This project investigates the feasibility of hydrogen refueling stations using photovoltaic systems for fuel cell scooters. The aim was to identify the solar electricity generation potential of noise barriers in the Netherlands and the connection between this and future theoretical hydrogen demand for fuel cell powered two wheelers (scooters or mopeds). Noise barriers block noise from inhabited areas, which is also where mopeds and their owners reside. Noise barriers retrofitted with solar panels, or Energy Walls, coupled with a hydrogen production, storage and dispensing system is proposed. This system would enhance sustainable, carbon neutral mobility as well as contribute renewable energy to the grid electricity mix. Particulate matter, greenhouse gases and noise pollution are reduced with electric drive vehicles such as fuel cell powered two wheelers. A case study of a noise barrier on the A20 in Rotterdam Noord was modeled; other suitable locations were identified as well, such as Amsterdam or Delft. The characteristics of the barrier in Rotterdam provided a lower levelized cost of PV electricity and was thus chosen to simulate specifically. A range of electrolyzer capacities was also simulated to understand the effect of the electrolyzer capacity and the final cost of hydrogen to the user. The demand base case was 100 scooters which traveled 2.5 km/day each on average. Provided a single metal hydride canister has a range of 25 km and capacity of 45 gH2 , this demand configuration means 0.45 kgH2/day needs to be produced for the users. An hourly simulation of the energy production was modeled using two control strategies, which focused on maximizing the Energy Wall input and the use of the electrolyzer respectively. The second strategy had higher electricity costs as compared to the first, however the cost of the electrolyzer was minimal regardless of the number of panels. The cost of hydrogen depends on the size of the Energy Wall and the electrolyzer capacity; for the lowest cost setup the price of hydrogen ranged between 14.3 and 7.2 EURO/kgH2 . The cost per kilometer traveled by fuel cell scooters was found to be between 2.57 and 1.4 ¢/km. Battery electric vehicles are still much cheaper to operate in this aspect, driving at a cost mileage of between 0.3 and 0.6 ¢/km, however gas powered scooters had the highest cost per kilometer (3 - 5¢/km). This model finds that the operational costs of a hydrogen fuel cell scooter are lower than its gas scooter counterpart, but not low enough to compete with battery powered scooters. The one advantage of gas scooters is the range on one refuel, however with their relatively high operational costs and emissions, electric drive engines become favorable. Battery electric have a good cost efficiency. Fuel cell scooters are quickly recharged with canisters, and are mid-cost range to operate compared to the other two technologies.

The rapid growth of the PV industry has meant that prices have been falling steeply, making them more accessible for a wider range of applications. The combined use of noise barriers as energy generating facilities, usually coined as PVNBs has been occurring since the late 80’s. It has not realised wide spread implementation, remaining mostly in the research domain. These structures are predominantly built in urban areas with high populations and energy consumption. In an attempt to shine light on the economic potential of these projects, an engineering design study has been carried out. By focusing on the retrofitting of current infrastructure it is hoped to stimulate investment for Energy wall projects.

The initial concept was coined as the Energy wall, which differed from PVNBs in that it consisted of a hybrid wind and solar system. However, in this thesis the focus remains on the solar aspect but the name Energy wall remains. An initial investigation of the infrastructure in the Netherlands was carried out using GIS, to determine what lengths of suitable noise barriers are currently available for conversion. A dataset gained from the Rijkswaterstaat, part of the Dutch ministry of infrastructure and the environment, was the base of this analysis. Calculations were made to add relevant attributes for determining the infrastructures suitability for conversion considering nearby spatially relevant data. The output of this, was used to develop a model that would perform a multi-criteria analysis, exploring PV types, configurations and PV systems costs. The focus was around the Rotterdam ring road to allow the methodology to be developed but with the hopes that the same process could be applied to a nationwide or multinational level. System costs were implemented based on interviews and research, which allowed for comparisons to be made and costs of energy to be found. A stakeholder analysis was also used to find the relevant parties for different system sizes, from here subsidies could be applied.

The model offers a guide for pricing of major components, highlighting critical parameters that could affect the success of the project. The main methodology for system comparisons is the LCOE. The application of subsidies can make project much more attractive to cooperation’s or nearby companies for tax refunds or self-consumption. While on a large scale, economies of scale make the project competitive with conventional PV systems. Installations of this scale are rarely seen in the built environment and could have a major contribution to grid balancing.