Sizing of Hybrid Energy Storage Systems Using Recurring Daily Patterns

A hybrid Energy Storage Systems (ESS) consists of two or more energy storage technologies, with different power and characteristics. Using a ESS, both high-frequency low-frequency variations can be addressed at the same time. For an accurate sizing use high-resolution data is required. However, over long periods leads to large sets, which are difficult handle. In this paper, improved motif discovery algorithm introduced find most recurring daily consumption patterns within time series interest. The pattern selected as representative for ESS. Next, simple optimization framework proposed selecting cut-off frequency low-pass filter, used allocating technologies. Finally, approach applied battery-flywheel ESS four low voltage distribution grids in southern Germany using real measurement data. It demonstrated that characteristics derived from only, effectively provide their intended grid services days during whole period series.