The hourly time series of differences between export and import of electrical energy for Germany have been studied. Difference between export and import is called net position, and it is one of the most important terms in the world trade of electrical energy. Net position follows the production and consumption of electrical energy, which vary according to hours, days and seasons. The objective of our research is short term forecast of net position. The analyzed data consists of the hourly values of the net position for the period between 1.6.2013 and 31.5.2014. The time series of hourly net position was modelled as a seasonal autoregressive integrated moving average process (SARIMA). Various one-month time intervals of the data were considered as training sets. The time series simulation was used for evaluation of the models and cross validation for accuracy of the forecasts of different models was analyzed and compared.
|