[L3munoz]

Dear all,

I am considering different criteria for vortex core identification. Now I have chosen lambda-2 and in the picture attached here you can see the resulting contour. The question is how to detect exactly the position (coordinates) of the core center, as well as the vortex core width?? Is there any accurate way or just an approximate one?

Thanks in advance,

/L.

[FMDenaro]

no, definition of vortical structure is somehow arbitrary...you can detect only in a qualitative way

[L3munoz]

Thanks a lot FMDenaro. However, I have found more than one journal paper where turbulence models are compared, and one of these comparisons concerns position (in terms of vortex core) and strength (width), so if this comparison is reliable, it should more than just an arbitrary criteria, doesn't it?

[FMDenaro]

Originally Posted by L3munoz

Thanks a lot FMDenaro. However, I have found more than one journal paper where turbulence models are compared, and one of these comparisons concerns position (in terms of vortex core) and strength (width), so if this comparison is reliable, it should more than just an arbitrary criteria, doesn't it?

Have a try, change the lambda-2 threshold ( zero cannot be used, generally some small negative value is fine) and measure the estimated variation in the width. I experienced strong variations in the identification, especially for confined flows.

[L3munoz]

Ok, I get your point, thanks a lot. I will try to get the center as accurate as possible but nevertheless seems only a rough approximation...