[hamid1]

Hi
Is it always correct to use a steady state result as initial condition for unsteady simulation?

thank you so much

[ghost82]

It depends if you want to study your system starting from the steady state...

[hamid1]

thanks for reply, i actually want to have a better convergence rather than starting from steady state, am I allowed to to do that? (i mean starting from steady state)

[ghost82]

Describe your system and what you have to study.

[hamid1]

Hi Daniele

Plz imagine a piping system with three inlets (pressure inlet, amount: P) and one outlet (pressure outlet, amount: -P),
There two phase of flow,
I initialize with pressure for all the domain (amount: P), I simulate with one phase flow in steady state and I take the result as the initial condition for the transient simulation, in transient simulation I introduce the second phase to solve it through VOF method,

P.S. the reason for the bc and initial condition is because I assume the fluids of both phases are initially stationary in the domain, then the suction (outlet pressure -P ) starts to suck the fluids and mainly the second phase.

Thanks for the helps

[ghost82]

Hi Amid,
if you initialize the transient simulation with the steady results of the primary phase and in transient simulation you introduce the secondary phase, you are simulating a sort of injection of a secondary phase during a steady flow of the primary phase, and your study starts from this.

If you want to study the flow of two phases both starting from static conditions I think you have not to initialize the transient simulation with steady results.
In addition, If you want to study the transient from static conditions take in mind that inlet pressure value will not be the steady value for t=0 s (for example 6 bar), but probably it will take some time (milliseconds) to reach 6 bar from atmospheric pressure; this can be easily implemented with an udf.

Daniele

[hamid1]

thanks so much for reply