欢迎访问《核聚变与等离子体物理》官方网站,今天是 分享到:

核聚变与等离子体物理 ›› 2020, Vol. 40 ›› Issue (4): 300-308.DOI: 10.16568/j.0254-6086.202004003

• 等离子体物理学 • 上一篇    下一篇

基于深度学习的ELM实时识别研究

黄 尧,夏 凡,杨宗谕,钟武律,刘春华   

  1. (核工业西南物理研究院,成都 610041)
  • 收稿日期:2019-03-21 修回日期:2020-04-08 出版日期:2020-12-15 发布日期:2021-06-03
  • 作者简介:黄尧(1995-),男,贵州黔东南人,硕士研究生,实习研究员,主要从事人工智能在等离子体控制和运行上的应用研究。
  • 基金资助:
    国家磁约束核聚变能发展研究专项(2018YEF0302104);国家自然科学基金(11875022)

ELM real-time recognition research based on deep learning

HUANG Yao, XIA Fan, YANG Zong-yu, ZHONG Wu-lü, LIU Chun-hua   

  1. (Southwestern Institute of Physics, Chengdu 610041)
  • Received:2019-03-21 Revised:2020-04-08 Online:2020-12-15 Published:2021-06-03
  • Supported by:
     

摘要: 基于深度学习的方法,在HL-2A装置上开发出了一套边缘局域模(ELM)实时识别算法。算法使用5200次放电数据(约24.19万数据切片)进行学习,得到一个深度为22层的卷积神经网络。为衡量算法的识别能力,识别了HL-2A装置自2009年实现稳定ELMy H模放电以来所有历史数据(约26000次放电数据),共识别出1665次H模放电,其中误识别35次,误报率为2.10%。在实际的1634次H模放电中,漏识别4次,漏识别率为0.24%。该误报率和漏报率可以满足ELM实时识别的精度要求。识别算法在实时控制环境下,对单个时间点的平均计算时间为0.46ms,可以满足实时控制的计算速度要求。

 

关键词: H模, 神经网络, 深度学习, HL-2A装置

Abstract: Based on the deep learning method, this paper introduced an ELM real-time recognition algorithm on HL-2A tokamak. The algorithm used data from 5200 shots (about 241,900 data slices) for learning and a 22-layer convolutional neural network was obtained. The algorithm has recognized all historical data of HL-2A since it achieved stable ELMy H-mode discharge in 2009. A total of 1665 shots of H-mode have been recognized, of which 35 shots were misidentified, with the false positive rate (FPR) of 2.10%. In the actual 1634 shots of H-mode, the system missed to recognize 4 of them, with the false negative rate (FNR) of 0.24%. The FPR and FNR fulfill the precision requirements of real-time ELM recognition. In the simulated real-time environment, the algorithm’s average calculation time of a single slice is 0.46 milliseconds, which satisfy the calculation speed requirements of the real-time ELM recognition.

Key words: H-mode, Neural network, Deep learning, HL-2A tokamak.

中图分类号: