The invention discloses an electricity optimizing configuration method based on big data. The method includes the following steps that 1, electricity consumption data of users is screened out from the electricity big data, dirty data in the electricity consumption data is deleted, and a model is established for the processed electricity data; 2, noise data existing in the electricity data is removed, and missing data existing in the electricity data is filled; 3, according to an autoregressive AR model of a time sequence prediction method and the processed electricity consumption data of the users, the electricity consumption of a load center in the next period is predicted; 4, the predicted electricity consumption is modified; 5, according to predicted electricity consumption, power transmission network deployment conditions and the installed capacity of each power station, the electric energy production amount of each power station is set; 6, according to the electric energy production amount, whether the electricity demands of the users are met or not is judged, if not, electricity optimizing configuration is performed, then whether the electricity demands of the users are met or not is judged again, and the operation is repeated so as to obtain a final electricity configuration scheme.