8.1 Modelling principles
The TYNDP 2026 scenario modelling is based directly on the methodological framework developed for the TYNDP 2024 top-down scenarios, ensuring continuity, transparency and comparability across scenario cycles. At the same time, selected refinements have been introduced, reflecting stakeholder feedback, evolving data collection methods, modelling limitations identified by the scenario team, and technical issues identified during the 2024 scenario cycle.
The core objective of the modelling framework is to represent the operational behaviour of the European energy system under different scenario assumptions, with a particular focus on supply–demand balances, hydrogen import, cross border exchanges and interactions between the electricity and hydrogen sectors.
The scenarios are designed to illustrate how the European energy system may operate under the assumptions defined in the National Energy and Climate Plans (NECPs) and under a selected set of weather conditions. They are not intended to assess network performance or to identify infrastructure investment needs, as these aspects are addressed through dedicated TYNDP network analyses and planning processes.
The modelling setup is fully aligned with the scenario grid assumptions and the input data provided by TSOs. The analysis does not rely on Monte Carlo simulations; instead, all runs are carried out using a single, fixed outage pattern for the power plants.
8.2 Modelling topology
The modelling framework represents the European energy system as an integrated set of electricity and hydrogen markets, with explicit interactions between both sectors.
The EU electricity sector is modelled with an hourly resolution, aiming to replicate day ahead market operation under a perfect foresight assumption, and is explicitly coupled with a detailed EU hydrogen market representation. In addition to electricity and hydrogen, part of the residential heat demand is modelled using Hybrid Heat Pumps. Synthetic fuel demand is introduced exogenously, while the model determines endogenously whether this demand is met through domestic production or through imports from outside the EU, depending on relative prices and availability.
General modelling topology
Figure 15 illustrates the generic modelling topology applied to represent the electricity and hydrogen systems, and their interactions with other carriers.
The topology is applied consistently across all countries where detailed input data were available, primarily EU27 countries and Switzerland, for which ETM-based datasets were used. In these cases, the full modelling structure is implemented.
For countries where input data were less granular, a simplified modelling representation is applied, with reduced nodal detail and functionality. These simplifications are introduced to ensure consistency in scenario coverage while maintaining alignment with available data.
A detailed overview of country-specific modelling configurations, including the level of simplification applied where relevant, is provided in Annex VI.
Electricity system structure
The electricity sector is represented by three distinct nodes per Bidding Zone, each capturing different components of demand flexibility:
- Electricity market (e-market): All generators connected to the transmission grid (e. g. thermal power plants, utility-scale solar PV, onshore and offshore wind, and nuclear power plants) are modelled within the e-market node. Large-scale flexibility assets, such as electrochemical batteries and hydropower plants with reservoirs, are also represented. Electricity demand in this node includes non-residential consumption, such as that of industry and data centres. The e-market is coupled with the hydrogen sector via electrolysers and hydrogen-fired power plants.
- Residential & tertiary sector (prosumer): Small-scale resources, including rooftop solar PV and residential batteries, are modelled separately in dedicated prosumer nodes. The electricity demand in this node reflects only the electricity used for heating. A wheeling charge is applied between the e-market and the prosumer nodes to represent distribution grid costs (this fee is applied to flows from the e-market to the residential sector).
- Electric vehicles (EVs): The passenger EV charging behaviour is modelled explicitly within the market tool, with charging decisions optimised to meet the mobility demand. The optimisation depends on the fleet characteristics (fixed vs. flexible), electricity prices, the availability of charging infrastructure over time, minimum battery state-of-charge requirements, and additional constraints representing the physical and operational limits of EVs. EVs are split into two distinct fleets based on their point of connection: when charging in public or on -street locations, they are linked to the e-market node, whereas when charging at home or at the workplace, they are linked to the prosumer node. Further details on EV modelling are given in the next section (8.3).
Figure 15: General model topology
Interconnections between Bidding Zones are represented in line with the electricity infrastructure assumptions for each target year (see Chapter 7). Electricity grid losses are incorporated into the demand timeseries of the corresponding e-market and prosumer nodes. Details are given in section 4.3 and Annex VIII.
Power plant efficiencies include own-consumption and, where applicable, CCS effects.
Hydrogen system structure
The hydrogen sector is represented using two main hydrogen zones, in continuity with the deviation scenarios’ topology adopted in the TYNDP 2024:
- A main integrated hydrogen market (Zone 2), representing future hydrogen system interactions.
