛𝑔𝑠Duration of driving time [h]
𝐷𝑈 𝑅𝐿𝑜𝑎𝑑𝑖𝑛𝑔𝑠Duration of loading and unloading [h]
𝐸𝐶 Electricity consumption [𝑘𝑊 ℎ𝑒𝑙 𝑘𝑔𝐻2]
𝐸𝐷 Energy density of hydrogen [𝑘𝑊 ℎ𝐻2𝑘𝑔𝐻2]
𝐸𝐸 Electric efficiency [𝑘𝑊 ℎ𝐻2𝑘𝑊 ℎ𝑒𝑙 ]
𝐸𝑃 Uniform single electricity price
[EUR/𝑘𝑊 ℎ𝑒𝑙]
𝐸𝑃𝑝Electricity price at p [EUR/𝑘𝑊 ℎ𝑒𝑙]
𝐹 𝐶𝑇 𝑟𝑢𝑐𝑘 Fuel consumption of delivery truck [𝑙𝑘𝑚]
𝐹 𝐿𝐻𝐸𝑙𝑒𝑐 𝑡𝑟𝑜𝑙𝑦𝑧𝑒𝑟 Full load hours of electrolyzers [h]
𝐹 𝑃 Fuel price [EUR𝑙𝑖𝑡𝑒𝑟]
𝑁𝐺𝐶𝑆𝑡𝑎𝑡𝑖𝑜𝑛𝑠Natural gas consumption of fuel station of
hydrogen state s [𝑘𝑊 ℎ𝑁𝐺 𝑘𝑔𝐻2]
𝐻𝐷𝑐Daily hydrogen demand at location c
[𝑘𝑔𝐻2/day]
𝐻𝐼𝐶 Hydrogen import costs [EUR𝑘𝑔𝐻2]
𝐼𝐶𝐶𝑜𝑛𝑣𝑒𝑟𝑠𝑖𝑜𝑛𝑠Capacity-specific investment costs of
conversion equipment [EUR/𝑘𝑔𝐻2]
𝐼𝐶𝐸𝑙𝑒𝑐 𝑡𝑟𝑜𝑙𝑦𝑧𝑒𝑟 Capacity-dependent investment costs of
electrolyzer [EUR/𝑘𝑊𝑒𝑙]
𝐼𝐶𝑆𝑡𝑎𝑡𝑖𝑜𝑛𝑠Investment cost per fuel station for state s
[EUR]
𝐼𝐶𝑇 𝑟𝑎𝑖𝑙𝑒𝑟𝑠𝑠Investment cost per trailer for state s
[EUR]
𝐼𝐶𝑇 𝑟𝑢𝑐𝑘𝑠 Investment cost per truck [EUR]
𝑀A very large positive number [-]
𝑁𝑆 Number of fuel stations [-]
𝑁𝐺𝑃 Natural gas price [EUR/𝑘𝑊 ℎ𝑁𝐺 ]
𝑁𝑇𝑠Number of trucks for product state s [-]
𝑁𝑇𝑇 𝑟𝑎𝑖𝑙𝑒𝑟𝑠𝑠Number of trailers for product state s [-]
𝑂&𝑀Operation and maintenance cost factor
[%]
𝑇Toll cost [EUR𝑘𝑚]
𝑊Wage [EURℎ𝑜𝑢𝑟]
𝑊 𝐴𝐶𝐶 Weighted average cost of capital [%]
After initial conversion, hydrogen is transported to the con-
sumption sinks. Hydrogen can be transported via tube trail-
ers mounted onto delivery trucks, or via pipelines. Since
related work indicates that transport via pipelines only be-
comes economically viable for high demand scenarios and
long transport distances (Reuß et al.,2019;Robinius et al.,
2017), our model focuses on transport via tube trailers on
delivery trucks.1Thus, transport operating costs consist of
1As hydrogen volumes grow, transport via dedicated hydrogen
pipelines could gain relevance. Future work could expand our model by
offering both truck based and pipeline based hydrogen transportation.
vom Scheidt et al.: Preprint submitted to Elsevier Page 4 of 19
Integrating Hydrogen in Single-Price Electricity Systems
costs for labor, fuel, and toll (13). Labor costs depend on
the time that drivers spend loading, driving, and unload-
ing the delivery trailers (14). A fixed loading and unload-
ing time per delivery is assumed. Round-trip driving time
is determined by the distance between connected produc-
tion plants and points of consumption, as well as the driv-
ing speed 𝐷𝑆. Transport distances are approximated via
air-line distance, multiplied with a detour factor of 1.3, in
line with Reuß (2019). Since the daily capacity of fueling
stations is assumed to be smaller than the capacity of one
delivery trailer, we multiply the distances to fueling stations
with a frequency factor 𝐻𝐷𝑐𝐶 𝐴𝑃𝑇 𝑟𝑎𝑖𝑙𝑒𝑟𝑠(15) simulating
that they are not provided with hydrogen on a daily basis.
This also applies to fuel and toll costs (16).
min
𝑋𝑝,𝐻𝑃𝑝,𝑌𝑝,𝑐
(
𝑝𝑃
𝑃 𝐶 𝐶𝑝+
𝑝𝑃
𝐶𝐶𝐶𝑝,𝑠 +𝑇 𝐶 𝐶𝑠+𝑆𝐶𝐶𝑠
+
𝑝𝑃
𝑃 𝑂𝐶𝑝,𝑠(𝑿𝒑, 𝑯 𝑷𝒑) +
𝑝𝑃
𝐶𝑂𝐶𝑝,𝑠(𝑿𝒑, 𝑯 𝑷𝒑)
+
𝑝𝑃
𝑐𝐶
𝑇 𝑂𝐶𝑝,𝑐,𝑠 (𝒀𝒑,𝒄 , 𝑯 𝑻𝒑,𝒄) + 𝑆 𝑂𝐶𝑠)
(1)
𝐴𝐹 =(1 + 𝑊 𝐴𝐶𝐶)𝑎𝑊 𝐴𝐶𝐶
(1 + 𝑊 𝐴𝐶𝐶)𝑎− 1 (2)
𝑃 𝐶 𝐶𝑝=𝑋𝑝𝐻𝑃𝑝𝐸𝐷 𝐼𝐶𝐸𝑙𝑒𝑐 𝑡𝑟𝑜𝑙𝑦𝑧𝑒𝑟
𝐹 𝐿𝐻 𝐸𝐸
∗ (1 + 𝑂&𝑀𝐸 𝑙𝑒𝑐𝑡𝑟𝑜𝑙 𝑦𝑧𝑒𝑟)
𝐴𝐹𝐸𝑙𝑒𝑐 𝑡𝑟𝑜𝑙𝑦𝑧𝑒𝑟 𝑝𝑃𝑃 𝑟𝑜𝑑𝑢𝑐 𝑡𝑖𝑜𝑛
(3)
𝑃 𝐶 𝐶𝐼𝑚𝑝𝑜𝑟𝑡 = 0 (4)
𝐶𝐶𝐶𝑠=𝐼𝐶𝐶𝑜𝑛𝑣𝑒𝑟𝑠𝑖𝑜𝑛𝑠 (1 + 𝑂&𝑀𝐶𝑜𝑛𝑣𝑒𝑟𝑠𝑖𝑜𝑛𝑠)
𝐴𝐹𝐶𝑜𝑛𝑣𝑒𝑟𝑠𝑖𝑜𝑛𝑠𝑠𝑆(5)
𝑇 𝐶𝐶𝑠=𝐼𝐶𝑇 𝑟𝑢𝑐𝑘𝑠 𝑁𝑇𝑠∗ (1 + 𝑂&𝑀𝑇 𝑟𝑢𝑐 𝑘𝑠)
𝐴𝐹𝑇 𝑟𝑢𝑐𝑘𝑠 +𝐼 𝐶𝑇 𝑟𝑎𝑖𝑙 𝑒𝑟𝑠𝑠𝑁𝑇𝑠
∗ (1 + 𝑂&𝑀𝑇 𝑟𝑎𝑖𝑙𝑒𝑟𝑠𝑠) ∗ 𝐴𝐹𝑇 𝑟𝑎𝑖𝑙 𝑒𝑟𝑠 𝑠𝑆
(6)
𝑆𝐶 𝐶𝑠=𝐼 𝐶𝑆 𝑡𝑎𝑡𝑖𝑜𝑛𝑠𝑁 𝑆 (1 + 𝑂&𝑀𝑆𝑡𝑎𝑡𝑖𝑜𝑛)
𝐴𝐹𝑆𝑡𝑎𝑡𝑖𝑜𝑛 𝑠𝑆(7)
𝑃 𝑂𝐶𝑝=𝐻 𝑃𝑝𝐸𝐶𝑃 𝑟𝑜𝑑𝑢𝑐 𝑡𝑖𝑜𝑛 𝐸 𝑃𝑝∗ 365
𝑝𝑃𝑃 𝑟𝑜𝑑𝑢𝑐 𝑡𝑖𝑜𝑛
(8)
𝑃 𝑂𝐶𝐼 𝑚𝑝𝑜𝑟𝑡 =𝐻𝑃𝐼𝑚𝑝𝑜𝑟𝑡 𝐻𝐼𝐶 ∗ 365 (9)
𝐶𝑂𝐶𝑝,𝐺𝐻 2=
𝑝𝑃𝑃 𝑟𝑜𝑑𝑢𝑐𝑡𝑖𝑜𝑛
𝐻𝑃𝑝𝐸𝐶𝐶 𝑜𝑚𝑝𝑟𝑒𝑠𝑠𝑖𝑜𝑛
𝐸𝑃𝑝 (1 + 𝐿𝑜𝑠𝑠𝐶𝑜𝑚𝑝𝑟𝑒𝑠𝑠𝑖𝑜𝑛) ∗ 365
(10)
𝐶𝑂𝐶𝑝,𝐿𝐻 2= (
𝑝𝑃𝑃 𝑟𝑜𝑑𝑢𝑐𝑡𝑖𝑜𝑛
𝐻𝑃𝑝𝐸𝐶𝐿𝑖𝑞 𝑢𝑒𝑓 𝑎𝑐𝑡𝑖𝑜𝑛
𝐸𝑃𝑝 (1 + 𝐿𝑜𝑠𝑠𝐿𝑖𝑞𝑢𝑒𝑓 𝑎𝑐𝑡𝑖𝑜𝑛 ))
+(
𝑝𝑃
𝐻𝑃𝑝𝐸𝐶𝐸 𝑣𝑎𝑝𝑜𝑟𝑎𝑡𝑖𝑜𝑛 𝐸𝑃
∗ (1 + 𝐿𝑜𝑠𝑠𝐸 𝑣𝑎𝑝𝑜𝑟𝑎𝑡𝑖𝑜𝑛)) ∗ 365
(11)
𝐶𝑂𝐶𝑝,𝐿𝑂𝐻 𝐶 = (
𝑝𝑃𝐼𝑚𝑝𝑜𝑟𝑡
𝐻𝑃𝑝𝐸𝐶𝐻 𝑦𝑑 𝑟𝑜𝑔𝑒𝑛𝑎𝑡𝑖𝑜𝑛
𝐸𝑃𝑝 (1 + 𝐿𝑜𝑠𝑠𝐻𝑦𝑑 𝑟𝑜𝑔𝑒𝑛𝑎𝑡𝑖𝑜𝑛))
+(
𝑝𝑃
𝐻𝑃𝑝∗ (𝐸𝑃 𝐸 𝐶𝐷𝑒ℎ𝑦𝑑 𝑟𝑜𝑔𝑒𝑛𝑎𝑡𝑖𝑜𝑛
+𝑁𝐺𝑃 𝑁 𝐺𝐶𝐷𝑒ℎ𝑦𝑑 𝑟𝑜𝑔𝑒𝑛𝑎𝑡𝑖𝑜𝑛 )
∗ (1 + 𝐿𝑜𝑠𝑠𝐷𝑒ℎ𝑦𝑑 𝑟𝑜𝑔𝑒𝑛𝑎𝑡𝑖𝑜𝑛 )) ∗ 365
(12)
𝑇 𝑂𝐶𝑠=𝐿𝑎𝑏𝑜𝑟𝐶𝑜𝑠𝑡𝑠 +𝐹 𝑢𝑒𝑙𝐴𝑛𝑑 𝑇 𝑜𝑙𝑙𝐶 𝑜𝑠𝑡𝑠 (13)
𝐿𝑎𝑏𝑜𝑟𝐶𝑜𝑠𝑡𝑠 = (𝐷𝑈 𝑅𝐷𝑟𝑖𝑣𝑖𝑛𝑔𝑠+𝐷𝑈 𝑅𝐿𝑜𝑎𝑑 𝑖𝑛𝑔𝑠𝑁𝑇𝑠)
𝑊(14)
𝐷𝑟𝑖𝑣𝑖𝑛𝑔𝑇 𝑖𝑚𝑒𝑠= (
𝑝𝑃
𝑐𝐶𝐼𝑛𝑑 𝑢𝑠𝑡𝑟𝑦
𝑌𝑝,𝑐 𝐷𝐼 𝑆𝑇 𝑝, 𝑐
+
𝑝𝑃
𝑐𝐶𝑆𝑡𝑎𝑡𝑖𝑜𝑛𝑠
𝑌𝑝,𝑐 𝐷𝐼𝑆𝑇𝑝,𝑐 𝐻𝐷𝑐
𝐶𝐴𝑃𝑇 𝑟𝑎𝑖𝑙𝑒𝑟𝑠
)
2 ∗ 𝐷𝐹
𝐷𝑆
(15)
𝐹 𝑢𝑒𝑙𝐴𝑛𝑑 𝑇 𝑜𝑙𝑙𝐶𝑜𝑠𝑡𝑠 = (𝐹 𝐶𝑇 𝑟𝑢𝑐𝑘 𝐹 𝑃 +𝑇)∗2
𝐷𝐹 ∗ ((
𝑝𝑃
𝑐𝐶𝐼𝑛𝑑 𝑢𝑠𝑡𝑟𝑦
𝑌𝑝,𝑐 𝐷𝐼 𝑆𝑇 𝑝, 𝑐)
+(
𝑝𝑃
𝑐𝐶𝑆𝑡𝑎𝑡𝑖𝑜𝑛𝑠
𝑌𝑝,𝑐 𝐷𝐼𝑆𝑇𝑝,𝑐 𝐻𝐷𝑐
𝐶𝐴𝑃𝑇 𝑟𝑎𝑖𝑙𝑒𝑟𝑠
))
(16)
𝑆𝑂𝐶𝑠= (𝐸𝐶𝑆 𝑡𝑎𝑡𝑖𝑜𝑛𝑠𝐸𝑃 +𝑁𝐺𝐶𝑆 𝑡𝑎𝑡𝑖𝑜𝑛𝑠𝑁 𝐺𝑃 )
∗ (1 + 𝐿𝑜𝑠𝑠𝑆 𝑡𝑎𝑡𝑖𝑜𝑛𝑠) ∗
𝑐𝐶𝑆𝑡𝑎𝑡𝑖𝑜𝑛𝑠
𝐻𝐷𝑐∗ 365 (17)
3.1.2. Constraints
The model includes both domestic production and an ex-
ogenously given import at one fixed node. The sum of daily
vom Scheidt et al.: Preprint submitted to Elsevier Page 5 of 19
Integrating Hydrogen in Single-Price Electricity Systems
domestic and imported hydrogen production 𝐻𝑃 must sat-
isfy the sum of the exogenously given demand 𝐻𝐷 (18).
