Application of AI/Deep Learning in the Forecast, Analysis and Control of Tokamak Plasmas
Speaker

Ge Dong, Neng Liang Qi Dian
Time

2022-03-16 10:00 ~ 11:00, in 2 days (Asia/Shanghai Time)
Venue

305, Science Building #5
Abstract

Fusion power is an ultimate solution to our energy and environment problem. In the past few years, private organizations such as CFS (U.S.) and NLQD (China) are collaborating with research institutions to speed-up the commercialization of magnetic fusion energy. The development of modernized control of plasma operation is a crucial task for the ultimate goal of fusion power production. Successful plasma stability forecast and control is essential for the steady state operations of next-generation tokamak devices to produce fusion power. Various AI/machine learning methodologies have been applied in tokamak instability forecast, real-time analysis, and equilibrium control. Here we will introduce some recent advancements in this cross-disciplinary area, including plasma disruption predictions, edge instability forecasts, and surrogate models of global gyrokinetic particle simulations. These on-going projects demonstrate promising application potential in future tokamak control systems.

Bio

Dr. Ge Dong is the head of plasma simulation at Neng Liang Qi Dian. She earned her B.S. in Physics from Peking University in 2013 and PhD in Astrophysical Sciences from Princeton University in 2019, before her post-doctoral appointment in the Princeton Plasma Physics Laboratory from 2019 to 2022. She is now mainly interested in developing plasma simulation and control tools.
--
修改:vinbo FROM 74.207.232.*
FROM 211.161.241.*