- A dedicated zone for grey SMR hydrogen production without CCS (Zone 1) in selected countries (Germany, Lithuania, Portugal), ensuring that grey hydrogen does not contribute to the integrated market supply.
Additional bottleneck nodes may be introduced to capture technical constraints limiting hydrogen flows within a country.
The modelling includes multiple hydrogen import routes, allowing differentiation between supply sources and enabling the model to determine the share of imports versus domestic production. The hydrogen network representation is aligned with the approved scenario grid and perimeter, and follows the same principles adopted for the electricity sector. (see Chapter 7).
Sector coupling and additional elements
Sector coupling between electricity and hydrogen is explicitly represented through:
- Electrolysers (power-to-hydrogen)
- Hydrogen-fired power plants
- Dedicated and shared renewable generation linked to hydrogen production
Additional boundaries are included as exogenous inputs, including:
- Synthetic fuel demand
- Hybrid heat pump thermal demand
- Fuel availability for thermal power plants
- Biogenic CO₂ availability for synthetic fuel production
Heat representation
A share of residential heat demand is modelled through hybrid heat pumps (HHPs), represented via dedicated demand nodes connected to:
- The electricity system (prosumer node)
- The hydrogen network (for H₂ HHP) or a methane fuel object (for CH4 HHPs)
Conversion efficiencies depend on the technology used (heat pumps vs boilers).
Offshore representation
Offshore wind generation, both for electricity and hydrogen production, can be connected either radially to onshore zones or to dedicated offshore nodes. Offshore grid assumptions fully follow TSO inputs, and no additional expansion is modelled within scenarios.
8.3 EV modelling approach
Passenger EVs are explicitly represented in the market model in order to capture the interaction between charging behaviour, flexibility provision, and electricity market. In contrast, electric trucks, buses, and vans are not modelled as individual flexible assets because of their driving schedules and driver’s work hours; their electricity consumption is included exogenously within the demand profiles.
Passenger EVs are modelled in continuity with the approach adopted in the deviation scenarios of TYNDP24. However, the 2026 update introduces significant improvements to address two key shortcomings of the previous formulation.
In TYNDP 2024, passenger EVs were represented using only two aggregated fleets, differentiated solely by charging location (home or street). As a result, all EV charging was optimised purely based on electricity prices, leading to an overestimation of EV charging flexibility. In addition, a modelling inconsistency in the previous cycle resulted in a significant portion of EV transport demand being supplied directly by the electricity grid, effectively bypassing the vehicle battery charging–discharging cycle. Both effects contributed to unrealistically peaky charging patterns and an excessive representation of EV flexibility.
The updated modelling framework corrects the accounting of EV energy flows and introduces a more granular fleet structure. This enables a more realistic representation of charging behaviour and flexibility provision by passenger EVs within the electricity system (Figure 16).
Figure 16: Electric vehicle (EV) modelling approach: E-market and prosumers (home)
Passenger EVs are represented as separate prosumer and street-connected fleets. Flexible fleets can shift charging over time in response to hourly price signals and, where enabled, can also provide vehicle-to-grid (V2G) services. In addition, the new formulation introduces fixed charging profiles and a further split between commuter and non-commuter fleets to reduce the overestimation of flexibility that can arise when large EV fleets are pooled into a single optimised asset. In particular, commuter EVs are assumed to be unavailable for charging during midday (09 – 16 h) on weekdays, which constrains charging optimisation and shifts charging activity to other time windows within the day or week. The revised formulation yields materially less system flexibility than the TYNDP 2024 setup, which had been considered too optimistic.
Flexibility in the model is driven primarily by smart charging, while V2G is treated as an additional but smaller source of flexibility because participation rates and technical uptake remain more uncertain. For the 2026 cycle, the share of flexible passenger EVs is parameterised through TSO survey-based trajectories, while the remaining vehicles follow fixed charging profiles (usually originating from DFT, for details see Section 4.3). In addition, the share of flexible EVs enabled to provide V2G services is defined based on the same TSO survey inputs. This allows the model to reflect different national assumptions while keeping the overall framework consistent across countries. The model parameters and survey results are described in the technical annex.
8.4 Hydrogen and synthetic fuel modelling
Hydrogen
Hydrogen modelling in TYNDP 2026 is designed to determine how hydrogen demand is met across Europe within a sector-coupled system, where domestic production, imports, infrastructure constraints and system balancing requirements are assessed simultaneously.
