Dear All,

How can I couple the turbulence-combustion in the modeling of turbulent diffusion flames if the k-epsilon model is used for the turbulence field and detailed chemistry used for the combustion with multi-compoment species?

How to use the presumed PDF with only one scalar such as mixture fraction and its variance to get the mean multi-species concentrations and temperature?

Best regards,

Sun Wen

For detailed chemistry you need much more than the mixture fraction. Depending on the fuel you need to calculate the balance equations for great number of species (15-30).

If you use the Eddy-Break-Up model or the assumption 'mixed equal burned' it is possible to use mixture fraction only.

Your question is very unspecific. If you write more about the case you want to calculate, i can perhabs give you a better answer.

Roland

Hi,

Sorry to write an unspesific quenstion.

I don't want to use the "simple" model such as EBU or fast chemistry assumption. The detailed chemistry is my concern for the combustion model. e.g. for H2/Air turbulent diffusion flames, I can solve 8 species with 40 chemical reactions to deal with the quenching phenomena. The k-epsilon model is used for the turbulence.

But my question is how to couple the combustion-turbulence in a simplest way. I can get the mixture fraction from the element such as H, O etc. Is it possible not to use the joint PDF to calculate the mean chemical reaction rate? Some people use the mixture fraction and the progress rate for the two-scalar PDF and use equilibrium assumption to obtain each species. And I also don't want to use 'flamelet' model, because I am not familiar with it.

Many thanks for your more advices.

Sun Wen

In general it is impossible to calculate the chemical source with the mean composition!

If you want to use detailed chemistry, you have use a joint pdf. You can solve the pdf transport equation either with a presumed pdf or with a monte carlo approach.

I don't know if somebody has combined presumed-pdf with detailed chemistry so far. The monte carlo approach was combined with detailed chemistry or reduced ILDM-chemistry in some papers.

The alternative is DNS.