Loading the lesson
Loading the lesson
Search AllRounder.ai
Search the courses, subjects, tracks, games and features, or jump straight to a page.
Search the courses, subjects, tracks, games and features, or jump straight to a page.
Learn it your way
A short class with Sarah and her students. The other ways are under More.
Sign up free to listen. Transcripts are free to read.
Today, we're going to explore Large Eddy Simulation, or LES. LES acts as a bridge between Direct Numerical Simulation and Reynolds Averaged Navier-Stokes equations. Can anyone tell me why DNS might be too computationally expensive?
Because DNS requires a lot of computational power to simulate all scales of turbulence, right?
Exactly! DNS is very accurate but consumes significant computational resources. LES, on the other hand, saves resources by modeling the small-scale structures. Can anyone guess why it's important to differentiate between large and small eddies?
Maybe because they behave differently in the flow?
Precisely! Large eddies are influenced heavily by the flow's geometry and boundary conditions.
What about the small eddies then?
Small eddies are more isotropic and have a universal behavior. They take energy from larger eddies during what we call an energy cascade. Can anyone remember what the name of the principle that describes this process is?
Is it the Kolmogorov hypothesis?
Well done! The Kolmogorov hypothesis is indeed key here. Let's move on to how we model the effects of these small eddies.
In the turbulent cascade, large eddies contribute energy to small eddies. Why do you think that transition is significant?
It shows how energy is shared across different scales in turbulence!
Exactly! This cascading process complicates turbulence modeling because small eddies behave differently than large ones. How does the distinction affect our computational models?
It means we need specialized models to handle the small scales.
Great insight! That's why we use subgrid scale models in LES to approximate the effects of small eddies. Let's examine how filtering works in this context.
In LES, we introduce a filtering operation. Can anyone explain what this means?
It probably means we filter out the smaller scales to focus on the larger ones?
Nice one! The filtering function helps isolate larger eddies. When we talk about grid size, does anyone remember how we differentiate between grid scales and subgrid scales?
Grid scales are for large eddies, and subgrid scales are for the small ones we don't directly compute!
Exactly right! Smaller eddies are modeled through turbulence models, while large eddies are solved directly. This approach simplifies our computations dramatically.
Lastly, let's touch upon the governing equations for LES. What do you think they include regarding the small eddy influences?
There must be terms related to subgrid scale stresses!
Correct! We'd have terms like the Leonard term, cross term, and SGS Reynolds stresses to represent these influences. Can anyone recall how these terms interact with the overall momentum equation?
They would be included to adjust the calculations to account for the missing small eddy effects.
Exactly! So, while we don't compute small eddies directly, their effects are critical for understanding the larger dynamics. Remember, a comprehensive turbulence model must capture the collective behavior of all eddies, which is a major challenge in turbulence research!
A method that resolves large eddies in turbulent flow while modeling the small ones.
Understanding the different behaviors between large (anisotropic) and small (isotropic) eddies.
The transition of energy from large eddies to smaller ones that complicates turbulence modeling.
A technique used to account for the effects of smaller scales not resolved directly in simulations.
Governing equations that represent large eddy dynamics with small-scale influences.
A theory stating a universal behavior for small eddies.
From large to small, energy flows, / In turbulent acts, the cascade glows.
Imagine a river flowing from a mountain. As it rushes down, it splits into smaller streams, just like how large eddies feed into smaller ones during turbulence.
LES = Large Eddies Simulated, SGS = Small Eddies Guesstimated.
LES stands for Large Eddy Simulation, where 'Eddy' is the main focus of modeling.
Term
What is Large Eddy Simulation?
Click to flip
Term
What does the Kolmogorov hypothesis state?
Click to flip
Term
What are subgrid scales?
Click to flip
Term
What is an energy cascade?
Click to flip
A simulation technique that resolves large turbulent eddies while modeling smaller ones.
A highly accurate computational method that solves the Navier-Stokes equations for all turbulence scales.
A method that averages the effects of turbulence, leading to less accurate but computationally cheaper solutions.
The process by which energy in turbulent flows transfers from larger eddies to smaller eddies.
A principle stating that small eddies have a universal behavior independent of the large-scale flow.
Eddy scales that are smaller than the grid size and are not directly resolved in simulations.
The scales of turbulence that are resolved directly by the computational grid.
The governing equations used in LES which account for large-scale dynamics while modeling small-scale effects.
The stress from subgrid scales that affects the flow's momentum.
A term in the LES equations associated with the energy transfer from resolved to unresolved scales.
This section discusses Large Eddy Simulation (LES) as a technique for modeling turbulent flows, striking a balance between accuracy and computational efficiency.
The section explores Large Eddy Simulation (LES) as an intermediate modeling approach between Direct Numerical Simulation (DNS) and Reynolds Averaged Navier-Stokes equations. It focuses on the distinctive behaviors of large and small eddies in turbulent flows and introduces subgrid scale modeling to handle the finer details not captured by the grid size.
This section details the Large Eddy Simulation (LES) method, which serves as a middle ground between the highly accurate but computationally expensive Direct Numerical Simulation (DNS) and the less accurate Reynolds Averaged Navier-Stokes (RANS) equations. LES effectively captures large-scale turbulent structures while modeling the influence of small-scale turbulence.
In turbulent flow, large eddies are anisotropic and predominantly governed by the geometry of the problem and boundary conditions.
The interaction between large and small eddies involves an energy cascade where large eddies feed energy into smaller eddies. This fundamental characteristic complicates the creation of universal turbulence models since large eddies exhibit dependency on various parameters that do not apply to smaller eddies.
LES employs a filtering function to separate the large and small eddies. The size of the mesh must be sufficiently small to accurately capture the large eddy dynamics, leading to the definitions of grid scales (GS) for large eddies and subgrid scales (SGS) for the smaller ones.
The filtered momentum equation in LES accounts for the influence of subgrid scales via subgrid scale stress terms. These terms include contributions such as Leonard, cross-term, and SGS Reynolds stresses, which factor into the overall turbulence modeling, thus allowing a more manageable computational workload than DNS while still delivering valuable insights into turbulent behavior.