Electric vehicles and the introduction of new energy sources, which are represented by wind power and optoelectronics, improve the impact of the power system on the environment. However, the randomness and the new energy electric vehicle network output uncertainty brings new problems to the system operation and scheduling. According to this background, a bi-level multi-objective optimization model for the joint scheduling of electric vehicles and new energy sources is set up, customer satisfaction index, pollution discharge as the upper level decision, the power generation cost of thermal power unit, the fluctuation of the electric vehicle and the new energy output as the lower decision-making. The improved NSGA2 algorithm is used to solve the problem of the upper and lower layers, and the Pareto non dominated solution set is obtained, determining the optimal compromise solution for uncertain decision making based on rough sets. The above mentioned model is applied to a practical example, the results not only verify the rationality of the proposed model, but also provide a reference for the power system in the new environment.