Register to reply

Need help on boundary layer meteorology

by lalula2
Tags: boundary, layer, meteorology
Share this thread:
lalula2
#1
May17-09, 11:06 PM
P: 2
Hi, i am mech undergrad student, currently i am doing my thesis for CFD simulation of wind flow in urban area by using FLUENT. I am trying to apply logarithmic velocity profiles for my inlet boundary condition.
My question is that from the formula of logarithmic wind profile, the velocity will become negative when it less than certain height due to the ln function. Do i need to use power law profiles to describe the velocity profile under certain height?
Thanks.

Best regards.
Phys.Org News Partner Science news on Phys.org
Experts defend operational earthquake forecasting, counter critiques
EU urged to convert TV frequencies to mobile broadband
Sierra Nevada freshwater runoff could drop 26 percent by 2100
Cyrus
#2
May18-09, 12:32 AM
Cyrus's Avatar
P: 4,777
This is a question you should find the answer to through journal papers.
minger
#3
May18-09, 07:21 AM
Sci Advisor
P: 1,498
Why are you applying a logarithmic inlet condition? It doesn't make much sense to me. If you are trying to get the inlet condition to be as close as the actual velocity profile, then don't worry about.

Expand your domain to include some space outside of your physical domain. Send in a uniform constant flow and let it do what i wants.

lalula2
#4
May18-09, 07:41 AM
P: 2
Need help on boundary layer meteorology

I trying so save my computational time by create my own udf for the boundary condition and based on those journals i read.
Beside that, my moderator criticized my velocity profiles and ask me to rectify the problem as he said my velocity above my study area are not logic.
For actual urban area, the building height is relatively low compare to the atmosphere boundary layer.
minger
#5
May18-09, 09:10 AM
Sci Advisor
P: 1,498
Well, having said that, you could just kind of substitute height for height above the surface, e.g.
Rather than
[tex] u(i,j,k) = log(z(i,j,k))[/tex]
Have something like:
[tex] u(i,j,k) = log( z(i,j,k) - z(i,j,1) )[/tex]

Do remember that boundary conditions are quite a problem in most computational codes, and many many problems can arise from them. Having said that, be very careful when analyzing results such that you ensure you're getting what you think you should.

I image there could be problems with reflections depending on the actual boundary condition implementation.


Register to reply

Related Discussions
Boundary layer in water General Physics 12
Boundary layer question Advanced Physics Homework 0
Boundary Layer (reposted) Introductory Physics Homework 0
Boundary Layer General Physics 3
Electromagnetic boundary layer. Mechanical Engineering 3