A novel capacity demand analysis method of energy storage system for peak shaving based on data-driven

With the large-scale integration of renewable energy into the grid, the peak shaving pressure of the grid has increased significantly. It is difficult to describe with accurate mathematical models due to the uncertainty of load demand and wind power output, a capacity demand analysis method of energy storage participating in grid auxiliary peak shaving based on data-driven is proposed in this paper. According to the statistical method, typical daily evaluation indexes with anti-peaking characteristics of wind power extracted from the operating data are regarded as the inputs of the back propagation (BP) neural network, and the corresponding fitness value calculated by the entropy weight and analytic hierarchy process (AHP) method is regarded as the output of the BP neural network. A typical daily mining model with anti-peaking characteristics of wind power based on data-driven is established, and the particle swarm optimization (PSO) algorithm is used to solve the model. In order to maximize the revenue of the system, an optimal capacity configuration model of energy storage participating in grid auxiliary peak shaving based on data-driven is established, and the artificial bee colony (ABC) algorithm is adopted to solve the model. The sensitivity of the energy storage capacity on grid auxiliary peak shaving under different fitness levels is analyzed. The correctness and effectiveness of the method proposed in this paper are verified by the simulation analysis of the actual operating data from a certain area power grid in China throughout the year. The simulation results show that this method provides a theoretical basis of energy storage participating in grid auxiliary services.

Automated summary

This paper proposes a data-driven method for analyzing the capacity demand of energy storage participating in grid auxiliary peak shaving, and studies the relationship between peak shaving pressure of the grid and energy storage capacity by establishing models and solving them using algorithms. The effectiveness of the method is verified through simulation analysis, providing a theoretical basis for energy storage participating in grid auxiliary services.