Introduction
I am Maud, research engineer at the Sciences Pour l'Environnement (SPE) laboratory in Ajaccio. Together with my colleague Chugoun, research engineer at the University of Corsica Pascal Paoli, we present the research work conducted at our laboratory on the integration of hydrogen as an energy vector in a non-interconnected zone, with applications in the electricity and transport sectors.
This work was funded by two ANR projects — the IL project and the UNITI project — in partnership with the University of Corsica, the University of Bourgogne-Franche-Comté, the FEMTO-ST laboratory in Belfort and the SPE laboratory in Ajaccio.
Summary of Previous Work — Chugoun
The work presented in the previous webinar focused on optimizing the hydrogen supply chain at the scale of Corsica by 2050, using a mathematical algorithm developed in the GAMS environment.
The model covers the entire chain: renewable energy sources (photovoltaic, wind) or local grid, hydrogen production via PEM or alkaline electrolyzers, storage in compressed or liquefied form, and distribution across nine distinct territorial zones. Modeled demand covers light mobility, heavy mobility and maritime shuttles.
Context: Non-Interconnected Zones
France has ten non-interconnected zones (ZNI) in the electrical sense — territories where electricity production is managed locally: Corsica and the overseas departments.
These zones share several characteristics:
— Electricity production that is predominantly or entirely fossil-based, relying on external imports
— Production costs three times higher than mainland France (~€80/MWh): between €250 and €650/MWh depending on the ZNI
— A CSPE compensation mechanism allowing local users to pay the same rates as mainland consumers
— A carbon intensity five times higher than mainland France
The Energy Transition Law for Green Growth sets an energy autonomy target of 2030 for all ZNIs, and 2050 specifically for Corsica, with fully renewable electricity production at that horizon.
The Corsican Electricity Mix Today
Since 2005, the Corsican electricity mix has rested on three pillars:
— One third renewable production, predominantly hydroelectric
— One third thermal production, from two power plants in the north and south
— One third electricity imports from Italy and Sardinia
In 2023, only 38% of the mix is renewable. Production remains heavily dependent on fossil resources and costly.
The Corsican Assembly has adopted three targets for 2050: an 89% reduction in greenhouse gas emissions from the energy sector, a 54% reduction in final energy consumption, and electricity production predominantly from local renewable sources.
Levers for Achieving Energy Autonomy
Two main levers have been identified.
The first is the decarbonization of production: developing renewable energy, decommissioning thermal power plants and commissioning a new bioliquid-fueled plant (the Ricanto plant). This lever requires the addition of storage systems to offset renewable intermittency.
The second is the reduction of final energy consumption: electrification of uses, development of electric vehicles and demand-side management measures such as building thermal renovation.
Our research focuses on both levers: adding storage systems and electrifying end uses.
Existing Hydrogen Deployments in Corsica
Two installations are already operational:
— The MYRTE platform, where the SPE laboratory is based, coupling a solar field with several hydrogen chains for grid services, with a distribution station supplying a hydrogen vehicle
— The Corse-Tyrrhène deployment in the south, supplying hydrogen-powered forklifts
Three projects are under development: two winners of the ADEME Territorial Hydrogen Ecosystems call for projects, targeting the decarbonization of the ports of Ajaccio and Bastia, and the Folleli project by Corsicasol, aiming to convert excess solar production into hydrogen.
Modeling Methodology
The Electricity Network
Given the complexity of the distribution network (over 6,000 source substations), we reduced the model to Corsica's high-voltage network at 90 kV, using the open-source tool PYPSA.
The Corsican network comprises 33 HTB source substations, divided into 28 distribution substations (connecting loads, generators and storage systems) and 5 injection substations (connecting large-capacity generators: hydroelectric dams and thermal power plants). Imports from Italy and Sardinia are modeled as a 150 kV high-voltage link.
The model was validated on 2018 and 2019 data, then extended to 2050 with additional renewable potentials, the new bioliquid plant and consumption trajectory scenarios.
Electricity Consumption
The consumption of Corsica's 360 municipalities was distributed across 20 distribution substations using a bottom-up approach. The consumption ratio per substation is assumed constant through 2050, with two evolution trajectories: trend-based or moderate (with stronger demand-side management actions).
The Hydrogen Network
The hydrogen network was modeled under two configurations depending on the application:
— Low-pressure storage for stationary applications (grid services)
— High-pressure storage for mobility (buses), requiring hydrogen compression
Electrolyzers are directly connected to the electricity network. Produced hydrogen is either compressed and stored at high pressure for distribution, or stored at electrolyzer outlet pressure and converted back to electricity via a fuel cell.
