Workers: Visualization Section

Notes on NIMROD Data Structure

Because the type of plots graphic packages are able to do often depend on the structure of the data being plotted, we briefly review a typical case of Nnimrod data. A dependent variable, for example pressure P, is typically viewed in three-dimensional space on a polar type grid: P(iradial,itheta,iphi)

The grid corresponding to this is a cylindrical grid R,Z,phi

R(iradial,itheta), Z(iradial,itheta), phi(iphi)

It is also common, especially for 1D and 2D plots, to also view the real and imaginary Fourier components of the data:

P(iradial,itheta,nmodes) For the grid R(iradial,itheta), Z(iradial,itheta), k(nmodes)

Here are plots of constant iradial and itheta:

The dependent variables are the vector fields V,B,J (velocity, magnetic field, current density) and the scalar fields P,P_e,ndens (pressure, electron pressure, number density).

For 3D plots, we often transform from the cylindrical R,Z,phi grid to a Cartesian X,Y,Z grid:

X= R cos(phi), Y=-R sin(phi), Z= Z

There are caveats to this (Nimrod can also run with unstructured grids, linear geometry is slightly different than the toroidal geometry shown, etc.), but this is a good start.

team contact info | about the fusion grid | fusiongrid research

Last modified 10/21/01. Comments? webmaster