Deep learning approaches to recover the plasma current density profile from the safety factor based on Grad–Shafranov solutions across multiple tokamaks  

作　　者：张瀚予 周利娜 刘钺强 郝广周 王硕 杨旭 苗雨田 段萍 陈龙 Hanyu ZHANG;Lina ZHOU;Yueqiang LIU;Guangzhou HAO;Shuo WANG;Xu YANG;Yutian MIAO;Ping DUAN;Long CHEN(College of Science,Dalian Maritime University,Dalian 116026,People’s Republic of China;General Atomics,San Diego 92186-5608,United States of America;Southwestern Institute of Physics,Chengdu 610041,People’s Republic of China;Chongqing Key Laboratory of Intelligent Perception and BlockChain Technology,Chongqing Technology and Business University,Chongqing 400067,People’s Republic of China)

机构地区：[1]College of Science,Dalian Maritime University,Dalian 116026,People’s Republic of China [2]General Atomics,San Diego 92186-5608,United States of America [3]Southwestern Institute of Physics,Chengdu 610041,People’s Republic of China [4]Chongqing Key Laboratory of Intelligent Perception and BlockChain Technology,Chongqing Technology and Business University,Chongqing 400067,People’s Republic of China

出　　处：《Plasma Science and Technology》2024年第5期17-28,共12页等离子体科学和技术（英文版）

基　　金：supported by National Natural Science Foundation of China (Nos. 12205033, 12105317, 11905022 and 11975062);Dalian Youth Science and Technology Project (No. 2022RQ039);the Fundamental Research Funds for the Central Universities (No. 3132023192);the Young Scientists Fund of the Natural Science Foundation of Sichuan Province (No. 2023NSFSC1291)

摘　　要：Many magnetohydrodynamic stability analyses require generation of a set of equilibria with a fixed safety factor q-profile while varying other plasma parameters.A neural network(NN)-based approach is investigated that facilitates such a process.Both multilayer perceptron(MLP)-based NN and convolutional neural network(CNN)models are trained to map the q-profile to the plasma current density J-profile,and vice versa,while satisfying the Grad–Shafranov radial force balance constraint.When the initial target models are trained,using a database of semianalytically constructed numerical equilibria,an initial CNN with one convolutional layer is found to perform better than an initial MLP model.In particular,a trained initial CNN model can also predict the q-or J-profile for experimental tokamak equilibria.The performance of both initial target models is further improved by fine-tuning the training database,i.e.by adding realistic experimental equilibria with Gaussian noise.The fine-tuned target models,referred to as fine-tuned MLP and fine-tuned CNN,well reproduce the target q-or J-profile across multiple tokamak devices.As an important application,these NN-based equilibrium profile convertors can be utilized to provide a good initial guess for iterative equilibrium solvers,where the desired input quantity is the safety factor instead of the plasma current density.

关 键 词：plasma equilibrium deep learning safety factor profile current density profile TOKAMAK 

分 类 号：TL631.24[核科学技术—核技术及应用] TL612