Deep Q Learning Based Economic Pinning Strategy for Microgrid Volt-VAR Control

To mitigate the voltage deviations caused by primary droop control, Deep Q Learning (DQL) Based Economic Pinning Strategy for Microgrid Volt-VAR Control is proposed to improves the economic performance of microgrids while effectively suppressing voltage fluctuations arising from load disturbances and changes in network topology. Specifically, the DQL is employed to enhance the adaptability of the pinning control mechanism. furthermore, a DQL-driven voltage pinning controller is designed to enable rapid and coordinated regulation among multiple distributed generators within the microgrid. Simulation results validate that the proposed method not only ensures voltage stability in isolated microgrids but also significantly reduces overall operational costs.