Plasma and Fusion Research

Volume 17, 1201048 (2022)

Rapid Communications

Prediction of Turbulence Temporal Evolution in PANTA by Long-Short Term Memory Network
Masaomi AIZAWACARANZA, Makoto SASAKI, Hiroki MINAGAWA, Yuuki NAKAZAWA, Yoshitatsu LIU, Yuki JAJIMA, Yuichi KAWACHI1), Hiroyuki ARAKAWA2) and Kazuyuki HARA
College of Industrial Technology, Nihon University, Narashino 275-8575, Japan
1)
Interdisciplinary Graduate School of Engineering Sciences, Kyushu University, Kasuga 816-8580, Japan
2)
Institute of Science and Engineering, Academic Assembly, Shimane University, Matsue 690-8504, Japan
(Received 9 March 2022 / Accepted 23 March 2022 / Published 13 May 2022)

Abstract

Prediction of time evolution of multi-scale turbulence is performed by using Long-short term memory networks. The time series data is obtained by Langmuir probes in a linear magnetized plasma device, PANTA. The simultaneous prediction of high and low frequency components of turbulence is shown to be possible within several tens percent accuracy. The prediction accuracy depends on the initial network, which can be controlled by reducing the learning rate.

Keywords

deep learning, plasma, turbulence, data driven science, prediction, LSTM

DOI: 10.1585/pfr.17.1201048

Full Text

PDF format (413Kbytes) PDF Download

References

  • [1] J.K. Harbeck et al., Nature 568, 526 (2019).
  • [2] A.W. Leonard et al., Phys. Plasmas 21, 090501 (2014).
  • [3] B.A. Carreras et al., Phys. Rev. Lett. 80, 4438 (1998).
  • [4] D.F. Kong et al., Nucl. Fusion 58, 034003 (2018).
  • [5] M. Sasaki et al., Plasma Phys. Control. Fusion 61, 112001 (2019).
  • [6] M. Sasaki et al., Plasma Phys. Control. Fusion 63, 025004 (2021).
  • [7] S. Inagaki et al., Sci. Rep. 6, 22189 (2016).
  • [8] F. Kin, Phys. Plasmas 26, 042306 (2019).
  • [9] Y. Kawachi et al., Phys. Plasmas 28, 112302 (2021).
  • [10] S. Hochreiter and J. Schmidhuber, Neural Computation 9 (8), 1735 (1997).