A Study on Forecasting Method for a Short-Term Demand Forecasting of Customer's Electric Demand

The traditional demand prediction was based on the technique wherein electric power corporations made monthly or seasonal estimation of electric power consumption for each area and subscription type for the next one or two years to consider both seasonally generated and local consumed amounts. Note, however, that techniques such as pricing, power generation plan, or sales strategy establishment were used by corporations without considering the production, comparison, and analysis techniques of the predicted consumption to enable efficient power consumption on the actual demand side. In this paper, to calculate the predicted value of electric power consumption on a short-term basis (15 minutes) according to the amount of electric power actually consumed for 15 minutes on the demand side, we performed comparison and analysis by applying a 15-minute interval prediction technique to the average and that to the time series analysis to show how they were made and what we obtained from the simulations.