Source
https://cordis.europa.eu/project/id/779541 — original source (opens in a new tab; the file is not redistributed)
REMOTE project Deliverable D2.2 — Technical specification of the four hydrogen-storage demonstrators
Official H2020 deliverable (Task T2.2, submitted 30/07/2018, dissemination level PU). Provides the detailed device specifications + a first-level Matlab operation-strategy model for the four off-grid/isolated-microgrid demonstrators. This is a pre-deployment technical-specification deliverable — its “results” are preliminary MODELLED simulations, not measured field data. It verifies CR_040’s Froan configuration figures from the primary source, but does not contain the achieved (measured) round-trip efficiency, real lifetime cost, or downtime RT_275 ultimately seeks — those are deferred to later work packages and to the economics deliverable.
Summary
REMOTE Deliverable D2.2 defines the technological specifications of the four EU demonstrators of fuel-cell-based hydrogen energy storage in isolated-microgrid or off-grid remote areas, and reports the outputs of a preliminary Matlab operation-strategy model used to justify the component sizing. For each site it gives the renewable sources, loads, and the power-to-power (P2P) storage hardware (electrolyser, fuel cell, hydrogen tank, battery), then models annual load coverage under two control strategies (battery-as-buffer vs battery-support-only) with an optional hysteresis band. The headline finding across sites: an oversized battery does most of the diurnal deficit-covering while the hydrogen pathway serves the longer-term / seasonal store, and the modelled systems cut external-source (diesel/grid) reliance to under ~5% of load (Ambornetti reaches full autonomy on solar + biomass). The deliverable is a design-and-simulation document; it explicitly states that real operational data and the economic/business cases will follow in subsequent work.
Key claims
- claim: "D2.2 is a PRE-DEPLOYMENT technical-specification deliverable (Task T2.2 'HOW to improve the local situations: detailed technical specification of the four demonstrators'; starting date 01/02/2018, actual submission 30/07/2018, due M4; dissemination level PU; 57 pages; version 1.0). Its reported load-coverage figures are outputs of 'a first-level operation strategy model, developed in Matlab' using 'nominal values for equipment sizes and efficiencies' — NOT measured operation: 'Real data from the P2P system operation in the various DEMOs will be then employed for a refinement of the model in the subsequent work packages.' The economics/business cases are explicitly deferred: 'The technical solutions derived in the current deliverable ... will allow for subsequently defining the economics (expected economic outcomes and the business cases) of the DEMOs.'"
source_location: "Cover page (Task, dates, dissemination, pages, version); Abstract; §3 P2P system model description (Matlab, nominal values); Conclusion (real data in subsequent WPs; economics deferred)"
- claim: "DEMO 4 Froan/Rye (Norway) — the community example CR_040/hydrogen_storage cite — verified from the primary source: RES = an 85 kW PV plant + a 225 kW wind turbine (sizes referred to the Rye test site); non-integrated P2P = a 55 kW PEM electrolyser (63% LHV efficiency, 10–100% modulation, 30 barg) + a 100 kW PEM fuel cell (50% LHV efficiency, 6–100%, 0.5 barg); hydrogen storage = 3333 kWh useful gross energy (LHV) ≈ 100 kg at 30 barg; Li-ion battery = 550 kWh (5 racks of 110 kWh) at 96% efficiency, SOC 20–90%. Annual Rye load ≈ 126.75 MWh; annual RES (PV+wind, TrønderEnergi maintenance) ≈ 284.68 MWh (209.7 wind + 74.9 PV)."
source_location: "§2.4 DEMO 4: Froan/Rye — RES sources; Non-integrated P2P system; Tables 12, 13, 14; Table 15 (load/RES balance)"
- claim: "Froan driver = grid-alternative avoidance (the 'Froan logic'): the PV/wind + P2P solution was chosen to 'avoid the expensive and invasive replacement of the outdated sea cable' (interconnector owned by TrønderEnergi), with diesel generation excluded because 'Froan is a natural reserve' and because of fuel transport/logistics cost. Verifies CR_040's claim that hydrogen there competed against a very expensive grid alternative, not against batteries alone."