Hydrogen output 𝐻𝑃𝑝of each electrolyzer depends on its in-
stalled capacity, which varies between a fixed minimum (19)
and maximum (20) value. The import nodes and their capac-
ity are exogenously set (21,22). In sum, the daily amount
of hydrogen transported from an electrolyzer must not ex-
ceed its production (23). The daily amount transported to a
consumer must meet its demand (24). Positive transport vol-
ume from a plant to a consumption location is only possible
if the delivery connection is established via binary variable
𝑌(25). Thus, one limitation of the model is that it does not
consider short term or long term temporal variations in hy-
drogen transportation or consumption and thus neglects stor-
age. While out of scope of this study, future work could at-
tempt to identify short term and long term temporal patterns
of hydrogen demand from industry and transportation.2We
do assess a sensitivity case "FlexOp" in which transporta-
tion and consumption remain continuous, but the produc-
tion is temporally flexible. This allows us to identify an op-
timistic estimate of the potential cost savings that flexible
electrolyzer operation can yield.
𝑝𝑃
𝐻𝑃𝑝=
𝑐𝐶
𝐻𝐷𝑐(18)
𝐻𝑃𝑝𝐶𝐴𝑃𝑃 𝑟𝑜𝑑𝑢𝑐 𝑡𝑖𝑜𝑛,𝑚𝑖𝑛 𝑋𝑝
𝑝𝑃𝑃 𝑟𝑜𝑑𝑢𝑐 𝑡𝑖𝑜𝑛
(19)
𝐻𝑃𝑝𝐶𝐴𝑃𝑃 𝑟𝑜𝑑𝑢𝑐 𝑡𝑖𝑜𝑛,𝑚𝑎𝑥 𝑋𝑝
𝑝𝑃𝑃 𝑟𝑜𝑑𝑢𝑐 𝑡𝑖𝑜𝑛
(20)
𝑋𝑝= 1 𝑝𝑃𝐼 𝑚𝑝𝑜𝑟𝑡 (21)
𝐻𝑃𝑝=𝐶𝐴𝑃𝐼𝑚𝑝𝑜𝑟𝑡 𝑋𝑝𝑝𝑃𝐼 𝑚𝑝𝑜𝑟𝑡 (22)
𝑐𝐶
𝐻𝑇𝑝,𝑐 𝐻𝑃𝑝𝑝𝑃(23)
𝑝𝑃
𝐻𝑇𝑝,𝑐 𝐻𝐷𝑐𝑐𝐶(24)
𝐻𝑇𝑝,𝑐,𝑠 𝑌𝑝,𝑐 𝑀𝑐𝐶 , 𝑝𝑃(25)
2Regarding long term storage, techno-economic parameters are pre-
sented by Reuß et al. (2019) and locations with high geological potential
for hydrogen storage are presented by Caglayan et al. (2020).
3.2. Electricity system model
Next, we model the electricity system to calculate elec-
tricity prices and congestion management costs, following
the approach introduced in vom Scheidt et al. (2021).
For the uniform price scenario, we adapt a stylized merit-
order and redispatch model from Staudt and Oren (2020).
For each hour, the model minimizes the marginal generation
costs for the entire single-price market area. The model’s
constraints ensure that demand and supply are balanced sub-
ject to the constraints that limit available generation capac-
ity (26). Complying with the market designs of single-price
markets, this model does not consider grid constraints. There-
fore, the resulting market allocation can be technically infea-
sible, in which case redispatch measures ensue. The cost-
based redispatch mechanism begins at the existing market
allocation and activates and deactivates generation capacity
in the system until the cost-minimal solution is found that
respects grid constraints which in the optimal case is equiva-
lent to the nodal pricing solution (Staudt,2019). Generators
that are activated through this procedure are compensated
based on their operating costs. The additional costs caused
by this procedure are referred to as redispatch costs. In the
considered idealized case, they are equivalent to the conges-
tion management costs (27). The annotation for uniform,
redispatch, and nodal pricing model are given in Table 3.
min (
𝑇
𝑡=1
𝑁
𝑛=1
𝐺
𝑔=1
𝑞𝑛,𝑔,𝑡 𝑝𝑛,𝑔 )
𝑠.𝑡.
𝑁
𝑛=1
𝑑𝑔,𝑡 =
𝑁
𝑛=1
𝐺
𝑔=1
𝑞𝑛,𝑔,𝑡 𝑡𝑇
𝑞𝑛,𝑔,𝑡 𝑐𝑛,𝑔,𝑡 𝑔𝐺𝑛𝑁 , 𝑡𝑇
(26)
min (
𝐺
𝑔=1
𝑞Δ
𝑔,𝑡 𝑝𝑛,𝑔 ) ∀𝑡𝑇
𝑠.𝑡.
𝐺
𝑔=1
𝑞Δ
𝑔,𝑡 = 0 𝑡𝑇
𝐺
𝑔=1
(𝑞Δ
𝑔,𝑡 +𝑞𝑔,𝑡 𝑐𝑔 ,𝑡) ∀𝑡𝑇
𝑞Δ
𝑔,𝑡 +𝑞𝑔,𝑡 0 ∀𝑔𝐺, 𝑡𝑇
|
𝑁
𝑛=1
𝐺
𝑔=1
(((𝑞𝑛,𝑔,𝑡 +𝑞Δ
𝑛,𝑔,𝑡) − 𝑑𝑛,𝑡) ∗ 𝐻𝑙,𝑛 ))|
𝜏𝑙𝑙𝐿, 𝑡𝑇
(27)
For the nodal price scenario, we use a nodal model with
a DC-load flow approximation. This model simultaneously
takes into account generation capacities and prices, as well
as transmission capacities (28). Both models optimize each
hour step-wise, independently of other hours. They thus ne-
glect ramping and storage.
vom Scheidt et al.: Preprint submitted to Elsevier Page 6 of 19
Integrating Hydrogen in Single-Price Electricity Systems
min (
𝑇
𝑡=1
𝑁
𝑛=1
𝐺
𝑔=1
𝑞𝑛,𝑔,𝑡 𝑝𝑛,𝑔 )
𝑠.𝑡.
𝑁
𝑛=1
𝑑𝑔,𝑡 =
𝑁
𝑛=1
𝐺
𝑔=1
𝑞𝑛,𝑔,𝑡 𝑡𝑇
𝑞𝑛,𝑔,𝑡 𝑐𝑛,𝑔,𝑡 𝑔𝐺𝑛𝑁 , 𝑡𝑇
|
𝑁
𝑛=1
𝐺
𝑔=1
((𝑞𝑔,𝑡 𝑑𝑛,𝑡) ∗ 𝐻𝑙,𝑛)|
𝜏𝑙𝑙𝐿, 𝑡𝑇
(28)
Table 3
Notation for electricity system model
𝑞𝑛,𝑔,𝑡 Generation of unit g at node n at time t
𝑞Δ
𝑛,𝑔,𝑡 Redispatch of unit g at node n at time t
𝑝𝑛,𝑔 Marginal generation costs of unit g at node n
𝑑𝑛,𝑡 Demand at node n at time t
𝑐𝑛,𝑔,𝑡 Generation capacity of unit g at node n (at time
t for renewables)
𝜏𝑙Transmission capacity of line l
𝐻Matrix of power distribution factors
𝑁Number of nodes n
𝐺Number of generation units g
𝐿Number of lines l
4. Case study
To demonstrate the functioning of the hydrogen model
and the electricity model, we apply it to a case study. For
this, we parametrize the models with data for the German
electricity system and hydrogen demand in 2030.
4.1. Hydrogen data
In this subsection, we present all data sources, prepro-
cessing steps, and assumptions used for creating the input
data sets for demand, production, conversion, and transporta-
tion of hydrogen.
4.1.1. Hydrogen demand
In the following paragraphs, we describe data acquisition
and preprocessing for the German hydrogen net demand in
2030. Hydrogen demand is assumed to come from the six
following sectors: steel, ammonia, methanol, refinery, road
transportation, and individual mobility. First, each demand
sector is presented with general assumptions about future
hydrogen demand and potential. Subsequently, the relevant
locations of the respective sector with hydrogen demand in
2030 are identified. The estimated total hydrogen net de-
mand amounts to 51.26 TWh. Table 11 in appendix Bshows
the numeric values and conversion factors used for the hy-
drogen demand calculations. For steel, ammonia, methanol,
and refineries, 100% availability of the production facilities
Table 4
Estimated hydrogen net demand of primary steel producers in
Germany, 2030
Steel producer Hydrogen net
demand [TWh]
ArcelorMittal Bremen 0.0
ArcelorMittal Duisburg 0.0
ArcelorMittal Eisenhüttenstadt 0.0
ArcelorMittal Hamburg 2.67
ROGESA (Dillinger & Saarstahl) 2.16
HKM Duisburg 0.0
Salzgitter Peine 2.25
Thyssenkrupp Steel Europe Duisburg 6.17
Total 13.25
is assumed. Correspondingly, quantities that have been cal-
culated down to hours are multiplied by 8,760 to get the re-
spective annual quantity. For details on data acquisition and
processing we refer to appendix A.
Steel Steel production in Germany offers a large potential
for the use of hydrogen in industry by switching to hydrogen-
based processes. In general, a distinction is made in steel
production between primary and secondary steel as well as
between blast furnace and electric arc routes (Hebling et al.,
2019). Today, primary steel production is mainly based on
coal- or coke-based processes to reduce iron ore in the blast
furnace, resulting in large amounts of carbon emissions
(Wilms et al.,2018). An alternative to the blast furnace is
direct reduction, in which the iron ore is reduced by natu-
ral gas or hydrogen and CO2emissions are directly avoided
(Hebling et al.,2019). The product "direct reduced iron"
(DRI) is further processed into steel in an electric arc fur-
nace. If hydrogen is produced by electrolysis with electricity
from renewable energies and used instead of coal in the di-
rect reduction process, up to 95 % of CO2emissions could be
avoided on the way to primary steel (Berger,2020). In addi-
tion to the possibility of switching to direct reduction, CO2
emission reductions can be achieved by blowing in hydrogen
as a substitute reducing agent. The basic idea is to reduce
the amount of injection coal required and to replace it with
hydrogen, in order to reduce CO2emissions (thyssenkrupp,
2019). Depending on the operating conditions, emissions
can be reduced by 21.4 - 28.5 % compared to a reference
case with today’s standard operating mode (Yilmaz,2018).
We identify all steel plants with potential for hydrogen
use in 2030. Table 4summarizes the hydrogen net demand
of the steel industry. Based on the assumptions made, the
total hydrogen net demand for 2030 amounts to 13.25 TWh
and is distributed over Hamburg, Dillingen/Saar, Peine and
Duisburg.
