Regular Articles

Full Text / References

Neural Network Data Analysis in the Large Helical Device Thomson Scattering System

Ichihiro YAMADA, Hisamichi FUNABA, Jong-ha LEE¹⁾, Yuan HUANG²⁾ and Chunhua LIU²⁾

National Institute for Fusion Science, Toki 509-5292, Japan

1)

Korea Institute of Fusion Energy, Daejeon 34133, Korea

2)

Southwestern Institute of Physics, P.O. Box 432, Chengdu, 610041, China

(Received 27 December 2021 / Accepted 8 March 2022 / Published 22 June 2022)

Abstract

In Thomson scattering diagnostics systems, a combination of the lookup table and the minimum χ² methods has been widely used to determine electron temperature. The concept of the minimum χ² method is based on clearly defined mathematical statistics. However, the minimum χ² method calculation requires a large amount of time because all χ² values have to be calculated at all temperatures included in the lookup table. Thus, this method is unsuitable for the real-time data analysis required for the next generation of fusion devices, e.g., the International Thermonuclear Experimental Reactor in France. To establish real-time data analysis for Thomson scattering diagnostics, we have developed a neural network program for the large helical device (LHD) Thomson scattering (TS) system. First, we systematically studied the number of nodes and training cycles required to obtain satisfactory results, and then applied them to the LHD TS system. The calculation time was successfully reduced by approximately 1/50 - 1/100 of the χ² method calculation time. In addition, experimental error estimation has been performed according to the concept of the neural network method used in this study.

Copyright (c) 2022 The Japan Society of Plasma Science and Nuclear Fusion Research

Keywords

Thomson scattering, Large Helical Device (LHD), neural network method, real-time data analysis, electron temperature, error estimation

DOI: 10.1585/pfr.17.2402061

Full Text

PDF format (435Kbytes)

References

• [1] K. Narihara et al., Design and performance of the Thomson scattering diagnostic on LHD, Rev. Sci. Instrum. 72, 1122 (2001).
• [2] I. Yamada et al., Recent Progress of the LHD Thomson Scattering System, Fusion Sci. Technol. 58, 345 (2010).
• [3] S.H. Lee et al., Development of a neural network technique for KSTAR Thomson scattering diagnostics, Rev. Sci. Instrum. 87, 11E533 (2016).
• [4] C. Liu et al., Artificial neural network approach applied to data processing of Thomson scattering on HL-2A, High Power Laser Part. Beams 31, 022003 (2019) (in Chinese).
• [5] J. Lee et al., Tangential Thomson scattering diagnostic for the KSTAR Tokamak, J. Instrum. 7, C02026 (2012).
• [6] J.H. Lee et al., Development of prototype polychromator system for KSTAR Thomson scattering diagnostic, J. Instrum. 10, C12012 (2015).
• [7] C.H. Liu et al., The progress in development of edge tangential Thomson scattering system on HL-2A tokamak, Rev. Sci. Instrum. 87, 11E555 (2016).
• [8] Y. Huang et al., Multipoint vertical-Thomson scattering diagnostic on HL-2A tokamak, Rev. Sci. Instrum. 89, 10C116 (2018).
• [9] Seung-Ju Lee et al., Design of GPU-based parallel computation architecture of Thomson scattering diagnostic in KSTAR, Fusion Eng. Des. 158, 111624 (2020).