The randomness and volatility of new energy bring difficulties to grid dispatching. Therefore, large-scale energy storage is introduced at the side of new energy stations, and the impact of the energy storage system on the operation simulation of power systems containing new energy is analyzed. This paper combines the characteristics of different modules with different functions during the operation of the power system, and proposes a simulation framework for the operation of new energy power systems with energy storage. It mainly includes: in the simulation of new energy output, using neural networks to predict the output of new energy; in the economy in terms of dispatching, an economic dispatch model for new energy systems with energy storage is established. Its objective function takes into account the cost of thermal power emissions, and considers the cost of reserve capacity compensation and the penalty cost of negative efficiency operation of wind farms. In terms of energy storage costs, the initial investment cost and operation and maintenance cost; in terms of operation analysis, in the ultra-short period, the two indicators of average energy change rate of new energy and peak inversion probability are used to evaluate the energy storage's ability to suppress power fluctuations and peak shaving; Operation simulation program for environment, suppression of fluctuations and peak shaving ability. Using genetic algorithm to take Xinjiang power grid as an example for simulation verification, the results show that the large-scale application of energy storage on the side of new energy stations can effectively suppress the fluctuation and reverse peaking of new energy output power.