School of Electrical Engineering, Southeast University, Nanjing 210096, China

In the context of energy transition, large-scale distributed photovoltaic (PV), energy storage systems and interactive loads are integrated into the distribution network. In dispatching and optimization models, due to differences in resource characteristics, various entities, such as distribution networks and end-users, possess the potential to act as flexible control variables or may still exhibit randomness. These entities may demonstrate flexibility when their load interaction capabilities are strong, energy storage capacities are high, and prediction errors are low; conversely, they may exhibit uncertainty. Firstly, this paper investigates transformation mechanisms of flexibility and uncertainty and derives general transformation conditions between control variables and parameters of various entities based on robust optimization results with aggregated power. Secondly, considering the optimal controllable or uncertain range of power aggregation under temporal coupling, a two-stage robust optimization approach is employed to delineate the controllable range of control variables and the uncertain interval of parameters. Finally, case studies are conducted to validate the effectiveness of the transformation conditions and characterization methods, which provides support for the modeling and rapid solving of multi-level coordinated optimization problems in the power grid.

This work is supported by National Natural Science Foundation of China (No. 52077035).