Ammonia Ammonia (NH3) is produced using the Haber-
Bosch process and requires the input components hydrogen
(H2) and nitrogen (N2) (Hermann et al.,2014). The po-
vom Scheidt et al.: Preprint submitted to Elsevier Page 7 of 19
Integrating Hydrogen in Single-Price Electricity Systems
Table 5
Estimated hydrogen demand of ammonia producers in Ger-
many, 2030
Ammonia producer Hydrogen net
demand [TWh]
BASF Ludwigshafen 5.18
INEOS Köln 2.25
SKW Stickstoffwerke Piesteritz 5.62
YARA Brunsbüttel 4.44
Total 17.49
Table 6
Estimated hydrogen net demand of methanol producers in Ger-
many, 2030
Methanol producer Hydrogen net
demand [TWh]
BASF Ludwigshafen 2.83
Shell Rheinland Raffinerie - Süd 2.74
Ruhr Oel - BP Gelsenkirchen 1.76
Total Raffinerie Mitteldeutschland 4.40
Total 11.73
tential for CO2emissions reduction lies in replacing fossil-
generated hydrogen with electricity-based hydrogen. To-
day, hydrogen is mostly produced from steam methane re-
forming, with the by-product CO2. This byproduct can be
used for processes in material composites, such as the pro-
duction of urea (Hebling et al.,2019). Nevertheless, our
estimation assumes a complete switch of ammonia produc-
tion to electricity-based hydrogen in order to define an upper
limit of hydrogen demand in the ammonia industry. Table 5
summarises the hydrogen demand of the ammonia industry.
Based on the assumptions made, the total hydrogen demand
is 17.49 TWh and is distributed over four plants.
Methanol Currently, methanol is commonly produced us-
ing synthesis processes with CO2emissions, which in fu-
ture can be switched to hydrogen based processes (Michal-
ski et al.,2019). Table 6summarises the hydrogen demand
of the methanol industry. Based on the assumptions made,
the total hydrogen demand is 11.73 TWh and is distributed
over four sites.
Refineries In refineries, hydrogen is used on a large scale
to desulfurize fuels and to refine heavy residues with hydro-
gen via hydrocracking (Hermann et al.,2014). The hydro-
gen needed for crude oil processing is supplied from internal
and external sources. This means that refineries are partly
self-sufficient, since hydrogen is a by-product of other pro-
cessing operations (ENCON.Europe GmbH,2018). In this
study, a 22 % net demand for hydrogen is assumed, analo-
gous to Wilms et al. (2018). This hydrogen net demand is as-
sumed to be entirely served by electricity-based hydrogen in
2030, in line with Prognos AG (2020b). Table 7summarises
Table 7
Estimated hydrogen net demand of refineries in Germany, 2030
Refinery Hydrogen net
demand [TWh]
Bayernoil Raffineriegesellschaft 0.19
BP Raffinerie Lingen 0.21
Gunvor Raffinerie Ingolstadt 0.22
Holborn Europa Raffinerie 0.23
MiRO Mineraloelraffinerie Oberrhein 0.66
Nynas 0.08
OMV Deutschland 0.16
PCK Raffinerie 0.51
Raffinerie Heide 0.19
Ruhr Oel - BP Gelsenkirchen 0.57
Shell Rheinland Raffinerie Werk Nord 0.41
Shell Rheinland Raffinerie Werk Süd 0.32
Total Raffinerie Mitteldeutschland 0.53
Total 4.29
the hydrogen net demand of the refineries in Germany 2030.
The estimated total hydrogen net demand is 4.29 TWh and
is distributed over thirteen sites in Germany.
Transportation sector In the first step, we estimate to-
tal national hydrogen demand in the transport sector and the
number of fueling stations required to satisfy the demand, in-
cluding both road transportation with trucks and individual
mobility with fuel cell passenger cars. In the second step,
we spatially disaggregate this total demand and determine
potential sites for fueling stations.
To determine the hydrogen demand for fuel cell trucks
and passenger cars in Germany in 2030, we use the mean
estimates from Fraunhofer-Institut (2019). We assume that
heavy-duty trucks with a total weight above 12,000 kg (Euro-
pean Alternative Fuels Observatory) will be responsible for
all truck based demand, because they have particularly high
carbon emission savings potential and the fuel cell based
version has stronger advantages over to their battery based
counterparts, i.e. heavier payloads, longer ranges, and shorter
recharging times (Weger et al.,2020). We assume the con-
sumption of trucks to decrease to 8 kg/100km until 2030, and
that of fuel cell passenger cars to decrease to 0.63 kg/100km,
in line with Grube and Stolten (2018); FCH-JU (2017);
Hyundai (2020). We assume that by 2030 all hydrogen sta-
tions will become L-size (IEA,2015) with a capacity of 1,000
kg/day. According to Reuß et al. (2019), station investment
cost is estimated considering scaling and learning effects,
based on Equation (29). With the total number of fuel sta-
tions (n) determined in our model, a capacity of each fuel
station C = 1,000 kg/day, and the exogenous parameters 𝛼,
𝛽, and 𝛾presented in Table 8, we derive the investment cost
per station for each hydrogen transportation state 𝑠(𝑠
vom Scheidt et al.: Preprint submitted to Elsevier Page 8 of 19
Integrating Hydrogen in Single-Price Electricity Systems
Table 8
Hydrogen station assumptions
GH2 LH2 LOHC
𝛼[−] 0.7 0.6 0.66
𝛽[−] 0.06 0.06 0.06
𝛾[−] 0.6 0.9 1.4
EC [𝑘𝑊 ℎ𝑒𝑙/𝑘𝑔𝐻2]1.6 0.6 4.4
NGC [𝑘𝑊 ℎ𝑁𝐺 /𝑘𝑔𝐻2]0 0 11.7
Depreciation years [𝑎]10 10 10
O&M [%] 5 5 5
{𝐺𝐻 2, 𝐿𝐻 2, 𝐿𝑂𝐻𝐶 }).
𝐼𝑆𝑠= 1.3 ∗ 600,000𝐸𝑈 𝑅 𝛾∗ ( 𝐶
212𝑘𝑔𝑑 𝑎𝑦 )𝛼
∗ (1 − 𝛽)log2(𝐶𝑛
212𝑘𝑔𝑑𝑎𝑦∗400 )
(29)
Next, we identify the number and locations of fueling
stations. Since passenger cars and trucks have different driv-
ing and refueling patterns, we separately select their fuel sta-
tion locations.
For passenger cars, we assume a utilization of 70% and
thus a turnover of 700 𝑘𝑔𝐻2per day, in line with Reuß et al.
(2019). This results in 412 fueling stations for cars. We
then first disaggregate the total demand to the >400 German
NUTS-2 regions proportionally to the NUTS-2 gross domes-
tic product (GDP). Since no more granular GDP data exists,
we further break down the hydrogen demand to the over
10,000 NUTS-3 regions in Germany proportionally to the
population in that NUTS-3 region. Currently, there are 72
hydrogen fueling stations (October 2019) in Germany (H2
MOBILITY,2019). Since these will not suffice to satisfy
demand in 2030, we assume that additional fueling stations
will be installed at the same locations as existing gasoline
stations. Therefore, we use the 11,285 gasoline stations from
OpenStreetMap as further potential sites (OpenStreetMap
Contributors,2020). For each of these stations, we calcu-
late the distance to the closest NUTS-3 region center. For
each NUTS-3 region, we then select stations with the short-
est distance to its center, until its demand is covered.
For trucks, Rose and Neumann (2020b) determine op-
timal hydrogen fuel station locations along highways under
consideration of traffic flow and capacity limits. From these
locations, we adopt those with highest utilization rate, which
leads to 97 stations. We assume all fuel stations have 1,000
kg/day capacity and have the same turnover. Thus, to meet
the demand from fuel cell heavy-duty trucks, the turnover of
each fuel station is 847.42 𝑘𝑔𝐻2per day.
Summary of hydrogen demand The total hydrogen net
demand in 2030 is estimated to be 51.26 TWh. Figure 2
displays the hydrogen net demands of the individual sectors.
The map in Figure 3shows the geographic distribution of the
hydrogen demand, with the size of the markers correspond-
ing to demand volume.
Steel
Ammonia
Methanol
Refinery
Trucks
Cars
0
5
10
15 13.25
17.49
11.73
4.29
1
3.5
Sectors
Hydrogen demand in TWh
Figure 2: Estimated hydrogen net demand per sector in Ger-
many, 2030
Figure 3: Spatial distribution of estimated hydrogen net de-
mand in Germany, 2030
4.1.2. Hydrogen production and import data
Electrolysis is the main pillar of political strategies for
hydrogen supply in Germany (Federal Government of Ger-
many,2020) and the European Union (European Commis-
sion,2020). Among the different electrolysis technologies,
proton exchange membrane (PEM) electrolysis is projected
to have the lowest CAPEX and highest efficiency in 2030
(Böhm et al.,2020). Therefore, we focus on PEM electroly-
vom Scheidt et al.: Preprint submitted to Elsevier Page 9 of 19
Integrating Hydrogen in Single-Price Electricity Systems
sis for hydrogen production. As input for the hydrogen sup-
ply chain model, we assume investment costs 𝐼𝐶𝐸 𝑙𝑒𝑐𝑡𝑟𝑜𝑙𝑦𝑧𝑒𝑟
of 604 EUR/𝑘𝑊𝑒𝑙, depreciation over 10 years, O&M costs
of 4% of investment costs and electricity consumption 𝐸𝐶
of 47.6 𝑘𝑊 ℎ𝑒𝑙 per 𝑘𝑔𝐻2, based on Schmidt et al. (2017);
Brown et al. (2018); Reuß et al. (2019). Electrolyzer effi-
ciency 𝐸𝐸 is set to 70% (Robinius et al.,2017;Reuß et al.,
2019). We set the minimum capacity 𝐶𝐴𝑃𝑃 𝑟𝑜𝑑 𝑢𝑐𝑡𝑖𝑜𝑛,𝑚𝑖𝑛 to
10 MW and the maximum capacity 𝐶𝐴𝑃𝑃 𝑟𝑜𝑑𝑢𝑐 𝑡𝑖𝑜𝑛,𝑚𝑎𝑥 to 100
MW. 3Regarding operation we analyze two different cases.
In the main case, all electrolyzers are assumed to operate
continuously at 70% of full capacity, which is within typical
ranges (Robinius et al.,2017;Guerra et al.,2019;Ruhnau,
2020)). In a sensitivity case, all electrolyzers are assumed to
have temporal flexibility which allows them to shift their op-
eration to hours with cheap wholesale electricity prices. In
this "FlexOp" case we assume they run at 100% during the
70% cheapest hours. Thus, in both cases, the total volume
of produced hydrogen is the same.
The potential locations for electrolyzers are equal to the
set of transmission grid nodes from our electricity system
model (compare Section 4.2.1).4Besides, we include hy-
drogen imports from overseas into our model, since they are
a key part of the German hydrogen strategy (Federal Gov-
ernment of Germany,2020). For these imports, we assume
a fixed, exogenous amount of daily available imported hy-
drogen of 27.40 GWh and costs of 3.48 EUR/𝑘𝑔𝐻2, in line
with the mean values reported by Runge et al. (2020). Fur-
thermore, we assume that all imports to Germany will occur
at one large port, i.e. Bremerhaven, in line with Runge et al.
(2020).
4.1.3. Hydrogen conversion data
Hydrogen can be converted to compressed state (GH2),
liquefied (LH2), or stored into chemicals (LOHC) for trans-
portation via tube trailers. Notably, for LH2 and LOHC,
there are capital and operating costs at the point of hydro-
gen production (for liquefaction, and hydrogenation, respec-
tively) and at the point of hydrogen consumption (evapora-
tion, and dehydrogenation, respectively).
The assumptions regarding investment costs, deprecia-
tion years, O&M costs, electricity and natural gas consump-
tion, and losses are displayed in Table 9.
4.1.4. Hydrogen transportation data
Transportation costs include costs for fuel, toll, and the
drivers’ wages. Fuel consumption of a delivery truck is as-
sumed to be 34.1 𝑙𝑑𝑖𝑒𝑠𝑒𝑙 /100km (Shell,2016). We assume a
diesel price in 2030 of 2.66 EUR/l, which includes the net
diesel price (1.73 EUR/l), mineral oil tax (0.47 EUR/l), oil
stockpiling fee (”Erdölbevorratungsbeitrag”) (0.30 EUR/l),
3This range was determined based on a review of 45 existing and cur-
rently planned projects in Germany.
4Such large-scale electrolyzers might be complemented by smaller, on-
site electrolyzers (see, e.g. Rose and Neumann (2020b); Golla et al. (2020))
in practice. Such on-site electrolyzers would be connected to the distribu-
tion grid. Analyzing congestion consequences at distribution grid level is
out of scope of this study.
and emission tax (0.16 EUR/l).
In line with Reuß (2019), we make the following cost as-
sumptions. Toll is set to 0.15 EUR/km. Drivers’ wage is set
to 35 EUR/h. Average driving speed is set to 50 km/h. Truck
investment costs are set to 160,000 EUR, with depreciation
over eight years and 12% O&M costs. For tube trailers, in-
vestment costs and capacities are technology specific. They
are set to 660,000 EUR and 1,100 𝑘𝑔𝐻2for gaseous hydrogen
(GH2), 860,000 EUR and 4,300 𝑘𝑔𝐻2for liquefied hydrogen
(LH2), and to 150,000 EUR and 1,620 𝑘𝑔𝐻2for LOHC. Be-
sides, we assume depreciation over twelve years and O&M
costs of 2%, adopted from Reuß et al. (2019).
