High renewable energy penetration poses significant challenges to coordinated optimization between distribution networks and microgrids

including hierarchical coupling

inadequate protection of stakeholder privacy

and the difficulty of balancing robustness and economic efficiency under uncertain wind and photovoltaic generation. To overcome the limitations of existing approaches

such as centralized models requiring full data sharing

conventional distributed methods with weak robustness

and overly conservative robust optimization techniques

this paper proposes a data‑driven distributed robust optimal scheduling method for distribution network‑microgrid coordination. First

distribution networks and microgrids are modelled as independent stakeholders. Analytical target cascading and Lagrangian penalty functions are employed to decouple tie‑line power and establish a privacy‑friendly distributed optimization model. Second

to characterize the uncertainty of wind and photovoltaic generation

a Gaussian mixture model is adopted to fit historical data

construct the joint probability distribution

and perform clustering analysis. A data‑driven ambiguity set based on the Wasserstein distance is then developed

and a two‑stage distributed robust optimization model is formulated. Finally

the proposed model is decomposed and solved using a column‑and‑constraint generation method in conjunction with strong duality theory. Simulations conducted on the IEEE141‑bus and 281‑bus systems verify the feasibility and effectiveness of the proposed data‑driven distributed robust optimal scheduling method for coordinated distribution network‑microgrid operation.