CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   FLUENT (https://www.cfd-online.com/Forums/fluent/)
-   -   Averaging in space (https://www.cfd-online.com/Forums/fluent/177839-averaging-space.html)

dedike September 22, 2016 06:30

Averaging in space
 
Dear all,

I'm a new member, so first of all hello to everyone :)

Currently, I'm writing my thesis and I'm facing with a problem which seems to be very easy, however I could not find a solution for it. I carried out Large Eddy Simulation in FLUENT and now I would like to postprocess the data. I have access to Tecplot via the university, but unfortunately it is very slow. I have no experience in Paraview just only in CFD-Post so that if someone knows a solution for my problem in CFD-Post that would be the best. However, if you have an idea for the realization in Paraview please share that one too.

About my simulation: I'm modeling a flow around a spinning cylinder in the transitional regime, Re = 140k. The base of the cylinder is placed to the xy plane and it is rotating around the z axis. I'm using an O-type grid which extends 20 diameter to the radial direction. The axial extension of my domain is 1 diameter. At the inlet an uniform horizontal velocity condition is imposed, whilst at the outlet outflow is used. I have 2 small symmetry boundary above and below the cylinder (so my domain in the xy plane almost looks like a big circle). In the axial direction periodicity is considered.

In order to draw the streamlines I need to average out my velocity field in time as well as in space. In FLUENT I turned on the 'data sampling for time statistic' command, thus I have the mean velocity field. The first problem came when I wanted to draw streamlines based on the mean velocity field. In CFD-Post there is only option for drawing streamlines based on the instantaneous field. After exporting the mean velocity data to a .cdat file and doing modification with the ASCII file (basically replacing the instantaneous values with the mean ones) I managed to draw the streamlines based on the mean field. I wanted to use the same idea for space averaging. I averaged out the u,v velocities in the z direction in MATLAB then I copied back the values. Obviously I set w to be 0. Unfortunately, this trick did not work. I got odd streamlines which are inclined next to my outlet boundary. The boundaries are far away from my cylinder so the streamlines should be horizontal close to them.

My question is that, is there anyone who had success in averaging vector field in space with CFD-post or Paraview? I can hardly imagine that these softwares do not have this feature.

Any help is highly appreciated :)


All times are GMT -4. The time now is 01:55.