Collaborative Optimal Configuration of Active–Reactive Flexible Resources Based on Wasserstein Confidence Set

Flexible resources (FRs) have significant potential in ensuring the dynamic balance between supply and demand as well as enhancing the security of active distribution networks (ADNs). However, determining the optimal FR capacity in an economically reasonable manner remains a challenging task. This paper addresses the lack of representativeness of wind turbine (WT) and photovoltaic (PV) power output scenarios in the planning stage by generating a basic set of joint WT-PV output scenarios using random sampling. Subsequently, a Wasserstein confidence set (WCS) is established based on data-driven technology to better represent the unknown distribution of the actual WT-PV joint fluctuations. This provides a more detailed description of the scenario set, enabling the precise quantification of the risk of resource allocation scenarios and enhancing the flexibility and rigor of the subsequent optimal configuration model (OCM). To improve the coordination of active–reactive FRs, a bi-level OCM with multi-timescale considerations is developed. Compared to traditional configuration methods, the proposed model not only improves economic efficiency but also ensures that system voltage remains within safe limits after configuration. The effectiveness and superiority of the proposed optimal configuration method are demonstrated through simulations on an improved 33-bus test system, where the model achieved a 9.208% reduction in annual cost compared to robust methods while maintaining voltage quality and avoiding overvoltage or equipment overloads.