Analytical studies of three different datasets for peak reductions with battery-based storage system

Abstract

Two basic controllers, namely threshold- and schedule-based controllers, have been used with different advanced features for battery-based energy storage system to perform peak reductions. However, the peak reductions are only influenced by the abilities of the newly featured controllers but also by the types of datasets being used to calculate the peak demands. There are three different datasets, namely instantaneous, average and energy load data. The main contribution of this paper is to present how different datasets can affect the peak reductions achieved by the two basic controllers without using any new features. The studies are extended to analyse how different datasets can influence the peak reductions by the controllers under different forecasting errors. The results show that threshold-based controller can achieve the highest peak reductions when the instantaneous load data is used. However, schedule-based controller is better than threshold-based controller when the levels of peak demands are highly unpredictable. These results are important because they can affect the prospective of energy storage system in the future.