Modelling the convective boundary layer at microscale using RANS
Fuente:
Zenodo
Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | Recurso digital |
| Publicado: |
Zenodo
2023
|
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866902058097967104 |
|---|---|
| author | Bleeg, James Del Hoyo, Mirko Montavon, Christiane |
| author_facet | Bleeg, James Del Hoyo, Mirko Montavon, Christiane |
| contents | <p>Computational fluid dynamics (CFD) models solving the Reynolds-Averaged Navier-Stokes (RANS) equations are often used in the wind industry to simulate the interaction of the atmosphere with terrain and wind turbine arrays. In practice, the atmospheric boundary layer (BL) is usually simulated as neutral, or sometimes stable; only rarely is it simulated as convective with capping stratification—this despite the fact that such conditions occur frequently at most wind projects.</p> <p>In many ways, RANS is not well-suited for simulating the unstable/convective BL at microscale. The RANS turbulence closure scheme is intended to represent all turbulent scales, yet RANS can also directly resolve thermal convections, resulting in the double-counting of turbulence effects. Worse, experience indicates that the resolved convective structures depart substantially from what occurs in an actual unstable BL; these spurious structures can have a very large impact on the quantities of interest, rendering the simulation results practically unusable. Furthermore, buoyancy production terms based on local gradients, common in microscale RANS models, are not able to reasonably represent the impact of buoyancy on turbulence in an unstable BL, particularly within the mixed layer.</p> <p>The wind energy community has pursued a few different options to deal with these challenges. Academic researchers typically avoid RANS altogether, preferring LES for microscale modelling. RANS, however, is much more computationally efficient than LES, and thus industry has not yet given up on its use. The industrial approach to simulating convective boundary layers with RANS often involves removing buoyancy from the momentum equation and adding source terms to the closure equations based on Monin-Obukhov (MO) similarity theory. Since the important influence of stratification above the BL acts primarily through the buoyancy term in the momentum equation, and since MO is only valid in the surface layer and not in the mixed layer, we explore a different approach in the research reported herein.</p> <p>The presented approach avoids resolving convections while still representing their impact on the mean flow—from the ground surface up through the capping inversion—while also directly capturing any inviscid effects related to stratification above the BL, such as gravity waves. The RANS simulations of the convective BL using this approach are validated against several field observations. </p> |
| format | Recurso digital |
| id | zenodo_https___doi_org_10_5281_zenodo_15039844 |
| institution | Zenodo |
| language | |
| publishDate | 2023 |
| publisher | Zenodo |
| record_format | zenodo |
| spellingShingle | Modelling the convective boundary layer at microscale using RANS Bleeg, James Del Hoyo, Mirko Montavon, Christiane <p>Computational fluid dynamics (CFD) models solving the Reynolds-Averaged Navier-Stokes (RANS) equations are often used in the wind industry to simulate the interaction of the atmosphere with terrain and wind turbine arrays. In practice, the atmospheric boundary layer (BL) is usually simulated as neutral, or sometimes stable; only rarely is it simulated as convective with capping stratification—this despite the fact that such conditions occur frequently at most wind projects.</p> <p>In many ways, RANS is not well-suited for simulating the unstable/convective BL at microscale. The RANS turbulence closure scheme is intended to represent all turbulent scales, yet RANS can also directly resolve thermal convections, resulting in the double-counting of turbulence effects. Worse, experience indicates that the resolved convective structures depart substantially from what occurs in an actual unstable BL; these spurious structures can have a very large impact on the quantities of interest, rendering the simulation results practically unusable. Furthermore, buoyancy production terms based on local gradients, common in microscale RANS models, are not able to reasonably represent the impact of buoyancy on turbulence in an unstable BL, particularly within the mixed layer.</p> <p>The wind energy community has pursued a few different options to deal with these challenges. Academic researchers typically avoid RANS altogether, preferring LES for microscale modelling. RANS, however, is much more computationally efficient than LES, and thus industry has not yet given up on its use. The industrial approach to simulating convective boundary layers with RANS often involves removing buoyancy from the momentum equation and adding source terms to the closure equations based on Monin-Obukhov (MO) similarity theory. Since the important influence of stratification above the BL acts primarily through the buoyancy term in the momentum equation, and since MO is only valid in the surface layer and not in the mixed layer, we explore a different approach in the research reported herein.</p> <p>The presented approach avoids resolving convections while still representing their impact on the mean flow—from the ground surface up through the capping inversion—while also directly capturing any inviscid effects related to stratification above the BL, such as gravity waves. The RANS simulations of the convective BL using this approach are validated against several field observations. </p> |
| title | Modelling the convective boundary layer at microscale using RANS |
| url | https://doi.org/10.5281/zenodo.15039844 |