Forecasting Method of Power Consumption Information for Power Users Based on Cloud Computing

In order to realize the real-time balance of power demand and effectively avoid the waste of power, it is necessary to forecast the power consumption. Under this background, a forecasting method of power consumption information for power users based on cloud computing is designed. The prediction model framework is designed based on cloud computing technology. Carry out abnormal data processing, missing data filling and normalization for power consumption data. Calculate the correlation degree and select the influencing factors of power consumption of power users. Combined with multiple regression analysis, the forecasting model of electricity consumption information of power users is constructed. The results show that the mean absolute percentage error (MAPS), the root mean square error (RMSE) and the equalization coefficient (EC) of the method are the minimum and the maximum, which proves the accuracy of the method.