Abstract: To address the risks arising from the uncertainties in renewable energy output and load demand, a robust optimal scheduling model of virtual power plant (VPP) integrating wind, photovoltaic, load and storage operation is proposed. A two-stage robust optimization model, constructed in the form of min-max-min, is designed based on the uncertainties of the source load, aiming to maximize system revenues and minimize penalty costs as well. Firstly, in the pre-scheduling stage, the output plan with the maximum VPP income is formulated according to the predicted value of the source load side. The second stage of VPP involves combining the decisions made in the previous stage and utilizing means such as power purchase, sales, and energy storage systems for rapid output regulation. These aforementioned measures are employed to manage the fluctuations of uncertain variables and ultimatelymaximize operating benefits even in the worst-case scenario. Thirdly, The algorithms utilized in the interaction iteration are based on pairwise transformations and columnand-constraint generation (C&CG) techniques. Finally, the experimental simulation results not only validate the economics, robustness and stability of the model, but also demonstrate that the optimized scheduling scheme contributes to reducing the fluctuations caused by uncertainty, which ultimately achieves a balance between the economic benefits and operational risks of the VPP.