| one publication added to basket [353431] | A new wake-merging method for wind-farm power prediction in the presence of heterogeneous background velocity fields
Lanzilao, L.; Meyers, J. (2022). A new wake-merging method for wind-farm power prediction in the presence of heterogeneous background velocity fields. Wind Energ. 25(2): 237-259. https://dx.doi.org/10.1002/we.2669 In: Wind Energy. Wiley: Chichester. ISSN 1095-4244; e-ISSN 1099-1824, more |   |
| Keyword | | | Author keywords | analytical wake model; coastal gradient; wake-merging method; wakes; wind-farm power prediction |
| Abstract | The difference in surface roughness between land and sea, and the terrain complexities, lead to spatially heterogeneous atmospheric conditions, and therefore affect the propagation and dynamics of wind-turbine and wind-farm wakes. Currently, these flow heterogeneities and their effects on plant aerodynamics are not modeled in the majority of engineering wake models. In this study, we address this issue by developing a new wake-merging method capable of superimposing the waked flow on a heterogeneous background velocity field. We couple the proposed wake-merging method with four different wake models, i.e. the Gaussian, super-Gaussian, double-Gaussian and Ishihara model, and we test its performance against LES results, dual-Doppler radar measurements and SCADA data from the Horns Rev, London Array, and Westermost Rough farms. The standard Jensen model with quadratic superposition is also included. In homogeneous conditions, the new method predicts slightly higher velocity deficits than the linear superposition method. Overall, the distributions of the difference in power ratio between the two wake-merging methods predictions and observations show a similar mean absolute error (MAE) and interquartile range (IQR) in such conditions. On the other hand, the new wake-merging method predictions display a lower MAE with a similar IQR in case of a spatially varying background velocity, being overall more accurate than the ones obtained with linear superposition. The most accurate estimates are obtained when the wake-merging methods are coupled with the double-Gaussian and Gaussian single-wake models. In contrast, the Jensen and super-Gaussian wake models overestimate the velocity deficits for the majority of cases analyzed. |
|
