Probabilistic upscaling and aggregation of wind power forecasts

Abstract

Abstract Background Wind power forecasts of the expected wind feed-in for the next hours or days are necessary to integrate the generated volatile wind energy into power systems. Most forecasting models predict in some sense the best value, but they ignore the other possible outcomes which may arise because of forecasting uncertainties. Probabilistic forecasts, on the other hand, also predict a distribution of possible outcomes with their respective probabilities that specific power values will occur and therefore have higher information content. In this work, we address two problems that hinder the introduction of probabilistic forecasts in practice: (1) no measurement data are available for some wind farms and (2) the flexible aggregation of probabilistic forecasts for changing wind farm portfolios. Methods We present an approach based on copulas that can solve both problems by modeling the spatial correlation structure between reference wind farms. By sampling from the resulting joint probability distribution, probability forecasts can be upscaled from reference wind farms to wind farms without power measurements. Furthermore, the results can be aggregated to probability forecasts of portfolios of arbitrary and changing size. Results We perform experiments by applying our procedure to three use cases. The results are quantitatively evaluated with different probabilistic scores. For single target wind farms, our approach is as good as a state-of-the-art reference even if no data are available for the wind farm under consideration. For portfolios, our approach also allows forecasts to be made if no data is available for some wind farms and also to aggregate flexible portfolios of changing sizes, which was not possible before. Conclusion Our work solves two problems that hindered the introduction of quantile probabilistic forecasts in the application. This work opens a pathway for many different applications, e.g., predictive grid securities with stochastic optimization, better marketing of renewable energies, or allow to compensate for forecast errors in various applications.