Abstract: In order to reduce the influence of the power fluctuation of distributed power supply and load on the tracking scheduling of microgrid, according to the energy storage characteristics of supercapacitors and batteries, a hybrid energy storage hierarchical optimization strategy based on hierarchical scheduling and model predictive control theory was proposed. Upper slow optimization model took the lowest operation cost as the optimization goal, converted the long-term cost model of energy storage system to short-term cost model associated with the real-time operation, comprehensively considered peak valley price difference of electricity price and storage life, and used the power, the response speed slower storage battery energy storage unit for energy scheduling with renewable energy and large power grid. In order to minimize the deviation from the reference value of the upper dispatching plan, the lower layer rapid optimization model used the supercapacitor energy storage unit with fast response speed to smooth the prediction error and load fluctuation of renewable energy, and the proposed control strategy was verified by a simulation example. The results show that the layered coordination of batteries and ultracapacitors can reduce the operating cost of the system, and can provide the optimal control scheme for the economic dispatching of microgrid.