Fully decentralized peer-to-peer energy sharing framework for smart buildings with local battery system and aggregated electric vehicles

Abstract

Decentralized peer-to-peer energy sharing techniques are highly promising to become the next-generation regime for smart building energy management, which can impulse the realization of nearly net-zero energy buildings. In this context, this paper proposes a comprehensive energy sharing framework for smart buildings in considering multiple dynamic components covering heating, ventilation, air conditioning (HVAC), battery energy storage systems (BESS) and electric vehicles (EVs). Specifically, both the power loss and shadow price of shared energy are explicitly modeled in a combined optimization framework, which is aimed to maximize the social welfare through peer-to-peer energy cooperation. Moreover, the role of agents acting as producers or consumers can be endogenously determined in the proposed model. In addition, distinguished with the classical distributed algorithms, we develop a fully decentralized algorithm based on dual-consensus version of alternating direction method of multipliers (DC-ADMM). The proposed algorithm avoids the need of coordinators at both the primal and dual variable updates in the iteration process, which suggests distinctive merits on high-level privacy protection as compared to most of the distributed optimization-based methods. Extensive case studies based on a smart building community demonstrate that the proposed peer-to-peer transactive framework can admirably improve the overall welfare for the involved smart buildings.