Nowcasting of wind speed using support vector regression. Experiments with time series from Gran canaria

Abstract

The aim of this paper is to describe and evaluate a proposal for nowcasting wind speed for wind farm locations from historical time series, based on the method of regression by support vectors. To show the improvement over other methods, we used the ANEMOS Project standard evaluation protocol for forecasting against three reference models to compare, referred to a statistical approach: persistence, autoregressive and autoregressive moving average models. The obtained results show a good performance of the proposed method and how beat the classical reference models.