Evolutionary battery scheduling optimization under variable electricity prices in micro-grids with renewable generation

Abstract

In this paper we propose an Evolutionary Algorithm (EA) to tackle the Battery Scheduling Optimization Problem (BSOP) in MicroGrids (MG). Specifically, we consider a MG that includes renewable generation, different load profiles, and is equipped with an energy storage device (battery) to address its scheduling (charge/discharge duration and occurrence) in a real scenario of variable electricity prices. We fully describe the proposed evolutionary algorithm, including its initialization and the different operators implemented to guide the search. Experiments in a MG with residential and industrial consumption profiles, photovoltaic and wind power generators’ profiles, together with an energy storage device are carried out. To show the good battery scheduling performance of the proposed EA, we have compared the results with what we called a deterministic scenario. It is defined as a fixed way of using the energy storage device, that only depends on the pattern of load and generation profiles considered. Hourly values of both generation and consumption have been considered, and the good performance of the proposed EA is shown for four different weeks of the year (one per season), where the effect of the battery scheduling optimization obtains savings up 10% of the total electricity cost in the MG, when compared with the deterministic approach.