Abstract: | In recent years some research towards developing forecasting models for wind power or energy has been carried out. In order to evaluate the prediction ability of these models, the forecasts are usually compared with those of the persistence forecast model. As shown in this article, however, it is not reasonable to use the persistence model when the forecast length is more than a few hours. Instead, a new statistical reference for predicting wind power, which basically is a weighting between the persistence and the mean of the power, is proposed. This reference forecast model is adequate for all forecast lengths and, like the persistence model, requires only measured time series as input. Copyright © 1998 John Wiley & Sons, Ltd. |