4.2. Electricity system data
We parametrize both, the uniform price and the nodal
price electricity model with data for generation, consump-
tion and the transmission grid in 2030. For this, we utilize
the data set published by vom Scheidt et al. (2020). In the
following, we briefly describe this data set. All data are more
elaborately documented and available for free use under a
Creative Commons license in vom Scheidt et al. (2020).
4.2.1. Transmission grid data
The transmission grid in 2030 is constructed from the
reference ELMOD model of the existing grid, which is en-
hanced with all the expansions and new installations until
2030 that have been announced by the German Federal Net-
work Agency. The resulting final grid representation con-
sists of 485 nodes and 663 lines. The transmission capacity
of all 220 kV lines is set to 490 MW, and that of all 380 kV
lines to 1700 MW, based on Egerer (2016); Kießling et al.
(2011).
4.2.2. Electricity demand data
For consumption, the hourly consumption forecast sce-
nario EUCO30 is used (European Network of Transmission
System Operators for Electricity,2018). To improve con-
sistency of grid and consumption data, these hourly values
are re-scaled so that the annual total (577 TWh) matches the
sum used in the official grid development plan (544 TWh)
by Bundesnetzagentur (2019a).
Next, these re-scaled hourly demand values are spatially
disaggregated to NUTS-3 levels. For this disaggregation,
the gross domestic product (GDP) and the population of a
region serve as proxies for its future electricity consumption.
The resulting NUTS-3 consumption time series are assigned
to the nearest grid node.
4.2.3. Electricity generation data
For generation, estimation is differentiated between re-
newable, i.e. non-dispatchable generation, and dispatchable
generation.
For renewable generation, i.e. solar and wind, historical
hourly generation data from the four national grid operators
(Bundesnetzagentur,2018) is used. These hourly values are
vom Scheidt et al.: Preprint submitted to Elsevier Page 10 of 19
Integrating Hydrogen in Single-Price Electricity Systems
Table 9
Conversion assumptions, based on Reuß et al. (2019); Nexant et al. (2008). 𝑥denotes
daily hydrogen output.
Investment [EUR] Depreciation
years
O&M 𝐸𝐶𝐶 𝑜𝑛𝑣𝑒𝑟𝑠𝑖𝑜𝑛
[𝑘𝑊 ℎ𝑒𝑙𝑘𝑔𝐻2]
𝑁𝐺𝐶𝐶𝑜𝑛𝑣𝑒𝑟𝑠𝑖𝑜𝑛
[𝑘𝑊 ℎ𝑁𝐺 𝑘𝑔𝐻2]
Loss [%]
Compressor 15 ∗ 103𝐸𝑈𝑅
𝑘𝑊 𝑥0.6089 ∗ 3 15 4% calculated 0 0.5
Liquefaction 105 ∗ 106𝐸𝑈 𝑅 ∗ ( 𝑥
50
𝑡𝐻2
𝑑𝑎𝑦
)0.66 20 4% 6.78 0 1.65
Evaporation 3 ∗ 103𝐸𝑈 𝑅 𝑥
1,000 10 3% 0.6 0 0
Hydrogenation 40 ∗ 106𝐸𝑈 𝑅 ∗ ( 𝑥
300
𝑡𝐻2
𝑑𝑎𝑦
)0.66 20 3% 0.37 0 1
Dehydrogenation 30 ∗ 106𝐸𝑈 𝑅 ∗ ( 𝑥
300
𝑡𝐻2
𝑑𝑎𝑦
)0.66 20 3% 0.37 11.7 1
re-scaled so that the annual total generation from each tech-
nology matches the sum used in the grid development plan
(Bundesnetzagentur,2019a). This results in an annual gen-
eration of 86.7 TWh from solar (compared to an mean of
35.34 TWh in 2016-2018), and of 247.4 TWh from wind
(compared to an mean of 108.6 TWh in 2016-2018). Next,
the re-scaled hourly generation values are spatially disag-
gregated. For this, we use the installed generation capacity
per zip code as provided by Deutsche Übertragungsnetzbe-
treiber (2018).
For dispatchable electricity generation capacity, all rele-
vant plants for 2030 from the power plant list of the German
grid regulator are used (Bundesnetzagentur,2019b). For
each plant, marginal costs are calculated, based on fuel type,
estimated efficiency, and emissions costs. A 𝐶 𝑂2price of
60 EUR/ton is assumed (Bundesregierung,2019).
Both renewable generation time series and dispatchable
power plants, along with their marginal costs, are assigned to
the nearest grid node. Note that this approach provides high
spatial granularity, but comes at the costs of treating Ger-
many as an isolated system without cross-border electricity
lines. This can affect the results for electricity prices and re-
dispatch in both directions (Xiong et al.,2021). Therefore,
a geographic expansion – e.g. a European model – can be
worthwhile future work, but requires substantial additional
data procurement efforts if the high spatial granularity (i.e.
485 nodes in the German system) is to be upheld. A starting
point could be the open network model PyPSA-Eur-Sec-30
that works with one node per country (Victoria et al.,2019).
5. Results and Discussion
Upon parametrizing the models presented in Chapter 3
with the case study data presented in Chapter 4, we run the
models in three steps. First, we derive baseline results for
the electricity system without hydrogen, including whole-
sale uniform prices, nodal prices and congestion manage-
ment costs. Second, based on the resulting electricity prices,
we derive information about the optimal hydrogen supply
chains, including total end-use costs of hydrogen, as well
as number, capacities and locations of electrolyzers. Third,
we observe the effects of integrating these hydrogen supply
chains in the electricity system, including changes in total
electricity demand, wholesale prices, and redispatch costs.
5.1. Baseline electricity system results
Without the integration of hydrogen the resulting annual
mean of the wholesale uniform price is 62.61 EUR/MWh.
Figure 4shows the price duration curve. The annual re-
dispatch costs are 6.16 Billion EUR. The resulting annual
means of nodal prices vary between -54.30 and +221.00
EUR/MWh, with a median value of 67.80 EUR/MWh. Fig-
ure 5shows the spatial distribution of nodal prices.
0
10
20
30
40
50
60
70
80
90
100
110
120
130
140
150
1 1001 2001 3001 4001 5001 6001 7001 8001
Price
[EURO/MWh]
70% quantile
[Hour]
Figure 4: Wholesale price duration curve in Germany, 2030
[EUR/MWh]
Low prices are predominantly found in the North-East
and North-West of the country, driven by high renewable
generation and low demand. This finding is in line with
Robinius et al. (2017) who analyze residual loads on county
level and find negative residual loads predominantly in the
North-East and North-West.
5.2. Hydrogen supply chain results
The resulting end-use costs for hydrogen are represented
in Figure 6. In the uniform pricing scenario, the final hy-
drogen costs for industry applications are 5.98 𝐸𝑈 𝑅𝑘𝑔𝐻2
for GH2, 4.30 𝐸𝑈 𝑅𝑘𝑔𝐻2for LH2, and 4.68 𝐸 𝑈 𝑅𝑘𝑔𝐻2
for LOHC. For trucks and cars, costs for fueling stations
need to be added, resulting in final hydrogen costs of 6.14
𝐸𝑈 𝑅𝑘𝑔𝐻2for GH2, 4.41 𝐸 𝑈 𝑅𝑘𝑔𝐻2for LH2, and 5.89
vom Scheidt et al.: Preprint submitted to Elsevier Page 11 of 19
Integrating Hydrogen in Single-Price Electricity Systems
Figure 5: Shadow nodal prices in Germany, 2030 [EUR/MWh]
𝐸𝑈 𝑅𝑘𝑔𝐻2for LOHC. The largest share of costs are caused
by hydrogen production in all cases.
For the cheapest form, i.e. LH2, we additionally com-
pute the FlexOp scenario, in which electrolysis is shifted
to hours with cheaper prices. As Figure 4shows, this in-
cludes prices in the range of 0 - 70 EUR/MWh. For LH2
this shift decreases total hydrogen costs by 5.30%, i.e. from
4.30 𝐸𝑈 𝑅𝑘𝑔𝐻2to 4.07 𝐸 𝑈 𝑅𝑘𝑔𝐻2.
In the scenario with nodal pricing costs are much more
reduced, to 3.55 (3.71 for trucks and cars) EUR/𝑘𝑔𝐻2for
GH2, 2.73 (2.84) EUR/𝑘𝑔𝐻2for LH2, and 3.32 (4.53)
EUR/𝑘𝑔𝐻2for LOHC.
Figures 7and 8depict the resulting locations of elec-
trolyzers. The size of the markers corresponds to production
volume. The largest marker in the North-East depicts over-
seas imports, which are exogenously determined (compare
Chapter 4.1.2) and thus occur equally in all scenarios.
Under uniform pricing with LH2, 102 domestic elec-
trolyzers are placed, all of which close to points of consump-
tion, in order to minimize transportation costs. Of these,
roughly the half (52) have the maximal possible capacity of
100 MW.
Under nodal pricing, electrolyzers are placed further away
from consumption, but at nodes with the cheapest electricity
prices. In the LH2 nodal case, 71 electrolyzers are placed,
of which 66 have maximal possible capacity. This indicates
that the cheaper electricity costs outweigh the higher trans-
portation operating costs. This effect is stable across the
three delivery states, in line with the preliminary findings
by vom Scheidt et al. (2021).
5.3. Integration results
From the above presented locations and capacities of elec-
trolyzers, we calculate the additional electricity demand from
hydrogen production at each grid node. With this new input,
we recalculate electricity prices and congestion management
costs to identify the effects of hydrogen on the electricity sys-
tem. For these calculations we assume LH2 delivery, since it
is the cost-minimal hydrogen supply chain set-up under both
price scenarios and for both industry and transportation ap-
plications.
Table 10 summarizes the key results. The electrolytic
production of hydrogen creates considerable new electric-
ity demand of 58,92 TWh per year that increases the to-
tal national electricity demand by about 11%. This lead to
rising wholesale prices. Compared to the benchmark sce-
nario without hydrogen, mean annual wholesale prices rise
by 10.33% with continuously operating electrolyzers, and by
10.76% with flexibly operation electrolyzers.
Under the uniform price, annual congestion management
costs increase by 11%. Interestingly, this increase occurs
for static as well as for flexible electrolyzers that respond to
uniform wholesale prices. This finding indicates that elec-
trolyzers which respond to wholesale prices, but are ineffi-
ciently placed from a system perspective might not be able
to realize the often expected positive impacts (e.g. regard-
ing the usable share of renewable energy Ruhnau (2020))
due to grid constraints. A key explanatory factor for this
might be that wind generation, which to a large extend is lo-
cated in the North of Germany (Deutsche Übertragungsnet-
zbetreiber,2018) has been shown to drive wholesale prices
down (Benhmad and Percebois,2018) and at the same time
drives congestion in the transmission grid (Staudt et al.,2019).
Nevertheless, dedicated analyses are needed to validate this
finding, as the focus of our study is the spatial dimension.
Therefore, we consider a more comprehensive combination
of spatial and temporal dimension a promising direction for
future work on hydrogen integration.
In contrast, when electrolyzers are placed under nodal
price signals, they decrease congestion management costs
by -23.71% or 1,462 million Euros. This represents a delta
of over 2 billion Euros per year between the two scenarios
of hydrogen integration. In other words, the production of
one 𝑘𝑔𝐻2on average creates additional congestion costs of
0.44 Euros under current regulation, whereas it reduces con-
gestion costs by 0.95 Euros under more efficient regulation.
This means a spatially differentiated subsidy for hydrogen
production (e.g. in the form of a per-kWh payment of the
spread between uniform prices and simulated nodal prices)
could largely be covered by saved redispatch costs.
One limitation of the generalizability our findings comes
from the focus on one technology for hydrogen production,
i.e. electrolysis. Hydrogen from steam methane reforming
with carbon capture and storage represents an alternative of
producing hydrogen with net neutral emissions and can be
an economic alternative to electrolytic hydrogen, depending
vom Scheidt et al.: Preprint submitted to Elsevier Page 12 of 19
Integrating Hydrogen in Single-Price Electricity Systems
0
1
2
3
4
5
6
7
Production Conversion Transport Fueling Station
[EUR/kg]
Figure 6: End-use hydrogen costs by component and scenario
(a) Delivery with GH2 Trailers (b) Delivery with LH2 Trailers (c) Delivery with LOHC Trailers
Figure 7: Optimal electrolyzer locations under uniform electricity prices
on political and geographic circumstances (see e.g. Bødal
et al. (2020)). In the German case, however, political action
is strongly focused on electrolysis (Federal Government of
Germany,2020).
Another interesting avenue for the expansion of this work
is the effect of electrolysis on system emissions. While quan-
tifying this effect is out of this study’s scope, we expect that
the decrease in congestion enabled by electrolyzers placed
according to nodal prices will also lead to a decrease in emis-
sions, since a considerable, and rising share of curtailed en-
ergy comes from renewable sources (Xiong et al.,2021).
