[suggalachaitu]

I am doing dynamic analysis for towers due to wind. The boundary condition for velocity changes every second . How can I apply such kind of boundary condition which varies with time.

[Komon]

hi to everyone,

to first reply to the question before-well, it depends on the type of simulations you do. If you do (unsteady I suppose) RANS, I'd suggest to make a function which changes the average velocity profile's direction...

going onto my problem now:

I might have a similar question, although from completely different branch of simulations. Namely, I'm trying to do some dns simulations in a venturi test section. I have generated some inflow velocities using periodic channel flow simulations (LES), but the matter is that I need a lot of data in order to simulate one-two flow through times in a venturi. which demands also a lot of time spent for LES of channel flow. Therefore I'm looking for something else. I've seen there is a chance to impose cyclic boundary conditions. If someone wouldn't mind sharing a nice manual/article to quickly look into this and try to impose it, it would be nice...but I have another question. Since I have some 3d velocity fields I saved from my LES simulations, which are also periodic, I'm wondering if I could try to impose inlet velocities by extracting 2d velocity planes from those 3d fields. Something like taking a 2d snapshot from 3d field as the time progresses. Since a 3d field I have is periodic, it means I have infinite data to be imposed on the inlet practically. But does anyone know if such a thing has been done before, and how? How would I progress/move across my 3d field to get good positions of my 2d snapshots for the inflow? with average velocity or sth else?

regards