In this new post on Computational Fluid Dynamics (CFD) we will see a simple CFD simulation case using a LES (Large Eddy Simulation) turbulence model. This will allow us to see the eddies and instabilities of turbulent flow in a pipe. In order to observe this phenomenon we will include some artificial turbulence in the inlet and analyse how these inlet conditions affect the flow.
In the tutorial available in our Downloads section you can see all the necessary steps to activate this model and include the conditions using Acusolve.
CFD model preparation
The model will be in this case very simple, a pipe with an inlet and an outlet to which we assign an average inlet velocity as shown in the following image.
The mesh has a boundary layer suitable for the expected phenomena and a central area with a structured grid. The details of the simulated cases are explained below.
CFD Simulation and Results
First, a stationary case with SST turbulence model is simulated. These results are used as the initial condition for the LES simulation, which must be transient. This initialisation favours the stability of the CFD simulation. For this example, a period of 0.01s is simulated with a very small time step to accurately capture the evolution of the flow.
The results of the transient analysis set up in this way are as expected for a flow through a pipe. They are also very similar to those of the stationary case. The velocity profile is shown in the following image.
Artificial Turbulence
If we want to simulate a flow coming from a disturbance zone, such as section changes or moving elements inside the pipe, we can activate the artificial turbulence option in the inlet.
The input values can come from experimental measurements or be results from CFD simulation of another part of the system. This is a useful tool for splitting simulations into subsystems.
With the data described in the tutorial, simulating again the transient model with LES, the unstable structures appearing in the flow can be visualised at the different instants. The following images show the results at the instants corresponding to one third, two thirds and the end of the simulated case.
Once again we note the importance of defining the simulation input conditions correctly in order to obtain results that match reality.