Distributed Energy Storage Optimal Scheduling in Distribution Network Based on the K-means++ Algorithm

Distributed energy storage technology can solve the problems of load peak-valley difference faced by distribution networks. Reasonable and efficient dispatch of distributed energy storage is a significant approach to play its performance in distribution network. However, the direct participation of large-scale distributed energy storages in distribution network will bring about many problems, such as the explosion of the dimension of decision variables and the difficulty of convergence of the solution results. In this paper, large-scale distributed energy storage is aggregated into a small number of characteristic clusters based on typical characteristic quantities, and an aggregation-scheduling model is established to deal with massive distributed energy storage resources participating in distribution network operation. Simulation results demonstrate that the proposed method reduces the dimension of variables and lower the level of the solving difficulty, which can proceed to schedule the distributed energy storage reasonably and intensify the feasibility in optimization.