Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

One characteristic of the ongoing transition towards a sustainable energy system is the emergence of distributed energy resources (DERs), e.g., rooftop photovoltaics (PV) and home batteries. Residential consumers are no longer passive end-users with an inelastic demand but they can also generate and store electricity, and thereby actively participate in the electricity system. This paradigm shift introduces new challenges for the design of distribution grid tariffs. Traditional volumetric tariffs (in €/kWh) do not reflect the underlying cost of the grid, which is governed by the grid capacity, and as such lead to cross-subsidies, e.g., between households with and without PV. Additionally, they do not provide an adequate response to the challenge of increasing peak loads, caused by the electrification of heating and transport, which may require significant grid reinforcements.To address these challenges, policy makers are looking to increase the economic efficiency of distribution tariffs. This requires tariff structures with larger shares of fixed and capacity-based (€/kW) components, time-varying tariffs, and finer locational granularity. Economic efficiency may also be enhanced by adopting marginal cost-based methods to allocate network costs to the tariff structure components, instead of the traditional historical cost-based methods.The first main objective of this dissertation is to investigate the direct impact of different tariff structures on the operational and investment decisions of active residential consumers, as well as the indirect impact on passive consumers and wholesale markets, via the interactions of these active consumers. To this end, equilibrium models are developed, capable of interlinking decision processes endogenously, leveraging concepts from game theory. The first equilibrium model represents a non-cooperative game between consumers shifting distribution costs to each other while cost recovery for the distribution system operator (DSO) is ensured. A case study illustrates that different fixed, volumetric, and capacity-based tariff structures induce varying operational schedules of a PV-battery system, thereby enabling active consumers to achieve cost savings at the expense of passive consumers.By incorporating the first equilibrium model in a wholesale market equilibrium model, the relationship between distribution tariffs and wholesale electricity markets is also endogenized. The developed model is used to investigate how different distribution tariff structures in one country affect the welfare and PV-battery investments in another country via coupled wholesale markets. A case study demonstrates that there can be both positive and negative welfare and storage investment spillovers. The underlying mechanisms, considering both sunk and variable network costs, are presented in depth. The results provide relevant insights for the policy discussion on the potential harmonization of tariff structures in Europe.The second main objective of this dissertation is to investigate the gap between the theory and practice of network cost allocation based on long-run marginal costs (LRMC). In an idealized setting, a coincident peak tariff signaling the LRMC has been shown to be efficient. In practice, however, the LRMC is not easy to define due to lumpy network investments. As a result, engineering estimates of the LRMC are required, using forward-looking cost models that combine network expansion planning models and peak demand forecasts.Considering investment lumpiness and elastic demand functions, the optimal coincident peak network tariff is analytically derived. The optimal tariff level, although it can be construed as the marginal network cost, is shown to be dependent on the characteristics of the network and the demand. This result demonstrates that optimal forward-looking network tariffs require information on the willingness to pay for peak demand. This is typically unknown to regulators and is thus not accounted for in practical forward-looking cost models.To gain further insights on the efficiency of forward-looking cost models in practice, a social welfare analysis of the specific Long-Run Incremental Cost (LRIC) methodology is performed. It is compared to a traditional historical cost allocation method and a theoretical benchmark. The results show that LRIC always achieves a social welfare gain compared to historical cost allocation but that the magnitude of this gain varies significantly with the demand growth rate, demand elasticity, and network upgrade cost. The mechanisms driving the social welfare gains are also analyzed, revealing that these are sometimes driven by network cost savings, and sometimes by an increase in consumer surplus.