Application of Grey Model Based on Twice Fitting in Short-term Power Load Forecasting

The accuracy of short-term load prediction is related to the safety planning and smooth operation of power system, and plays an important role in the safety planning of power system and power supply of power grid. In order to solve the problems of the quadratic fitting grey prediction model, such as weak noise reduction ability and strict requirement on the stability of historical data, this paper proposes an improved method to combine the quadratic fitting grey prediction model with equal dimension and new information technology, and to modify the prediction results by using Fourier series. Problems existed in the work of basic data noise, the historical data preprocessing, considering the number of data pretreatment could lead to input new data equals the number of old data don't delete (called the new interest is unequal), analysis in the case of the new interest is unequal, based on Fourier residual correction of quadratic fitting grey prediction precision of the model. The simulation results show that the accuracy test level of the improved model is level 1, and the prediction model has good stability and the accuracy meets the expected requirements in the case of different dimensions of new information.