Abstract
In recent years, the penetration of solar and wind generators has increased drastically, becoming one of the main power sources in several electric systems. Solar and wind generators are climate dependent, and therefore characterized for its uncertainty and variability. The exposure to climate uncertainty suppose a great challenge for market agents, who depend heavily on climate forecasts to predict market prices and to adequately develop and distribute generation resources. Therefore, it is of great importance to develop forecasting models that confront different possible scenarios, supporting risk-analysis and decision-making processes. However, the literature on predictive models is mainly focused on short-term horizons and single-valued expectations, which do not provide any information on risk exposure. In this research, we propose a statistical long to medium-term forecasting model that allows capturing and analyzing different climate scenarios in a probabilistic way, helping to deal with the inherent risk of renewable energy sources.
Probabilistic energy forecasting: using statistical methods to predict wind and solar medium-term output
