Comparative Techno-economic Assessment of Wind-Powered Green Hydrogen Pathways

Benoit Couraud; David Flynn; Jamie Blanche; Merlinda Andoni; Satria Putra Kanugrahan; Si Chen; Sonam Norbu; Valentin Robu

REVIEW 5 major objections 7 minor 14 references

Comparative Techno-economic Assessment of Wind-Powered Green Hydrogen Pathways

T0 review · 5 major / 7 minor · reviewed 2026-08-05 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Behind-the-meter, electrolyser-first wind hydrogen is cheapest, at £4.62/kg

desk verdict Useful but not fully reproducible comparison of UK wind-hydrogen configurations; the missing wind data and high implied capacity factor are the main issues. read the letter →

arxiv 2509.00136 v1 pith:GHSMR36U submitted 2025-08-29 eess.SY cs.SYecon.GNq-fin.EC

Merlinda Andoni , Benoit Couraud , Valentin Robu , Jamie Blanche , Sonam Norbu , Si Chen , Satria Putra Kanugrahan , David Flynn This is my paper · ORCID

classification eess.SYcs.SYecon.GNq-fin.EC

keywords greenhydrogenLCOHPEMelectrolyserwindenergytechno-economicanalysisbehind-the-meterUKpolicysensitivity

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper builds a techno-economic framework to compare five ways of pairing a 10 MW wind plant with a PEM electrolyser in the UK, estimating the Levelised Cost of Hydrogen (LCOH) for each. It finds that the cheapest configuration is a behind-the-meter system where the electrolyser has priority over grid exports and no grid back-up is used, giving an LCOH of £117.34/MWh (£4.62/kg). Off-grid wind-electrolyser systems and grid-connected electrolysers buying wind via a PPA are close behind, while grid-only electrolysis is the most expensive. Across all configurations, electricity cost is the dominant share of LCOH, followed by electrolyser capital cost. The paper argues that policy should focus on lowering electricity costs and network charges, and on enabling co-located wind-electrolyser projects.

What carries the argument

The central object is the Levelised Cost of Hydrogen (LCOH), computed from annualised capital and operating costs divided by annual hydrogen output, with costs annualised through the capital recovery factor. Wind power is allocated at each half-hour step among hydrogen production, grid export, and curtailment; hydrogen output follows an iterative electrolyser-compressor power split that converges on a consistent allocation. The framework separates costs into electrolyser, compressor, interconnection (grid access and private wire), and electricity, allowing each use case's cost structure to be compared directly.

What would settle it

Re-run the same cost equations using a publicly available UK half-hourly wind generation profile (e.g., from a metered 10 MW wind site or the National Grid ESO data) and check whether the behind-the-meter electrolyser-first configuration still has the lowest LCOH. If another configuration wins with that wind data, the paper's central ranking claim does not generalise beyond its unspecified input.

Watch

Extended reading notes

Core claim

The central claim is that for UK wind-powered green hydrogen, the economic outcome is determined less by the electrolyser technology itself and more by the commercial and physical arrangement between the wind plant, the electrolyser, and the grid. Using a half-hourly simulation of a 10 MW reference wind plant and a PEM electrolyser, the paper computes LCOH for five use cases: grid-only, off-grid, grid-connected with wind PPA, grid-connected using curtailed wind, and behind-the-meter with partial grid connection. The lowest LCOH, £117.34/MWh (£4.62/kg), is achieved by the behind-the-meter configuration where the electrolyser is served first and there is no grid back-up (V-b-i). The off-grid s

Load-bearing premise

The half-hourly wind power time series is not specified or sourced in the paper, and every LCOH number in the results depends on it, so a different wind profile could materially change the ranking of the use cases.

