[sasha-]

hi there, I hope you are having a great day!
I wanted to apply the variation of air properties with the altitude in my simulation(a wind turbine) in cfx pre, I know I have to use a CCL code but have no idea how to do it. I would appreciate any help on this topic.

ps: I'll attach an image of the expressions.

Screenshot (996).png

[Opaque]

May I ask what your goal is? Forcing the density that way will break all the thermodynamics within the software.

You are making pressure and temperature a function of h(m), and not a function of the flow.

Alternatively, if you substitute T and P as given into density, it means the properties are fixed as a function of h(m) and independent of the flow conditions.

[sasha-]

Hi, thanks for your reply!
actually, I simulated my turbine without applying these expressions, and the results have almost 15% error from the paper's results. I already have checked mesh and everything else and they seem to be ok, so thought not applying these expressions(which are mentioned in the paper as CCL code in cfx pre) would be the reason. Do you think it's not going to be effective on the results?

[Opaque]

Have you verified your results are mesh independent already?


May I ask about the context of the solution in the paper? Experimental results? Numerical calculations? Error bands for the paper's results? Model differences between your model and theirs?

w/o that information from both simulations: yours and the papers, any of them can be correct. The fact those results are published does not make them correct.

You need to systematically analyze both: the paper's approach, and your approach to obtaining the results and understanding the scope of each model

[sasha-]

hi, thank you for your answer.
yes, my results are mesh independent.
it is a simulation of an airborne wind turbine consists of a rotor and a balloon.
the only difference between my model and the paper's is that my balloon does not have wings but the paper's has.
in the paper, just the results of the rotor without the outer shell (balloon) have been compared to the available experimental results of the rotor. I guess I can validate my results in this way as well.

[ghorrocks]

If you are intending to have the simulation run at constant density, temperature and pressure; but with the ambient density, temperature and pressure defined by your equations then this is very easy.

If you are intending to have a simulation with variable temperature, density and pressure then, as Opaque says, your functions are not going to work.

You should only need to use variable temperature, density and pressure if they vary significantly over the domain. I would think that unlikely for a turbine.

A while ago when I did some CFD of air flow over a mountain range, where the density, temperature and pressure did vary due to the huge altitude range I was modelling - the best way to simulate it is to model potential temperature (https://en.wikipedia.org/wiki/Potential_temperature) rather than temperature. The modelling is here: https://www.computationalfluiddynami...ation-mystery/