Event-based state-space model predictive control of a renewable hydrogen-based microgrid for office power demand profiles

Abstract

This paper focuses on the design and implementation of an event-based control architecture to manage a renewable-based microgrid. This microgrid has renewable-energy generation and a hybrid energy storage system that uses electricity and hydrogen. The main load of the microgrid is the energy demand of an office. The primary control objective is to satisfy this load using the available renewable generation and stored energy while reducing the amount of energy purchased from the Utility Power Grid and the degradation of the electromechanical storage devices. To do that, the control architecture defined within an event framework, makes use of a set of state-space model predictive controllers which are selected as a function of a variable sampling period. To evaluate the performance of the proposed architecture, simulation tests for a summer day as well as an analytical study is performed. The obtained results show that the use of the event-based control architecture allows a significant reduction of the number of changes in the control action at the expense of an acceptable deterioration of set-point tracking for a microgrid with several types of electrochemical storage.