As wind energy continues to be a crucial part of sustainable power generation, the need for precise and efficient modeling of wind turbines, especially under yawed conditions, becomes increasingly significant. Addressing this, the current study introduces a machine learning-based symbolic regression approach for elucidating wake dynamics. Utilizing WindSE’s actuator line method (ALM) and Large Eddy Simulation (LES), we model an NREL 5-MW wind turbine under yaw conditions ranging from no yaw to 40 degrees. Leveraging a hold-out validation strategy, the model achieves robust hyper-parameter optimization, resulting in high predictive accuracy. While the model demonstrates remarkable precision in predicting wake deflection and velocity deficit at both the wake center and hub height, it shows a slight deviation at low downstream distances, which is less critical to our focus on large wind farm design. Nonetheless, our approach sets the stage for advancements in academic research and practical applications in the wind energy sector by providing an accurate and computationally efficient tool for wind farm optimization. This study establishes a new standard, filling a significant gap in the literature on the application of machine learning-based wake models for wind turbine yaw wake prediction.

This work presents a comparison study of the CFD modeling with two different turbulence modeling approaches viz. unsteady RANS and LES, on a full-scale model of the (New) MEXICO rotor wind turbine. The main emphasis of the paper is on the rotor and wake aerodynamics. Simulations are carried out for the three wind speeds considered in the MEXICO experiment (10, 15, and 24 ms⁻¹). The results of uRANS and LES are compared against the (New) MEXICO experimental measurements of pressure distributions, axial, radial, and azimuth traverse of three velocity components. The near wake characteristics and vorticity are also analyzed. The pressure distribution results show that the LES can predict the onset of flow separation more accurately than uRANS when the turbine operates in the stall condition. The LES can compute the flow structures in wake significantly better than the uRANS for the stall condition of the blade. For the design condition, the mean absolute error in axial and radial velocity components along radial traverse is less than 10% for both the modeling approaches, whereas tangential component error is less than 2% from the LES approach. The results also reveal that wake recovers faster in the uRANS approach, requiring further research of the far wake region using both CFD modeling approaches.