Plasma Physics (Physics/ECE/NE 922) Seminars |
Events on Monday, November 7th, 2016
- Development and applications of Veri cation and Validation procedures
- Time: 12:00 pm - 1:00 pm
- Place: 2241 Chamberlin Hall
- Speaker: Fabio Riva, Ecole Polytechnique Federale de Lausanne
- Abstract: The methodology used to assess the reliability of numerical simulation codes constitutes the
Veri cation and Validation (V&V) procedure. V&V is composed by three separate tasks: the code
veri cation, which is a mathematical issue targeted to assess that the physical model is correctly
implemented in a simulation code; the solution veri cation, which evaluates the numerical errors
a ecting a simulation; and the validation, which determines the consistency of the code results,
and therefore of the physical model, with experimental data.
To perform a code veri cation, we propose to use the method of manufactured solutions, a method-
ology that we have generalized to PIC codes, overcoming the diculty of dealing with a numerical
method intrinsically a ected by statistical noise. The solution veri cation procedure we put for-
ward is based on the Richardson extrapolation, used as higher order estimate of the exact solution.
These veri cation procedures were applied to GBS, a three-dimensional
uid code for SOL plasma
turbulence simulation based on a nite di erence scheme, and to a unidimensional, electrostatic,
collisionless PIC code. To perform a detailed validation of GBS against experimental measure-
ments, we generalized the magnetic geometry of the simulation code to include elongation and
non-zero triangularity, and we investigated theoretically the impact of plasma shaping e ects on
SOL turbulence. An experimental campaign to validate our ndings against experimental mea-
surements in tokamak limited con gurations has just been completed on TCV. - Host: Paul Terry