Application of machine learning and artificial intelligence to extend EFIT equilibrium reconstruction

被引：43

作者：

Lao, L. L. ^{[1
]}

Kruger, S. ^{[2
]}

Akcay, C. ^{[1
]}

Balaprakash, P. ^{[3
]}

Bechtel, T. A. ^{[1
,4
]}

Howell, E. ^{[2
]}

Koo, J. ^{[3
]}

Leddy, J. ^{[2
]}

Leinhauser, M. ^{[5
]}

Liu, Y. Q. ^{[1
]}

Madireddy, S. ^{[3
]}

McClenaghan, J. ^{[1
]}

Orozco, D. ^{[1
]}

Pankin, A. ^{[6
]}

Schissel, D. ^{[6
]}

Smith, S. ^{[1
]}

Sun, X. ^{[1
,4
]}

Williams, S. ^{[7
]}

机构：

[1] Gen Atom, San Diego, CA 92121 USA

[2] TechX, Boulder, CO USA

[3] Argonne Natl Lab, Lemont, IL USA

[4] Oak Ridge Associated Univ, Oak Ridge, TN USA

[5] Univ Delaware, Newark, DE USA

[6] Princeton Plasma Phys Lab, POB 451, Princeton, NJ 08543 USA

[7] Lawrence Berkeley Natl Lab, Berkeley, CA USA

来源：

PLASMA PHYSICS AND CONTROLLED FUSION | 2022年 / 64卷 / 07期

关键词：

tokamak equilibrium reconstruction; machine learning; artificial intelligence; Gaussian process; model order reduction; neural network; 3D perturbed equilibrium; REVERSED MAGNETIC SHEAR; ENHANCED CONFINEMENT; MHD EQUILIBRIUM; TOKAMAK; DISCHARGES; STABILITY; PLASMAS; MODES;

D O I：

10.1088/1361-6587/ac6fff

中图分类号：

O35 [流体力学]; O53 [等离子体物理学];

学科分类号：

070204 ; 080103 ; 080704 ;

摘要：

Recent progress in the application of machine learning (ML)/artificial intelligence (AI) algorithms to improve the Equilibrium Fitting (EFIT) code equilibrium reconstruction for fusion data analysis applications is presented. A device-independent portable core equilibrium solver capable of computing or reconstructing equilibrium for different tokamaks has been created to facilitate adaptation of ML/AI algorithms. A large EFIT database comprising of DIII-D magnetic, motional Stark effect, and kinetic reconstruction data has been generated for developments of EFIT model-order-reduction (MOR) surrogate models to reconstruct approximate equilibrium solutions. A neural-network MOR surrogate model has been successfully trained and tested using the magnetically reconstructed datasets with encouraging results. Other progress includes developments of a Gaussian process Bayesian framework that can adapt its many hyperparameters to improve processing of experimental input data and a 3D perturbed equilibrium database from toroidal full magnetohydrodynamic linear response modeling using the Magnetohydrodynamic Resistive Spectrum - Feedback (MARS-F) code for developments of 3D-MOR surrogate models.

引用

页数：16

共 61 条

[1]

Appel L C., 2006, 33 EPS C PLASM PHYS, pp2.184

[2] Equilibrium reconstruction in the START tokamak [J].