Cluster Sampling for the Demand Side Management of Big Data

In view of the DSM under the big data environment, an improved FCM clustering is proposed, and the daily load curve of the whole study area was obtained with the electricity data. According to the formulation of the TOU price, which is consistent with the characteristics of local users is given. The electricity suggestions based on the specific user load curve is provided, including the return of the DR. Subsequently, the sampling division is put forward to expand the improved model. Finally, the method is tested by the actual data, and the results show that it has a processing speed 10 times of the direct processing when the data is more than 10000.