6. Conclusions and Policy Implications
Policymakers in dozens of countries are currently plan-
ning public funding for the development of future hydrogen
infrastructure. They can expect that the integration of hy-
drogen into electricity systems will have a large effect on the
operation of these systems. Our study sheds some light on
the effects of hydrogen integration, and the role of spatial
economic signals.
For this, we propose a three-step methodology based on
linking an electricity system dispatch model and a hydrogen
supply chain model, both with granular spatial resolution.
vom Scheidt et al.: Preprint submitted to Elsevier Page 13 of 19
Integrating Hydrogen in Single-Price Electricity Systems
(a) Delivery with GH2 Trailers (b) Delivery with LH2 Trailers (c) Delivery with LOHC Trailers
Figure 8: Optimal electrolyzer locations under nodal electricity prices
Table 10
Electricity demand, wholesale price and congestion management costs in 2030
Total electricity demand
[TWh/year]
Mean wholesale price
[EUR/MWh]
Congestion management
costs [MEUR/year]
Baseline without 𝐻2543.90 62.61 6,163.96
With 𝐻2under Uniform Price 602.82 (+10.83%) 69.07 (+10.33%) 6,839.06 (+10.95%)
With 𝐻2under Uniform Price (FlexOp) 602.82 (+10.83%) 69.34 (+10.76%) 6,852.58 (+11.17%)
With 𝐻2under Nodal Price 602.82 (+10.83%) 69.07 (+10.33%) 4,702.22 (-23.71%)
We apply this methodology for a case study of the German
system in 2030.
In the first step, we use the electricity system dispatch
model to simulate uniform electricity prices – representing
current regulation – and nodal prices, without considering
hydrogen demand and production.
In the second step we feed those prices into the hydrogen
model, together with additional techno-economic parame-
ters for capital and operation costs. This way, we determine
the optimal spatial design of hydrogen supply chains under
current uniform regulation, and a regulation with efficient
spatial price signals. We identify liquefied hydrogen as the
most economical form of truck based hydrogen delivery in
all scenarios. Furthermore, we find that under current uni-
form prices, electrolyzers are cost-minimally placed close to
consumption points, such as industry plants and large cities.
In the alternative nodal pricing scenario, we find that the
price differences among nodes are large enough to move hy-
drogen production to low-cost nodes that are further away
from consumption points and closer to low-cost electricity
generation capacity.
In the third step, we feed back the resulting electric loads
from electrolyzers into the electricity system dispatch model.
The results show that the integration of hydrogen under cur-
rent uniform prices causes a large increase in congestion man-
agement costs of about 11%, or 675 million Euros per year.
Thus, our analysis shows that the existing inefficiencies of
single-price zonal markets can be strongly aggravated by hy-
drogen. Given efficient spatial signals, electrolyzers are inte-
grated in a much more system-friendly way and actually de-
crease congestion management costs by 24%, or about 1,462
million Euros per year, compared to the benchmark scenario
without hydrogen. This is important information for pol-
icy makers in single-price electricity markets that intend to
subsidize hydrogen, as our results demonstrate the consid-
erable benefits of spatially differentiated subsidies. In fact,
the subsidies a regulator would have to pay to mimic nodal
prices for hydrogen within the existing single-price market
design could almost entirely be covered from avoided redis-
patch costs.
Given prevailing political barriers to introducing nodal
pricing markets in Europe (European Network of Transmis-
sion System Operators for Electricity,2021) it is important
to note that policy makers can still incorporate our findings
within the existing single-price markets. For instance, they
could offer a specific nodal tariff which bills electrolyzers
based on shadow nodal prices instead of wholesale prices.
Alternatively, if subsidies in the form of reduced per-kWh
vom Scheidt et al.: Preprint submitted to Elsevier Page 14 of 19
Integrating Hydrogen in Single-Price Electricity Systems
prices are planned (like in the German Hydrogen Strategy
(Federal Government of Germany,2020)), those can be lim-
ited to electrolyzers that are connected to nodes with low
shadow nodal prices. Another approach could be to allow
grid operators in case of congestion to curtail electrolyz-
ers before performing the regular redispatch measures, thus
creating an incentive for electrolyzer investors to avoid fre-
quently curtailed nodes of the grid. Given the large potential
benefits identified in this study, an interesting avenue for fu-
ture work certainly lies in investigating the solution space of
economically efficient, and politically feasible mechanisms
for the integration of hydrogen in single-price electricity sys-
tems.
CRediT authorship contribution statement
Frederik vom Scheidt: Conceptualization, Methodol-
ogy, Software, Formal analysis, Investigation, Data Cura-
tion, Writing - Original Draft, Writing - Review & Editing,
Visualization, Project administration. Jingyi Qu: Soft-
ware, Resources, Data Curation, Visualization, Writing - Re-
view & Editing. Philipp Staudt: Conceptualization, Writ-
ing - Review & Editing, Supervision, Funding acquisition.
Dharik S. Mallapragada: Conceptualization, Writing - Re-
view & Editing. Christof Weinhardt: Resources, Super-
vision, Funding acquisition.
7. Acknowledgements
We thank Julian Huber and Marc Schmidt for their ad-
vice on efficiently deploying the electricity system dispatch
model. We thank Jürgen Beck for assistance with the pro-
curement of hydrogen demand data.
References
Agora Energiewende, Wuppertal Institut, 2019. Klimaneutrale Industrie:
Schlüsseltechnologien und Politikoptionen für Stahl, Chemie und Ze-
ment. Berlin, Wuppertal: Agora Energiewende .
Antenne Brandenburg, 2020. Arcelormittal verpflichtet
sich zur klimaneutralen Stahlproduktion. URL:
https://www.rbb24.de/studiofrankfurt/beitraege/2020/07/
eisenhuettenstadt-arcelormittal- stahl-wasserstoff.html.
ArcelorMittal, 2017. Wasserstoff-Stahl: ArcelorMit-
tal und HAW Hamburg legen Studie vor. URL:
https://hamburg.arcelormittal.com/icc/arcelor-hamburg- de/
broker.jsp?uMen=9d0f6fbb-a799- 5199-f8b4- 947d7b2f25d3&
uCon=7c010c15-a102- cd51-db2a- 9147d7b2f25d&uTem=
aaaaaaaa-aaaa- aaaa-aaaa- 000000000011.
ArcelorMittal, 2020a. Hamburg: Wasserstoff-
Projekt mit konkreter Planung. URL: https://
hamburg.arcelormittal.com/icc/arcelor-hamburg- de/
broker.jsp?uMen=9d0f6fbb-a799- 5199-f8b4- 947d7b2f25d3&
uCon=efe407a1-7e2f- 4171-5933- 6eb4ba5c485d&uTem=
aaaaaaaa-aaaa- aaaa-aaaa- 000000000011.
ArcelorMittal, 2020b. Pressemitteilung: ArcelorMittal
Europe will noch 2020 den ersten grünen Stahl pro-
duzieren. URL: https://bremen.arcelormittal.com/icc/
arcelor-bremen- de/med/454/454205d9-81ba- 1571- 0202-ac33ef90c5de,
11111111-1111- 1111-1111- 111111111111.pdf.
BASF, 2020. Interview Methanpyrolyse. URL: https:
//www.basf.com/global/de/who-we- are/sustainability/
we-produce- safely-and- efficiently/energy- and-climate- protection/
carbon-management/interview- methane-pyrolysis.html.
Bazzanella, A., Ausfelder, F., 2017. Low carbon energy and feedstock for
the European chemical industry. DECHEMA, Gesellschaft für Chemis-
che Technik und Biotechnologie eV.
Benhmad, F., Percebois, J., 2018. Photovoltaic and wind power feed-in im-
pact on electricity prices: The case of germany. Energy Policy 119,
317–326. URL: https://www.sciencedirect.com/science/article/pii/
S0301421518302581, doi:https://doi.org/10.1016/j.enpol.2018.04.042.
Berger, C., 2020. 95 Prozent weniger CO2 bei der Stahlpro-
duktion. URL: https://www.springerprofessional.de/anlagenbau/
rohstoffe/95-prozent- weniger-co2- bei- der-stahlproduktion/18437776.
Bezirksregierung Köln, 2017. Genehmigungsbescheid: Wesentliche
Änderung der Anlage zur Herstellung von Ammoniak (Ammoniak-
Anlage) auf dem Werksgelände Köln der Firma INEOS Köln GmbH.
URL: https://www.bezreg-koeln.nrw.de/brk_internet/verfahren/52_53_
industrieanlagen_genehmigungen/bekanntmachungen_koeln/ineos_koeln_
gmbh_201612021/bescheid.pdf.
Böhm, H., Zauner, A., Rosenfeld, D.C., Tichler, R., 2020. Projecting
cost development for future large-scale power-to-gas implementations
by scaling effects. Applied Energy 264, 114780.
BP, 2019. Zahlen & Fakten: Produktion. URL: https://www.bp.com/de_
de/germany/home/wo-wir- sind/raffinerie-gelsenkirchen/wer- wir- sind/
zahlen-und- fakten.html.
Brown, T., Schlachtberger, D., Kies, A., Schramm, S., Greiner, M., 2018.
Synergies of sector coupling and transmission reinforcement in a cost-
optimised, highly renewable european energy system. Energy 160, 720–
739. doi:https://doi.org/10.1016/j.energy.2018.06.222.
Bundesnetzagentur, 2018. Marktdaten. Retrieved from
https://www.smard.de/home/downloadcenter/download_marktdaten/
726#!?downloadAttributes=%7B%22selectedCategory%22:1,
%22selectedSubCategory%22:8,%22selectedRegion%22:%22DE%22, %22from%
22:1546297200000,%22to%22:1577833199999, %22selectedFileType%22:
%22CSV%22%7D.
Bundesnetzagentur, 2019a. Bestätigung des Netzentwicklungs-
plans Strom für das Zieljahr 2030. Retrieved from https://www.
netzentwicklungsplan.de/sites/default/files/paragraphs-files/
NEP2019-2030_Bestaetigung.pdf.
Bundesnetzagentur, 2019b. Bnetza-power-plant-list. Re-
trieved from https://www.bundesnetzagentur.de/DE/
Sachgebiete/ElektrizitaetundGas/Unternehmen_Institutionen/
Versorgungssicherheit/Erzeugungskapazitaeten/Kraftwerksliste/
kraftwerksliste-node.html.
Bundesregierung, 2019. Klimaschutzprogramm 2030 der Bun-
desregierung zur Umsetzung des Klimaschutzplans 2050. Retrieved
from https://www.bundesregierung.de/resource/blob/975226/1679914/
e01d6bd855f09bf05cf7498e06d0a3ff/2019-10- 09-klima- massnahmen- data.
pdf?download=1.
Bødal, E.F., Mallapragada, D., Botterud, A., Korpås, M., 2020. De-
carbonization synergies from joint planning of electricity and hydro-
gen production: A texas case study. International Journal of Hydro-
gen Energy 45, 32899–32915. URL: https://www.sciencedirect.com/
science/article/pii/S0360319920335679, doi:https://doi.org/10.1016/
j.ijhydene.2020.09.127.
Caglayan, D.G., Weber, N., Heinrichs, H.U., Linßen, J., Robinius, M.,
Kukla, P.A., Stolten, D., 2020. Technical potential of salt caverns for
hydrogen storage in europe. International Journal of Hydrogen Energy
45, 6793–6805. URL: https://www.sciencedirect.com/science/article/
pii/S0360319919347299, doi:https://doi.org/10.1016/j.ijhydene.2019.
12.161.
Deutsche Übertragungsnetzbetreiber, 2018. EEG-Anlagenstammdaten. Re-
trieved from https://www.netztransparenz.de/EEG/Anlagenstammdaten.
Dillinger, 2016. Produktion. URL: https://www.dillinger.de/d/de/
corporate/dillinger/produktion/.
Dillinger, 2019. Nachhaltige Stahlproduktion an der
Saar. URL: https://www.dillinger.de/d/de/aktuelles/news/
vom Scheidt et al.: Preprint submitted to Elsevier Page 15 of 19
Integrating Hydrogen in Single-Price Electricity Systems
nachhaltige-stahlproduktion- an-der- saar- 88575.shtml.
Egerer, J., 2016. Open source Electricity Model for Germany (ELMOD-
DE). DIW Data Documentation 83. German Institute for Economic Re-
search (DIW). Berlin. URL: http://hdl.handle.net/10419/129782.
Emonts, B., Reuß, M., Stenzel, P., Welder, L., Knicker, F., Grube,
T., Görner, K., Robinius, M., Stolten, D., 2019. Flexible sec-
tor coupling with hydrogen: A climate-friendly fuel supply for
road transport. International Journal of Hydrogen Energy 44,
12918 – 12930. URL: http://www.sciencedirect.com/science/article/
pii/S0360319919312121, doi:https://doi.org/10.1016/j.ijhydene.2019.
03.183.
ENCON.Europe GmbH, 2018. Potentialatlas für Wasserstoff: Analyse
des Marktpotentials für Wasserstoff, der mit erneuerbaren Strom
hergestellt wird, im Raffineriesektor und im zukünftigen Mobilitätssek-
tor. URL: https://www.dwv-info.de/wp- content/uploads/2018/04/
Potentialstudie-f%C3%BCr- gr%C3%BCnen-Wasserstoff- in- Raffinerien.
pdf.