Hydrogen supply is represented through a structured merit order that combines domestic production and imports, while reflecting different economic and contractual conditions. Hydrogen storage complements this supply stack as a flexibility option, allowing temporal decoupling between production and demand and contributing to system balancing and price formation.
Hydrogen supply can be conceptually structured into three main components.
Domestic production:
- Electricity-based production via electrolysers, including both grid-connected and dedicated-renewables configurations;
- Natural-gas-based production via Steam Methane Reformers (with or without CCS), representing both blue and grey hydrogen pathways depending on the scenario.
Non-European imports under long-term contractual arrangements: representing lower-cost supply with limited short-term flexibility. For modelling purposes, the low-cost band is assigned a cost of 0 € / MWh H2 to ensure dispatch as a priority within the optimisation. This assumption should not be interpreted as a zero economic cost, but rather as a modelling construct used to reflect the inelastic nature of committed import volumes under long-term obligations.
Additional non-European imports at market-based prices (marginal-cost band): the residual share of import potential beyond the long-term contracted volumes, dispatched as flexible supply that responds to system conditions. Pipeline and shipping routes both add to this band.
However, as the model operates with a limited nodal topology, where each country is represented as a single node, cross-border flows supplying hydrogen from one region to another within the same country, but passing through another country, are not reflected in the model, even though these flows generate significant transport volumes in the transit country – such hydrogen flows are hidden in the respective country node.
In addition, hydrogen storage is represented as a flexibility option, allowing temporal decoupling between production and demand, and contributing to system balancing and price optimisation. Different storage types from steel tanks to salt caverns enter the merit order at different cost levels, depending on their flexibility characteristics.
Hydrogen imports: multi-band pricing and availability
Imported hydrogen is modelled using a marginal cost approach with differentiated import cost bands rather than as a single uniform import block. Different import routes and supply types, including renewable-linked pipeline imports, other pipeline imports, and ship-based imports such as ammonia, are represented with different marginal costs and, where relevant, specific availability profiles, elaborated in Table 23. A detailed explanation of import potentials, infrastructure assumptions and cost methodology is provided in Section 5.7.
| Pricing band | Profile | Modelling treatment |
|---|---|---|
| Long-Term Contract (LTC) band | Flat, must-run profile | Represents firm “take-or-pay” infrastructure investments. Modelled with a zero marginal cost so that the contracted capacity is dispatched as a priority within the optimisation, irrespective of hourly system conditions. |
| Flexible / marginal band | Hourly availability, route-specific | Captures the residual import potential beyond LTC volumes. Subject to higher market prices and to availability constraints; for green hydrogen corridors (North Africa and Ukraine) the profile is derived from daily averages of the renewable generation profile in the exporting country. |
| Shipped ammonia / hydrogen carriers | Slower modulation, security-of-supply role | Modelled as providing flexibility and security of supply. The ammonia pathway is parameterised to include regasification. The LTC band approach is also considered for Ammonia. |
Table 23: Modelling of Imported hydrogen (H₂ )
Domestic production and coupling with the electricity system
Domestic electrolytic hydrogen production is fully coupled to the electricity system. Electrolysers and related conversion assets operate subject to electricity availability, variable costs, network constraints and the wider cross-sector merit order. Hydrogen production is therefore determined by the coupled dispatch rather than by an exogenous flat output assumption.
Electrolysers can be powered through three distinct configurations, which are treated differently in the dispatch:
- E-market. Electrolysers are directly connected to the electricity grid and procure power from the wholesale electricity market. Their dispatch is fully driven by marginal electricity prices, grid constraints and the cross-sector merit order, allowing them to operate flexibly in response to system conditions.
- Dedicated RES (DRES). The renewable asset is physically islanded and supplies hydrogen production exclusively. Power generated by DRES units does not enter the public electricity market and is converted into hydrogen up to the rated capacity of the connected electrolyser; any excess generation that cannot be absorbed is curtailed at source.
- Shared RES (SRES). The renewable asset operates in effect as a virtual Power Purchase Agreement (PPA): its hourly profile strictly prioritises the associated electrolyser, and only the energy left over after electrolyser uptake is allowed to spill into the electricity market. This preserves the contractual logic of dedicated supply while keeping the renewable surplus available to the wider system.
