[zhubohong]

Hi, everyone. Recently, I tried to use the CFX-CEL tool to modify the material property. The modified equation involves the integral equation, as shown in the following figure. Dose someone know what is the function symbol of intergration in CFX-CEL?

[Opaque]

Would you mind describing in more detail?

There is no integral in the figure shown.

[zhubohong]

Originally Posted by Opaque

Would you mind describing in more detail?

There is no integral in the figure shown.

I want to simulate the density change of gas phase as a result of the temperature difference when it is injected in the high temperature liquid phase.
It means I don't use the energy equation to model it, just predict it by modifying the material property of gas phase. As shown in Fig., the temperature change of gas phase is a function of the time that gas phase contacts with liquid phase. This is a intergral equation. Therefore, my question is what is symbol of intergration in CFX-CEL. Just like, in CFX-CEL, the symbol of multiplication is *, the symbol of exponentiation is ^, etc.

[ghorrocks]

CEL has no integral function like you describe. It has some line, area and volume integral functions - see the CFX documentation reference guide for that.

Also - rather than trying to make your own model to get the temperature why not consider the built in lagrangian temperature model? Again, have a look at the documentation to see if the built in temperature model will do what you want.

[zhubohong]

Originally Posted by ghorrocks

CEL has no integral function like you describe. It has some line, area and volume integral functions - see the CFX documentation reference guide for that.

Also - rather than trying to make your own model to get the temperature why not consider the built in lagrangian temperature model? Again, have a look at the documentation to see if the built in temperature model will do what you want.

Thanks! It sounds the above equation cannot be modeled using the CFX. BTW, could I ask an another question about the CFX simulation method. My model uses the k-e turbulence model. I want to change the density of gas phase, and it is a function of e (turbulence dissipation). It means I should evaluate the value of e at initial time, and the later value of e calculated from the flow field can be automatically fed back until all the parameters reach the stable value. Is it possible to be acheived in CFX?

[ghorrocks]

That is not what I said at all.

First of all, there already is a model for temperature in CFX. You should look at this model to see if it does what you want. You don't need to make your own model for it.

And if that does not work then you can probably do it using additional variables. But I am hopeful the built in temperature model will work for you, so look at that first.

[zhubohong]

Originally Posted by ghorrocks

That is not what I said at all.

First of all, there already is a model for temperature in CFX. You should look at this model to see if it does what you want. You don't need to make your own model for it.

And if that does not work then you can probably do it using additional variables. But I am hopeful the built in temperature model will work for you, so look at that first.

My present model is a little bit complex now. I am not just simulating heat transfer process, it also involves the multiphase flow. So I think it is hard to add more models, such as temperature model, and compressible gas flow.


As I said that before, if the density property of gas phase is a function of e, I can't directly modify the gas property using this equation. This is because the flow field is not calculated, and the value of e cannot be fed back. As you said, if I use a additional variable to read the value of e and change the equation, which is a fucntion of additional variable, to describe the density of gas phase, can this way work?

[ghorrocks]

I have just checked the Lagrangian model and I think the temperature model built in will do what you are looking for.

Alternately, you may be able to model this with a Eularian model and I think you will also be able to model the effects you are talking about there as well.

You should not build your own model for something which looks like it is already in the software. If the built in models are not suitable please explain why they are not suitable.