Tool for optimization of sale and storage of energy in wind farms

Abstract

In this work we address the problem of energy management in a wind farm supported by an Energy Storage System (ESS) that operates in an electricity market with six intraday sessions and with penalty policies for imbalances between commitments and the energy really injected. We face it through a cascade of model predictive controllers that also require the design of predictors for wind and electricity market price forecasts. The master controller is executed synchronously with the market sessions and decides the commitments. The slave controller is executed each hour and decides the energy that should be sold to minimize the economical penalties if the commitment is not achievable. Finally, a real-time controller decides how to manage the energy storage in the ESS to sell the desired energy when possible. We use historical real data for the design and validation of the approach and show its benefits. The results show that the cascade structure helps to adequately adapt the energy committed in the intraday market. We also obtain the necessary prices on batteries so that their use is profitable.