Abstract: The environmental externality produced in the construction and operation of virtual power plants（VPPs） affects the benefits of the economic and environmental operation. First, the unit environmental externality cost of coal power, wind power and photovoltaic power is calculated separately and included in the unit power generation cost of VPPs. On this basis, a two-stage distributed robust optimization model of the prediction error fuzzy set for wind and photovoltaic power output is built based on the Wasserstein distance. The model enables the VPP to adjust the flexible resources under the condition that the prediction errors of wind and photovoltaic output satisfies the worst distribution of the fuzzy set, so as to realize the minimum operating cost,and the scheduling and decision problems of the VPPs are studied. Finally, the example simulation and comparative analysis verify the positive effect of considering environmental externality on the economic and low-carbon operation of VPPs, the applicability and the computing simplicity of the two-stage distributed robust optimization model.