The large-scale integration of variable renewable energy sources (VRES) brings about challenges related to the stochastic characteristics in supply. In this light, a recent and promising approach which has received increasing attention is that of multi-energy systems (MES). A MES is defined as an energy system whereby different energy sectors interact with each other, and builds on the notion of considering the optimization of the whole energy system instead of focussing on specific energy sectors. The increase in substitutions between different energy carriers offer increases in overall efficiency and system flexibility, compared to non-integrated energy system (NIES). This thesis presents a MES investment model. The model is able to systematically find the cost-optimal investment portfolio for meeting three types of demand (electricity, heating, and hydrogen). The outcomes consist of the capacities and locations of generation and conversion technologies, as well as electricity and hydrogen network lines and storage technologies, simultaneously, whilst adhering to balance and RES target constraints. The model is an extension of the current cost-optimal energy investment Greenfield Renewables Investment Model (GRIM) and allows the incorporating of spatial characteristics of VRES supply, leading to a more realistic representation of the real energy system. Furthermore, the model is able to account for limitations on maximum available land for onshore wind and solar photovoltaics (PV).
A case study was conducted on the energy system of the Netherlands. Fifteen scenarios were run, for increasing RES targets, different levels of sector integration and land-use restriction for solar and onshore. The results show that increasing sector integration can lead to a cost reduction of 25% in a fully renewable scenario. Increasing connections between electricity, heating and hydrogen sectors provides system flexibility, resulting in a decrease in generation, network, and storage investment. Furthermore, the results highlight the importance of considering the spatial components: both limitations on land-use and spatial-temporal VRES supply profiles have significant impact on the optimal solution. For the Netherlands, imposing land-use constraints resulted in a decrease in market share from onshore and an increase in solar, since the maximum amount of available land for onshore wind is reached, when including three types of demands. The academic relevance of this thesis is found in the modelling approach developed. In addition, this thesis builds on existing energy planning literature by providing a case study of the future energy system of the Netherlands. Future research should analyse different objectives and include additional generation technologies.