Stabilized Control Strategy of ST-Fed Grid Based on Reinforcement Learning for More Sustainable Airports

Smart transformer (ST) is used in more sustainable airport power supply system due to its excellent characteristics. However, with the increasing installation of renewables in grid, the interactions between ST and renewables could incur stability problems. In this paper, a small-signal model of the ST-based ground power system (GPU) is established to analyze the system stability. Then, a Biquad filter-based stabilized control is proposed to improve the stability of the ST-fed grid. A reinforcement learning (RL) algorithm is used to facilitate the control design and enhance the system robustness when system parameters change.