Forecasting Electricity Demand in Australian National Electricity Market

Load forecasting is a key task for the effective operation and planning of power systems. It is concerned with the prediction of hourly, daily, weekly, and annual values of the system demand and peak demand. Such forecasts are sometimes categorized as short-term, medium-term and long-term forecasts, depending on the time horizon. Long-term load forecasting is an integral process in scheduling the construction of new generation facilities and in the development of transmission and distribution systems, while short-term forecasting provides essential information for economic dispatch, unit commitment and electricity market. A comprehensive forecasting solution developed by Monash University is described in this paper. The semi-parametric additive models based forecasting system has been used to forecast the electricity demands for regions in the National Electricity Market. The forecasting system covers the time horizon from hours ahead up to years ahead, and provides both point forecasts (i.e., forecasts of the mean or median of the future demand distribution), and density forecasts (providing estimates of the full probability distributions of the possible future values of the demand). The performance of the methodology have been validated through the developments of the past years, and the forecasting system is currently used by the Australian Energy Market Operator (AEMO) for system planning and schedule.