Abstract: The widespread integration of distributed resources and the volatility of uncontrollable resources such as wind and solar power pose significant challenges to the operation and control of power systems. This paper investigates optimal scheduling strategies for distributed cluster resources under the uncertainty of renewable energy, exploits the flexibility of various controllable resources, and enhances the utilization rate of centralized uncontrollable resources. To achieve the flexible control of distributed controllable resources, this paper first proposes a time-variant adjustable potential boundary computation method for aggregating data centers and photovoltaic-energy storage clusters, ensuring their aggregation optimality and decomposition feasibility. Furthermore, the ∞-Wasserstein fuzzy set is constructed based on historical data to capture the stochastic characteristics of wind power, providing reliable out-of-sample guarantee. Finally, an adaptive polyhedral approximation method is proposed to address the inherent infinite-dimensional solving challenge in the distributionally robust optimization problem. The two-stage distributionally robust scheduling model is transformed into a finite-dimensional problem to achieve the rapid solution. The effectiveness of the proposed method is verified based on the simulation on the modified IEEE-RTS 24-bus system.