Application of machine learning for optical emission spectroscopy data in NAGDIS-II

Shin Kajita, Takehiro Sakakibara, Hideki Kaizawa, Hiroki Natsume, Hirohiko Tanaka, Keisuke Fujii, Noriyasu Ohno

Research output: Contribution to journalArticlepeer-review

2 Scopus citations

Abstract

In this study, we applied machine learning to optical emission spectroscopy (OES) data and device parameters from the linear plasma device NAGDIS-II to explore the potential application of machine learning for predicting electron density, ne, and temperature, Te. The covered ranges of ne and Te, which were measured by an electrostatic probe, are 3.6 × 1017–2.4×1019 m−3 and 0.3–7.1eV, respectively. A three hidden layer neural network (NN) is introduced to model the relationship between ne/Te and the combination of line intensities, radial position, and device parameters. It is shown that the errors in ne and Te become 18.0 and 18.8%, respectively, which were almost the same level for the electrostatic probe, using all available data. Lasso regression and greedy algorithm are used to select the necessary line emissions. It is shown that four or five line intensities are sufficient to obtain almost the same quality as the one with all the other lines.

Original languageEnglish
Article number114012
JournalFusion Engineering and Design
Volume196
DOIs
StatePublished - Nov 2023

Funding

This work was supported in part by a Grant-in-Aid for Scientific Research 19H01874 , 21K18617 and 20H00138 , Fund for the Promotion of Joint International Research 17KK0132 and 21KK0048 , and NIFS collaboration research program, Japan ( NIFS22KIPP002 , NIFS23HDAF011 , NIFS23KIPH026 ).

Keywords

  • Diagnostics
  • Machine learning
  • Optical emission spectroscopy
  • Plasma

Fingerprint

Dive into the research topics of 'Application of machine learning for optical emission spectroscopy data in NAGDIS-II'. Together they form a unique fingerprint.

Cite this