Optimal bilevel scheduling of electric vehicles in distribution system using dynamic pricing

The integration of distributed renewable energy sources (RES) and the electrification of devices raise new challenges for the distribution system operator (DSO). This paper assesses PAR versus cost tradeoff when scheduling an electric vehicle (EV) fleet. We formulate a bilevel Mixed-Integer Linear Programming optimization problem. At the lower level, we minimize individual household electricity bills using dynamic pricings. At the upper level, we aim to smooth the power load curve of a typical Brussels MV/LV transformer. We show that a small deviation from cost-only optimization can reduce significantly the Peak-to-Average Ratio of the power load curve of a transformer. Harnessing load flexibility from EV allows the DSO to manage the transformer load to avoid grid congestion and also incentivizes load aggregators to participate in the ancillary services market.