As the energy transition transforms power systems towards decentralised systems dominated by renewable energy sources, the electrification of other energy vectors drives the development of integrated multi-energy systems (MESs), especially at the local level. Local MESs coupling electricity and heat can improve energy efficiency and autonomy, reduce carbon emissions, and minimise transmission losses by matching local generation and demand. However, during the design stage of local MESs, the technical and economic role of the coupled thermal vector, particularly for system sizing and leveraging its flexibility capability to provide ancillary services, remains unclear. This thesis focuses on the optimal design of local MESs with coupled electricity and heat. A two-stage stochastic optimisation framework is developed which is adaptable and comprehensive, allowing to study several important aspects of local MES design: impact of component modelling choices on electrical storage systems; impact of individual and interdependent component sizing on technical system flexibility; integration of ancillary services and their role on optimal design; and uncertainty in future forecasting and market prices. Moreover, a novel method was developed that allows for constant flexibility calculation in relation to a time-varying reference schedule. The framework is applied to real-world case studies to provide techno-economic insights in local MES design. The explicit modelling of the thermal vector avoids oversizing of the battery energy storage system (BESS) and reduces overall costs, highlighting the importance of incorporating coupled energy vectors in real-world electrical storage design. Moreover, the developed framework enables constant flexibility provision based on timevarying reference schedules, while accounting for internal grid constraints via a convex relaxation, which represents a suitable compromise between model accuracy and computational tractability. This allows energy system operators to assess the technical flexibility potential of their MESs across multiple market products. Furthermore, incorporating market participation for local MESs demonstrates their ability to provide multiple grid services, whose revenues shape the design of the BESS and solar PV. Despite the thermal vector participating in electricity markets, only modest oversizing of thermal storage is profitable. Additionally, risk-neutral investment positions favour large BESS capacities, while riskaverse positions prefer smaller BESSs to limit high-cost tail risks under uncertainty. Finally, the frameworks are adaptable to be used by system planners of MESs in techno-economic design studies, and can be extended to incorporate additional energy vectors.
Philipp Glücker (Wed,) studied this question.