European Alternative Fuels Observatory, . European classification for ve-
hicle category. Retrieved from https://www.eafo.eu/knowledge-center/
european-vehicle- categories. Accessed 25.03.2020.
European Commission, 2020. A hydrogen strategy for a climate-neutral
europe. Retrieved from https://ec.europa.eu/energy/sites/ener/files/
hydrogen_strategy.pdf. Accessed 09.07.2020.
European Network of Transmission System Operators for Electricity,
2018. Transparency Platform. Retrieved from https://tyndp.entsoe.
eu/maps-data.
European Network of Transmission System Operators for Electricity,
2021. Options for the design of European Electricity Markets in 2030.
Retrieved from https://eepublicdownloads.entsoe.eu/clean-documents/
Publications/Market%20Committee%20publications/210331_Market_
design%202030.pdf.
European parliament and European council, 2003. Regulation (ec) no
1059/2003.
FCH-JU, 2017. Development of business cases for fuel cells and
hydrogen applications for regions and cities. Retrieved from
https://www.fch.europa.eu/sites/default/files/171121_FCH2JU_
Application-Package_WG1_Heavy%20duty%20trucks%20%28ID%202910560%
29%20%28ID%202911646%29.pdf. Accessed 28.03.2020.
Federal Government of Germany, 2020. Die Nationale Wasserstoff-
strategie. Retrieved from https://www.bmwi.de/Redaktion/DE/
Publikationen/Energie/die-nationale- wasserstoffstrategie.pdf?__
blob=publicationFile&v=12. Accessed 10.06.2020.
Fleiter, T., Schlomann, B., Eichhammer, W., 2013. Energie-
verbrauch und CO2-Emissionen industrieller Prozesstechnologien:
Einsparpotenziale, Hemmnisse und Instrumente. Fraunhofer-Verlag.
URL: https://www.isi.fraunhofer.de/content/dam/isi/dokumente/ccx/
2013/Umweltforschungsplan_FKZ-370946130.pdf.
Fraunhofer-Institut, 2019. Eine Wasserstoff-Roadmap für Deutsch-
land. Retrieved from https://www.ise.fraunhofer.de/content/
dam/ise/de/documents/publications/studies/2019-10_Fraunhofer_
Wasserstoff-Roadmap_fuer_Deutschland.pdf. Accessed 28.03.2020.
Fröhlich, T., Blömer, S., Münter, D., Brischke, L.A., 2019. CO2-Quellen
für die PtX-Herstellung in Deutschland: Technologien, Umweltwirkung,
Verfügbarkeit. URL: https://www.ifeu.de/wp- content/uploads/ifeu_
paper_03_2019_CO2-Quellen- f%C3%BCr-PtX.pdf.
Golla, A., vom Scheidt, F., Röhrig, N., Staudt, P., Weinhardt, C., 2020.
Vehicle scheduling and refuelling of hydrogen buses with on-site elec-
trolysis. Jahrestagung der Gesellschaft für Informatik 2020 .
Grube, T., Stolten, D., 2018. The impact of drive cycles and auxiliary power
on passenger car fuel economy. Energies 11. URL: https://www.mdpi.
com/1996-1073/11/4/1010, doi:10.3390/en11041010.
Guerra, O.J., Eichman, J., Kurtz, J., Hodge, B.M., 2019. Cost competitive-
ness of electrolytic hydrogen. Joule 3, 2425 – 2443. URL: http://www.
sciencedirect.com/science/article/pii/S2542435119303228, doi:https:
//doi.org/10.1016/j.joule.2019.07.006.
H2 MOBILITY, 2019. Tankstellen. Retrieved from https://h2.live/en/
tankstellen.
Hebling, C., Ragwitz, M., Fleiter, T., Groos, U., Härle, D., Held, A., Jahn,
M., Müller, N., Pfeifer, T., Plötz, P., et al., 2019. Eine Wasserstoff-
Roadmap für Deutschland. Fraunhofer-Institut für System-und Innova-
tionsforschung ISI, Karlsruhe .
Hermann, H., Emele, L., Loreck, C., 2014. Prüfung der klimapolitis-
chen Konsistenz und der Kosten vonMethanisier ungsstrategien. Techni-
cal Report. Öko-Institut e.V. URL: https://www.oeko.de/oekodoc/2005/
2014-021- de.pdf.
HKM, 2020. Die Gesellschafter der HKM. URL: https://www.hkm.de/
unternehmen/gesellschafter/.
Hofbauer, H., Kaltschmitt, M., Keil, F., Neuling, U., Wagner, H., 2016.
Vergasung in der Gasatmosphäre, in: Kaltschmitt, M., Hartmann, H.,
Hofbauer, H. (Eds.), Energie aus Biomasse: Gr undlagen, Techniken und
Verfahren. Springer, Berlin, Heidelberg, pp. 1059–1182. doi:10.1007/
978-3- 662-47438- 9_13.
Hölling, M., Weng, M., Gellert, S., 2017. Bewertung der Herstellung von
Eisenschwamm unter Verwendung von Wasserstoff. Stahl Und Eisen
137, 47–53.
Hydrogen Council, McKinsey & Company, 2021. Hydrogen Insights. Re-
trieved from https://hydrogencouncil.com/wp-content/uploads/2021/02/
Hydrogen-Insights- 2021-Report.pdf. Accessed 15.04.2021.
Hyundai, 2020. Erste Brennstoffzellen-Lkw Hyundai Xcient Fuel Cell
kommen nach Europa. URL: https://www.hyundai.news/de/unternehmen/
erste-brennstoffzellen- lkw-hyundai- xcient- fuel-cell- kommen-nach- europa/.
IEA, 2015. Large-scale hydrogen delivery infrastructure. Retrieved from
http://ieahydrogen.org/Activities/Task-28/Task- 28-report_final_v2_
ECN_12_2_v3.aspx. Accessed 28.03.2020.
IEA, 2019. The Future of Hydrogen: Seizing today’s opportuni-
ties. URL: https://webstore.iea.org/download/direct/2803?fileName=
The_Future_of_Hydrogen.pdf.
IKTS, 2020. CO2-Emissionen bei der Stahlproduk-
tion: Von 100 auf 5 Prozent! URL: https://
www.ikts.fraunhofer.de/de/presse/pressemitteilungen/
29-9- 2020-co2- emissionen- bei-der- stahlproduktion-- von- 100-auf- 5-.
html.
IVA, 2018. Wichtige Zahlen Düngemittel: Produktion, Markt
und Landwirtschaft: Hersteller von Ammoniak, Stickstoff- und
Phosphat-Düngemitteln (auf Rohphosphatbasis). URL: https:
//www.iva.de/sites/default/files/benutzer/%25uid/publikationen/
wichtige_zahlen_2017-2018.pdf.
Jendrischik, M., 2020. Total setzt in Leuna auf synthetis-
ches Methanol. URL: https://www.cleanthinking.de/
synthetisches-methanol- total-sunfire- wasserstoff/.
Kießling, F., Nefzger, P., Kaintzyk, U., 2011. Freileitungen: Planung,
Berechnung, Ausführung. Springer-Verlag.
Michalski, M., Altmann, M., Bünger, U., Weindorf, W., 2019. Wasser-
stoffstudie Nordrhein-Westfalen. Eine Expertiese für das Ministerium
für Wirtschaft, Innovation, Digitalisierung und Energie des Landes
Nordrhein-Westfalen, Düsseldorf .
MWV, 2020. Jahresbericht 2020. URL: https://www.mwv.de/
wp-content/uploads/2020/09/MWV_Mineraloelwirtschaftsverband- e.
V.-Jahresbericht- 2020-Webversion.pdf.
Nexant, I., Liquide, A., Laboratory, A.N., Venture, C.T., Institute, G.T.,
Laboratory, N.R.E., Laboratory, P.N.N., LLC, T., 2008. Wasser-
stoff: Bundesregierung schlägt Kriterien zur Umlagebefreiung
vor. Retrieved from https://www.energate-messenger.de/news/210637/
wasserstoff-bundesregierung- schlaegt-kriterien- zur- umlagebefreiung-vor.
Accessed 15.04.2021.
OpenStreetMap Contributors, 2020. Openstreetmap data for germany. Re-
trieved from https://download.geofabrik.de/europe/germany.html. Ac-
cessed 28.03.2020.
Peters, K., Thumann, R., 2016. Yara: Aus Dithmarschen-Wiki. URL:
https://www.dithmarschen-wiki.de/Yara.
Prognos AG, 2020a. Energiewirtschaftliche Projektionen und Folgeab-
schätzungen 2030/2050. URL: https://www.bmwi.de/Redaktion/DE/
Publikationen/Wirtschaft/klimagutachten.pdf?__blob=publicationFile&
v=8.
Prognos AG, 2020b. Kosten und Transformationsp-
fade für strombasierte Energieträger: Studie im Auf-
vom Scheidt et al.: Preprint submitted to Elsevier Page 16 of 19
Integrating Hydrogen in Single-Price Electricity Systems
trag des Bundesministeriums für Wirtschaft und Energie.
URL: https://www.bmwi.de/Redaktion/DE/Downloads/Studien/
transformationspfade-fuer- strombasierte-energietraeger.pdf?__
blob=publicationFile.
Rechenberger, D., 2020. BASF SE Produktion Ammoniak: E-Mail.
Redenius, A., 2020a. SALCOS, WindH2, GrInHy – Wasserstoffpro-
jekte bei der Salzgitter AG. URL: https://www.prozesswaerme.net/
fileadmin/Prozesswaerme/Dateien_Redaktion/Ausgewaehlte_Beitraege/
13_sonderteil_salzgitter.pdf.
Redenius, A., 2020b. Vortrag auf der WerkstoffPlusAuto 2020: SALCOS
- nachhaltige, flexible, CO2-arme Stahlproduktion.
Reuß, M., 2019. Techno-ökonomische Analyse alternativer Wasserstoffin-
frastruktur. Ph.D. thesis. RWTH Aachen. doi:10.18154/RWTH-2019- 07432.
Reuß, M., Grube, T., Robinius, M., Stolten, D., 2019. A hydrogen sup-
ply chain with spatial resolution: Comparative analysis of infrastructure
technologies in germany. Applied Energy 247, 438–453. doi:https:
//doi.org/10.1016/j.apenergy.2019.04.064.
Robinius, M., Otto, A., Syranidis, K., Ryberg, D.S., Heuser, P., Welder,
L., Grube, T., Markewitz, P., Tietze, V., Stolten, D., 2017. Linking the
power and transport sectors—part 2: Modelling a sector coupling sce-
nario for germany. Energies 10. URL: https://www.mdpi.com/1996- 1073/
10/7/957, doi:10.3390/en10070957.
ROGESA, 2016. Daten und Fakten. URL: http://www.rogesa.de/rogesa/
produktion/daten/index.shtml.de.
Rose, P.K., Neumann, F., 2020a. Hydrogen refueling station networks for
heavy-duty vehicles in future power systems. Transportation Research
Part D: Transport and Environment 83, 102358. URL: http://www.
sciencedirect.com/science/article/pii/S1361920920305459, doi:https:
//doi.org/10.1016/j.trd.2020.102358.
Rose, P.K., Neumann, F., 2020b. Hydrogen refueling station networks for
heavy-duty vehicles in future power systems. Transportation Research
Part D: Transport and Environment 83. doi:https://doi.org/10.1016/j.
trd.2020.102358.
Ruhnau, O., 2020. Market-based renewables: How flexible hydrogen elec-
trolyzers stabilize wind and solar market values. Technical Report. Her-
tie School. Kiel, Hamburg. URL: http://hdl.handle.net/10419/227075.
Runge, P., Sölch, C., Albert, J., Wasserscheid, P., Zöttl, G., Grimm, V.,
2019. Economic comparison of different electric fuels for energy sce-
narios in 2035. Applied Energy 233–234, 1078–1093. doi:https:
//doi.org/10.1016/j.apenergy.2018.10.023.
Runge, P., Sölch, C., Albert, J., Wasserscheid, P., Zöttl, G., Grimm, V.,
2020. Economic comparison of electric fuels produced at excellent lo-
cations for renewable energies: A scenario for 2035. Working paper
doi:http://dx.doi.org/10.2139/ssrn.3623514.
vom Scheidt, F., Müller, C., Staudt, P., Weinhardt, C., 2020. The German
electricity system in 2030: data on consumption, generation, and the
grid. doi:10.5445/IR/1000124167.
Schmidt, O., Gambhir, A., Staffell, I., Hawkes, A., Nelson, J., Few, S., 2017.
Future cost and performance of water electrolysis: An expert elicita-
tion study. International Journal of Hydrogen Energy 42, 30470–30492.
doi:https://doi.org/10.1016/j.ijhydene.2017.10.045.
Schweer, D., Scholz, G., Heisel, M., 2002. On-site-Versorgung von Erdöl-
raffinerien mit technischen Gasen. Erdöl, Erdgas, Kohle 118, 115–120.
