Wind forecasting models are divided in two main categories, physical and statistical. The former are based on Numerical Weather Prediction (NWP). The statistical models, on the other hand, use on-line measurements. In this paper we use hybrid models, which combine elements of both types. In particular three forecast systems based on Artificial Neural Networks have been developed in order to predict power production of a wind farm in different time horizons: 1, 3, 6, 12 and 24 hours. In the first forecast system, the neural network has been used only as a statistic model based on time series of on-line measured wind power, while in the second and third forecast systems different combinations of measured data and numerical weather predictions have been used, improving the performance in the predictions, especially over long time horizons. The error of the different forecast systems is investigated for various forecasting horizons and statistical distributions of the error are calculated and presented.

This content is only available via PDF.
You do not currently have access to this content.