Using computational fluid dynamics, we analyze the error trends of an analytical shear stress distribution model used to drive aeolian transport for coastal dunes, which are an important line of defense against storm-related flooding hazards. Additionally, two data-driven approaches (i.e., symbolic regression and projection-based, non-intrusive reduced-order modeling) are proposed for reducing the error while maintaining computational efficiency.
Simulation | 2D OpenFOAM simulations of flow over several idealistic dune profiles using the RNG k-epsilon turbulence model
Cite This Data:
Cecil, O., N. Cohn, M. Farthing, S. Dutta, A. Trautz (2024). "2D OpenFOAM simulations of flow over several idealistic dune profiles using the RNG k-epsilon turbulence model", in Examination of analytical shear stress predictions for coastal dune evolution. DesignSafe-CI. https://doi.org/10.17603/ds2-4w1m-7998
Hide Data
Simulation Type
Wind
Author(s)
; ; ; ;
Date Published
2024-12-13
DOI
10.17603/ds2-4w1m-7998
License
Open Data Commons Attribution
Description:
This dataset curates a series of 2D OpenFOAM simulations for 4 idealized dune profiles and 21 height-to-characteristic length ratios where the characteristic length is taken as the half-length at half-height. This data was used to examine error trends of an analytical surface shear stress model commonly used to drive coastal dune evolution models. Additionally, this data was used in the training of symbolic regression models and projection-based, non-intrusive reduced-order models.
This data accompanies a the paper published with the title "Examination of analytical shear stress predictions for coastal dune evolution".
Report | Associated Publication
Description:
This dataset accompanies the paper published with the title "Examination of analytical shear stress predictions for coastal dune evolution."
File Name
README
Simulation Model | 2D OpenFOAM
Description:
This dataset curates a series of 2D OpenFOAM v2006 simulations for 4 idealized dune profiles and 21 height-to-characteristic length ratios where the characteristic length is taken as the half-length at half-height. For all cases, the characteristic length was set to 25 m and the surface roughness to 1e-3 m. SimpleFOAM was used to solve the steady-state RANS equations using the RNG kEpsilon turbulence model. Atmospheric boundary layer profiles were prescribed for the inlet, the top boundary enforces a fixed shear stress condition on the velocity, and atmospheric boundary layer wall functions were used at the bottom surface. The overall domain was 425 m long and extends 200 m vertically above the top of the dune profile under consideration. The overall domain is symmetric about x=0 m and the dune is positioned within the domain such that the crest is at x=30 m.
File Name
Simulation Input | 2D OpenFOAM Case Files
Description:
The OpenFOAM case files can be made available upon request on a case-by-case basis.
File Name
Simulation Output | Processed 2D OpenFOAM Output
Description:
Contains the processed model output collected into two netCDF files. The OFMedium file contains height-to-characteristic length ratios 0.1 to 0.3 and the OFLarge file contains height-to-characteristic length ratios 0.3 to 0.5. Each case is included as group with the naming convention "[profile]_HL[height-to-characteristic length ratio]_RNGkEpsilon_WallSize[nominal first cell height]". Each group contains the processed physical and Fourier transformed dune profile and surface shear stress perturbation.
