Research on Mid-long Term Load Forecasting based on Gray Fourier Series Residual Correction Model

摘要

The purpose of power load forecasting is to provide the development status and level of the region's future load, providing the basis for the electric power production department and management department to make production and development plans. This paper puts forward the grey prediction model modified by Fourier series residual. First of all, the moving average method is used to preprocess the original data and weaken the burr of the original data. Secondly, the GM (1,1) is used to estimate the load of the selected sample area. On this basis, use the Fourier series to revise the residual error of the grey forecasting model, making the model fitting of historical data as much as possible. The example analysis results show that the grey prediction model modified by Fourier series residual has the higher prediction accuracy compared with the general GM (1,1), proving the validity of the model.