Structured / Unstructured
 

What is the difference really? Some people say unstructured grids give more accurate solution to the fluid flows. Is that true? We know that it is more difficult to write codes based on unstructured grids, but what is the most important benefit of unstructured grids? Thanks.

Re: Structured / Unstructured
 

In a lot of case, interests are local, so we want to resolve the local area by putting a lot of mesh points in there. This can be naturally done with unstructured grids.

On the other hand, if we use structure grids, we may waste computational effort in unnecessary parts. But if you use adaptive structured grids, we can relieve this and get accurate calculation also.

Re: Structured / Unstructured
 

I don't think unstructured grids will give more accurate solutions than structured grids. There are many other factors that affect the accuracy more. The major advantage of using unstructured grids is certainly the ease of meshing domains of various shapes.

Re: Structured / Unstructured
 

Really agree with chen25. Easy meshing is THE advantage of unstructured mesh.

Re: Structured / Unstructured
 

Hi Junseok,

I am interested in your comment on local area need to be given more node points. I agree with you, but am not sure how important the upstream contribute to local flow. Could you please give me some references on this.

Thanks, Thomas

Re: Structured / Unstructured
 

What I meant by local area is in the multi-phase flow, in many cases, we want to track the interface location and most of computational details are needed at that interface region such as velocity, pressure, and surface tension.

Would you please visit our group's research publication page.

http://math.uci.edu/~cristini/publications.php

Thanks

Re: Structured / Unstructured
 

I'd say meshing is no problem in a commercial cfd code using structured meshing. The answer is STL files and a CAD program...