Flamelet Progress Variable (FPV)

[SofT]

Hello,
I'm trying to implement the "Flamelet/Progress Variable" model, that known as "FPV" and is based on Laminar Diffusion Flamelet model, in the Cantera to create table of laminar counter flow diffusion flame structure, but I think it is  Not developed in Cantera yet.
Has anyone ever worked on this model in Cantera Code?
Thanks all.
SofT

[Chris N]

It isn't implemented with respect to the transformed flamelet equations. There is a stagnation flow code functionality. It's not as straight-forward as just marching through scalar dissipation rates to get the counterflow diffusion flamelets, but I think you can vary the strain rate(proportional to scalar dissipation rate) to create flamelet solutions. There is no arclength continuation method, so you have to manually march down the unstable branch.

[SofT]

Thanks Chris;

As I learned yet, the scalar dissipation rate can not provide a unique mapping of all of the chemical states in the flamelet tables. so this method fails in regions far from steady flamelet assumption that is "near equilibrium" chemical condition. then we have to use a new scalar as "Progress Variable" to overcome this problem.
Finally I think (upon my friend experience and as you say) the widely change of strain rate may lead to abnormal answers rather than truly prediction of unsteady flame condition.
Thanks anyway.

SofT

[Chris N]

That's true. So far as I know(been generating FPV flamelet tables for about 2 years now), you need a series of counterflow diffusion flame 1D laminar solutions for different values of the scalar dissipation rate. You fill out the S-curve of solutions encompassing the burning branch, the unstable branch, and the pure-mixing branch. One you have these you can generate a flamelet table that is parameterized by the progress variable. Do you have a reference for a paper that you're basing your work on?

[SofT]

Chris,

Thanks for the answer and I apologize for the long absence.

Available codes in the 1D flame model in Cantera, such as "diffusion_flame_batch.py" and "diffusion_flame_extinction.py" maximum can automatically go up to the extinction point of the s-curve due to absence of arclength continuation method.
Now my question is how it is possible to generate (even manually) flamelet tables of the unstable burnin branch, using Cantera?

Regards,
SofT