Distributed Network Control by Multi-agent System

This paper proposes a distributed control framework for managing the charging/discharging actions of energy storage systems (ESS), which are used with a huge number for Peer-to-Peer energy trading in wide area distributed grid. The framework consists of a consensus building algorithm and a power limited algorithm for the cooperation of the distributed grid operator and ESSs. In particular, the consensus building algorithm can tolerate the non-uniform, asymmetrical, and time-varying communication delays. The effectiveness of the proposed system was evaluated by numerical tests, which are based on the actual setting of a real peer-to-peer power trading system developed at Toyota Frontier Research Center(1)～(6). The test results show that, even with a maximum delay steps of 12, the demand-supply balance of the distributed grid can be successfully assured in the tolerance range ±α=0.08. Thanks to the distributed structure, the proposed control framework can be conveniently applied to large-scale power networks.