Abstract: Wind power and photovoltaic grid-connected significantly increase the uncertainty of power system operation scheduling. The maximum power that it can emit often fluctuates randomly according to a certain probability distribution around a certain value in a certain time slot, and there is a prediction error between the predicted power and the actual power. When the maximum power that the wind power and photovoltaic can actually emit is less than the power scheduled in the scheduling plan, it will affect the scheduling strategy formulated, and may even cause a substantial increase in operating costs. In order to studying the influence of the uncertainty of renewable energy output on microgrid dispatching, this paper divided microsources into two types of basic power supply and requency modulation (FM) power supply according to the frequency modulation characteristics, and proposed a two-step optimal scheduling method based on power classification. Using Monte Carlo lestimated the expected and variance of the cost function value considering the uncertainty of renewable energy and obtained a certain number of samples through sampling. Using the particle swarm optimization (PSO) algorithm found the FM power installation plan and the basic load power supply scheduling plan that minimize the average sample cost. In the context of a large number of sampling samples, the rationality and effectiveness of the proposed two-step scheduling method were verified through example.