source_location: "§2.4 DEMO 4: Froan/Rye — Site description and drivers"
- claim: "Design round-trip efficiency of the hydrogen (P2G→G2P) pathway is the product of the nominal stack efficiencies used as model inputs: 63% LHV electrolyser × 50% LHV fuel cell = ~31.5% LHV, BEFORE DC/DC (95–98%) and DC/AC (94–97%) converter losses, compression, and auxiliary loads. Auxiliary loads are non-trivial: a constant ~1 kW control-and-gas-unit draw plus fuel-cell (AUXFC = 2 + 4·PFC/PFC,NOM kW) and electrolyser (AUXEL = 2 + 7·PEL/PEL,NOM kW) ventilation/cooling loads; at Ginostra these add up to 8.76 MWh/yr (control) + 4.04 MWh/yr (electrolyser aux) + 1.11 MWh/yr (fuel-cell aux). These are nameplate design values, not measured achieved efficiency. Corroborates the low round-trip in [[cr_030_nz-offgrid-backup-firming-reliability-2026|CR_030]] (~35–48%) and [[cr_040_nz-community-green-hydrogen-viability|CR_040]] (~30–45%)."
source_location: "Tables 1/5/8/12 (63% P2G, 50% G2P LHV); §1.1 (DC/DC 95–98%, DC/AC 94–97%); §2.1 (constant ~1 kW; AUXFC/AUXEL formulae); §4.1 (8.76 / 4.04 / 1.11 MWh/yr auxiliary loads)"
- claim: "Across the modelled sites the BATTERY does most of the deficit-covering and hydrogen serves the longer-term/seasonal store. DEMO 4 Froan/Rye modelled annual load coverage (strategy 2 + hysteresis): 61.2% directly from RES, 25.8% from battery, 8.4% from fuel cell, 4.5% from external source. DEMO 1 Ginostra (170 kW PV; integrated P2P; 600 kWh battery; 1793 kWh gross H2 tank): 47.8% RES-direct, 44.3% battery, 3.5% fuel cell, 4.4% external. The deliverable states 'the battery pathway is generally more favoured than the hydrogen one ... considering the higher efficiency of the battery ... However, hydrogen is still necessary because of its capability to perform a longer term storage' — the H2 tank state-of-charge fills over spring/summer and empties over the deficit season."
source_location: "§4.3 DEMO 4 Table 25; §4.1 DEMO 1 Table 17; §4.1 discussion + Figure 21 (H2 SOC over the year); §2.1 Tables 1–3 (Ginostra sizes)"
- claim: "DEMO 3 Ambornetti (off-grid Italian hamlet; 40 kW PV + 50 kWe biomass-gasification CHP; integrated P2P with a 498 kWh gross H2 tank, 30 kWh battery) reaches COMPLETE energy autonomy in the model — 0 MWh from any external source — with biomass, not the fuel cell, as the real backstop: modelled load coverage 55.1% direct PV, 8.3% fuel cell, 36.6% biomass, 0% external (biomass covers ~4× the fuel-cell contribution). The H2 pathway's value there is maximising RES exploitation (near-zero curtailment ~2.8%), not being the primary firm supply."
source_location: "§2.3 DEMO 3 (RES sources, Tables 8–11); §4.2 DEMO 3 Tables 21–22 + discussion ('integration of the biomass source is more consistent (around four times more) than the usage of the fuel cell')"
- claim: "DEMO 2 Agkistro (Greece) uses the P2P/H2 system purely as a 1–2-day BACKUP behind a 0.9 MW hydroelectric plant, sized to a ~282.5 kWh/day load — which requires ~615 kWh of hydrogen (LHV) per day to satisfy — with a 996 kWh (useful 854 kWh) hydrogen store 'to guarantee backup energy for 1-2 days'; its battery is only 30 kWh (system start-up/auxiliaries, not an energy buffer). Driver = grid-connection avoidance: 'Avoid an expensive investment cost for connection to the grid (20 km away)'. DEMO 1 Ginostra is an off-grid island (Stromboli) currently on three 48 kW + one 160 kW diesel gensets; the P2P solution is modelled to cut diesel-genset reliance to ~4.4% of load (target: full diesel-off during experimental operation)."
