Abstract: Response to the increasing impact of PV uncertainty on the grid,this paper proposes a source-grid-load economic dispatching method considering demand response. Firstly,with PV historical data,non-parametric estimation is used to fit the PV output,while Monte Carlo is used for scenario characterization and scenario reduction is performed by using the Marxian distance-based scenario reduction technique. Secondly,a source-grid-load economic dispatch model for power system considering the minimum system operation cost under demand response is established by combining stochastic programming methods. Finally,the IEEE39 node model is used for analysis,and the results show that the proposed optimization model can realize the uncertainty dispatch of PV and reduce the cost of electricity for customers,which provides theoretical support for the future multi-energy complementation under demand response of power system.