Spatially-explicit optimization of an integrated wind-hydrogen supply chain network for the transport sector: The case study of Sicily

Abstract

Almost half of the European railways are served by diesel-powered trains used for both cargo and passenger transport. Similar considerations apply to the bus sector in urban and suburban areas. The decarbonization of the transport sector can be achieved by using hydrogen as a fuel for trains and buses. However, the main barrier to adopting hydrogen fuel cell trains and buses is the construction of adequate infrastructure. Thus, to foster a sustainable energy transition, existing renewable plants should be considered when designing new hydrogen supply chains. In the proposed Sicilian case study, existing wind farms are chosen as electricity production plants. Most of them have been shut down due to grid unbalancing issues; still, the curtailed renewable electricity could be used to produce zero-carbon hydrogen via water electrolysis, significantly reducing costs. To account for this opportunity, this study models the hydrogen production, transportation, and refueling stages, employing a mixed-integer linear programming approach to find the optimal location and capacities of hydrogen infrastructure, while minimizing the total daily cost of the supply chain. The model is applied to Sicily and different scenarios for varying hydrogen demands for trains and buses are analyzed and discussed. Results show that capital expenses cover more than 90% of the total cost of the supply chain, with a hydrogen cost dispensed at the filling station ranging from 6.32 to 9.02 €/kgH2. The environmental analysis shows that as hydrogen demand rises, so do the carbon emissions from the distribution stage. However, the impact of using wind-generated electricity in terms of avoided emissions is notably greater. Finally, it is shown that transporting hydrogen in gaseous form using tube trailers is economically more attractive than transportation in liquid form with tanker trucks in the early stages of the network.