首页|基于有限元法与机器学习相结合研究高温超导带材的动态电阻

基于有限元法与机器学习相结合研究高温超导带材的动态电阻

扫码查看
原文链接
  • NETL
  • NSTL
  • 万方数据
高温超导体在直流输运过程中，当受到外部交变磁场的影响时，会产生高温超导体独有的动态电阻现象。在高温超导带材应用的仿真研究中，使用有限元方法可以快捷地计算和分析动态电阻。然而，由于生产工艺等因素的差异，实际高温超导带材的临界电流受磁场影响程度不同，表现出不同的输运特性，会给不同带材的有限元建模带来挑战。与使用经验公式或插值方法处理实际带材的临界电流数据不同，本文采用机器学习来处理三种不同商业带材的临界电流数据，并建立有限元模型来分析动态电阻特性。结果表明，机器学习具有建模高度非线性和多因素的超导临界电流特性的能力，结合了机器学习工具的有限元模型对超导动态电阻的讨论同样具有可靠的精确度及可行性。
Analyzing dynamic resistance in high-temperature superconducting tapes by combining finite element method with machine learning
When an external alternating field reaches a threshold value,high-temperature superconductors(HTS)that are carrying direct current can exhibit dynamic resistance phenomenon.This phenomenon,often observed in tape applications,can be effectively studied using finite element methods(FEM).However,due to differences in production processes,HTS tapes have varying parameters,including magnetic-dependent critical current.This can pose a significant challenge when comparing dynamic resistance differences among HTS tapes.Due to the capability of machine learning to conveniently handle the nonlinear characteristics of superconductors and adapt to multivariate function fitting,this paper employs machine learning for fitting the critical current characteristics of tapes and applies it to calculate dynamic resistance in the FEM model.By employing machine learning to handle the critical current characteristics of various tapes,the FEM model showcases both feasibility and accuracy in the results.

dynamic resistancefinite element methodcritical currentmachine learning

肖书良、曾志刚、周迪帆、贾焯越、颜志超、李奇展、宋世恒、蔡传兵

展开 >

Department of Physics,Shanghai University,Shanghai 200444,China

Shanghai Key Laboratory of High Temperature Superconductors,Shanghai University,Shanghai 200444,China

动态电阻 有限元法 临界电流 机器学习

国家自然科学基金国家自然科学基金国家自然科学基金国家重点研发计划Shanghai Science and Technology Innovation Program,China中国科学院战略规划重点项目

5217227112374378523070262022YFE0315020022511100200XDB25000000

2024

10.1007/s11771-024-5614-7

中南大学学报(英文版)
中南大学

中南大学学报(英文版)

CSTPCDEI
影响因子：0.47
ISSN：2095-2899
年,卷(期)：2024.31(3)
  • 33