A privacy-preserving approach to grid balancing using scheduled electric vehicle charging

The introduction of renewable energy generation (e.g. solar and wind) in the energy distribution infrastructure makes balancing the total energy load and production in the grid more challenging due to the weather-dependent nature of these energy sources. One approach to mitigate the issue is to use weather forecasts to predict the production and then offer incentives to electric vehicle users (EVUs) to charge their vehicles during the times of energy surplus. However, doing this without leaking sensitive information about the EVUs location and identity presents challenges to the system design. This thesis proposes a privacy-preserving architecture that allows the grid operator to offer incentives for contributing to the grid stability, and to reliably and automatically quantify the extent of each contribution while still maintaining the privacy of the EVUs. Furthermore, the architecture enables decentralised privacy-preserving dispute resolution without leaking any personally identifiable information (PII). The architecture fulfils the goal by utilising self-sovereign identity technologies, such as decentralised identifiers (DIDs), and privacy-preserving digital credentials solutions, such as verifiable credentials (VCs). They allow the solution to utilise ephemeral identifiers and to compartmentalise the information into three different knowledge domains to ensure that only the minimum amount of information needed crosses any domain border. An analysis of the solution indicates that the architecture ensures relatively strong privacy guarantees to the EVUs and solves the grid balancing problem while reducing the number of assumptions to the minimum. This makes the architecture applicable to a wide set of use cases in the EV charging field. Future work includes a detailed performance analysis of a proof-of-concept (PoC), although the information available from related research already indicates relatively low latency and a good level of deployability even on resource-constrained Internet-of-things (IoT) devices