Hydrogen Demand for Mobility
Demand was modeled from the conversion of bus fleets in Ajaccio and Bastia:
— Ajaccio: 6 buses, 502 liters of diesel/day, 5.3 MWh to supply → 162 kg H₂/day
— Bastia: 5 buses → 110 kg H₂/day
Hydrogen Demand for Electricity Production
Two stationary applications were modeled: consumption at the Ajaccio port zone (buildings and quays) and power supply for ferries at berth, currently running their onboard diesel engines during stopovers that can last up to 10 hours.
Results for the 2050 Electricity Network
The retained simulation integrates the new bioliquid plant, PV and wind renewable capacities raised to their maximum, without reinforcing import capacity.
Results:
— Total electrical capacity: 1,650 MW
— Annual production: ~2,900 GWh, split between bioliquid (25%), photovoltaic and hydroelectric
— Production cost: €89/MWh, a 61% reduction compared to today
— CO₂ emission reduction from electricity production: 70%
— No reinforcement of electrical lines required
The addition of 700 MW of PV was made possible by the installation of hydrogen equipment across the island.
Results for the Hydrogen Network
— Annual hydrogen consumption for stationary use: 13 kT
— Annual consumption for mobility (two bus stations): 0.1 kT
— Hydrogen production cost: €117/MWh, or €5.8/kg
This cost remains high compared to global projections, but is explained by the need to install the entire chain from scratch on an island with no existing hydrogen infrastructure.
Required equipment includes 150 MW of electrolyzers distributed across the island (operating 5 to 12 hours per day, primarily on solar surplus), 49 tonnes of low-pressure storage per day, and 183 MW of fuel cells distributed across the island (operating in the evening and early morning).
Focus on the Ajaccio Station
For a fleet of 6 buses and a daily demand of 161 kg H₂, the station requires:
— A 1.5 MW electrolyzer consuming 2.4 GWh/year, covering 225 m²
— A 0.5 MW compressor consuming 2 GWh/year, covering 156 m²
— An estimated total installation cost of €1.5 million
The electrolyzer operates mainly during the day using solar surplus from the Ajaccio source substation. The compressor runs in the early morning and late afternoon to fill buses before and after their service runs.
Non-Economic Impacts Assessed
— Land area required for electrolyzers: a significant constraint, especially in densely populated areas
— Water consumption for electrolysis: 17,500 m³ — assessed as having no impact on the island's drinking water supply
— Storage volume at source substations: 1,743 m³, concentrated mainly in Ajaccio and Bastia, with safety regulatory constraints to anticipate
— Life Cycle Assessment (LCA) conducted to evaluate the overall environmental impact of adding hydrogen chains across the island
Conclusions
Hydrogen can make a significant contribution to Corsica's renewable production mix by 2050, by valorizing excess solar production and restoring it during periods of low availability. It also enables a reduction in greenhouse gas emissions from mobility and grid services.
Key barriers identified include land area constraints in high-density zones, regulatory constraints (nature reserves, protected areas), and installation costs.
Perspectives
— Conversion of the bioliquid plant into a hydrogen-fueled plant, leveraging the tri-fuel engines planned (gas, light fuel oil, bioliquid), to build a complete hydrogen ecosystem
— Integration of climate change effects into models, as historical solar radiation and wind data may no longer accurately reflect future conditions
— Refinement of the model to incorporate geographical and regulatory constraints such as nature reserves and national parks
Q&A
On stations open to light mobility — The model currently covers heavy mobility only, but can be extended to include light hydrogen vehicles or electric vehicles.
On the tri-fuel engines at the Ricanto plant — The planned engines (total capacity ~112 MW) can run on gas, light fuel oil or bioliquid. Gas was ruled out due to the absence of a gas network in Corsica (estimated cost: €1 billion). Bioliquid (imported rapeseed oil) was ultimately selected, with zero direct emissions.
On the €5.8/kg price — This is the production cost, including the electrolyzer, compressor and operating costs. It does not represent the pump price paid by end users.
On hydrogen vs battery storage — Battery storage modules were considered but not retained, as Corsica's energy plan prioritizes green hydrogen. Both technology costs are expected to decrease by 2050, including battery decommissioning costs that should be factored into network optimization.
On the carbon intensity of produced hydrogen — The model is projected to 2050 with a mix free of thermal plants and imports. Hydrogen produced will therefore be low-carbon, derived primarily from renewable surpluses and bioliquid with zero direct emissions.
On distribution pressure — Stations are sized at 350 bar for heavy mobility. Adaptation to 700 bar for private vehicles is possible by modifying the compressor configuration.
On equipment lifespan — An average lifespan of 7 to 10 years was assumed, with investment renewal at the end of each cycle.
On links with the business ecosystem — Research results are integrated into Corsica Energia's EPFH 2A and 2B projects, winners of the ADEME Territorial Hydrogen Ecosystems call for projects.
