- Posts: 14
- Thank you received: 1
Problem on simulating (polimi) soot model
4 years 2 months ago #36
by Hafiz
Problem on simulating (polimi) soot model was created by Hafiz
I was working on ethylene as a fuel so I made a C2H4 folder and make it go through a fuelInlet in my case. Spark duration is set to 0.01s and in that duration, the temperature peaked to the spark temperature. After that, the temperature dropped to an undesirable temperature. I am using soot model available on creck kinetic page. I didn't make any significant changes from the CH4 case but the solverOptions and radiationProperties. In solverOptions, I just added sootClasses off, (I'm thinking of not using the laminarSMOKE post processor just yet) and in radiationProperties, I'm not sure by the changes I made, so I just add C2H4 high temperature coefficients and got rid of CH4. I understand that there is a lot of species dangling in this case but should we add all the coefficient species inside the radiationProperties? My focus is on using C2H4 as a fuel for visualising soot distribution.
To clarify myself, my questions are down to these 3 things:
(1) Why is my temperature dropping?
(2) Should I add all the species in radiationProperties?
(3) In post-processing, is it possible to collectively name BINs as soot?
To clarify myself, my questions are down to these 3 things:
(1) Why is my temperature dropping?
(2) Should I add all the species in radiationProperties?
(3) In post-processing, is it possible to collectively name BINs as soot?
Please Log in or Create an account to join the conversation.
4 years 1 month ago #40
by acuoci
Replied by acuoci on topic Problem on simulating (polimi) soot model
In radiationProperties dictionary you have to specify ONLY the species who actively contribute to radiative heat transfer. Tipically you always include CO2 and H2O, which are the most relavant species with radiative properties. Additionally you may consider CO and CH4. Other species have no relevant role in radiative heat transfer. For numerical reasons, especially if you use the P1 model, it is convenient to provide N2 too, using a very small absorption coefficient (I think you can find an example in the tutorial files), just to be sure that a non-zero Planck mean absorption coefficient is always present in your computational domain (even in regions were there are no CO2, neither H2O).
In case you want to simulate flames including soot from the POLIMI mech, since the total number of species and reactions is huge (more than 300 species and more than 15000 reactions), it is better to use the following approach:
1) Simulation using a simple mechanism with a reduced number of species, to initialize the temperature and major species field (for example GRI 2.1)
2) Refinement of the simulation, using a larger mechanism, including larger species (C6H6 and C10H8, for example). A good candidate is the POLIMI C1C3HT mechanism. Of course you restart from the previous simulation.
3) Additional refinement of the simulation, using a larger mechanism including the formation of PAH (you may consider one of the POLIMI mechanisms for real fuels)
4) Simulation with the complete POLIMI soot mechanism. If you restart from solution (3) you already have a very good approximation of temperature and major species field and in addition you have also a good estimation of PAHs, which are the soot precursors. This simplifies a lot the calculations. Actually, this last step #4 can be split in additional sub-step:
4.1. At the beginning it is convenient to freeze the temperature field (i.e. switch off the energy equation)
4.2. Then you can turn on the energy equation, including the radiative heat transfer from gas phase only
4.3. Then you can turn on the radiation from soot too
As you can see, the simulation requires a lot of time and effort from your side. But this is not so surprising, since the problem you are trying to solve is really huge.
In case you want to simulate flames including soot from the POLIMI mech, since the total number of species and reactions is huge (more than 300 species and more than 15000 reactions), it is better to use the following approach:
1) Simulation using a simple mechanism with a reduced number of species, to initialize the temperature and major species field (for example GRI 2.1)
2) Refinement of the simulation, using a larger mechanism, including larger species (C6H6 and C10H8, for example). A good candidate is the POLIMI C1C3HT mechanism. Of course you restart from the previous simulation.
3) Additional refinement of the simulation, using a larger mechanism including the formation of PAH (you may consider one of the POLIMI mechanisms for real fuels)
4) Simulation with the complete POLIMI soot mechanism. If you restart from solution (3) you already have a very good approximation of temperature and major species field and in addition you have also a good estimation of PAHs, which are the soot precursors. This simplifies a lot the calculations. Actually, this last step #4 can be split in additional sub-step:
4.1. At the beginning it is convenient to freeze the temperature field (i.e. switch off the energy equation)
4.2. Then you can turn on the energy equation, including the radiative heat transfer from gas phase only
4.3. Then you can turn on the radiation from soot too
As you can see, the simulation requires a lot of time and effort from your side. But this is not so surprising, since the problem you are trying to solve is really huge.
Please Log in or Create an account to join the conversation.
4 years 1 month ago #41
by Hafiz
Replied by Hafiz on topic Problem on simulating (polimi) soot model
I used GRI-2.11 and the temperature rose to 2300K which is what we hope to see.. so thats the major species done but when I changed to kinetic mechanism to C1C3HT, I got this error:
I've copied Ydefault from 0 folder to the latest time folder and simulate it both on parallel and serial (just to be sure) but still get the error. Usually, from what I've experienced before, this is because of boundary conditions but I can't seem to figure out which went wrong. I've attached the log file here.
#0 Foam::error::printStack(Foam::Ostream&) at ??:?
#1 Foam::sigFpe::sigHandler(int) at ??:?
#2 ? in "/lib64/libpthread.so.0"
#3 log in "/lib64/libm.so.6"
#4 ? at ??:?
#5 ? at ??:?
#6 ? at ??:?
#7 ? at ??:?
#8 ? at ??:?
#9 ? at ??:?
#10 ? at laminarBuoyantPimpleSMOKE.C:?
#11 ? at ??:?
#12 __libc_start_main in "/lib64/libc.so.6"
#13 ? at /home/abuild/rpmbuild/BUILD/glibc-2.26/csu/../sysdeps/x86_64/start.S:122
Floating point exception (core dumped)I've copied Ydefault from 0 folder to the latest time folder and simulate it both on parallel and serial (just to be sure) but still get the error. Usually, from what I've experienced before, this is because of boundary conditions but I can't seem to figure out which went wrong. I've attached the log file here.
Please Log in or Create an account to join the conversation.
Time to create page: 0.165 seconds