Michele Domenico Fasciana
Investigation of a saturation rule for experimental fluxes using a Neural Network approach.
Rel. Fabio Subba. Politecnico di Torino, Corso di laurea magistrale in Ingegneria Energetica E Nucleare, 2019
|
Preview |
PDF (Tesi_di_laurea)
- Tesi
Licenza: Creative Commons Attribution Non-commercial No Derivatives. Download (4MB) | Preview |
Abstract
An investigation on the saturation rule of the Trapped Gyro-Landau Fluid (TGLF) quasi-linear turbulent transport model has been performed. The turbulent fluxes obtained with TGLF was compared with the experimental fluxes based on experiment performed with DIII-D tokamak, at DIII-D National Fusion Facility. The gyrokinetic transport solvers are powerful models that solves non-linear simulations to inform and help the researchers involved in thermonuclear fusion development. One of the limitations of these code is the computational cost. To face this problem, reduced models have been developed. TGLF is a reduced model that compute a saturation rule calibrated to fit a set of nonlinear gyrokinetic simulation.
A neural network based model for the TGLF saturation rule that has been developed to links the quasi-linear quantities computed with TGLF to the experimental fluxes
Relatori
Tipo di pubblicazione
URI
![]() |
Modifica (riservato agli operatori) |
