Learning to Optimise Wind Farms with Graph Transformers
This work addresses the need for efficient optimization of wind farm operations, offering a data-driven solution that reduces computational overhead for energy companies.
The authors tackled the problem of predicting power generation in wind farms under varying layouts and conditions by developing a graph transformer model, which achieved similar accuracy to standard simulation tools at a fraction of the computational cost.
This work proposes a novel data-driven model capable of providing accurate predictions for the power generation of all wind turbines in wind farms of arbitrary layout, yaw angle configurations and wind conditions. The proposed model functions by encoding a wind farm into a fully-connected graph and processing the graph representation through a graph transformer. The graph transformer surrogate is shown to generalise well and is able to uncover latent structural patterns within the graph representation of wind farms. It is demonstrated how the resulting surrogate model can be used to optimise yaw angle configurations using genetic algorithms, achieving similar levels of accuracy to industrially-standard wind farm simulation tools while only taking a fraction of the computational cost.