Neural Network Model of Neutral Beam Injection on EAST to Enable Fast Transport Simulations

Z. Wang, S. Morosohk, E. Schuster, T. Rafiq and M.D. Boyer

Symposium on Fusion Technology (SOFT)

September 18-23, Dubrovnik, Croatia, 2022

Abstract

The neutral beam injection (NBI) system on EAST injects energetic neutral particles, which collide with electrons and ions in tokamak plasmas and heat the plasma through Coulomb collisions. It also introduces a non-inductive current drive, due to the charge-exchange collision between neutral particles and ions. Moreover, it injects toroidal torque, which generates a toroidal rotation of the plasma. The effects caused by the NBI system, such as plasma heating, power deposition, current drive, total neutron rate, momentum transfer, and shine-through, are modeled in simulations by using a comprehensive module called NUBEAM [1]. However, NUBEAM is computationally intensive since it relies on Monte Carlo methods. In this work, a neural network model has been developed as a surrogate model for NUBEAM on EAST. The database for neural-network model training and validation is generated by running TRANSP simulations with the NUBEAM module for experimental discharges from recent EAST campaigns (after the latest NBI upgrade). Simulation results illustrate that the trained neural network has the capability of replicating the predictions by NUBEAM while demanding a significantly shorter execution time.

[1] A. Pankin, D. McCune, R. Andre, G. Bateman, A. Kritz, “The tokamak Monte Carlo fast ion module NUBEAM in the National Transport Code Collaboration library,” Computer Physics Communications, 159(3):157–184, 2004.

*Supported by the US DOE under DE-SC0010537.