Fast modeling of turbulent transport in fusion plasmas using neural networks
| Autor: | F. J. Casson, Victor I. Dagnelie, Jet Contributors, Federico Felici, Yann Camenen, Clarisse Bourdelle, Simon Van Mulders, A. Ho, Jonathan Citrin, Karel Lucas van de Plassche |
|---|---|
| Přispěvatelé: | Dutch Institute for Fundamental Energy Research [Eindhoven] (DIFFER), Association EURATOM-CEA (CEA/DSM/DRFC), Commissariat à l'énergie atomique et aux énergies alternatives (CEA), Physique des interactions ioniques et moléculaires (PIIM), Aix Marseille Université (AMU)-Centre National de la Recherche Scientifique (CNRS), Eindhoven University of Technology [Eindhoven] (TU/e), Joint European Torus (JET-EFDA), Culham Science Centre [Abingdon], Control Systems Technology, Science and Technology of Nuclear Fusion, Camenen, Yann |
| Jazyk: | angličtina |
| Rok vydání: | 2020 |
| Předmět: |
Physics
Jet (fluid) Tokamak Artificial neural network Turbulence FOS: Physical sciences Mechanics Solver Condensed Matter Physics 7. Clean energy 01 natural sciences Physics - Plasma Physics 010305 fluids & plasmas law.invention Plasma Physics (physics.plasm-ph) Surrogate model Orders of magnitude (time) law Physics::Plasma Physics [PHYS.PHYS.PHYS-PLASM-PH]Physics [physics]/Physics [physics]/Plasma Physics [physics.plasm-ph] [PHYS.PHYS.PHYS-PLASM-PH] Physics [physics]/Physics [physics]/Plasma Physics [physics.plasm-ph] 0103 physical sciences Uncertainty quantification 010306 general physics |
| Zdroj: | Physics of Plasmas Physics of Plasmas, 2020 Physics of Plasmas, 27(2):022310. American Institute of Physics Physics of Plasmas, 27, 022310 Physics of Plasmas, American Institute of Physics, 2020 HAL |
| ISSN: | 1089-7674 1070-664X |
| DOI: | 10.1063/1.5134126 |
| Popis: | We present an ultrafast neural network (NN) model, QLKNN, which predicts core tokamak transport heat and particle fluxes. QLKNN is a surrogate model based on a database of 300 million flux calculations of the quasilinear gyrokinetic transport model QuaLiKiz. The database covers a wide range of realistic tokamak core parameters. Physical features such as the existence of a critical gradient for the onset of turbulent transport were integrated into the neural network training methodology. We have coupled QLKNN to the tokamak modelling framework JINTRAC and rapid control-oriented tokamak transport solver RAPTOR. The coupled frameworks are demonstrated and validated through application to three JET shots covering a representative spread of H-mode operating space, predicting turbulent transport of energy and particles in the plasma core. JINTRAC-QLKNN and RAPTOR-QLKNN are able to accurately reproduce JINTRAC-QuaLiKiz T i,e and n e profiles, but 3 to 5 orders of magnitude faster. Simulations which take hours are reduced down to only a few tens of seconds. The discrepancy in the final source-driven predicted profiles between QLKNN and QuaLiKiz is on the order 1%-15%. Also the dynamic behaviour was well captured by QLKNN, with differences of only 4%-10% compared to JINTRAC-QuaLiKiz observed at mid-radius, for a study of density buildup following the L-H transition. Deployment of neural network surrogate models in multi-physics integrated tokamak modelling is a promising route towards enabling accurate and fast tokamak scenario optimization, Uncertainty Quantification, and control applications. 18 pages, 11 figures, Physics of Plasmas, ICDDPS 2019 conference paper |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|