source_location: "§2.2 DEMO 2: Agkistro (drivers; Tables 5–7; 282.5 kWh/day load → ~615 kWh H2/day → 996 kWh store for 1–2 days); §2.1 DEMO 1 (diesel gensets; drivers); §4.1 ('drastically decrease the use of ... diesel generators to a value of around 4.4%')"Neobiome Intelligence relevance
This is the primary REMOTE deliverable behind the community-hydrogen coverage already in the wiki (hydrogen_storage, URL_016, CR_040, CR_030). It strengthens the existing verdict rather than changing it, and does so from an authoritative primary source rather than an AI synthesis:
- Verifies CR_040’s Froan configuration (85 kW PV + 225 kW wind + 3333 kWh H₂ ≈ 100 kg + 550 kWh Li-ion) directly from D2.2 — closing the “cited via secondary” gap on the wiki’s headline community-hydrogen example.
- Confirms the “battery does the diurnal work, hydrogen is the seasonal store” architecture with hard modelled numbers (Froan: 25.8% battery vs 8.4% fuel cell of load), consistent with the D01 finding that an oversized battery + a dispatchable backstop beats hydrogen for diurnal firming.
- Confirms biomass, not hydrogen, as the effective backstop at Ambornetti (36.6% biomass vs 8.3% fuel cell; full autonomy) — corroborating the NI “biomass/diesel backstop” framing over community hydrogen.
- Supplies a design-level round-trip proxy (~31.5% LHV nameplate before converter/BoP/auxiliary losses), at or slightly below the low end of the CR_030 (~35–48%) / CR_040 (~30–45%) range — a caution, not a contradiction: it is a nameplate figure, and real delivered efficiency is lower.
- Keeps hydrogen a V1+/research-horizon option in the engine (unchanged). ⚠ All figures are EU demonstrator DESIGN/MODELLED values (a €6.7M subsidised R&D project), not unsubsidised community-scale operational evidence, and not NZ-specific — the NZ CAPEX and operational-benchmark gaps (RT_276 / RT_178) remain open.
Research targets
Documents to retrieve
- RT_275 (advanced, kept open) — this deliverable (D2.2) supplies the REMOTE demonstrator technical specifications and preliminary modelled coverage, but not the measured achieved round-trip efficiency, real lifetime cost, or downtime originally sought. Residual = the REMOTE operational-results deliverables and the economics/business-case deliverable (D2.1 economic-context and the later WP outputs referenced in this document’s conclusion).
Research gaps
- No new gaps. The NZ-specific community off-grid hydrogen techno-economic benchmark (RT_276) and the NZD community electrolyser+fuel-cell+storage CAPEX (RT_178) remain the binding gaps; this EU deliverable does not fill either.
Notes
Downloaded authoritative PDF (official H2020 public deliverable) — not an AI-prepared capture, so no retrieval-provenance block applies; data_quality: high (authoritative primary source, every figure traced verbatim to the deliverable’s tables, but the coverage “results” are preliminary Matlab-modelled design values, not measured field data). Sibling of URL_016 (the CORDIS fact sheet for the same project) — this is the underlying deliverable that fact sheet pointed to. Cross-referenced from CR_040 and hydrogen_storage. Contributors span the REMOTE consortium (BPSE, EGP/ENEL Green Power, EPS/Electro Power Systems, HOR/Horizon, HYG/Hydrogenics, IRIS, POW/Powidian, TE/TrønderEnergi).
Connections
Links to
Sources (3): CR_030 · CR_040 · URL_016
Technologies (1): Hydrogen Storage (community-scale, seasonal)
Referenced by
EDT domains (1): D01: Renewable Energy & Storage Systems
Technologies (1): Hydrogen Storage (community-scale, seasonal)