Two-stage spatiotemporal decoupling configuration of SOP and multi-level electric-hydrogen hybrid energy storage based on feature extraction for distribution networks with ultra-high DG penetration

Abstract

Driven by the “dual carbon” goals, some county-level distribution networks (DN) in China have exhibited ultra-high penetration of distributed generation (DG), resulting in operational issues such as power and energy imbalances and voltage violations. These issues place greater demands on the coordinated utilization and optimal configuration of flexibility resources across feeders. Moreover, the strong spatiotemporal coupling of these resources significantly increases the complexity of system modeling and solution. To address these challenges, this paper proposes a two-stage spatiotemporal decoupling approach for configuring a multi-level electric–hydrogen hybrid energy storage and multi-port soft open point (MEH-SOP) system, based on feature extraction. First, an MEH-SOP system is developed to enable energy coordination across intra-day to inter-week timescales and resource sharing among feeders. Second, a multi-scale spatiotemporal coordination mechanism is established for the MEH-SOP system. A two-stage spatiotemporal decoupled configuration model is then formulated, where the interactive power between the multi-level electric–hydrogen hybrid energy storage (MEH) and soft open point (SOP), along with time-of-use (TOU) pricing, serves as a linkage between the two stages. This enables coordinated configuration of MEH in the temporal dimension and SOP in the spatial dimension. Meanwhile, to further reduce the computational complexity of the MEH configuration model, a trend feature extraction strategy based on seasonal-trend decomposition using loess (STL) is introduced. The charging/discharging states (0/1) of seasonal hydrogen storage (SHS) are preset to simplify the model. Finally, simulations are conducted on a modified 31-bus Taiwan distribution system. The results show that the proposed approach not only ensures high configuration accuracy but also significantly improves computational efficiency. It effectively mitigates energy imbalances and voltage violations under multi-scale spatiotemporal conditions, showing strong engineering applicability and promising prospects for practical deployment.