Water cost on electrolyser generation
To accurately reflect operating expenses, an explicit real-world water-usage cost is applied to electrolyser generation, calibrated at 0.33 € / MWh of hydrogen produced. While modest in absolute terms, this term ensures that the operating cost stack of electrolysers reflects all incremental variable costs and is internally consistent with the broader cost framework.
Hydrogen storage modelling
Hydrogen storage modelling moves away from generic representations to reflect actual geological and operational realities. Storage assets are characterised by varying flexibility levels typically described as daily, weekly or seasonal derived from TSO-submitted data covering injection capacities, withdrawal capacities and working-gas volumes. This explicitly differentiates fast-acting facilities from slower seasonal reserves.
Storage parameters are set on an asset-by-asset basis where data are available. Round-trip efficiency, ramp rates and minimum stock levels are reflected so that storage cycles compete realistically with imports and domestic production within the merit order.
Synthetic fuel Modelling
Synthetic fuels (or synfuels) are modelled as an integrated component of the system under the assumption of carbon neutrality at system level. This is achieved by enforcing that all CO₂ emissions from synthetic fuel combustion are offset by biogenic carbon capture and storage (BECCS) elsewhere in the system; therefore, synthetic fuels do not add net CO₂ to the atmosphere within the scenario framework.
From a modelling perspective, the origin of CO₂ used in synthetic fuel production (fossil, biogenic or from direct air capture) is not differentiated, provided that the overall system-wide CO₂ balance is satisfied via BECCS.
An EU-level carbon constraint is implemented whereby total CO₂ removed via BECCS must equal or exceed total CO₂ consumed in synthetic fuel production. National BECCS capacities are derived from a dedicated data collection and aggregated at EU level for modelling purposes. As a result, synthetic production is not geographically constrained by the location of CO₂ capture facilities.
To operationalise this constraint, annual synthetic fuel production capacities (provided in TWh) are converted into corresponding CO₂ feedstock requirements using molecular stoichiometry. This step bridges raw capacity data and the CO₂ budget constraint embedded in the model.
Representation of synthetic fuel production and aggregation
In the model, the EU synthetic fuel sector is represented through two aggregate nodes:
- Synthetic natural gas (SNG)
- Liquid synthetic fuels (e liquids)
Because different synthetic fuels have different chemical compositions, their energy volumes cannot be directly aggregated. Therefore, all synthetic fuel flows are converted into hydrogen-equivalent terms based on their synthesis requirements.
This hydrogen-equivalent representation allows:
- Aggregation of different synthetic fuel types
- Consistent coupling with the hydrogen system
- Integration into the overall system optimisation
The conversion to hydrogen-equivalent values is applied dynamically for each target year, reflecting changes in the composition of liquid synthetic fuels over time.
Stoichiometry and efficiency assumptions
The hydrogen and CO₂ inputs required for synthetic fuel production are determined through a two-step approach:
1// Calculation of theoretical minimum reactant quantities based on stoichiometry of industry standard synthesis reactions (e. g. Sabatier reaction for SNG).
2// Inclusion of industrial plant inefficiencies.
The first step defines the absolute physical limits of each synthesis route, assuming perfect (100 %) conversion efficiency, with no mechanical losses and no unrecovered heat. For complex liquid fuels such as e-kerosene and e-diesel, simplified proxy molecules were selected to reproduce the average carbon-to-hydrogen ratio and energy density of the real fuel blends.
These reactions are inherently exothermic, implying that the hydrogen energy input exceeds the final fuel energy output (e. g., approximately 1.23 TWh of hydrogen is chemically required to produce 1 TWh of e-kerosene).
The second step incorporates plant-level inefficiencies, which capture additional energy requirements associated with:
- Compression
- Recycling loops
- Product separation
- Heat integration
Efficiency losses vary depending on process complexity:
- Simple gaseous fuels (e. g. methane, ammonia): lowest losses (~5 %)
- Intermediate fuels (e. g. methanol): moderate losses (~10 %)
- Multi-step liquid fuels (e. g. diesel, kerosene): higher losses (~15 %)
- Complex pathways (e. g. ethanol): highest losses (~30 %)
These values are treated as representative modelling assumptions rather than plant-specific parameters. A full carbon recovery rate is assumed, meaning that CO₂ losses within the process are negligible at model level.
