Hyper-Heuristics Meet Controller Design: Improving Electrical Grid Performance through Microgrids

Microgrids stand as an alternative for incorporating Renewable Energy Sources into the electrical grid, but they require an adequate control scheme. Although the literature contains plenty of alternatives, it lacks implementations of hybrid controllers based on Hyper-Heuristics (HHs). Hence, we analyze whether they are of benefit. Our goal is simple: to alternate through diverse controllers as the simulation progresses. To this end, we consider some simple sequence-based selection HHs and test them across 13 scenarios. Instead of the customary low-level heuristics, we use predefined controllers that were previously tuned through a Genetic Algorithm. For the most part, at least one of the proposed models outperforms the best available controller. Thus, using HHs as an advanced control scheme seems feasible and should be explored more deeply in future works.