Decentralized State Estimation-Based Optimal Integral Model Predictive Control of Voltage and Frequency in the Distribution System Microgrids

The power distribution system is undergoing a shift from centralized to decentralized operation due to the large-scale integration of distributed energy resources (DER). To increase the scalability and reduce the dependency on communication networks, there is a strong need for a decentralized control framework that can maintain the nominal voltage and frequency in a microgrid. In this article, a novel decentralized state estimation-based optimal voltage and frequency restoration strategy is proposed. The framework computes the optimal integral model predicted control reference signal to the primary controller, that is unaffected by the local measurement noise, hence, ensuring stable power-sharing among the DERs. The proposed framework is built on a first-order DER model and a local/global measurements-based local state estimation technique, facilitating deployability to the grid edge devices. The capabilities of the proposed framework are tested and validated on a real-time simulator with a full system dynamic model considering an islanded section of the IEEE 123 node distribution network for the various grid events. It is observed that the framework offers significant transient performance improvement in comparison to the Linear Quadratic Regulator method, with around 70% faster voltage restoration when the communication latency is considered.