The resulting conversion efficiencies are summarised in Table 24.
| E-fuel | Our model | Literature range* |
|---|---|---|
| e-methane | 79 % | 70 %–83 % |
| e-diesel | 70 % | 59 %–78 %** |
| e-kerosene | 71% | |
| e-ethanol | 65% | n.a. |
| e-methanol | 80% | 69%–89 % |
| * Review of electrofuel feasibility—cost and environmental impact. Maria Grahn et al. (2022) Progress in Energy 4: 3. | ||
| ** Hydrogen to Fischer–Tropsch | ||
Table 24: Conversion efficiencies of different e-fuels used in our model and literature data
Demand and supply representation
Synthetic fuel demand is modelled as an exogenous input and expressed in hydrogen-equivalent terms. Demand is assumed to be constant over time, with annual demand distributed uniformly across all hours of the year. This reflects the high storability of synthetic fuels and is consistent with the approach adopted in previous TYNDP Scenario cycles.
On the supply side, synthetic fuel production is linked to national hydrogen systems. Each country can supply synthetic fuels up to its reported production capacity, converted into maximum hourly flows.
When domestic production is insufficient to meet EU-level total demand, imports are used to cover the remaining demand. Imports are modelled as unconstrained supply at higher marginal cost than domestic production.
Synthetic fuel import costs are converted into hydrogen-equivalent terms based on the hydrogen content of each synthetic fuel (LHV).
CO₂ accounting and BECCS constraint
To enforce carbon‑neutrality, annual BECCS‑availability is represented through annual constraint objects that track the total volume of biogenic CO₂ sequestered each year.
During synthetic fuel production, these objects are “consumed” based on the CO₂‑to‑H₂ ratio of the fuel being produced. For SNG, that ratio remains constant over time, as the product is a fixed molecule. For aggregated liquid synthetic fuels, the ratio varies annually with the evolving mix of individual liquids, weighted by their respective capacities and CO₂ / H₂ stoichiometry.
Security-of-supply adequacy loop
To ensure that hydrogen demand is met in every hour of the simulation and to avoid failures resulting in Energy Not Served (ENS), an adequacy loop is deployed on top of the base dispatch and proceeds in clearly ordered steps:
1// If the base configuration leads to hydrogen curtailment in one or more hours, the model first releases unconnected ammonia import potentials at the highest market-band price. These volumes act as an additional security-of-supply layer, dispatched only when no cheaper option is available.
2// If a shortfall remains after these unconnected potentials have been exhausted, the model dispatches technology-neutral virtual units at the Value of Lost Load (VoLL). These virtual units carry no physical realism but ensure that demand can always be balanced and that the resulting cost signal correctly reflects the scarcity revealed by the simulation.
In this cycle, no additional supply of NH3 was needed to address adequacy issues.
| Elec Merit Order | Elec Merit Order – Priority | Technology | H₂ merit order | H₂ merit Order – Priority |
|---|---|---|---|---|
| LTC – LOW band (nonEU Pipelines; NH3 imports) | 1 | |||
| RES | 1 | P2G | RES through P2G | 2 |
| Hydro | 2 | Hydro through P2G | 3 | |
| Nuclear | 3 | Nuclear through P2G | 4 | |
| Biofuels | 4 | Biofuels through P2G | 5 | |
| SMR | 6 | |||
| nonEU imports | 7 | |||
| NH3 imports | 8 | |||
| Other nonRES | 5 | P2G to avoid H₂ curtailment | Other nonRES through P2G | 9 |
| Gas plants | 6 | Gas through P2G | 10 | |
| Lignites | 7 | Lignite through P2G | 11 | |
| Coal | 8 | Coal through P2G | 12 | |
| Heavy oil | 9 | Oil through P2G | 13 | |
| DSR | 10 | 14 | ||
| VOLL Elec, VOLL Heat | VOLL H₂, VOLL synfuels, VOLL Heat | |||
Table 25: Electricity and hydrogen merit order
8.5 Hybrid Heat Pump (HHP) modelling
HHPs are modelled as the only endogenous component of the heat system within the TYNDP 2026 Scenario modelling framework.
Each hybrid unit can generate heat through two alternative pathways:
- An electric heat pump, linked to the prosumer electricity node
- A boiler, supplied either by hydrogen or methane depending on the technology type
The efficiency of the electric heat pump component varies with weather conditions through time-dependent coefficient of performance (COP) profiles, reflecting the impact of ambient temperature on heat pump performance.
In the network representation, dedicated heat nodes are introduced at national level for each hybrid technology (one for hydrogen HHPs and one for methane HHPs). The load at each heat node is defined by the corresponding heat demand profile. The heat pump component links the prosumer electricity node and the heat node, with the conversion efficiency given by the time-varying COP profile.
