What are the advantages and disadvantages of LES for naval architecture?

1 What is LES?

LES is a type of computational fluid dynamics (CFD) technique that aims to capture the most important features of turbulent flows, such as vortices, shear layers, and separation. LES divides the flow into two scales: the large eddies, which are resolved by the numerical grid, and the small eddies, which are filtered out and modeled by a subgrid-scale model. The subgrid-scale model accounts for the effects of the unresolved eddies on the resolved ones, such as energy dissipation, momentum transfer, and turbulence production. LES can provide more detailed and realistic results than other turbulence models, such as Reynolds-averaged Navier-Stokes (RANS) or detached-eddy simulation (DES), which either average or switch between the large and small scales.

2 Why use LES for naval architecture?

LES offers several advantages for naval architecture applications, such as capturing the unsteady and three-dimensional nature of marine flows which can improve the accuracy and reliability of force and power predictions. Additionally, LES can resolve complex flow phenomena around marine structures, such as vortex shedding and cavitation, that can enhance the understanding and optimization of hydrodynamic performance. Lastly, LES can provide high-fidelity data to validate, calibrate, and improve other turbulence models, such as RANS or DES, which can help reduce uncertainties and errors associated with them.

3 What are the limitations of LES for naval architecture?

LES has some disadvantages for naval architecture applications, such as its computational expense and time-consuming nature, which can limit its feasibility and efficiency for large-scale and long-duration problems. Additionally, LES is sensitive to the choice of subgrid-scale model, filter size, and boundary conditions, with no universal or optimal model for all flows and geometries. Furthermore, LES is not a fully predictive tool, as it still requires empirical or experimental data to initialize, calibrate, or validate the simulation. In addition, the boundary conditions must be compatible with the filter size and subgrid-scale model, which can be challenging for complex and moving boundaries. Finally, LES results may need to be compared with experimental data to assess accuracy and uncertainty.

4 Here’s what else to consider

This is a space to share examples, stories, or insights that don’t fit into any of the previous sections. What else would you like to add?