Energy storage systems can provide a solution for the current challenges derived from the increasing penetration of renewable energies. Each energy storage system has different characteristics so their combination can be the best solution to achieve the requirements of a given scenario. To achieve the maximum potential of the Energy storage system they must be supplied with an optimal control strategy. Traditional control strategies only focus on increasing self consumption and do not take into consideration future generation and load. Model predictive control can use load and generation forecasts to provide a multi-objective solution which takes into consideration energy storage system degradation, grid congestion and self consumption between others. Neural networks are used to obtain the generation and load forecast, trained with empirical data from real households. An online model based predictive controller implemented for a grid composed by one lithium-ion battery, one vanadium redox flow battery, photovoltaic generation and electric consumption of 14 households. Finally the results of the classical method of maximizing self consumption, the ideal predictive controller considering perfect forecast and the real predictive controller are shown and discussed.

This research is part of the CSIC program for the Spanish Recovery, Transformation and Resilience Plan funded by the Recovery and Resilience Facility of the European Union, established by the Regulation (EU) 2020/2094, CSIC Interdisciplinary Thematic Platform (PTI+) Transicion Energetica Sostenible+ (PTI-TRANSENER+ project TRE2103000), the Spanish Ministry of Science and Innovation under projects MAFALDA (PID2021-126001OB-C31 funded by MCIN/AEI/ 10.13039/501100011033 / ERDF,EU), and MASHED (TED2021-129927B-I00), and by the Spanish Ministry of Universities funded by the European Union - NextGenerationEU (2022UPC-MSC-93823)

Trabajo presentado en el 28th International Conference on Emerging Technologies and Factory Automation (ETFA), celebrado en Sinaia (Rumanía), del 12 al 15 de septiembre de 2023

Peer reviewed