Neural Network PID-Based Frequency Control Strategy for Energy Storage Participating Loads

With the goal of “double carbon” and the deepening of China’s power system reform, the operating characteristics of the power system are more complex and variable. The demand for frequency control of traditional thermal power units and hydropower units is growing. Load frequency control (LFC) is one of the important means of frequency control in the power system. Its primary goal is to keep the system frequency within zero steady-state error. A neural network proportional-integral-derivative (PID)-based energy storage participation LFC strategy is proposed for the load frequency problem of two types of units. First, the traditional closed-loop LFC models of thermal and hydroelectric units are established according to the frequency response characteristics of the traditional units; then, the energy storage system used for the frequency regulation of the power system is selected, and the corresponding frequency response model is established. Finally, an LFC controller design method based on a neural network PID algorithm is studied, and the simulation results show that the proposed neural network PID control method has better response characteristics compared with the traditional PI control.