The invention relates to the field of intelligent power grids, and discloses a day-ahead energy scheduling method based on a virtual power plant, which comprises the following steps of: 1, performingmodeling processing on uncertainty encountered in scheduling by using a point estimation method; 2, establishing a virtual power plant overall framework in the day-ahead energy scheduling method, andconstructing a cost model of the virtual power plant overall framework; and step 3, according to the models in the step 1 and the step 2, establishing a scheduling model of day-ahead energy schedulingon the premise of considering constraints. Compared with the prior art, according to the method, the uncertainty in the virtual power plant is processed by using a point estimation method, the uncertainty part is stripped and is divided into a deterministic part and an uncertainty part for separate modeling, the day-ahead energy scheduling mode of the virtual power plant is constructed by takingthe maximum profit as a target and considering various constraints, and an optimal strategy scheme can be better provided for a decision maker.