P. Potočnik, E. Govekar, and I. Grabec (Slovenia)
Forecasting, short-term, energy consumption, natural gas, consumption cycles.
The natural gas market requires forecasting for the optimisation of leasing additional storage capacities. Consequently, natural gas distribution companies have an economic stimulus to accurately forecast their future gas consumption. A method for short-term forecasting of natural gas consumption is presented in this paper. The method consists of analysing natural gas consumption cycles (yearly, weekly, daily) and then constructing a forecasting model for hourly forecasting for the next day. We propose a forecasting model with the following properties: (1) two submodels for the winter and summer seasons; (2) input variables including past consumption data, weather data, weather forecasts and basic cycle indexes; (3) a hierarchical forecasting structure: a daily model as the basis, with the hourly forecast obtained by modeling the relative daily profile. This proposed method is illustrated by a forecasting example for Slovenia’s natural gas market.
Important Links:
Go Back
