Mathematical Analysis of Time-Series & Meteorological Factors in Electricity Demand Forecasting Based on Carbon Neutrality

Due to the modernization of the electric power system and the increase of renewable energy, there is a significant need for various energy management systems (EMSs), in which electricity demand forecasting technology plays a remarkable role. Moreover, the load dispatching center operator regulates the supply-demand balance in the service area to meet the ever-changing electricity demand based on the demand forecasting results. Unfortunately, climatic and natural disaster factors bring uncertainties in the demand forecast process, which leads to the low prediction accuracy of the existing forecast techniques with failing to fully consider these factors affecting load fluctuations. In this paper, the importance of major time-series, meteorological and business-categorical energy consumption influencing factors applied to electricity demand forecasting is thus comprehensively and systematically illustrated through the correlation and regression analysis methods based on the real electricity demand data and various influencing factors data affecting electricity demand variation. The simulation results analysis in this study can assist load forecasters in more rationally identifying the influencing factors of maximum electricity consumption at various periods, thereby improving the overall accuracy of load forecasting.