Scenario-based Robust Scheduling for Electric Vehicle Charging Games

摘要

We study the decentralised charge scheduling problem for an electric vehicle (EV) fleet participating in a demand response (DR) scheme, in the presence of price uncertainty. To cope with the unmodeled externalities acting on the price, we adopt a data-driven paradigm and represent uncertainty by means of scenarios. We enforce a partially-cooperative robust approach among the EVs, and we show how the solution of this problem can be formulated as a Nash equilibrium of a noncooperative minmax game. Due to this approach, the game is characterised by nondifferentiable objective functions: by resorting to a "lifted" game involving the DR aggregator as an additional player, we are able to recover a solution by means of a decentralised iterative algorithm. The performance of the proposed mechanism is evaluated by means of a detailed simulation analysis.