While doing simulation for unsteady state fluid problem, we use time intervals and time steps in fluent. What is the concrete meaning of these terms?
Hi,
The time step is the incremental change in time for which the governing equations are being solved. It is advised to select small time-step size to capture all the flow fluctuations.
Hope it helps.
Mohsen
Hi, The time step is the incremental change in time for which the governing equations are being solved. It is advised to select small time-step size to capture all the flow fluctuations.
Hope it helps.
Mohsen
Unsteady means that the fluid flow is time varying. Time derivatives are approximated by finite differences in time, this finite difference is delta v/ delta t (for example)... Then the time step is this delta t that comes from the time discretization of the problem.
Time steps are used for advancing real time in small steps to compute the solution for an unsteady problem. The minimum required time step is calculated using the CFL number (also specified in Fluent) for an explicit and semi implicit condition so that the numerical dissipation through convection terms does not affect our solution.
Hello Punet
In unsteady flow, conditions vary with time. So Time interval is the period of time elapsing while the phenomenon is happening (from beginning to end) and time step is the incremental change in time for solving the governing equations.
Regards
the time step delta_t must be small enough to resolve time-dependent features; make sure that the convergence is reached within the number of Max Iterations per Time Step. Its size estimate can also be chos en so that the transient characteristics of the flow can be resolved. delta_t can be estimated by: "the small discretization size / a characteristic velocity of the flow". Hope it helps.
How to calculate the number of time steps from the time step size and time...?
when you discretize the term of evolution (unsteady) in the conservation equations (of continuity, momentum and energy) example in the energy equation dT / dt + u.gradt (T) = ....
Here dT / dt = ~ (3 * T (t ") - 4 * T (t ') + T (t)) / (2 * dt) + something in the order of dt².
t '= t + 2 * dt, t' = t + dt, t=n*dt (the nth itertaion)
If you know that your problem is steady you can omit dT / dt, else if unsteady you coose a time step dt which depends on the problem. Normally it should be less than 0.1 and can go to 1E-7.
Simply put, time steps are used to specify the increase in interval for your analysis/simulation, which help the equations to solve. ANSYS as well as other softwares like Orcaflex, use them to solve the equation for any given problem
- Badges
- Science topic