Highlights:

• An improved formulation for the wake-added turbulence is presented.

• The new formulation is integrated into a widely-used Gaussian wake model. 

• Diverse intra-farm and farm-to-farm scenarios are analyzed.

• Enhanced power prediction for both isolated and downstream-influenced wind farms is achieved.   

Source: N. Zehtabiyan-Rezaie, J. P. Justsen, and M. Abkar, “Wind-farm power prediction using a turbulence-optimized Gaussian wake model”, Wind Energy and Engineering Research, vol. 2, p. 100007, 2024. https://doi.org/10.1016/j.weer.2024.100007.   

Highlights:

• A progressive data-augmented model is developed for Reynolds-averaged simulation of wind farms.

• Unlike common RANS methods, the new model captures secondary flows of Prandtl’s second kind in the wake.

• Important features of wake flow like velocity deficit and turbulence intensity are satisfactorily predicted.

Code: https://github.com/AUfluids/k-wSST-PDA-WindFarm 

Source: A. Amarloo, N. Zehtabiyan-Rezaie, and M. Abkar, “A progressive data-augmented RANS model for enhanced wind-farm simulations”, Energy, vol. 313, p. 133762, 2024.  https://doi.org/10.1016/j.energy.2024.133762.   

Highlights:

• This work sheds light on the formation of secondary flows of Prandtl's second kind in turbine wakes, unveils the physical mechanisms behind their emergence, and measures their influence on wake characteristics.

• This study highlights the significant influence of spatial gradients of Reynolds stresses in the wake downstream of turbines, particularly under the influence of ground effects.

• This analysis underscores the pivotal role of secondary flows in shaping wake structures, especially the upward shift observed in the wake center.

Source: N. Zehtabiyan-Rezaie, A. Amarloo, and M. Abkar, “Secondary flows in the actuator-disk simulation of wind-turbine wakes”, Physics of Fluids, vol. 36, p. 045142, 2024.  https://doi.org/10.1063/5.0203068   

Highlights:

• An extended k-ɛ model is proposed for simulations of wakes and power losses in wind farms.

• An analytical term linked to turbine forces is incorporated into the transport equation of turbulent kinetic energy. 

• The extended k-ɛ model outperforms the standard version in all considered wind-farm cases.

Code: https://github.com/AUfluids/k-epsilon-Sk 

Source: N. Zehtabiyan-Rezaie, M. Abkar, “An extended k-ɛ model for wake-flow simulation of wind farms”, Renewable Energy, vol. 222, p. 119904, 2024. https://doi.org/10.1016/j.renene.2023.119904. 

Highlights:

• A comprehensive review of wind-farm flow control via reinforcement learning is provided.

• Challenges in implementing reinforcement learning for wind-farm flow control are unveiled.

• Potential opportunities for future endeavors in wind-farm flow control with reinforcement learning are identified.

Source: M. Abkar, N. Zehtabiyan-Rezaie, A. Iosifidis, “Reinforcement learning for wind-farm flow control: Current state and future actions”, Theoretical and Applied Mechanics Letters, vol. 13, p. 100475, 2023. 10.1016/j.taml.2023.100475. 

Highlights:

• Discussed an issue propagating in the literature related to turbine-induced added turbulence in the wake region.

• Quantified the impact of improper parametrization of turbine-induced added turbulence on wind-farm modeling.

Source: N. Zehtabiyan-Rezaie, M. Abkar, “A short note on turbulence characteristics in wind-turbine wakes”, Journal of Wind Engineering and Industrial Aerodynamics, vol. 240, p. 105504, 2023.  10.1016/j.jweia.2023.105504. 

Highlights:

• Traditionally, engineering wake models are used for power-output prediction for optimization and real-time applications. These models -being based on simplistic assumptions- obtain fast predictions with sacrificed accuracy. 

• Machine learning models show a high potential to serve as efficient tools for studying the fluid mechanics of wind farms.

• Leveraging the physics-guided machine-learning approach results in turbine-level power prediction models with enhanced levels of generalizability and explainability.

Source: N. Zehtabiyan-Rezaie, A. Iosifidis, and M. Abkar, “Physics-Guided Machine Learning for Wind-Farm Power Prediction: Toward Interpretability and Generalizability”, PRX Energy, vol. 2, p. 013009, 2023.  10.1103/PRXEnergy.2.013009. 

Highlights:

• Model-form uncertainty in RANS simulation of wind farms is quantified.

• A novel data-driven approach is used for the Reynolds stress perturbation.

• A surrogate machine-learning model is trained to predict target quantities.

• The trained model performs well when utilized for two other unseen wind farms.

Source: A. Eidi, N. Zehtabiyan-Rezaie, R. Ghiassi, X. Yang, and M. Abkar, “Data-driven quantification of model-form uncertainty in Reynolds-averaged simulations of wind farms”, Physics of Fluids, vol. 34, p. 085135, 2022. 10.1063/5.0100076. 

Highlights:

• Growing wind farms drive research towards data-driven techniques for optimizing design and operation.

• Current algorithms struggle with fluid dynamics intricacies in real wind farms, posing unique modeling challenges.

• Data-driven models need interpretability for user trust; incorporating physics enhances generalizability in wind-farm flow modeling.

Source: N. Zehtabiyan-Rezaie, A. Iosifidis, and M. Abkar, “Data-driven fluid mechanics of wind farms: A review”. Journal of Renewable and Sustainable Energy, vol. 14, p. 032703, 2022. 10.1063/5.0091980.