The aim of this thesis is to find a model for Electric power load with help of Ident-GUI in Matlab for the period of 1999(winter period), i.e., from the 17th of September 20:00 hrs to the 31st of December 00:00 hrs for the year 1999. It is clear, however, that the process displays different properties depending on the time of day and day of week. The GUI system identification toolbox is used to find a suitable model and for prediction. (Algorithms have been developed to predict the power load with the use of time-local SARMAX (Seasonally differentiated ARMAX) which didn’t give satisfied results. Electric power load is forecasted using temperature, wind velocity, Sun intensity as internal variables. In the thesis a new method to develop a model for the power load has been used, i.e., the dependence upon temperature has been used as an effecting variable for the consumption of power. When examining the developed models, measured data from 1999 and 2000 are used for testing and verifying the results. Matlab is used for implementing the forecasting algorithms. When comparing different forecasting methods, measured data from 1999 and 2000 are used.