Research of Control Strategy of Power System Stabilizer Based on Reinforcement Learning

Power system stabilizer (PSS) is used to generate excitation system auxiliary control signals which can suppress low frequency oscillation in power system. It has the ability of self-learning and parameter online tuning, which is a development trend of smart grid PSS controller in the future. This paper presents a design method of power system stabilizer based on reinforcement learning. Q-learning algorithm is one of reinforcement learning, and is used to PSS as the additional control. The simulation results show that the PSS based on Q-learning can effectively improve the ability of suppressing low frequency oscillation in power system, and the robustness of the system is significantly enhanced.