Editorial extensions

If this is right

If the ranking holds, investors seeking the lowest green hydrogen cost should prioritise behind-the-meter, electrolyser-first configurations over grid-connected or grid-back-up designs.
Grid back-up, despite enabling full electrolyser utilisation, is economically counterproductive under current UK retail electricity prices; this challenges the common assumption that high utilisation is always desirable.
Curtailment-based hydrogen production is only competitive if curtailed energy is available at very low or zero cost; otherwise low electrolyser utilisation dominates the cost.
Policy measures that reduce the electricity price, network charges, or environmental levies will have a larger effect on LCOH than electrolyser cost reductions, across all configurations.
Co-located wind-electrolyser systems outperform virtual PPAs or grid-only procurement on cost, so regulatory frameworks that ease private wire connections and shared grid access would support cheaper green hydrogen.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

The LCOH ranking is likely sensitive to the wind resource profile and its correlation with electricity prices; a different wind time series, especially one with less midday peaking or different seasonal pattern, could change which use case is cheapest, so the ranking should not be generalised beyond the specific wind data used.
The analysis assumes onshore wind costs but states the framework applies to offshore projects; offshore wind's higher capacity factor might narrow the gap between behind-the-meter and grid-connected configurations, a directly testable extension.
A testable implication of the 'electrolyser-first' rule is that it increases electrolyser load factor by sacrificing export revenue; the paper's chosen export price of £0.044/kWh partly drives the result, so the ranking would shift if export prices rise.
The result that free curtailed energy gives an LCOH of £121.68/MWh suggests that a targeted policy creating zero-cost access to curtailed renewable energy could make curtailment-based hydrogen competitive with off-grid systems, a scenario worth quantifying with real curtailment data.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Quick take: this is a solid, incremental techno-economic comparison of five UK wind-to-hydrogen configurations. The LCOH ranking is plausible and the cost breakdown—electricity dominates, then electrolyser CAPEX—matches the broader literature. The main thing to check before relying on it is the wind resource data, which is never described.

What the paper does well: it maps five realistic deployment models side by side, including behind-the-meter dispatch priorities and curtailment-based operation, and it uses a standard LCOH framework with clearly stated assumptions. The sensitivity analysis on electricity price, electrolyser cost, efficiency, and stack lifetime is useful. The results in Table II are internally consistent with the stated equations, and the qualitative conclusion that electricity cost is the dominant driver holds across configurations.

The soft spots: the biggest one is the missing wind time series. The paper never says where the half-hourly wind data comes from, what turbine was assumed, or which site. This matters because Table II implies a wind capacity factor of roughly 49% for the off-grid and PPA-only cases (II and III-a). That is high for UK onshore, which typically sits around 25–35%. If the actual wind profile is closer to 30%, annual H2 production drops by ~40% and LCOH for every wind-only case rises substantially; the ordering of V-b-i, II, and III-a could shift, though the general pattern—behind-the-meter/off-grid beats grid-only—would probably survive. The stress-test note is accurate on this point, and the lack of any uncertainty quantification means the reader can't tell how fragile the ranking is. Relatedly, the curtailment use case assumes zero or low-cost curtailed energy, which they acknowledge depends on regulatory mechanisms that don't yet exist. These are addressable weaknesses, not fatal flaws.

Who it's for: hydrogen investors, UK policy analysts, and anyone working on LCOH benchmarks. It's not a methodological breakthrough, but it's a competent, relevant comparison. It deserves a serious referee, but the authors need to provide the wind data and add a sensitivity analysis over the resource profile before I'd trust the specific ordering.

Recommendation: engage with it; send to peer review with a request for full data and uncertainty analysis.

Referee Report

5 major / 7 minor

Summary. The paper develops a techno-economic framework for estimating the Levelised Cost of Hydrogen (LCOH) from wind-powered PEM electrolysis, and applies it to five UK use cases: grid-only electrolysis, off-grid wind–PEMEL, grid-connected PEMEL with a wind PPA, grid-connected PEMEL using curtailed wind energy, and behind-the-meter partial grid connection with grid-first or electrolyser-first dispatch. A 10 MW reference wind plant and 10 MW (or 5 MW) electrolyser are simulated at half-hourly resolution. The headline result is that the behind-the-meter, electrolyser-first, no-backup configuration (V-b-i) achieves the lowest LCOH (£117.34/MWh or £4.62/kg), followed by the off-grid system (II) and the virtual-PPA case (III-a). Electricity purchase cost is identified as the dominant LCOH component, followed by electrolyser CAPEX. A sensitivity analysis for one use case and a comparison with literature ranges are also provided.

