2.5. Overall Summary of LES

Another such technique is called Large Eddy simulation. See in the DNS one important thing to note was that we had the best accuracy but lot of computational time is required. LES is sort of a tradeoff between the Reynolds average and DNS.

Large Eddy Simulation (LES) is a computational fluid dynamics (CFD) technique that aims to accurately simulate turbulent flows while balancing the computational costs. Unlike Direct Numerical Simulation (DNS), which provides high accuracy at the expense of significant computational power, LES offers a compromise. It maintains higher fidelity compared to Reynolds averaging methods by simulating the large, energy-carrying eddies directly while modeling the smaller eddies.

Consider the difference between trying to capture every detail of a busy street (like DNS) versus focusing on the larger vehicles and their movement while estimating the smaller pedestrian movements (like LES). The former is exhaustive and time-consuming, while the latter gives a reasonable approximation without needing to track every small detail.

So there is a big difference in the behaviors of large and small eddies in turbulent flow fields. Large eddies are more anisotropic and their behavior is dictated by the geometry of the problem domain and the boundary conditions. Small eddies are nearly isotropic.

In turbulent flows, there are two types of eddies: large eddies and small eddies. Large eddies are influenced by the geometry and boundary conditions within the flow, making them anisotropic, or directionally dependent. In contrast, small eddies are more uniform in behavior (isotropic) and do not vary significantly based on their surroundings. Understanding the differences between these two types of eddies is crucial for accurately modeling turbulence in simulations.

Think of large eddies as large ships that navigate through large harbors, influenced by the harbor's shape and conditions (the boundary conditions). In contrast, small eddies are like small boats, which move relatively freely and uniformly within the water, less affected by the harbor's shape.

The important thing to remember is that the large eddies extract energy from the mean flow, while small eddies take energy from slightly larger eddies. This process is referred to as the energy cascade.

In turbulent flow, energy is transferred from larger eddies to smaller eddies in a cascading manner. The large eddies extract energy from the main flow, providing energy to smaller eddies. These smaller eddies, in turn, interact and gain energy from slightly larger ones, nurturing a continuous cycle of energy distribution among different scales of eddies. This phenomenon illustrates the complexity of turbulence and poses challenges for modeling.

Imagine pouring water from a large bucket (large eddies) into smaller cups (small eddies). The larger bucket has more water, allowing it to fill the smaller cups. As the larger bucket empties, it can affect the flow, similar to how large eddies impact smaller ones.

In LES, the larger eddies are computed with a time-dependent simulation where the influence of the small eddies is incorporated through a turbulence model.

Large Eddy Simulation involves resolving the large eddies directly while using a turbulence model to account for the effects of small eddies, which are not directly simulated due to their fine scale. This approach allows for a more computationally efficient simulation while still capturing the essential dynamics of turbulence influenced by larger eddies.

Think of it like a camera capturing a large scene (large eddies) while using a blur effect to represent the details of small movements (small eddies). Instead of focusing on every tiny detail, the camera captures the essence of the scene while applying a filter to represent the smaller movements.

The scales that are directly solved for on the grid are called the grid scales for large eddies, and for the smaller ones, the subgrid scales (SGS).

In the context of LES, the simulation grid is designed to resolve the large-scale features of turbulence directly (grid scales), while smaller features that cannot be resolved are modeled (subgrid scales). This distinction is critical for ensuring that the simulation captures significant flow structures while still being computationally feasible.

Imagine a map where major highways (grid scales) are clearly represented, but smaller local roads (subgrid scales) are not shown. The major highways give you a good understanding of the overall traffic flow without needing to detail every single road.

--