Vortex-Stretching based Large Eddy Simulation Framework for Wind Farms

TL;DR

This paper introduces a novel LES framework incorporating vortex-stretching and an improved actuator disk model to accurately simulate wind turbine wakes and predict flow statistics in wind farms.

Contribution

It presents a new LES methodology with vortex-stretching and a feedback-aware actuator disk model, enhancing wake prediction accuracy.

Findings

Vertical wind profiles match experimental data

Reynolds stresses are well predicted, with dispersive stresses constituting 40%

Accurate flow statistics in near- and far-wake regions

Abstract

In large wind farms, wake distribution behind a wind turbine causes a considerable reduction of wind velocity for downstream wind turbines, resulting in a significant amount of power loss. Therefore, it is very crucial to predict wind turbine wakes efficiently. Thus, we propose a large-eddy simulation (LES) methodology, which takes the vorticity stretching to model transients in wind turbine wakes. In addition, we present an improved actuator disk model, which accounts for two-way feedback between the atmosphere and the wakes. First, we show that the vertical profile of the mean wind predicted with the new model has an excellent agreement with experimental measurements. Next, we validate the predicted Reynolds stresses against wind tunnel data and show that the dispersive stresses account for about 40% of Reynolds stresses. Finally, we show that the proposed LES method accurately predicts the characteristics of wind turbine wakes. Comparing the LES results with previously reported data, we have found that the new LES framework accurately predicts the flow statistics in both the near-wake and the far-wake regions.