Dear all,

We are working in Grid generation of complex geometries. we generated many complex grids. In addition we can able to do clustering (grid adaptation) along the middle portion in both axial and radial directions. But we are struggling to cluster along the boundary where we want to capture the sharp gradients.

If anyone knows kindly reply..

Thanks in Advance,

Yours sincerely,

Arul Prakash K

What type of grid generation are you using?.

We are generating Structured Grid generation using Multi Block Technique

We are using Structured grid generation technique. The outer boundary was created using algebraic method and finally smoothed using elliptic grid generation technique.

Suggestions regarding clustering along the boundaries are most welcomed.

If you are using elliptic grid generation, there are the "forcing functions". I will assume that you know these functions, but you can read about it in many texts. If your answer is: hey! I know the forcing functions P and Q yet, but they only clusters over radial coordinates!. Then I don't know the answer. I am nowadays traying to cluster the iso-eta lines (circumferential lines) over the boundary, but the forcing function requires a iso-xi line of reference to cluster around too.

Dear Doctor Blade,

Thank you very much for your response. I adopted P and Q functions. I am using both iso-eta and iso-zhi lines. But still I am facing problem to cluster near the boundaries. Whether U manage to cluster near the boundary??

Thanks, Arul.

Well, if you are using P and Q functions, I congratulate you if your iteration converges to a solution. I've not been capable of it in my trials. But I can remember P or Q functions have a iso-eta line of clustering reference inside the exponential factor, haven't they?. What happens if you fix this iso-eta line of reference to be the boundary one?.

And another question: are you using Matlab?