Abstract: High proportion of new energy grid-connected brings new challenges to the dispatch and operation of power systems. In order to alleviate the reserve pressure of power systems, a stochastic optimal scheduling model considering multiple flexible reserve resources on both source and load sides is proposed. Firstly, a model of variable scenario is established on the basis of scenario generation method, which considers the influence of the capacity of grid-connected wind power and the area of grid-connected photovoltaic power on the uncertainty of new energy generation. Then, the reserve model of various flexible resources in the system is established. On the source side, the reserve models of conventional units, photovoltaic and wind power are established, respectively, and the uncertainty of wind power and photovoltaic reserve is considered; On the load side, the reserve model is established on the basis of the incentive demand response. Then, the reserve dispatching model is established on the basis of the two-stage stochastic optimization model. The model considers the day-ahead operation and reserve decisions, as well as wind power curtailment, photovoltaic power curtailment, and load shedding risks in uncertain scenarios within the day. Finally, the case studies based on the modified IEEE RTS-24 system verify the effectiveness of the proposed model.