Data-driven analyses for offshore wind turbine wake characteristics

Background and significance.

In offshore wind farms, turbulence generated in turbine wakes impact the farm layout planning, the output power and turbine load estimation [1]. 10-20% power could be lost due to wake effects in wind farms [2, 3]. Understanding power losses and turbulence increase due to wind turbine wake interactions in offshore wind farms is crucial to optimizing power production. Traditional computational/experimental fluid dynamics (CFD/EFD) techniques can provide reliable predictions of the turbine wake characteristics but are valid within limited spatio-temporal resolutions and under idealized conditions. Data-driven models in complex dynamic systems have great potential for reduced-order approximation and optimization [4]. In this project, a well-defined CFD model will be used to explore the impact of upstream flow conditions on the near-/far-wake characteristics of offshore wind turbines. The computational platform will be either ANSYS CFD package or OpenFOAM. Wake velocity fields will serve as inputs to develop data-driven models. These models will be employed to analyze the turbine wake characteristics for different inflow conditions [5, 6].

REU student involvement. 

REU students gain hands-on experience in using CFD models to analyze wake velocity fields around offshore wind turbines, export datasets for further analyses using machine learning, predict velocity vector fields when inflow conditions change, and suggest modifications that support enhancing efficiency/capacity factor of offshore wind turbines.

References.

1. D. Ziaei and N. Goudarzi, "A take on wake modeling of turbines based on deep learning," in ASME Power Conference, vol. 83747.  American Society of Mechanical Engineers, 2020, p. V001T10A017.

2. R. J. Barthelmie, S. T. Frandsen, M. Nielsen, S. Pryor, P.-E. Rethore, and H. E. Jørgensen, "Modelling  and measurements of power losses and turbulence intensity in wind turbine wakes at middelgrunden offshore wind farm," Wind Energy: An International Journal for Progress and Applications in Wind Power Conversion Technology, vol. 10, no. 6, pp. 517–528, 2007.

3. ​​​​​​​M. F. Howland, S. K. Lele, and J. O. Dabiri, "Wind farm power optimization through wake steering," Proceedings of the National Academy of Sciences, vol. 116, no. 29, pp. 14 495–14 500, 2019.

4. ​​​​​​​S. L. Brunton, B. R. Noack, and P.  Koumoutsakos, "Machine Learning for Fluid Mechanics," Annu.   Rev. Fluid Mech., no. 52, pp. 477–508., 2020.

5. ​​​​​​​Y. Wang, Y. Yu, S. Cao, X. Zhang, and S. Gao, "A review of applications of artificial intelligent algorithms in wind farms," Artificial Intelligence Review, vol. 53, no. 5, pp. 3447–3500, 2020.

6. ​​​​​​​Z. Deng, Y. Chen, Y. Liu, and K. C. Kim, "Time-resolved turbulent velocity field reconstruction using     a long short-term memory (lstm)-based artificial intelligence framework," Physics of Fluids,  vol. 31,  no. 7, p. 075108, 2019.