A Distributed Architecture for Robust and Optimal Control of DC Microgrids

This paper presents a distributed, robust, and optimal control architecture for a network of multiple dc-dc converters. The network of converters considered form a dc microgrid in order to regulate a desired dc bus voltage and meet prescribed time-varying power sharing criteria among different energy sources. Such coordinated microgrids provide an important framework for leveraging the benefits of distributed power generation and consumption. The proposed control design seamlessly accommodates communication architectures that range from the centralized to decentralized scenarios with graceful degradation of the performance with lessened communication ability. Moreover, the methods developed are applicable to the case where the desired proportion in which the sources provide the power varies with time. A distinguishing feature of the control design approach is that it regards the net load current as a disturbance signal, lending itself to tractable analysis with tools from the robust and optimal control theory. A quantifiable analysis of the closed-loop stability and the performance of the network of converters is performed; the analysis simplifies to studying the closed-loop performance of an equivalent single-converter system. The control approach is demonstrated through simulations and experiments.