Comparative analysis of wind speed short term forecasts for wind farms

Article history: Received: August 2014 Received in the revised form: September 2014 Accepted: October 2014 The purpose of study was verification regarding quality of wind speed forecasts used during designing the wind farm capacity, with AAN [artificial neural network] methods and Brown, Holt, Winters and ARIMA time models. Analysis included results of forecasts for December, namely a month with the biggest wind speed amplitude changes, considering data for period of 2008-2009. Analysis of results confirmed that appropriate linear models and artificial neural methods for the period of wind speed forecast may ensure good results regarding forecasts of wind power output generated by wind farms.