Shell, 2016. Shell nutzfahrzeug-studie diesel oder alternative antriebe
– womit fahren lkw und bus morgen? Retrieved from https://
www.shell.de/promos/media/shell-goods- vehicle-study/_jcr_content.
stream/1466682556570/006b9c62dcca41b86d0adafc3ee2ad4fa14ef4d3/
shell-nutzfahrzeugstudie.pdf. Accessed 28.03.2020.
skw Piesteritz, 2015. Sicherheitsdatenblatt: Ammoniak. URL: https://
www.skwp.de/media-center/broschueren/reach/.
Stagge, M., 2020. Wasserstoffbedarf thyssenkrupp 2030: E-Mail.
Staudt, P., 2019. Transmission Congestion Management in Electric-
ity Grids-Designing Markets and Mechanisms. Ph.D. thesis. KIT-
Bibliothek.
Staudt, P., Oren, S., 2020. A merchant transmission approach for uniform-
price electricity markets, in: Proceedings of the 53rd Hawaii Interna-
tional Conference on System Sciences, pp. 1–10.
Staudt, P., Rausch, B., Gärttner, J., Weinhardt, C., 2019. Predicting trans-
mission line congestion in energy systems with a high share of renew-
ables, in: 2019 IEEE Milan PowerTech, pp. 1–6. doi:10.1109/PTC.2019.
8810527.
Staudt, P., Wegner, F., Garttner, J., Weinhardt, C., 2017. Analysis of re-
dispatch and transmission capacity pricing on a local electricity market
setup, in: 2017 14th International Conference on the European Energy
Market (EEM), pp. 1–6.
swb, 2020. Nachhaltigkeit: Wasserstoff: Mit grünem Stahl Emis-
sionen reduzieren. URL: https://www.swb.de/ueber- swb/unternehmen/
nachhaltigkeit/wasserstoff/elektrolyseur.
Thoma, H.J., 2020. Methanol Produktionskapazität: E-Mail.
thyssenkrupp, 2019. Wasserstoff statt Kohle. thyssenkrupp Steel startet
wegweisendes Projekt für eine klimafreundliche Stahlproduktion am
Standort Duisburg. URL: https://www.thyssenkrupp-steel.com/de/
newsroom/pressemitteilungen/wasserstoff-statt- kohle.html.
thyssenkrupp, 2020a. Die Klimastrategie von thyssenkrupp Steel zur nach-
haltigen Stahlproduktion. URL: https://www.thyssenkrupp-steel.com/
de/unternehmen/nachhaltigkeit/klimastrategie/.
thyssenkrupp, 2020b. Grüner Wasserstoff für die Stahlproduktion:
RWE und thyssenkrupp planen Zusammenarbeit. URL: https:
//www.thyssenkrupp-steel.com/media/content_1/presse/dokumente/
2020_1/juni_2/20200610_pm_thyssenkrupp_steel_loi_rwe_pre-final.pdf.
VCI, 2020. Chemiewirtschaft in Zahlen 2020. URL: https://www.vci.
de/vci/downloads-vci/publikation/chemiewirtschaft- in-zahlen- print.
pdf.
Victoria, M., Zhu, K., Brown, T., Andresen, G.B., Greiner, M., 2019. The
role of storage technologies throughout the decarbonisation of the sector-
coupled european energy system. Energy Conversion and Management
201, 111977. URL: https://www.sciencedirect.com/science/article/
pii/S0196890419309835, doi:https://doi.org/10.1016/j.enconman.2019.
111977.
vom Scheidt, F., Qu, J., Staudt, P., Mallapragada, D., Weinhardt, C., 2021.
The effects of electricity tariffs on cost-minimal hydrogen supply chains
and their impact on electricity prices and redispatch costs. Proceedings
of the 54th Hawaii International Conference on System Sciences .
Warscheid, L., 2020a. Fahrplan der Saar-Stahlkocher zum
grünen Stahl. Saarbruecker Zeitung 2020. URL: https:
//www.saarbruecker-zeitung.de/saarland/saar- wirtschaft/
stufenplan-zu- gruenem-stahl- aus- dem-saarland- bis-2050_
aid-53012835.
Warscheid, L., 2020b. In Dillingen wird jetzt mit Wasser-
stoff gekocht. Saarbruecker Zeitung 2020. URL: https:
//www.saarbruecker-zeitung.de/saarland/saar- wirtschaft/
neue-anlage- von-rogesa- saarstahl- dillinger-nutzt- wasserstoff_
aid-52906817?utm_source=mail&utm_medium=referral&utm_campaign=
share#successLogin.
Weger, L.B., Leitão, J., Lawrence, M.G., 2020. Expected impacts on
greenhouse gas and air pollutant emissions due to a possible transi-
tion towards a hydrogen economy in german road transport. Interna-
tional Journal of Hydrogen Energy URL: http://www.sciencedirect.com/
science/article/pii/S0360319920341847, doi:https://doi.org/10.1016/
j.ijhydene.2020.11.014.
Wilms, S., Schäfer-Stradowsky, S., Jahnke, P., 2018. Heutige Ein-
satzgebiete für Power Fuels—Factsheets zur Anwendung von
klimafreundlich erzeugten synthetischen Energieträgern. URL:
https://www.dena.de/fileadmin/dena/Publikationen/PDFs/2019/181123_
dena_PtX-Factsheets.pdf.
WV Stahl, 2020. Statistisches Jahrbuch der Stahlindustrie 2019/2020.
Xiong, B., Predel, J., Crespo del Granado, P., Egging-Bratseth, R., 2021.
Spatial flexibility in redispatch: Supporting low carbon energy systems
with power-to-gas. Applied Energy 283, 116201. URL: https://www.
sciencedirect.com/science/article/pii/S0306261920315981, doi:https:
//doi.org/10.1016/j.apenergy.2020.116201.
Yilmaz, C., 2018. Maßnahmen zur Dekarbonisierung des Hochofen-
prozesses durch Einsatz von Wasserstoff. Cuvillier Verlag.
Zhang, C., Greenblatt, J.B., Wei, M., Eichman, J., Saxena, S., Muratori, M.,
Guerra, O.J., 2020. Flexible grid-based electrolysis hydrogen produc-
tion for fuel cell vehicles reduces costs and greenhouse gas emissions.
vom Scheidt et al.: Preprint submitted to Elsevier Page 17 of 19
Integrating Hydrogen in Single-Price Electricity Systems
Applied Energy 278, 115651. URL: https://www.sciencedirect.com/
science/article/pii/S0306261920311491, doi:https://doi.org/10.1016/
j.apenergy.2020.115651.
A. Appendix A: Hydrogen demand data
Steel To identify all steel plants with potential for hydrogen
use in 2030, we use the statistical report of the steel indus-
try (WV Stahl,2020). Looking at future hydrogen demand,
only those 70 % of steel producers who manufacture via the
blast furnace route are relevant, as large quantities of CO2
are emitted here and can be avoided by switching to the di-
rect reduction route. In addition, the ArcelorMittal plant in
Hamburg is included, as it already uses a direct reduction
approach (Hölling et al.,2017). Table 4lists the eight iden-
tified steel production sites.
The production volumes and relative shares of primary
and secondary steel in Germany have been approximately
constant since 2012 (WV Stahl,2020). Therefore, and in
line with Hebling et al. (2019), we use past production vol-
ume and distribution as 2030 estimates. In particular, we use
2017 values, as only those are available in (WV Stahl,2020).
Table 4shows the crude steel quantities produced in 2017 for
each identified site with potential hydrogen demand.
However, it can be assumed that not all steel produc-
ers will switch to direct reduction by 2030, due to various
reasons. For instance, the switch is associated with high
investment costs, is technically demanding (IKTS,2020),
and comes with new uncertainties like future hydrogen costs
(Agora Energiewende and Wuppertal Institut,2019). Corre-
spondingly, steel producers are planning individual solutions
for medium-term CO2emission reduction to achieve reduc-
tion goals. Therefore, all relevant plants must be analyzed
individually.
ArcelorMittal Hamburg has been operating a direct re-
duction plant since the mid-1970s (Hölling et al.,2017). The
reduction gas used today consists of about 60 % hydrogen
(ArcelorMittal,2017). By 2030, steel production is planned
to be completely CO2-neutral (ArcelorMittal,2020a). Ac-
cordingly, we assume that there will be a complete switch
to the direct reduction route with 100 % hydrogen input by
2030. For the direct reduction route, we assume the specific
hydrogen demand factor 80𝑘𝑔𝐻2𝑡𝑠𝑡𝑒𝑒𝑙, based on Michalski
et al. (2019). The hydrogen demand of ArcelorMittal Ham-
burg for the year 2030 is estimated with equation 30.
𝐻𝐷 =𝑂𝑢𝑡𝑝𝑢𝑡𝑡𝑆𝑡𝑒𝑒𝑙 𝑠𝑝𝑒𝑐 𝑖𝑓 𝑖𝑐𝐷𝑒𝑚𝑎𝑛𝑑 𝐹 𝑎𝑐 𝑡𝑜𝑟
∗ 33.33𝑘𝑊 ℎ𝐻2𝑘𝑔𝐻2
(30)
ArcelorMittal Eisenhüttenstadt and ArcelorMittal Duis-
burg have not publicly announced any plans to use hydrogen
until 2030, but it has been indicated that long-term adoption
of hydrogen for the former plant will depend on the results
of current pilot projects of the ArcelorMittal group (Antenne
Brandenburg,2020;ArcelorMittal,2020b). Therefore, we
assume that these plants do not have any hydrogen demand
in 2030.
ArcelorMittal Bremen is focusing on the use of hydro-
gen via the blast furnace route to achieve the medium-term
goals. However, the company already plans to construct an
electrolyser on-site (swb,2020) that will be sufficient to fully
meet the hydrogen demand in 2030. Thus, the plant does not
have any net demand for hydrogen.
ROGESA, a subsidiary of Dillinger and Saarstahl, pro-
duces pig iron, which is supplied to Dillinger and Saarstahl
for the subsequent crude steel production (Dillinger,2016).
Therefore, Dillinger and Saarstahl are considered collectively
for further calculations. ROGESA operates two blast fur-
naces and plans to optimise both by blowing in hydrogen as
a reducing agent in order to achieve a reduction in CO2emis-
sions (Dillinger,2019). According to a step-by-step plan of
the Saarland-based steel industry, both blast furnaces are to
remain in operation until 2031 (Warscheid,2020a). In addi-
tion, an electric furnace and a direct reduction plant are to be
built, which will initially only use natural gas to produce di-
rectly reduced iron from iron ore (Warscheid,2020a). There-
fore, we assume that by 2030, both blast furnaces will use
the maximum amount of hydrogen. Both blast furnaces are
technically able to use a maximum of approximately 3,700
𝑘𝑔𝐻2(Warscheid,2020b;ROGESA,2016). Thus, the hy-
drogen demand of ROGESA (Dillinger and Saarstahl) for the
year 2030 is estimated to be 2.1606 𝑇 𝑊 ℎ𝐻2, based on equa-
tion 31.5
𝐻𝐷 = 2 ∗ 3700𝑘𝑔𝐻2∗ 8760
∗ 33.33𝑘𝑊 ℎ𝐻2𝑘𝑔𝐻2
(31)
Hüttenwerke Krupp Mannesmann (HKM) is owned 50
% by Thyssenkrupp Steel Europe AG, 30 % by Salzgitter
Mannesmann GmbH and 20 % by the French company Val-
lourec Tubes S.A.S (HKM,2020). Regarding the use of hy-
drogen in production, no press reports were found that were
published by HKM. Consequently, it is assumed that due to
the structure of the company, no hydrogen will be used un-
til 2030, as the shareholders might primarily concentrate on
their own production facilities and their optimisation.
Salzgitter is pursuing a gradual conversion to hydrogen-
based steel production via the direct reduction/electric arc
furnace route. In the first stage of expansion, a direct re-
duction plant and an electric arc furnace will be built (Rede-
nius,2020a). This expansion stage will lead to a hydrogen
use of 81,332 Nm3/h and a specific hydrogen demand factor
of 12.27 𝑘𝑔𝐻2𝑡𝑠𝑡𝑒𝑒𝑙 (Redenius,2020b) for the overall plant
output. Thus, the hydrogen demand for 2030 can be calcu-
lated with equation 30.
Thyssenkrupp plans to replace two blast furnaces with
two direct reduction plants, and to optimize one blast furnace
by blowing in hydrogen until 2030 (thyssenkrupp,2020a).
Current estimations indicate that around 200,000 tons of hy-
drogen per year will be needed from 2030 (Stagge,2020). A
share of this will be supplied through a long-term contract
5To validate the results, we also estimate the demand with equation 30,
which returns 2.0668 𝑇 𝑊 ℎ𝐻2and thus confirms the calculations. For all
further calculations, we use 2.1606 𝑇 𝑊 𝐻2as demand for the Dillinger
and Saarstahl steel plants.
vom Scheidt et al.: Preprint submitted to Elsevier Page 18 of 19
Integrating Hydrogen in Single-Price Electricity Systems
with RWE, from a 100 MW electrolyzer capable of supply-
ing 1.7 tons of hydrogen per hour (thyssenkrupp,2020b).
This supply is deducted from the total demand in order to cal-
culate the hydrogen net demand for 2030 as shown in equa-
tion 32.
