Gaussian process surrogate modeling for efficient controller tuning and fatigue load prediction of the helix wake-mixing method

Daan van der Hoek, Tim Dammann, and Jan-Willem van Wingerden

Data sets

Dataset accompanying the publication "Gaussian process surrogate modeling for efficient controller tuning and fatigue load prediction of the helix wake-mixing method" D. van der Hoek et al. https://doi.org/10.5281/zenodo.21397322

Short summary

Wind farms suffer power losses and increased structural loading due to wake interactions. The helix method mitigates this by continuously moving upstream turbine blades to accelerate wake recovery. We developed two surrogate models from high-fidelity simulations: one identifying optimal pitch settings, achieving 7.5 % power gain, and one predicting fatigue loads under various conditions. This enables joint evaluation of power gains and load penalties, supporting informed wind farm control design.