Hello all
I am trying to validate a hydrogen leakage problem. I am using Composition PDF transport model and using hydrogen-air mixture in Fluent. There are two inlets in the geometry, one for wind and one is the hydrogen leak point. While specifying the boundary conditions at the inlet, there is an option for initial species fraction. I have specified it to be 1 for hydrogen at the leakage inlet. But my solution is not converging.
Precisely, there is problem with the convergence of continuity equation. I have tried altering the time step size, but nothing works. At this point, I am not sure where I am wrong.
If somebody has any idea regarding this, please shed some light.
I am following this paper,
Article CFD modelling of accidental hydrogen release from pipelines
The components of the mixture should sum 1 at both inlets and in the domain initialization.
there may be a few reasons.. but of course I am guessing what the problem maybe.. so you may have tried all these before..
1- make sure you are using gaseous H2. Boiling temp for H2 is about 20K and the wind? will be at a much higher temp. it will boil off when it is exiting... or you may go with some other mechanisms later..
2- define the H2 inlet as if it were travelling in a short pipe and that it was then exiting into the atmosphere. and put inflation layers in the short pipe.
3- make sure the cell sizes are small enough for the pipe and that the there is no abrupt change from the pipe exit to the domain.
4 - make the H2 inlet a mass-flow inlet..
these came to my mind.
Thank you for your suggestions.
The simulation is running now, not smoothly but satisfactorily.
What I did differently now is that I switched from k-epsilon to k-omega, decreased mesh size at the leakage point and kept the sum of species mass fraction at both gas and wind inlets as 1.
- Badges
- Science topic
- Similar topics
- Heat