Significance. The paper addresses a relevant policy and investment question: which wind-hydrogen business model is most cost-effective in the UK context. The use-case taxonomy is comprehensive and the LCOH equations are, in large part, standard. The work explicitly maps grid connection options, PPA structures, and control strategies onto LCOH, which is a useful contribution. If the numerical results and ranking were robust, the paper would provide actionable guidance for developers and policymakers. However, the current manuscript has several load-bearing deficiencies: the wind resource time series is not specified, the electrolyser efficiency curve is not defined, the stack replacement costing is incomplete, and the sensitivity analysis does not cover the most uncertain inputs. These issues prevent full confidence in the central ranking and must be addressed before the paper can be recommended for publication.

major comments (5)

[§II, §V, Table II] The half-hourly wind time series PW,t is never specified or sourced. The implied wind capacity factor for Use Cases II and III-a is approximately 48.8% (electrolyser load factor equals 48.83% when all wind power is dedicated to hydrogen production and no grid import is used). This is high for UK onshore wind, where typical capacity factors are 25–35%; a 30% capacity factor would reduce annual H2 output by roughly 38% and raise LCOH correspondingly, potentially changing the ordering of the main use cases. Since Eq. (2) and the LCOH denominator in Eq. (4) depend linearly on PW,t, the ranking is not reproducible without this input. The authors must provide the data source, site, turbine characteristics, and a sensitivity analysis over plausible capacity factors.
[§III, Eq. (2)] The electrolyser efficiency η(PPEM,t) is not defined anywhere. No functional form, baseline efficiency value, or reference is given in Table I or the text. The sensitivity analysis in Fig. 5 varies 'efficiency' by +3% to +10%, but the baseline is ambiguous. Hydrogen production in Eq. (2) and electricity consumption in Eq. (8) depend directly on this curve; without it, all LCOH values in Table II are unreproducible. The authors should provide the efficiency curve (e.g., as a piecewise-linear or polynomial function of load) or at minimum cite an empirical model and state the baseline efficiency.
[§III, Eqs. (5)–(8), Table I] The cost equations omit the wind plant CAPEX and OPEX, despite these being listed in Table I. Eq. (8) uses pPPA for wind electricity; in the 'LCOH Free' cases (pPPA=0) the wind generation cost is zero, so the wind CAPEX/OPEX are not recovered in any term. This makes the 'LCOH Free' values a hydrogen-production-and-interconnection cost, not a full system LCOH. For sole-investor or co-located cases this understates the true cost and can bias the comparison. Either include wind CAPEX/OPEX as an explicit cost component in Eq. (5)–(8), or clearly label LCOH Free as excluding wind generation costs and revise any conclusions drawn from it.
[§III, Eq. (5)] The stack replacement cost is levelised using only a single discounted replacement event at nS,y. Over the 30-year electrolyser lifetime, the stack will need to be replaced multiple times: at 48.8% load factor roughly twice (≈14 and 28 years), and at 100% load factor about four times (≈6.9, 13.7, 20.5, and 27.4 years). The formula should sum over all replacement events, e.g., LCreplace,S = Σ_j C_replace,S / (1+d)^(j·nS,y). The current treatment understates replacement cost more for high-load-factor cases, which could change the relative LCOH of Use Case V-b-i (63.5% load factor) versus II/III-a (48.8% load factor).
[§V, Table II and Fig. 5] The sensitivity analysis is reported only for Use Case III-a and does not include the two most significant uncertain inputs: the wind capacity factor and the wind CAPEX. Because the central claim is a cost ranking across use cases, the paper should demonstrate that the ranking is robust to plausible ranges of these inputs (e.g., capacity factor 25–40%, efficiency ±5%, PPA price ±20%, electrolyser CAPEX ±30%). A tornado diagram or a range table for all use cases would be sufficient; without it, the ranking remains a point-estimate result sensitive to unquantified assumptions.

