Energy efficiency management and automation scheduling strategies in microgrids

This study is going to enhance the energy efficiency and scheduling of microgrids and will propose a Q-learning based optimization scheme that is based on the complexity and uncertainties of microgrid operation. The main focus of our research is to find the best strategy for energy dispatch by discretizing the microgrid operation modes and employing reinforcement learning techniques. As the various energy systems are integrated and the energy exchange processes between the distribution network, electric vehicles and energy storage systems are experimentally simulated. The outcome indicates that the microgrid optimized by Q-learning can reduce power fluctuation and improve system stability in an effective manner. Moreover, Q-learning is an effective method that can increase the long-term gains and decrease the carbon emissions as opposed to the conventional experience-based learning strategies. In summary, reinforcement learning is the best approach for energy management of microgrids and it is also a new view for microgrids operation in the most efficient manner and less impact on the environment.