𝐻𝐷 = (200,000,000𝑘𝑔𝐻2− 1,700𝑘𝑔𝐻2
∗ 8,760) ∗ 33.33𝑘𝑊 ℎ𝐻2𝑘𝑔𝐻2
(32)
Ammonia The hydrogen demand from the German ammo-
nia industry can be estimated as follows. The ideal specific
hydrogen demand for ammonia synthesis is 3 moles of H2
for 2 moles of NH3(Hermann et al.,2014), or 177.55 kgH2
per ton of ammonia.
We acquire a list of all ammonia producers in Germany
from the Industrial Association Agrar (IVA,2018). The pro-
duction volumes of ammonia in Germany have been approx-
imately constant since 2012 (VCI,2020). While, to the best
of our knowledge, no information on site-specific current
ammonia production is publicly available, we identify site-
specific production capacities based on skw Piesteritz (2015);
Peters and Thumann (2016); Bezirksregierung Köln (2017);
Rechenberger (2020). The sum of these capacities (2,955,000
t/a) is somewhat higher than the current total ammonia pro-
duction (2,415,327 in 2019). However, global ammonia de-
mand is assumed to increase by 2030 (Hebling et al.,2019;
IEA,2019). Therefore, in the following, the production ca-
pacities are assumed as basis for the site-specific hydrogen
demand estimation.
Regarding self-supply, no information on large-scale elec-
trolysers at the identified ammonia plants was found. While
BASF (2020) is building its own electrolysis plant for re-
search purposes, no details are available about any large scale
operation. Correspondingly, the total demand is assumed
to be equal to the net demand. With the assumptions made
above, the site-specific demand can be estimated with equa-
tion 33.
𝐻𝐷 =𝑡𝐴𝑚𝑚𝑜𝑛𝑖𝑎 ∗ 177.55𝑘𝑔𝑡𝐴𝑚𝑚𝑜𝑛𝑖𝑎
∗ 33.33𝑘𝑊 ℎ𝐻2𝑘𝑔𝐻2
(33)
Methanol The specific hydrogen demand is estimated as 2
moles of H2for 1 mole of CH3OH (Hofbauer et al.,2016), or
188.73 kgH2 per ton of methanol. This is consistent with the
assumptions of Bazzanella and Ausfelder (2017) and Michal-
ski et al. (2019). Currently, there are five relevant methanol
plants in Germany (Fröhlich et al.,2019). However, one
of them has terminated production and is being liquidated
(Thoma,2020), and therefore is disregarded for 2030.
In the next step, production capacities of the individual
plants are identified (Fleiter et al.,2013;Jendrischik,2020;
BP,2019). The sum of current production of 1,398,146 t/a
(VCI,2020) is lower than the total production capacity of
1,865,000 t/a. However, production has been rising in recent
years, and global methanol demand is assumed to increase
by 2030 (Hebling et al.,2019;IEA,2019). Correspondingly,
as with the hydrogen estimate for ammonia, the production
capacity is used as basis for further calculations.
Regarding self-supply, there are smaller electrolyzers for
research purposes (Jendrischik,2020;BASF,2020), but no
information on large-scale electrolysers at the identified am-
monia plants was found. Correspondingly, the total demand
is assumed to be equal to the net demand. With the assump-
tions made above, the site-specific demand can be estimated
with equation 34.
𝐻𝐷 =𝑡𝑀𝑒𝑡ℎ𝑎𝑛𝑜𝑙 ∗ 188.73𝑘𝑔 𝑡𝑀 𝑒𝑡ℎ𝑎𝑛𝑜𝑙
∗ 33.33𝑘𝑊 ℎ𝐻2𝑘𝑔𝐻2
(34)
Refineries We use the list of all refineries and their output
capacities from the German Petroleum Industry Association
(MWV,2020). Mineral oil consumption will decrease by
varying degrees by 2030, depending on assumptions about
the demand for liquid fuels (Michalski et al.,2019). Cor-
respondingly, the current production volume of 87,013,000
tons is distributed across the sites in proportion to their pro-
cessing capacity. Then, derived from the results of Prognos
AG (2020a), the assumption is made that the demand for
mineral oil will decrease by about 20 % until 2030.
The specific hydrogen net demand is assumed to be ap-
proximately 100 m3H2 per ton crude oil, based on Schweer
et al. (2002). Thus, the site-specific hydrogen demand for
refineries can be estimated with equation 35.
𝐻𝐷 =𝑡𝑂𝑖𝑙,𝑝𝑞,2030 ∗ 100𝑚3𝑡𝑂 𝑖𝑙,𝑝𝑞,2030
∗ 0,0841 ∗ 𝑘𝑊 ℎ𝐻2𝑘𝑔𝐻2∗ 22% (35)
B. Appendix B: Conversion factors
Table 11
Numeric values and conversion factors for H2
Lower heating value of hydrogen 33.33 kWh/kg
Conversion factor kg in m³11.89
Conversion factor m³in kg 0.0841
vom Scheidt et al.: Preprint submitted to Elsevier Page 19 of 19
ResearchGate has not been able to resolve any citations for this publication.
Article
Full-text available
Transitioning German road transport partially to hydrogen energy is among the possibilities being discussed to help meet national climate targets. This study investigates impacts of a hypothetical, complete transition from conventionally-fueled to hydrogen-powered German transport through representative scenarios. Our results show that German emissions change between −179 and +95 MtCO2eq annually, depending on the scenario, with renewable-powered electrolysis leading to the greatest emissions reduction, while electrolysis using the fossil-intense current electricity mix leads to the greatest increase. German energy emissions of regulated pollutants decrease significantly, indicating the potential for simultaneous air quality improvements. Vehicular hydrogen demand is 1000 PJ annually, requiring 446–525 TWh for electrolysis, hydrogen transport and storage, which could be supplied by future German renewable generation, supporting the potential for CO2-free hydrogen traffic and increased energy security. Thus hydrogen-powered transport could contribute significantly to climate and air quality goals, warranting further research and political discussion about this possibility.
Article
Full-text available
The energy transition faces the challenge of increasing levels of decentralised renewable energy injection into an infrastructure originally laid out for centralised, dispatchable power generation. Due to limited transmission capacity and flexibility, large amounts of renewable electricity are curtailed. In this paper, we assess how Power-to-Gas facilities can provide spatial and temporal flexibility by shifting pressure from the electricity grid to the gas infrastructure. For this purpose, we propose a two-stage model incorporating the day-head spot market and subsequent redispatch. We introduce Power-to-Gas as a redispatch option and apply the model to the German electricity system. Instead of curtailing renewable electricity, synthetic natural gas can be produced and injected into the gas grid for later usage. Results show a reduction on curtailment of renewables by 12 % through installing Power-to-Gas at a small set of nodes frequently facing curtailment. With the benefits of decentralised synthetic natural gas injection and usage, we exploit the advantages of coupling the two energy systems. The introduction of Power-to-Gas provides flexibility to the electricity system, while contributing to a higher effective utilisation of renewable energy sources as well as the natural gas grid.
Conference Paper
Full-text available
Hydrogen fueled transportation can contribute substantially to the reduction of global carbon emissions. However, the production of hydrogen through electrolysis creates interdependencies with electricity systems. Therefore, we present a new model which couples the hydrogen supply chain with the electricity system. We use this model to analyse a case study of Germany in 2030. We find that if efficient spatially resolved electricity tariffs are applied instead of existing uniform tariffs, electrolyzers are placed primarily at low-cost nodes and farther away from consumption centers. For hydrogen, this leads to higher transportation costs, but lower production costs, and lower total costs. Moreover, costs for congestion management decrease substantially.
Conference Paper
Full-text available
In this paper, we present the first model for hydrogen-fueled bus fleet operation with on-site hydrogen production. To support the planning and operation of municipal hydrogen bus fleets, we transfer and adapt a vehicle scheduling algorithm for conventional buses. We optimize vehicle routes with regard to minimizing operational costs. Based on that, we implement a hydrogen refueling strategy suited to minimize the peak demand for a hydrogen refueling station with on-site electrolysis. To demonstrate the functionality of the approach, it is applied to a real-world scenario for public transport in Karlsruhe, Germany. The findings show that the scheduling strategy for vehicle routing and fueling is suited to reduce peak hydrogen demand and thus the required electrolyser capacity and associated costs. In comparison to a naive benchmark refueling scenario, the peak demand is reduced by 20%, lowering investment costs by 400,000 e.
Data
Full-text available
This documentation describes a data set for the German electricity system in 2030 and the methodology behind certain figures. The data can be used under a Creative Common license. The data set includes 1.) topological data on the transmission grid, the generation infrastructure and the demand, 2.) a time series of electricity demand in 2030 in an hourly resolution, 3.) a time series of renewable generation in 2030 in an hourly resolution, and 4.) dispatchable generation capacity and costs in 2030. All data is spatially resolved, at the level of 485 nodes.
Article
Full-text available
The rapid transformation of the electricity sector increases both the opportunities and the need for Data Analytics. In recent years, various new methods and fields of application have been emerging. As research is growing and becoming more diverse and specialized, it is essential to integrate and structure the fragmented body of scientific work. We therefore conduct a systematic review of studies concerned with developing and applying Data Analytics methods in the context of the electricity value chain. First, we provide a quantitative high-level overview of the status quo of Data Analytics research, and show historical literature growth, leading countries in the field and the most intensive international collaborations. Then, we qualitatively review over 200 high-impact studies to present an in-depth analysis of the most prominent applications of Data Analytics in each of the electricity sector’s areas: generation, trading, transmission, distribution, and consumption. For each area, we review the state-of-the-art Data Analytics applications and methods. In addition, we discuss used data sets, feature selection methods, benchmark methods, evaluation metrics, and model complexity and run time. Summarizing the findings from the different areas, we identify best practices and what researchers in one area can learn from other areas. Finally, we highlight potential for future research.
Article
Full-text available
A potential solution to reduce greenhouse gas (GHG) emissions in the transport sector is to use alternatively fueled vehicles (AFV). Heavy-duty vehicles (HDV) emit a large share of GHG emissions in the transport sector and are therefore the subject of growing attention from global regulators. Fuel cell and green hydrogen technologies are a promising option to decarbonize HDVs, as their fast refueling and long vehicle ranges are consistent with current logistic operational requirements. Moreover, the application of green hydrogen in transport could enable more effective integration of renewable energies (RE) across different energy sectors. This paper explores the interplay between HDV Hydrogen Refueling Stations (HRS) that produce hydrogen locally and the power system by combining an infrastructure location planning model and an electricity system optimization model that takes grid expansion options into account. Two scenarios – one sizing refueling stations to support the power system and one sizing them independently of it – are assessed regarding their impacts on the total annual electricity system costs, regional RE integration and the levelized cost of hydrogen (LCOH). The impacts are calculated based on locational marginal pricing for 2050. Depending on the integration scenario, we find average LCOH of between 4.83 euro/kg and 5.36 euro/kg, for which nodal electricity prices are the main determining factor as well as a strong difference in LCOH between north and south Germany. Adding HDV-HRS incurs power transmission expansion as well as higher power supply costs as the total power demand increases. From a system perspective, investing in HDV-HRS in symbiosis with the power system rather than independently promises cost savings of around seven billion euros per annum. We therefore conclude that the co-optimization of multiple energy sectors is important for investment planning and has the potential to exploit synergies.
Article
Driven by the need for decarbonizing energy carriers across sectors, and the increasing availability of low-cost renewable electricity generation future energy systems will see a rise of power-to-gas technology. For example, hydrogen and its derivates can make enable the usage of carbon-neutral electricity for hard-to abate industry sectors and serve as long-term seasonal storage. Given recent drafts of ambitious political hydrogen strategies around the world, the question arises which power-to-gas configurations provide the highest value for money from a power system perspective. This work provides a flexible framework to compare the performance of current power-to-gas sites all over the world. Power-to-gas technologies are assessed with a benchmarking framework based on Composite Indicators to compare the system value of existing conversion technologies, plant sizes, cost structures, and configurations. Our analysis confirms recent research that suggests that plant performance is higher for larger projects and improves as projects move from research stage over pilot stage to commercial stage. Our findings inform policy makers and electricity system planners who aim to identify the economically and technically most promising solutions for investment.
Article
Hydrogen fuel cell electric vehicles (FCEVs) have been proposed as an option for lowering carbon dioxide (CO2) and pollutants emissions from the transportation sector, when implemented in combination with green hydrogen production methods such as water electrolysis powered by renewable electricity. FCEVs also have the added advantages of high specific energy density and rapid refueling, two important challenges that battery electric vehicles have not yet fully overcome. Moreover, flexible operation of electrolysis could support the grid and lower electricity costs. In this paper, we simulate time-varying FCEV hydrogen refueling demand for light, medium- and heavy-duty vehicles met using electrolysis systems distributed throughout the Western U.S. power system. We find that by oversizing electrolyzers the resulting load flexibility results in different hydrogen generation temporal profiles, average electricity costs, renewable curtailment levels, and CO2 emissions. Our results indicate that increasing hydrogen production flexibility lowers hydrogen and electricity generation cost and CO2 emissions, but there is a tradeoff between lowering operational cost and increasing electrolyzer capital cost, yielding a minimum total system cost at a size corresponding to between 80% and 90% annual capacity factor assuming a future electrolyzer cost of $300/kW.