minor comments (7)

[Abstract] Typo: 'determined by the its production pathway' should read 'determined by its production pathway.'
[§II] State explicitly that PW,t is the average wind power over a 0.5 h interval, and specify the meteorological year or data period used. The current notation is clear but the temporal aggregation is not defined.
[§IV-D] The description of Use Case IV ('2x5=10 MW') is confusing. Clarify that five 10 MW wind farms each contribute 2 MW of curtailment to form the 10 MW electrolyser feed, and explain how the curtailment time series is constructed.
[§V] The sentence 'e.g. when running the simulations for 2 wind farms with a 5 MW partial connection each, the LCOH reduces to £158.41/MWh or £6.24/kg' appears to be an incomplete example. Please either complete the explanation or remove it, as it is not tied to the defined use cases.
[Table I] The notation for line capacity 'PLine [0,5,8,10]^d' and wire capacity 'Wire 0–10 MW' would benefit from explicit mapping to the use cases in a separate column rather than only in footnotes; this would improve readability.
[Fig. 5] In the caption, 'ranges from [-100%-+100%]' should be written as 'from –100% to +100%.'
[§V, Table III] When comparing reported LCOH values from the literature, indicate whether these values include compression, storage, and grid connection costs. The current comparison is not fully apples-to-apples.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: LCOH estimates are computed from stated cost and operating assumptions, with no fitted input renamed as a prediction.

full rationale

The paper's central result (Table II) is a set of LCOH values computed by the definitional identity LCOH = C_tot^(α) / M_H2^(α) (Eq. 4), where the annual hydrogen mass follows from Eq. (2) applied to an exogenous wind time series and the annual cost follows from Eqs. (5)-(8) with parameter values listed in Table I. There is no parameter fitted to a target LCOH or to the ranking of use cases; the ranking emerges from applying the same formula across configurations. The sensitivity analysis (Fig. 5) varies stated inputs and reports output changes, which is the opposite of a fitted-input/prediction loop. The comparison with literature LCOH ranges (Table III) is an external benchmark, not an input to the model. The main weakness—the half-hourly wind time series P_W,t is neither sourced nor characterized, and the 48.83% PEMEL load factor in Cases II/III-a acts as an implicit wind capacity factor—is a data transparency / reproducibility concern, not a circularity: a missing input cannot make a derivation circular unless the conclusion is built into that input. The paper also notes its own scope limitations (onshore costs assumed; offshore may differ), and no load-bearing step reduces to a self-citation. A citation to prior work [3] supplies cost parameters (e.g., p_PPA, discount rate), which are independent inputs rather than the target result. Therefore, no circular step is present; the appropriate score is 0.

Assumptions & free parameters 5 free parameters · 5 assumptions · 0 invented entities

The analysis is a standard techno-economic model; all inputs are economic assumptions from literature or chosen scenarios. The central claim rests on the exogenous wind profile and the PEMEL efficiency curve, neither of which is fully specified.

free parameters (5)

Wind-PEM PPA price pPPA = 0, 0.057 £/kWh
Chosen contractual price between wind and electrolyser investors, varied across use cases, not fitted to LCOH.
Grid export price pexp = 0.07, 0.044 £/kWh
Assumed export tariffs for behind-the-meter scenarios, chosen per use case, not fitted.
Grid import price pimp,r = 0.184 £/kWh
Average industrial retail price, used for grid backup and grid-only cases.
Grid access cost CLine = 100 £/kW
Chosen annualised grid connection cost per kW of line capacity.
Private wire cost CWire = 15 £/kW
Chosen cost per kW for private wire between wind and electrolyser.

