To deal with the randomness and volatility of renewable energy and ensure the economic and stable operation of microgrid

an improved energy management strategy for microgrid based on multi-time scale is proposed under the consideration of the characteristic that the prediction accuracy of renewable energy increases with the decrease of the time scale and the coupling relationship between different time scales. Firstly between long-term and short-term scheduling

buffer boundary constraints are put forward to simplify the constraint to speed up the solving; then is added between ultra short-term and real-time scheduling to make the scheduling more refined step by step; and the SOC penalty factor of energy storage is settled to coordinate the global and local optimality of the system. Finally

experiments in an actual microgrid show that the proposed model has higher accuracy and efficiency than traditional scheduling methods. This work is supported by National Natural Science Foundation of China (No. 51307023) and State Grid Corporation of China.