To better understand the effect of the clearing size and the forest characteristics on the wind flow, two-dimensional numerical simulations have been performed with a Reynolds-Averaged Navier Stokes (RANS) approach. A modified k – e model taking into account the forest drag was used after being validated against wind tunnel experiments from two authors. A neutrally stable horizontally homogeneous atmospheric surface layer has been modelled, driven both by a pressure gradient and a shear stress and using original top boundary conditions for the turbulent quantities, and validated against field measurements. Clearing sizes ranging from 1h to 100h were investigated with a reference forest representative of a typical boreal forest as well as with three other Leaf Area Density (LAD) distributions and three other Leaf Area Indexes (LAI) ranging from 2.5 to 10. Assuming that a wind turbine around 3MW was placed in the clearing, spatial variations of available wind Energy (E), Average Wind Shear (AWS) and Cumulative Turbulent Kinetic Energy (cTKE) in layers where the wind flow intersects with the swept area were estimated from the simulations. Three hub heights were used to investigate its influence on the distribution of the variables. Unexpectedly, the results show that the clearing in the forest does not necessarily generates a gain in E or a reduction of cTKE or AWS compared to a completely forested area, and enlarging the gap does not necessarily improve these variables. For small clearing sizes, around 4h or less, the clearing has almost no effect. "Medium" sizes (around 10h to 40h) are the least suitable, providing the greatest reductions in E, and the greatest increases in cTKE and AWS. Continuing to enlarge the clearing can eventually progressively increases E but also AWS. Increasing the hub height does not reduce the relative changes in the flow generated by the clearing. Although increasing forest density, as well as concentrating the LAD more at the top of the trees, does intensify the effect of the clearings, the changes are not significant and the distributions remain largely the same.
| Date | 18 Aug 2021 |
|---|
| Original language | French |
|---|
| Awarding Institution | - École de technologie supérieure
|
|---|
| Supervisor | Christian Masson (Supervisor), Antoine Tahan (Supervisor) & Jonathon Sumner (Co-supervisor) |
|---|
Faurie, E. (Author), Masson (Supervisor),
Tahan (Supervisor) & Sumner (Co-supervisor),
18 Aug 2021Student thesis: Master's thesis › Master in Engineering: Engineering