Battery time of discharge setting for maximum effectiveness in a distribution smart grid application

Distributed Generation (DG) is a feature of smart grids in power distribution networks. The DG comprises of various types of renewable energy. Battery storages may be used along with the DG sources to store their produced energy and then release it at a proper time. Most of the current schemes discharge the stored energy based on a timer, which normally start the discharging cycle at a fixed expected peak time. But, the peak time in a distribution network does not remain at a fixed time. This paper proposes a novel intelligent method to determine a suitable time for discharging a battery based on a dynamic forecast of the peak time. A combination of fuzzy logic and artificial neural network has been used to forecast electrical power load up to four hours ahead. Another FLS is used to estimate the possibility of the current time being close to a peak period, which is represented by a factor called peak possibility factor (PPF). Based on the maximum forecasted power output of the ANN among the four outputs, i.e. 1 hour ahead to 4 hours ahead forecasts, and the calculated PPF, the starting time of the discharge cycle will be decided. © 2014 IEEE.