Is it preferrable to use LARGE EDDY SIMULATION over the K-e model for analysing the flow pattern in 3d street canyon.

my project mainly deals with analysing the flow of pollutants released by the automobiles by changing the wind direction normal to the street canyon.

I would be thankful to you if u send the url's of any related papers to 3D street canyon and some related results for validation.

yes..LES is always prefered ..for urban flows because RANS model overpredicts TKE. For experimental work on street canyons..you can read Brown et. al.(2001), Steve Hanna, Petra Kastner klein, Eric pardyjak, S. Pal Arya, Sugrammond..etc...you can search these name in you library catolog. i dont have the urls. For numerical simulations..Moin et. al,...

Its a very popular problem..so you wont find any difficulty in getting the literature

akshay

thanks akshay

now actually i have performed the simulation using K-epsilon and large eddy both of them but what i got are two extremely different flow patterns the one with k-e has got vortex flow and the one with LES has not got the vortex(with the same grid and boundary conditions), Is it that LES used for unsteady and 3D models for large scale makes the difference

if u would like to view my results i would be happy to send you which igot working on fluent to you email ID with your permission

thank you.

go to www.sciencedirect.com and search for "street canyon" you should hit at least 50.

the results should match, I have personaly reviewed this topic and according to the literature the results match (with sligth differences) but noting meaningful... check your model

[fek66]

I have problems to make geometry to study the effect of wind in 2D idealized street canyon. If any one experimented in ICEM CFX ?