Cluster-Based Hierarchical Coordination Planning and Control Strategy for Distribution Network Generation-Load-Storage

The increasingly prominent high intermittency characteristics of a high proportion of distributed energy sources pose new challenges for the rational planning of generation-storage and the coordination control between devices in distribution networks. In light of this, this paper proposes a multi-level coordination planning and control strategy for distribution networks based on cluster partitioning method. Firstly, it introduces a comprehensive index cluster partitioning method for distribution networks that takes into account electrical distance under load forecasting. secondly, building upon the cluster framework, a two-level joint planning model is proposed. In the upper level, an addressing and sizing model for sources and storage is developed specifically for clusters. In the lower level, the objective function is set to minimize node voltage deviations, and a voltage control model is established. Then. the upper and lower-level models are solved using a particle swarm algorithm with a simulated annealing strategy and a deep deterministic policy gradient algorithm. Finally, the effectiveness of the two-tier optimization model for generation-load-storage was validated in a specific 35 kV/10 kV distribution network.