Short-term Load Forecasting Based on Grey Model and Correction System

A novel short-term forecasting method is presented on the basis of GM (1,1) model and the application on the correcting latest historical load data in this paper. And the traditional input method of the historical data has been improved, that is, the difference value (D-value) of the same type of load data with same instant for the adjacent two days has been regarded as input sampling sequence, and the class ratio out of the feasible has been processed using the method of square root. Moreover, the relative factors that influencing load changes has been taken into consideration, and the latest historical load data has been employed to correct the forecasting value, for improving the precision of the forecasting. The efficiency of the forecasting precision of the proposed method has been verified by the demonstrated example