Cost-Aware Dynamic Bayesian Coalitional Game for Energy Trading among Microgrids

Abstract

The future electricity distribution system will be highly impacted by the emergence of peer-to-peer energy trading within microgrid (MG) communities. The idea of peer-to-peer energy trading is to export the surplus energy of a MG to a nearby MG or a group of MGs whose electrical load exceeds their generation. The variations in demand and generation, and the dynamic nature of these communities result in uncertainty on whether MGs will be able to satisfy their trading commitment or not. In this paper, the problem of energy trading among MGs is addressed with the objective of minimizing the cost under uncertainty. A Bayesian coalitional Game (BCG) based scheme is proposed, which helps the MGs to minimize the overall cost by forming stable coalitions. The results show 15% to 30% improvement in terms of cost minimization compared to an existing Q-learning based scheme and a conventional coalitional game theory (CG)-based approach from the literature.