Two-stage mechanism design for energy trading of strategic agents in energy communities

This paper proposes a mechanism design for energy trading in energy communities. Each community member is modeled as a strategic agent making strategic actions in interaction with other agents. The proposed mechanism includes two stages, in which in the first stage agents iteratively bid in a day-ahead market, and schedule their energy profiles for the whole operation horizon. Then, in the second stage, agents participate in a one-shot auction for energy trading in each time slot. The two-stage design allows agents to reduce their energy costs by having access to better prices in ahead markets, and also, by granting agents the ability to adjust their bids near real-time, based on their updated demand/generation profile. It is shown that, under mild assumptions on bidding of agents in stage two, the proposed mechanism is Bayes–Nash incentive compatible. Moreover, it is proven that the mechanism is ex-post budget balanced, and individually rational. Case studies demonstrate that the proposed mechanism can reduce the energy costs of the community and helps to reduce the reliance of the community on upstream networks.