Mesh study

[ggege0524]

Hi you all, 

I really need your help, regarding a mesh study. I have a domain that consists of three parts (see figure below), each of them has a different mesh refinement (1 is the finest, 3 the most coarse). There is also the boundary layer and for that I have used element sizing and inflation layers, so i can get a low y+ value and good mesh quality. I want to do a mesh study, according to this paper:"Procedure for Estimation and Reporting of Uncertainty Due to Discretization in CFD Applications " but I am kinda lost. I dont know how to treat the boundary layer, should I change anything, since it makes the most out of all cells, should I treat each region seperately, how  should I choose the quantity I want to investigate and , where, at a point at a line? I would be very greatful for you advice. help me pleeeeeeeeeeeeease! :)

 

[Rob]

Maybe? Having not read the paper (and also being unable to follow most links or download files) it's difficult to comment. What does your supervisor think about it?