ARIMA Residual Correction Energy System Grey Prediction Model and its Application

In view of China's current consumption of energy is greater than energy production, in order to more precisely predict China's consumption of energy, better understand the trend of China's energy development. We combine nonlinear and linear forecasting models with a new skill to build a metabolic model. This model is a nonlinear grey model and this model is also a Combination of Grey System Model and ARIMA Model(NMGM-ARIMA). The hybrid prediction technique proposed in this paper uses the linear model to modify the nonlinear prediction, effectively synthesizes the advantages of the linear model and the nonlinear model, and alleviates their limitations. The feasibility of NMGM-ARIMA technology was verified by sliding validation experiments based on China's energy consumption data from 1990 to 2020. The results display that the new proposed model can produce tetter reliable prediction results with an average percentage error of 0.041%. So, So, using the same dataset, we run ARIMA model and grey theory models proposed by other scholars. The results of the comparison display that mean absolute percentage error of the new model is smaller than other models. This shows that the new proposed model can Obtain more precise prediction results. Therefore, we believe that the new technology can better predict energy consumption in China.