CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > Main CFD Forum

What's the difference in formulation for steady state and time-dependent modeling?

Register Blogs Members List Search Today's Posts Mark Forums Read

Reply
 
LinkBack Thread Tools Display Modes
Old   January 22, 2010, 19:49
Default What's the difference in formulation for steady state and time-dependent modeling?
  #1
Member
 
bearcat
Join Date: Jul 2009
Location: Ohio
Posts: 35
Rep Power: 7
bearcat is on a distinguished road
Send a message via MSN to bearcat
Suppose we solve the hyperbolic eq(good for all cases). We have the time term and the computations match in time, explicitly or implicitly. The convergence process will look like a flow field varying in "time" from initial to final steady state. But we don't call it time-dependent or time-accurate modeling. Transient modeling should be something else.

When modeling transient flow on codes like Fluent, one easily notices that the solver performs a few dozen sub-iterations after each time step advancement. But very few CFD books explain how this works, they are all for steady problems.

I know it's a stupid question. But can anyone please explain a little their formulation difference?

Thank you.
bearcat is offline   Reply With Quote

Old   January 23, 2010, 07:32
Default
  #2
New Member
 
Farshad
Join Date: Jan 2010
Location: Iran
Posts: 4
Rep Power: 6
f_panahizadeh is on a distinguished road
Quote:
Originally Posted by bearcat View Post
Suppose we solve the hyperbolic eq(good for all cases). We have the time term and the computations match in time, explicitly or implicitly. The convergence process will look like a flow field varying in "time" from initial to final steady state. But we don't call it time-dependent or time-accurate modeling. Transient modeling should be something else.

When modeling transient flow on codes like Fluent, one easily notices that the solver performs a few dozen sub-iterations after each time step advancement. But very few CFD books explain how this works, they are all for steady problems.

I know it's a stupid question. But can anyone please explain a little their formulation difference?

Thank you.
please explain your thread more.
f_panahizadeh is offline   Reply With Quote

Old   January 23, 2010, 08:13
Default
  #3
Senior Member
 
Claus Meister
Join Date: Aug 2009
Location: Wiesbaden, Germany
Posts: 241
Rep Power: 8
idrama is on a distinguished road
I don't understand you question too, but could your question related to Navier-Stokes-Equation or/and turbulent flow.

Cheers
idrama is offline   Reply With Quote

Old   January 23, 2010, 11:46
Default
  #4
Senior Member
 
Ahmed
Join Date: Mar 2009
Location: NY
Posts: 241
Rep Power: 8
Ahmed is on a distinguished road
Quote:
Originally Posted by bearcat View Post
Suppose we solve the hyperbolic eq(good for all cases). We have the time term and the computations match in time, explicitly or implicitly. The convergence process will look like a flow field varying in "time" from initial to final steady state. But we don't call it time-dependent or time-accurate modeling. Transient modeling should be something else.

When modeling transient flow on codes like Fluent, one easily notices that the solver performs a few dozen sub-iterations after each time step advancement. But very few CFD books explain how this works, they are all for steady problems.

I know it's a stupid question. But can anyone please explain a little their formulation difference?

Thank you.
1- write down the complete NS equations, the classical way
2- move the advection terms to the right hand side
3- Read the equation now
the time rate of change of phi = the total sum of the advection, diffusion and pressure gradients
4- when the left hand side is set equal to zero, then you have reached the steady state solution
In order to reach this numerical zero value, Fluent (or any other programme) discretise the left hand side by , say a first order Runge Kutta technique, and marching in small steps of time reaches the steady state....etc
the rest is explained in most books
good luck
Ahmed is offline   Reply With Quote

Old   January 23, 2010, 13:11
Default
  #5
Member
 
bearcat
Join Date: Jul 2009
Location: Ohio
Posts: 35
Rep Power: 7
bearcat is on a distinguished road
Send a message via MSN to bearcat
Thank you for all your reply. I think I should make the question more clear.

Actually I am talking about the unsteady flow simulation, particularly for incompressible flow. People may use the techniques like "dual-time steps" or "pseudo-time-level" to compute the time-accurate flow field. This is a common topic in CFD applications but very few CFD books talk about it in detail.
bearcat is offline   Reply With Quote

Old   January 25, 2010, 20:30
Default
  #6
Member
 
ganesh
Join Date: Mar 2009
Posts: 40
Rep Power: 7
ganesh is on a distinguished road
Dear Bearcat,

When you solve the unsteady flow equations, you attempt to update the solution at each time step, from day u^n to u^(n+1). Calculating this solution increment dU itself is a system of equations when considering the more popular and fast implicit schemes. You end up in solving linear systems using iterative solvers (such as Krylov solvers in every Newton iteration or relaxation schemes like SGS after approximating the Jacobian) which are seen as sub iterations at every iteration in physical time. With a suitable accuracy for the time--stepping scheme, this provides for a time-accurate computation, that is of importance for accurate transient simulations.

Hope this helps

Regards,

Ganesh
ganesh is offline   Reply With Quote

Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
Simulation of a single bubble with a VOF-method Suzzn CFX 18 October 2, 2009 05:18
mass flow in is not equal to mass flow out saii CFX 2 September 18, 2009 09:07
Transient simulation not converging skabilan OpenFOAM Running, Solving & CFD 12 September 17, 2007 18:48
Two-Phase Buoyant Flow Issue Miguel Baritto CFX 4 August 31, 2006 13:02
time dependent curve faiz rauf FLUENT 0 August 31, 2004 14:02


All times are GMT -4. The time now is 03:39.