Design and Implementation of a True Decentralized Autonomous Control Architecture for Microgrids

Abstract

Microgrids can serve as an integral part of the future power distribution systems. Most microgrids are currently managed by centralized controllers. There are two major concerns associated with the centralized controllers. One is that the single controller can become performance and reliability bottleneck for the entire system and its failure can bring the entire system down. The second concern is the communication delays that can degrade the system performance. As a solution, a true decentralized control architecture for microgrids is developed and presented. Distributing the control functions to local agents decreases the possibility of network congestion, and leads to the mitigation of long distance transmission of critical commands. Decentralization will also enhance the reliability of the system since the single point of failure is eliminated. In the proposed architecture, primary and secondary microgrid controls layers are combined into one physical layer. Tertiary control is performed by the controller located at the grid point of connection. Each decentralized controller is responsible of multicasting its status and local measurements, creating a general awareness of the microgrid status among all decentralized controllers. The proof-of concept implementation provides a practical evidence of the successful mitigation of the drawback of control command transmission over the network. A Failure Management Unit comprises failure detection mechanisms and a recovery algorithm is proposed and applied to a microgrid case study. Coordination between controllers during the recovery period requires low-bandwidth communications, which has no significant overhead on the communication infrastructure. The proof-of-concept of the true decentralization of microgrid control architecture is implemented using Hardware-in-the-Loop platform. The test results show a robust detection and recovery outcome during a system failure. System test results show the robustness of the proposed architecture for microgrid energy management and control scenarios.