[pagru]

Hello everyone,

I've already searched the forum, but couldn't find exactly what I'm searching for, so I hope somebody can help me!


My problem/situation is:

I'm simulating a system where the flow inlet is a simple channel (4.0 m X 5.6 m).
The walls of the channel are declared as "wall", hence they do have friction.
The boundary layer resolution consists of:
- 5 Layers
- expansionRatio = 1.25
- finalLayerThickness = 0.4
- base mesh size is approx. 25 mm

Simulation type is DDES with Spalart-Allmaras turbulence model.

When running the simulation and controlling the static pressure (the pressure coefficient), there are pseudo stagnation points right at the edges between the inlet face and the walls. I assume that they are the result of the beginning wall friction.

Since I need a certain mass flow into my system, my inlet BC for the velocity is "surfaceNormalFixedValue" with "refValue" approx. 40 m/s.

So I'm forcing a uniform velocity profile into the channel, but due to the boundary layer evolution there are stagnation points falsifying the desired pressure distribution.


My idea:
My idea is to run a seperate simulation with an elongated inlet channel (same cross section and meshing).
In that simulation I'd search for a cross section with the right velocity/boundary layer profile and an appropriate nuTilda to match my desired inlet profile.
Then I'd like to write the data of this cross section to an .csv-file and read that in as my Inlet-BC for the actual simulation.

Unfortunately, I don't really know which functionality of OF I might use for that. I read about "mapFields" and "TimeVaryingMappedFixedValue" in similar problems.
Could somebody give me a hint on that and/or explain how to do this properly?

If so, many many thanks in advance!!!

Regards,
Philipp

[yongxiang]

Hi Philipp,
I am also running into such problem. That I also need to read from an external file, interpolate it and map it onto the inflow patch. Do you solve this problem now? would you please share you idea?
regards
Yong