Hydrogen gas production from offshore wind

Abstract

The decarbonisation of the Dutch power system cannot be accomplished by electrification alone. Although offshore wind power is to play a leading part in the future Dutch power system, long-term energy storage and hard-to-electrify energy consumers are in need of other energy carriers as well. Hydrogen gas is able to provide this system essential. For the production of green hydrogen gas, offshore wind is a promising option because it has a high capacity factor, relative to other sustainable energy technologies. Vice versa, hydrogen production can strengthen the business case of offshore wind as the average power price will decreases as a result of increasing power grid penetration from renewable energy power generators. Furthermore, the expansion of offshore wind in the Netherlands may lead to high cable costs with increasing distances to shore and onshore grid congestion. Hydrogen production can circumvent these problems while continuing the decarbonisation of the Dutch energy system. This study aims to determine the most profitable offshore hydrogen wind hub configuration for 2030 and 2040. To determine the optimal configuration, this thesis evaluates two main considerations. First, whether onshore or offshore electrolysis is more profitable. Second, whether a bidirectional or unidirectional grid connection may increase the profitability of respectively the onshore or offshore electrolysis configuration. For this evaluation, six configurations are examined that aim to represent the spectrum of possibilities on how a 12 GW hydrogen wind hub could be integrated in the energy system. These configurations vary in their positioning of the electrolyser (onshore/offshore), electrolyser capacity (11.5/9.5/5.6 GW) and grid connection capacities (0/2/6 GW). To compare the profitability of each configuration, this thesis evaluates their cost and revenues. The cost analysis is conducted through a literature study. The revenue analysis is performed by modeling the hydrogen production of each configuration for a given weather year in order to establish the hydrogen production cost. Additionally, for the grid-connect configurations, the power revenues are estimated using the Plexos power market model. This evaluation reveals whether the grid connection configurations can profit from both volatile power prices (sell \& buy) and a stable hydrogen price. All analyses use a screenshot approach on the year 2030 and 2040. Thus, annual cost and annual revenues are expressed for these years. In the cost analysis, cost variations are taken into account for the HVDC-components, offshore installation factor and WACC. Also, different distances to shore of the wind hub are taken into account (i.e. 88, 209 \& 330 km). For the power revenue analysis, the power system state data on 2030 and 2040 are used from the 'Ten-year network development plan' of ENTSOG/ENTSO-E, which span over more than ten years. From this data, the 'Global ambition' scenario and the weather year 1982 are used. This 'Global ambition' scenario represents a pathway of centralised innovation. In order to establish future power prices, the Plexos power market modelling tool is used to solve an hourly Unit Commitment Economic Dispatch problem by Mixed Integer Linear Programming. This nodal model consist of 59 EU- and surrounding non-EU countries. Additionally, a hydrogen market price is established in which the feedstock sector is assumed to be price-setter as an international hydrogen market is non-existent for now. In 2030, onshore electrolysis is more economical than offshore electrolysis as the electrolyser investment cost and accompanying offshore installation cost are high. The dedicated hydrogen configurations and the grid connection configurations have equal hydrogen production cost; 3.10 €/kg. In 2040, onshore and offshore electrolysis are equally economical up to about 200 km from shore. At larger distances, offshore electrolysis is more economical. Uncertainty in the system cost and differences in system cost between onshore and offshore electrolysis and per distance to shore are found to mainly result from uncertainty in the offshore installation factor, the cost of HVDC component and the WACC. The bidirectional 2 GW grid connection of onshore electrolysis and the unidirectional 2 GW grid connection of offshore electrolysis both result in the same hydrogen production cost as the dedicated hydrogen configuration; 2.50 €/kg. Overall, considering the operation of a hydrogen wind hub through 2030 and 2040 within one life cycle, there is no significant advantage of grid connection as the break-even prices of the 2 GW grid connected configurations and the dedicated hydrogen configurations are similar. Nevertheless, a 2 GW grid connection might provide risk spreading for the investor and a lower total hydrogen subsidy amount to achieve a profitable configuration. The larger grid connection of 6 GW is definitely less profitable operating through 2030 and 2040. When assuming the future technological maturity of turbines with DC output, efficient electronic power converters (DC-DC) and electrochemical compression combined with optimistic cost reduction of the electrolyser, the hydrogen production cost decrease to 2.65 €/kg in 2030 and 2.30 €/kg in 2040. Given these production cost, the 2 GW grid connected configuration and dedicated hydrogen configuration become competitive with grey hydrogen production in 2040 based on natural gas price of 7.31 €/GJ and 80 €/tCO\textsubscript{2}. An important note is that the current CO\textsubscript{2} price already surpassed the price trajectory for 2030 as used in this study. Variations in turbine cost will affect the production cost as well. However, in this thesis the emphasis is more on the comparison of the configurations than on the absolute level of the production cost. For different turbine cost, the comparison remains similar as the turbine capacity is equal for all configurations. Therefore, a fixed cost reduction is used for the wind turbines. Future techno-economic development of in-turbine electrolysis will determine whether decentralised electrolysis becomes beneficial over centralised electrolysis. It is recommended that the North Sea offshore wind infrastructural planning must take into account the broader potential of 60 GW Dutch North sea wind in 2050 (+ 450 GW onshore in Europe) and the forthcoming power grid challenges in all North sea countries. For specifically offshore wind-based hydrogen production, electrolyser capacity planning towards and beyond 2050 should be considered as the economies of scale of hydrogen pipelines might favour offshore electrolysis as the overall cost might be lower. Future onshore electrolysis provides possible symbiosis with solar PV, nuclear \& natural gas power generation and industry thereby limiting grid congestion and increasing the electrolyser load factor. Future modelling efforts that aim to value the benefits of combined hydrogen and power production should focus on different demand-side-response and power-to-gas capacities as both affect the volatility of the power prices. Also, alternative trajectories of the CO\textsubscript{2} price and energy system scenarios should be incorporated in the power price modelling. Furthermore, the power market model can be improved by including the recovery of long run marginal cost, strategic bidding behaviour and sector coupling of power and heat.