For H₂ HHPs, the boiler component links the relevant hydrogen market node and the heat node, with a fixed boiler efficiency. For CH₄ HHPs, the boilers are not linked to a zonal gas market (since natural gas is not modelled as a market) but instead draws fuel from a single EU-wide natural gas fuel object dedicated to the hybrid heating system.
Installed capacities of both boilers and heat pumps are not constrained in the model. It is assumed that sufficient capacity is always available to fully meet the thermal energy demand, ensuring that capacity does not become a limiting factor. This formulation allows the optimisation to determine, at each point in time, whether heat is provided by the electric heat pump or by the boiler. The decision depends on the availability and marginal cost of electricity and hydrogen at their respective market nodes, the COP profile of the heat pumps, and, for CH₄ HHPs, the assumed natural gas cost and boiler efficiency.
In contrast, fully electric heat pumps are not modelled as separate assets in the system. Instead, their electricity consumption is implicitly included within the prosumer demand profiles, as described in Section 4.3, and therefore does not appear as a distinct technology or decision variable in the optimisation.
8.6 Dedicated RES and Shared RES modelling
Dedicated RES (DRES) and Shared RES (SRES) are explicitly represented in the modelling framework to capture the interaction between renewable generation and hydrogen production under different contractual agreements.
DRES refers to renewable generation that is physically co-located with electrolysers under Physical Power Purchase Agreements (PPAs). This collocation may result either from contractual arrangements between renewable asset owners and electrolyser operators or from common ownership of both assets. While the latter configuration does not formally constitute a PPA, it is treated equivalently within the modelling framework and included under the DRES category.
As for other technologies, installed capacities for DRES and their associated electrolysers are derived from TSO data. Production profiles follow the same RES datasets used elsewhere in the model and are used exclusively to meet hydrogen demand.
SRES are intended to represent renewable generation (solar PV, onshore wind, and offshore wind) contractually linked to electrolysers through virtual PPAs, without direct physical co‑location. Electrolysers are required to prioritise the absorption of renewable generation when available, while any residual electricity can be sold to the electricity market.
Compared to the representation used in the TYNDP 2024 deviation scenarios, the current modelling of SRES introduces two key methodological changes. First, SRES capacities and the associated electrolyser capacities are defined ex‑ante, based on NECP targets and national policy data, whereas in previous studies they were derived endogenously through expansion model optimisation. Second, hydrogen production is prioritised by construction: renewable generation is first allocated to electrolysers, and only surplus electricity is made available to the power market.
In the model implementation, SRES are treated similarly to DRES in order to decouple hydrogen production decisions from electricity market price signals. Renewable generation profiles are consistent with the PECD datasets used for the other renewable sources. Hydrogen production profiles are derived by combining RES generation with electrolyser capacity: when renewable production is lower than electrolyser capacity, the entire output is used for hydrogen production and no electricity is exported; conversely, when renewable generation exceeds electrolyser capacity, the surplus is supplied to the electricity market at the e‑market node. Separate input profiles are therefore provided for hydrogen production and residual electricity generation.
8.7 Offshore modelling
Offshore modelling in TYNDP 2026 covers both electricity and hydrogen and is based on a conservative representation of offshore assets. The methodology approach follows TSO submissions, Scenario Grid inputs and validated project information, with datasets cross-checked against modelling inputs and project collections to ensure consistency.
For the electricity system, offshore generation is primarily modelled as being radially connected to the onshore e-market node of the corresponding bidding zone. However, in cases where TSOs have reported more detailed offshore configurations through the data collection, dedicated nodes and corresponding links are introduced. This allows the model to represent specific offshore structures – such as non-radial connections or internal offshore constraints – where these are considered material for system behaviour.
For the hydrogen system, the methodology distinguishes between electricity-led offshore configurations, hydrogen-led offshore production, and hybrid concepts. Fully off-grid hydrogen production and shared renewable pathways links are not collapsed into one generic offshore node when doing so would hide materially different infrastructure interactions.
Dedicated offshore nodes are introduced where required to preserve internal offshore bottlenecks, including cases where offshore wind capacity, electrical capacity and offshore electrolysis capacity are not interchangeable within a single aggregate node. This is particularly important where offshore hydrogen production or export depends on infrastructure constraints that differ from those governing electricity flows to shore.
