A Statistical Model for Wind Power Forecast Error and its Application to the Estimation of Penalties in Liberalized Markets

Abstract

The problem of accurately forecasting wind energy has garnered a great deal of attention in recent years. There are always some errors associated with any forecasting methodology. Although it is sometimes assumed that the forecast errors are Normally distributed, it is a special case arising from the geographical dispersion of wind resources, as shown in this paper. The distribution of the forecast error needs to be examined individually for every wind farm to determine the impact of this error on trading energy in electricity markets. This paper addresses the problem of modeling the distribution of the forecast errors associated with Persistence forecasts at the level of a single wind farm, and develops a novel, mixed distribution-based model to approximate the distribution of these errors. The model is then used to estimate the penalties for imperfectly forecast energy injections in the short-term markets. The results from the application of this model to trading are further used to assess the feasibility of energy storage in hedging against imperfect forecasts.