A distributed constraint satisfaction approach for reactive power sharing in microgrids

Decentralized droop-based control has gained enormous attention in the operation of islanded microgrids (IMGs) over the last decade due to its superiority over centralized control schemes. However, poor reactive power sharing has been identified as a major drawback that limits the implementation of the communicationless droop control in IMGs. In order to preserve the decentralized control structure in IMGs, distributed communication-based reactive power sharing has been recently mentioned as a complementary process for the decentralized droop control. This paper proposes a scalable and solid mathematical approach for distributed reactive power sharing in IMGs. First, the problem of reactive power sharing in IMGs has been formulated mathematically as a distributed constraint satisfaction problem in a multi-agent environment, where the local controller of each distributed generation is defined as a control agent. The proposed formulation aims to achieve the desired reactive power sharing among the control agents while ensuring the satisfaction of other IMG operation requirements including bus voltages limits and line current capacities. Second, an asynchronous weak commitment (AWC) technique has been proposed to solve the formulated distributed constraint satisfaction problem. The proposed technique tends to search for voltage droop parameter settings of the control agents based on an asynchronous peer-to-peer cooperative protocol. Several case studies have been carried out to validate the effectiveness of the proposed algorithm and test its performance and convergence characteristics. The results have shown that the proposed approach can achieve accurate reactive power sharing and satisfy the IMG operation constraints under different operating conditions. Further, the proposed algorithm has shown fast convergence properties.