Numerical simulation of a turbulent plume
 

Hi, Foamers, I am running a large eddy simulation of a turbulent plume in a density varying medium (stratified medium). I am comparing mean axial velocity with an experimental study. I found that qualitatively I m getting a good agreement. However quantitatively, There are differences that can be clearly seen in the velocity. I am trying to figure out why this difference or deviation is coming.

For details: I am using the dynamic k-equation model in OpenFOAM-5.0. I am using buoyantBoussinesqPimplefoam solver (which I modified for the stratified medium). Apart from that, the plume is a forced plume with initial buoyancy and momentum. I am using around 10 million cells for the computation.

I have also plotted turbulent kinetic energy (LES) with the experimental value on the centerline. I observed that resolved Kinetic energy is more as compared to experiment, which quite strange. I am not able to figure out what is the issue. I WILL BE very grateful if any experienced user help.

Also, I observed that the entrainment is less from the velocity contours.

From your second chart, it looks like the entrainment rate is higher in your LES than in the experiment, which matches the observation that the TKE levels are higher.

I know that this is an old post, but did you find a cause for the difference?