reement, adopted by virtually all 195 countries of the United Nations, is a binding deal with targets to keep global warming of the earth beneath 2∘C, and reduce emissions of greenhouse gases in 2050 by at least 80% compared to 1990. These targets require significant reductions in energy consumption and switch to carbon free renewable energy sources. Changing to renewable energy sources brings several complications to the electricity system. Supply and demand of electricity need to match at any time. The intermittent nature of renewable energy sources, such as solar and wind, requires energy storage and need to be balanced with dispatchable power generation. Passenger cars could offer dispatchable power as they are parked for 95% of the time. Especially fuel cell electric vehicles (FCEVs) connected to the grid could offer this service in a clean and efficient way. Could parked and unused grid connected FCEVs replace the positive dispatchable power plants to balance 100% renewable energy systems? This research investigates how future 100% renewable national electricity, heating and transport systems can be balanced with the use of hydrogen production and storage, and grid connected FCEVs. A model is developed that simulates the future energy systems of Germany, France, Spain, Great Britain, Denmark and Belgium. The energy systems include electricity generation and consumption, road transport, hot water and space heating. Road transport vehicles are battery electric vehicles (BEVs), FCEVs or a combination of both. Electricity and hydrogen are the only energy carriers. Electricity is mainly supplied by solar and wind power. Hot water and space heating is mainly supplied by solar thermal energy and electric heat pumps. Electricity generation and consumption profiles and temperature data of 2014, 2015 and 2016 serve as inputs. The future 100% renewable energy scenarios are based on scenarios published by government agencies, research institutions or transmission system operators. Demand response heating (DRH) is analysed and applied to all cases. Interconnecting the electricity grids of Germany and France is also investigated.
Hydrogen production and grid connected FCEVs can balance national electricity grids. Electrolysers can act as negative balancing power consuming excess electricity of intermittent renewable energy sources to produce hydrogen. Roughly 0.4-0.6 GW of electrolyser capacity is required to balance 1 GW of renewables in the investigated countries. This requirement can be lower when curtailment is applied and the installed capacity of renewables is slightly increased. Hydrogen can be stored locally in high pressure storage tanks or at large scale in underground salt caverns. A typical salt cavern can store around 6 million kg of working gas. Per TWh of final energy consumption approximately 1-2.5 million kg of hydrogen storage capacity is required. Conventional positive balancing plants such as gas turbines could be replaced by FCEVs connected to the electricity grid. Peak backup demands can be balanced with 25 - 50% of the FCEV passenger fleet, which corresponds to 12.5 - 25% of the total passenger car fleet. The utilisation of FCEVs for V2G varies from 2.5 to 8%. Between 7% and 14% of the electricity consumption in a country is supplied by grid connected FCEVs. Hydrogen can be locally produced at hydrogen fuelling stations or electrolysers can be installed near large scale electricity generation sites or salt caverns where hydrogen can be produced and stored directly. Hydrogen fuelling stations need an average dispensing capacity of 3000 kg/day (∼600 passenger FCEVs/day) to cover all fuelling demands except peak demands.

In the Netherlands the main sources of electricity are currently coal and gas fired power plants. Due to the increasing share of electricity that is produced from renewable energy sources, the operational hours of these conventional power plants are decreasing. However, these power plants are also the main source of frequency reserves, which are required to guarantee the stability of the electricity grid. Stability is maintained if there is a real time balance between the electricity production and consumption. Due to the decreasing availability of conventional power plants, the possibility to offer frequency reserves with other power sources must be explored. A possible provider is the Car Park as Power Plant (CPPP). This is an aggregation of Fuel Cell Electric Vehicles (FCEV) parked in a car park and operating in Vehicle to Grid (V2G) mode. This thesis contains a technical and economic feasibility assessment of a Car Park as Power Plant offering frequency reserves in a future power system with a low share of conventional power plants.

The dynamics of the frequency was analysed considering the reduction of the operational hours of the conventional power plants. This causes the inertia in the system to decrease. As a consequence, this increases the rate of change of frequency (RoCoF). A high RoCoF makes the frequency react faster which will make it more feasible that the maximum instantaneous frequency deviation will be reached. When this maximum deviation is reached, the system enters the alert state, which could endanger the global security of the system. To prevent the frequency from reaching the maximum instantaneous frequency deviation under conditions of a high RoCoF, the full activation time (FAT) of the frequency reserves could be reduced. The FAT is the maximum time between the moment that the signal for a change in power output is given and the actual moment that the required power output is reached. Reserves with a reduced FAT are referred to as fast frequency reserves. By measuring the FAT of the FCEV in V2G mode with an experimental setup, it was tested if the FCEV could offer the fast frequency reserves. The FCEV appeared to be a suitable power source to offer fast frequency reserves. When the power output of the battery and the fuel cell are combined, an even higher power gradient and thus a shorter full activation time, can be obtained. However, improvements must be implemented in the V2G discharge unit and the energy management system of the FCEV to optimise the operation in V2G mode. The economic feasibility of the CPPP as provider of frequency reserves was then analysed. The factor that has the highest impact, is the position of the CPPP on the bid ladder. Only when the price for frequency reserves is relatively high, the reserves of the CPPP will be activated. This is caused by the high price for hydrogen, which is the dominant factor in the marginal costs of the CPPP. Most of the time the prices for frequency reserves and for electricity sold on the spot market are lower than the marginal costs of the CPPP.

From a technical point of view the CPPP is a suitable power source to offer fast frequency reserve. The profitability of the CPPP is, however, strongly related to the occupation pattern of the car park and the price for frequency reserves. The occupation pattern can be influenced by the aggregator by giving incentives to the car owners, combining different car parks with deviating occupation patterns or by operating company car parks with, for example, autonomous driving cars. The price for frequency reserves is dependent on the quickly evolving market and can not be influenced by the aggregator of the CPPP. The aggregator can influence the marginal costs of the CPPP. By adding relevant components to the system, like hydrogen production units or storage facilities, the aggregator can offer reserves with different power sources, which will have different marginal costs. From the point of view of the transmission system operator, it could be possible to add an extra product for fast frequency reserves instead of changing the requirements of the existing reserves. This could have a positive effect on the profitability of the CPPP. The specific requirements such as the minimum bid size, validity period and payment mechanism for this product should be evaluated in further studies.