2.3. Governing Equations of LES

LES is a technique that offers a compromise between Direct Numerical Simulation (DNS) and Reynolds Average Navier-Stokes (RANS) equations. It involves solving large eddies directly while modeling the effect of smaller eddies.

Large Eddy Simulation (LES) is a computational fluid dynamics technique that balances accuracy and computational workload. Unlike DNS, which provides high fidelity results but is very resource-intensive, and RANS, which uses many approximations resulting in less accurate data, LES aims to solve the dynamics of large-scale turbulence directly. It models the smaller scales, which saves computational resources while still capturing significant flow features.

Imagine trying to understand how a river flows by watching both the large currents (large eddies) and the tiny ripples on the water's surface (small eddies). Rather than focus on every tiny ripple, which would take a lot of effort to track, you can study the major currents to understand the general flow pattern and rely on simplified rules to understand the smaller ripples.

In turbulent flow, large eddies are more anisotropic and influenced by the problem’s geometry and boundary conditions, while small eddies are nearly isotropic and demonstrate more universal behavior.

Eddies in turbulent flow can be categorized into large and small. Large eddies, because of their size and the forces acting on them, tend to be affected significantly by the surrounding environment such as the shape of the container they flow through. They can be quite chaotic. In contrast, small eddies are more uniform in behavior, similar across different environments, and can be modeled using the theories of turbulence developed by researchers like Kolmogorov.

Think of large eddies as people in a crowd where every person moves differently based on their surroundings and interactions. In contrast, small eddies are like grains of sand in a sandbox, which tend to move in similar ways regardless of the particular conditions around them—they just shift and settle into place.

In the energy cascade phenomenon, large eddies extract energy from the mean flow, which is then transferred to smaller eddies. This process forms part of the Kolmogorov hypothesis.

The energy cascade refers to how turbulence spreads energy from larger scales to smaller scales. Large eddies take energy from the general flow of the fluid, and as they interact and break down into smaller eddies, this energy is passed down through the smaller scales. The Kolmogorov hypothesis suggests that there is a statistical similarity in small eddies, allowing for universal modeling of turbulence on those scales.

Imagine a large waterfall (the large eddy) that splashes water into smaller streams (the small eddies) downstream. The waterfall brings in a lot of energy, but as it hits the rocks and spreads out, that energy is divided among the tiny trickles of water that now flow in different directions, each carrying a part of the larger flow's energy.

In LES, large eddies are resolved using time-dependent simulations, while the effects of small eddies are accounted for through turbulence models. The grid size must be fine enough to capture the largest eddies.

LES employs a time-dependent simulation to compute large eddies directly, while small eddies are included in the overall flow behavior using turbulence models. Grid size is crucial; it needs to be small enough to resolve those large eddies accurately. If the grid is too coarse, it may miss crucial dynamics of the larger scales, affecting the accuracy of the simulation.

Think about trying to take a detailed photo of a cityscape. You need a high-resolution camera (a fine grid) to capture the buildings and landmarks (the large eddies). If you use a low-resolution camera (a coarse grid), you might just get a blurry image with no details. But even if you can’t capture the subtle nuances of every leaf on a tree (the small eddies), you still want to portray the overall architecture of the city correctly.

The filtering operation differentiates between large and small eddies in LES. The filtered Navier-Stokes equations account for large eddies, while the influence of small eddies is represented through subgrid-scale stresses.

In LES, a filtering function is used to separate the effects of different scales. The filtered Navier-Stokes equations handle the larger eddies while including a term (SGS stress) to account for the smaller eddies’ influence. This approach allows for a simplified model that still captures key dynamics without needing to resolve every small fluctuation.

Imagine sifting flour through a sieve. The sieve only allows the fine flour (the large eddies) through while leaving behind the larger clumps (the small eddies). By using the flour that passes through, you're still able to bake a great cake (the overall flow dynamics) without needing to focus on what’s left behind.

--