中国物理B ›› 2019, Vol. 28 ›› Issue (12): 125204-125204.doi: 10.1088/1674-1056/ab55d1

• PHYSICS OF GASES, PLASMAS, AND ELECTRIC DISCHARGES • 上一篇    下一篇

Estimation of plasma equilibrium parameters via a neural network approach

Zi-Jian Zhu(朱子健), Yong Guo(郭勇), Fei Yang(杨飞), Bing-Jia Xiao(肖炳甲), Jian-Gang Li(李建刚)   

  1. 1 Department of Engineering and Applied Physics, University of Science and Technology of China, Hefei 230026, China;
    2 Institute of Plasma Physics, Chinese Academy of Sciences, Hefei 230031, China;
    3 Department of Medical Information Engineering, Anhui Medical University, Hefei 230026, China
  • 收稿日期:2019-08-25 修回日期:2019-10-22 出版日期:2019-12-05 发布日期:2019-12-05
  • 通讯作者: Fei Yang E-mail:yangfei@ahmu.edu.cn
  • 基金资助:
    Project supported by the National Magnetic Confinement Fusion Energy R&D Program of China (Grant No. 2018YFE0302100), the National Key Research and Development Program of China (Grant Nos. 2017YFE0300500 and 2017YFE0300501), the National Natural Science Foundation of China (Grant Nos. 11575245, 11805236, and 11905256), and Young and Middle-aged Academic Back-bone Finance Fund from Anhui Medical University.

Estimation of plasma equilibrium parameters via a neural network approach

Zi-Jian Zhu(朱子健)1,2, Yong Guo(郭勇)2, Fei Yang(杨飞)3, Bing-Jia Xiao(肖炳甲)1,2, Jian-Gang Li(李建刚)1,2   

  1. 1 Department of Engineering and Applied Physics, University of Science and Technology of China, Hefei 230026, China;
    2 Institute of Plasma Physics, Chinese Academy of Sciences, Hefei 230031, China;
    3 Department of Medical Information Engineering, Anhui Medical University, Hefei 230026, China
  • Received:2019-08-25 Revised:2019-10-22 Online:2019-12-05 Published:2019-12-05
  • Contact: Fei Yang E-mail:yangfei@ahmu.edu.cn
  • Supported by:
    Project supported by the National Magnetic Confinement Fusion Energy R&D Program of China (Grant No. 2018YFE0302100), the National Key Research and Development Program of China (Grant Nos. 2017YFE0300500 and 2017YFE0300501), the National Natural Science Foundation of China (Grant Nos. 11575245, 11805236, and 11905256), and Young and Middle-aged Academic Back-bone Finance Fund from Anhui Medical University.

RichHTML

0

PDF (PC)

161

赞

摘要： Plasma equilibrium parameters such as position, X-point, internal inductance, and poloidal beta are essential information for efficient and safe operation of tokamak. In this work, the artificial neural network is used to establish a non-linear relationship between the measured diagnostic signals and selected equilibrium parameters. The estimation process is split into a preliminary classification of the kind of equilibrium (limiter or divertor) and subsequent inference of the equilibrium parameters. The training and testing datasets are generated by the tokamak simulation code (TSC), which has been benchmarked with the EAST experimental data. The noise immunity of the inference model is tested. Adding noise to model inputs during training process is proved to have a certain ability for maintaining performance.

关键词: neural network, fusion, plasma equilibrium, noise immunity

Abstract: Plasma equilibrium parameters such as position, X-point, internal inductance, and poloidal beta are essential information for efficient and safe operation of tokamak. In this work, the artificial neural network is used to establish a non-linear relationship between the measured diagnostic signals and selected equilibrium parameters. The estimation process is split into a preliminary classification of the kind of equilibrium (limiter or divertor) and subsequent inference of the equilibrium parameters. The training and testing datasets are generated by the tokamak simulation code (TSC), which has been benchmarked with the EAST experimental data. The noise immunity of the inference model is tested. Adding noise to model inputs during training process is proved to have a certain ability for maintaining performance.

Key words: neural network, fusion, plasma equilibrium, noise immunity

中图分类号:  (Tokamaks, spherical tokamaks)

  • 52.55.Fa
07.05.Mh (Neural networks, fuzzy logic, artificial intelligence) 52.70.-m (Plasma diagnostic techniques and instrumentation) 52.25.Xz (Magnetized plasmas)