With the advancement of energy transformation, a new type of power system is urgently required to be built. Due to the strong volatility of wind and photovoltaic power generation, the stability of the power system is affected by the large-scale integration of wind and photovoltaic power. In order to reduce the damage caused by the volatility of new energy to the power system and to improve the economic efficiency of the power system, a scheduling method based on interval prediction of new energy output is proposed. First, the Particle Swarm Optimization Back Propagation Neural Network (PSO-BP) algorithm is used to predict the point of new energy output, and then the Bootstrap method is used for interval prediction on this basis, so that the fluctuation of new energy output is taken into account as much as possible in the scheduling system. In addition, in order to improve the consumption capacity of new energy, a pumped storage energy system is added to the system. To improve the economic benefits brought by pumped storage energy to the power system, the capacity of the pumped storage energy system is planned. The power system in a certain region of Xinjiang is taken as an example and simulation analysis is conducted. The results show that reasonable capacity planning can bring more than one million yuan in economic benefits to the power system and improve its ability to absorb new energy. Compared with point prediction, interval prediction not only improves the stability of the power system, but also increases the total revenue of the power system by 13% and reduces the wind and solar power curtailment rate by 36%.