assumptions (5)

domain assumption Wind power output follows a half-hourly time series PW,t that is exogenous and not specified in the paper.
All LCOH results depend on the temporal profile of wind output, used in Eq. (1) and the simulation.
domain assumption PEMEL efficiency is a function of input power, η(PPEM,t), but the efficiency curve is not defined.
Eq. (2) requires η(PPEM,t) to compute hydrogen output, yet the paper does not provide this curve or its source.
standard math The iterative electrolyser/compressor power allocation converges to a unique solution.
The paper states the allocation is recomputed iteratively until convergence, but provides no proof of convergence or uniqueness.
domain assumption Electricity prices are fixed at the stated averages with no temporal variation.
Table I lists fixed prices for grid import, export, and PPA, neglecting real-time market dynamics.
domain assumption Stack lifetime is 60,000 hours and replacement cost is 48% of CAPEX.
Used to levelise stack replacement cost in Eq. (5) and affects LCOH across all use cases.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Comparative Techno-economic Assessment of Wind-Powered Green Hydrogen Pathways." pith.science (2026). https://pith.science/paper/GHSMR36U

@misc{pith2026250900136,
  author       = {Pith},
  title        = {Pith review of: Comparative Techno-economic Assessment of Wind-Powered Green Hydrogen Pathways},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GHSMR36U}},
  note         = {Machine review of arXiv:2509.00136}
}

read the original abstract

Amid global interest in resilient energy systems, green hydrogen is considered vital to the net-zero transition, yet its deployment remains limited by high production cost. The cost is determined by the its production pathway, system configuration, asset location, and interplay with electricity markets and regulatory frameworks. To compare different deployment strategies in the UK, we develop a comprehensive techno-economic framework based on the Levelised Cost of Hydrogen (LCOH) assessment. We apply this framework to 5 configurations of wind-electrolyser systems, identify the most cost-effective business cases, and conduct a sensitivity analysis of key economic parameters. Our results reveal that electricity cost is the dominant contributor to LCOH, followed by the electrolyser cost. Our work highlights the crucial role that location, market arrangements and control strategies among RES and hydrogen investors play in the economic feasibility of deploying green hydrogen systems. Policies that subsidise low-cost electricity access and optimise deployment can lower LCOH, enhancing the economic competitiveness of green hydrogen.

Figures

Figures reproduced from arXiv: 2509.00136 by the authors.

**Figure 2.** Key features considered when conceptualising potential use cases [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗

**Figure 3.** Illustration of use cases assumed in this study [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗

Figures from the paper (2 more)

**Figure 4.** Figure 4: Cost distribution as percentage of total cost of hydrogen production [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]

**Figure 5.** Figure 5: Use case III-a: Sensitivity of LCOH: electricity price between investors [PITH_FULL_IMAGE:figures/full_fig_p005_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

14 extracted references · 14 canonical work pages

[1]

Electrification versus hydrogen for UK road freight: Conclusions from a systems analysis of transport energy transitions,

M. J. Haugen et al., “Electrification versus hydrogen for UK road freight: Conclusions from a systems analysis of transport energy transitions,” Energy Sustain. Dev. , vol. 68, pp. 203–210, 2022

work page 2022
[2]

UK Hydrogen Strategy,

HM Government, “UK Hydrogen Strategy,” tech. rep., 2021

work page 2021
[3]

Techno-economic analysis of large-scale green hydro- gen production and storage,

A. Vives et al. , “Techno-economic analysis of large-scale green hydro- gen production and storage,” Appl. Energy, vol. 346, p. 121333, 2023

work page 2023
[4]

Assessment of hydrogen production costs from elec- trolysis: United States and Europe,

A. Christensen, “Assessment of hydrogen production costs from elec- trolysis: United States and Europe,” Int. Counc. Clean Transp. , vol. 1, no. 1, pp. 1–64, 2020

work page 2020
[5]

Current status of water electrolysis for energy storage, grid balancing and sector coupling via power-to-gas and power-to- liquids: A review,

A. Buttler et al., “Current status of water electrolysis for energy storage, grid balancing and sector coupling via power-to-gas and power-to- liquids: A review,” RSER, vol. 82, pp. 2440–2454, 2018

work page 2018
[6]

Electricity generation costs 2023,

DESNEZ, “Electricity generation costs 2023,” tech. rep., 2023

work page 2023
[7]

