Forecasting models for daily natural gas consumption considering periodic variations and demand segregation

Due to expensive infrastructure and the difficulties in storage, supply conditions of natural gas are different from those other traditional energy sources like petroleum or coal. To overcome these challenges, supplier countries require take-or-pay agreements for requested quantities. These contracts have many pre-clauses; even if they not met due low/high consumption external factors, buyers must completely fulfill them. A similar contract is then imposed on distributors wholesale consumers. It is, thus, important all parties forecast their daily, monthly, annual demand minimize risk. In this paper, a model consisting modulated expansion Fourier series, supplemented by deviations comfortable temperatures as regressor proposed monthly weekly over one-year horizon. This day-ahead feedback mechanism daily consumption. The method applied study major residential areas Turkey, yearly, weekly, basis. shown that heating dominates winter masks variations. On hand, weekend holiday effects visible summer provide an estimate industrial use. advantage capability long term projections, reflecting causality, providing accurate forecasts with minimal information.