Model predictive control of energy systems with hybrid storage

In this work an algorithm to control the power flow of an electric power system with two integrated energy storage systems is investigated. The power system under consideration consists of a conventional distribution feeder that supplies the power to satisfy the customers' demand, a set of photovoltaic (PV) panels that also contribute to the power generation, one unit of Lithium-Ion battery storage for the intra-day use and a combined power-to-gas (PtG) and gas-to-power installation that converts the power excess in the summertime into hydrogen and injects power back to the system in the wintertime. The proposed control algorithm is based on model predictive control tailored for the energy system under investigation. To demonstrate the performance of the proposed control, a set of synthetic PV and demand profiles representing future conditions in Switzerland were created and used as input data to the system model. The synthesized generation and consumption data span a whole year of operation. A number of detailed simulations performed in the framework of the study reported in this paper demonstrate the effectiveness of the proposed control algorithm and provided invaluable insights into the optimum operation of the complex integrated power system.