An Unsupervised Learning Approach for Predicting Wind Farm Power and Downstream Wakes Using Weather Patterns

Wind energy resource assessment typically requires numerical modeling at fine resolutions, which is computationally expensive for multi-year timescales. Increasingly, unsupervised machine learning techniques are used to identify representative weather patterns that can help simulate long-term behavior. Here we develop a novel wind energy workflow that combines the weather patterns from unsupervised clustering with a numerical weather prediction model (WRF) to obtain efficient long-term predictions of wind farm power and downstream wakes, which provide a good approximation to full WRF simulations at vastly reduced computational cost. We use ERA5 reanalysis data and compare clustering on low altitude pressure and wind velocity, a more relevant variable for wind resource assessment. We also compare varying domain sizes for the clustering. A WRF simulation is run at each cluster center and the results aggregated into a long-term prediction using a novel post-processing technique. We consider two case study regions and show that our long-term predictions achieve good agreement with a year of WRF simulations in 2% of the computational time. Moreover, clustering over a Europe-wide domain produces good agreement for predicting wind farm power output, but clustering over smaller domains is required for downstream wake predictions which agree with the year of WRF simulations. Our approach facilitates multi-year predictions of power output and downstream farm wakes, by providing a fast, reliable, and flexible methodology applicable to any global region. Moreover, this constitutes the first tool to help mitigate effects of wind energy loss downstream of wind farms.