Economic analysis of a high-pressure urban pipeline concept (HyLine) for delivering hydrogen to retail fueling stations,

M. Penev et al. , “Economic analysis of a high-pressure urban pipeline concept (HyLine) for delivering hydrogen to retail fueling stations,” Transp. Res. D: Transp, Environ. , vol. 77, pp. 92–105, 2019

work page 2019
[8]

Quarterly energy prices: March 2025,

DESNEZ, “Quarterly energy prices: March 2025,” tech. rep., 2025

work page 2025

Show all 14 references

[9]

Splitting the difference — reducing the cost of electrolytic hydrogen,

Renewable UK & Hydrogen UK, “Splitting the difference — reducing the cost of electrolytic hydrogen,” tech. rep., 2025

2025
[10]

Power to hydrogen: A geospatial and economic analysis of green hydrogen for UK high-heat industry,

G. De Castro et al. , “Power to hydrogen: A geospatial and economic analysis of green hydrogen for UK high-heat industry,” Fuel Process. Technol., vol. 273, p. 108221, 2025

2025
[11]

The cost of clean hydrogen from offshore wind and electrolysis,

S. J. P. Hill et al., “The cost of clean hydrogen from offshore wind and electrolysis,” J. Clean. Prod. , vol. 445, p. 141162, 2024

2024
[12]

Hydrogen insights 2023 the state of the global hydrogen economy with a deep dive into renewable hydrogen cost evolution,

Hydrogen Council and McKinsey & Company, “Hydrogen insights 2023 the state of the global hydrogen economy with a deep dive into renewable hydrogen cost evolution,” tech. rep., 2023

2023
[13]

Clean hydrogen production cost scenarios with pem electrolyzer technology,

M. Hubert et al. , “Clean hydrogen production cost scenarios with pem electrolyzer technology,” tech. rep., DOE, 2024

2024
[14]

Global hydrogen review 2023,

IEA, “Global hydrogen review 2023,” tech. rep., 2023

2023

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.

[1] [1]

Electrification versus hydrogen for UK road freight: Conclusions from a systems analysis of transport energy transitions,

M. J. Haugen et al., “Electrification versus hydrogen for UK road freight: Conclusions from a systems analysis of transport energy transitions,” Energy Sustain. Dev. , vol. 68, pp. 203–210, 2022

work page 2022

[2] [2]

UK Hydrogen Strategy,

HM Government, “UK Hydrogen Strategy,” tech. rep., 2021

work page 2021

[3] [3]

Techno-economic analysis of large-scale green hydro- gen production and storage,

A. Vives et al. , “Techno-economic analysis of large-scale green hydro- gen production and storage,” Appl. Energy, vol. 346, p. 121333, 2023

work page 2023

[4] [4]

Assessment of hydrogen production costs from elec- trolysis: United States and Europe,

A. Christensen, “Assessment of hydrogen production costs from elec- trolysis: United States and Europe,” Int. Counc. Clean Transp. , vol. 1, no. 1, pp. 1–64, 2020

work page 2020

[5] [5]

Current status of water electrolysis for energy storage, grid balancing and sector coupling via power-to-gas and power-to- liquids: A review,

A. Buttler et al., “Current status of water electrolysis for energy storage, grid balancing and sector coupling via power-to-gas and power-to- liquids: A review,” RSER, vol. 82, pp. 2440–2454, 2018

work page 2018

[6] [6]

Electricity generation costs 2023,

DESNEZ, “Electricity generation costs 2023,” tech. rep., 2023

work page 2023

[7] [7]

Economic analysis of a high-pressure urban pipeline concept (HyLine) for delivering hydrogen to retail fueling stations,

M. Penev et al. , “Economic analysis of a high-pressure urban pipeline concept (HyLine) for delivering hydrogen to retail fueling stations,” Transp. Res. D: Transp, Environ. , vol. 77, pp. 92–105, 2019

work page 2019

[8] [8]

Quarterly energy prices: March 2025,

DESNEZ, “Quarterly energy prices: March 2025,” tech. rep., 2025

work page 2025