Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.
Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.
Wind flow model
Posted Jun 11, 2012, 12:07 p.m. EDT Version 4.3 1 Reply
Please login with a confirmed email address before reporting spam
Hi,
I'm currently tying to create a wind flow model of a cluster of buildings in order to analyze their effect on a urban wind turbine. I'm having a hard time understanding how to use the software. I have done a few tutorials and have tried finding applicable tutorials to what I'm trying to do. I have not successfully found a large scale unenclosed flow model. I have only found models with boundaries such as pipes or enclosed tube flows. I haven't found anything that is an open flow such as a car in the outdoors or a wind turbine. 1. Where can I find information on creating such model? 2. Do I have to create a 'fictitious boundary' in order to simulate my flow? 3. Does COMSOL treat this kind of analysis as if it was inside a wind tunnel? 4. Does anyone have a model that they could share?
I did find an animation that would be helpful to have the .mph file. www.comsol.com/showroom/animations/963/ This is the kind of model I'm looking for.
Thank you for your time.
Sincerely,
Matt
I'm currently tying to create a wind flow model of a cluster of buildings in order to analyze their effect on a urban wind turbine. I'm having a hard time understanding how to use the software. I have done a few tutorials and have tried finding applicable tutorials to what I'm trying to do. I have not successfully found a large scale unenclosed flow model. I have only found models with boundaries such as pipes or enclosed tube flows. I haven't found anything that is an open flow such as a car in the outdoors or a wind turbine. 1. Where can I find information on creating such model? 2. Do I have to create a 'fictitious boundary' in order to simulate my flow? 3. Does COMSOL treat this kind of analysis as if it was inside a wind tunnel? 4. Does anyone have a model that they could share?
I did find an animation that would be helpful to have the .mph file. www.comsol.com/showroom/animations/963/ This is the kind of model I'm looking for.
Thank you for your time.
Sincerely,
Matt
1 Reply Last Post Jun 11, 2012, 3:05 p.m. EDT
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
Indeed I havent seen very many "large" CFD models, but from my understading one often need clusters to analyse such "large" model, because of the mesh density and RAM sizes required.
If you want 3 sides "open": use symmetry conditions for the wind flow, at least these will then force a parallel flow, while leaving 3 sides fully open require that you know how to balance the boundary crossing flow exchange correctly
If I understand you correctly you need essentially the lower shape of your buildings as a boundary, so one domain with a rather complex lower boundary is all that is requried for the geometry, perhaps you need to add gravity for boyance forces, and some inlet velocity. What I do not know (CFD is not really my field) is how to estimate the turbulence length and therefore the mesh density. But you have theaverage Reynolds cell number to give you some indications about the laminar - turbulent limit of the mesh used.
If you want to add air mosture and temperature fluctuations you will probably have to use a cluster as these add a few factors in RAM requirements, but that is also possible and made easier in 4.3 I have heard
--
Good luck
Ivar
Indeed I havent seen very many "large" CFD models, but from my understading one often need clusters to analyse such "large" model, because of the mesh density and RAM sizes required.
If you want 3 sides "open": use symmetry conditions for the wind flow, at least these will then force a parallel flow, while leaving 3 sides fully open require that you know how to balance the boundary crossing flow exchange correctly
If I understand you correctly you need essentially the lower shape of your buildings as a boundary, so one domain with a rather complex lower boundary is all that is requried for the geometry, perhaps you need to add gravity for boyance forces, and some inlet velocity. What I do not know (CFD is not really my field) is how to estimate the turbulence length and therefore the mesh density. But you have theaverage Reynolds cell number to give you some indications about the laminar - turbulent limit of the mesh used.
If you want to add air mosture and temperature fluctuations you will probably have to use a cluster as these add a few factors in RAM requirements, but that is also possible and made easier in 4.3 I have heard
--
Good luck
Ivar