Cluster-based Analysis of Feasible Operating Regions for Aggregated Energy Storage Systems

This paper presents a cluster-based analysis of the time-coupled feasible operating region of aggregated energy storage systems in context of upstream flexibility provision from distribution grids. We utilize a method based on markov-chains to sample the solution space of available flexible power and a k-means algorithm to geometrically cluster the time-series of flexibility demands. The results are analysed in terms of the deviation of realized to planned flexible active power. Our method is applied to an exemplary scenario of aggregated storage systems, which exhibits aggregation errors within the feasible solution space due to loss of information in aggregation.