CFD Online Discussion Forums

CFD Online Discussion Forums (
-   Main CFD Forum (
-   -   gradient at boundary(FVM) ? (

Yim February 10, 2003 01:26

gradient at boundary(FVM) ?
Hello all.

I study FVM with unstructured grid(tetrahedron). In some references, normal gradient at cell face is (d@/dn)=(@N-@P)*G +{(grad@)f .n-G*(grad@)f .d(PN)} [.:inner product,n:face normal vector,d(PN):vector between P&N, G:geometrical factor]

At interior cell face, (grad @)f,gradient@ at cell face,is interpolated from the two adjacent cell gradients or vertex gradients(the average of the surrounding cell gradients). however, I don't know how can define gradient at boundary face. Is it also the average of the surrounding cell gradients(half of above case)? I think it's not correct.. Any referenses would be helpful !!!!

thank you !!

Praveen February 17, 2003 04:06

Re: gradient at boundary(FVM) ?
One way is to use the least squares (also called Energy) method which basically fits a polynomial to the local function data. Differentiating the polynomial gives you the gradient. See the VKI lecture notes of Barth. You can get these from his website.

All times are GMT -4. The time now is 21:21.