| 作成者 |
|
|
|
| 本文言語 |
|
| 出版者 |
|
| 発行日 |
|
| 収録物名 |
|
| 巻 |
|
| 号 |
|
| 開始ページ |
|
| 出版タイプ |
|
| アクセス権 |
|
| 利用開始日 |
|
| 権利関係 |
|
| 関連DOI |
|
| 関連HDL |
|
| 概要 |
We propose a large eddy simulation (LES) model for offshore wind farms that combines the lattice Boltzmann method (LBM) and actuator line model. The lattice Boltzmann computations are accelerated and ...parallelized using GPU computing and supercomputing, thereby addressing the high computational cost associated with the LES of wind farms. A mesh convergence study for the Vestas V80 wind turbine demonstrates that the power curve converges at a grid with 2 m resolution. The characteristics of mesh convergence for the wake profiles vary depending on the inflow conditions. Under uniform inflow, high-resolution grids promote turbulence in the wake, and the mesh convergence is not substantiated. In contrast, with turbulent inflow, typical of real wind conditions, even low-resolution 4 m grids produce reasonable results for the wake velocity and turbulence intensity. The proposed model is validated by comparing it with the observed data of the Horns Rev 1 wind farm. The comparison indicates that the simulations reasonably reproduce the observed data for the power output of the downstream turbines influenced by the wind direction. The proposed LBM framework completes a wind farm LES with 1.65 billion grid points in only 12.5 h using 32 GPUs, demonstrating its practicality.続きを見る
|
Mendeley出力
