Plasma and Fusion Research

Volume 13, 3405021 (2018)

Regular Articles


Disruption Prediction by Support Vector Machine and Neural Network with Exhaustive Search
Tatsuya YOKOYAMA, Takamitsu SUEYOSHI, Yuya MIYOSHI1), Ryoji HIWATARI1), Yasuhiko IGARASHI2), Masato OKADA and Yuichi OGAWA
Graduate School of Frontier Science, The University of Tokyo, Kashiwa 277-8561, Japan
1)
Rokkasho Fusion Institute, QST, Rokkasho 039-3212, Japan
2)
Japan Science and Technology Agency, PRESTO, 4-1-8 Honcho, Kawaguchi, Saitama 332-0012, Japan
(Received 27 December 2017 / Accepted 19 February 2018 / Published 10 April 2018)

Abstract

A disruption is an event in which the plasma current suddenly shuts down in a tokamak reactor. Establishing methods to predict, mitigate, and avoid disruptions may be indispensable for realizing a tokamak reactor. In the present study, we have used the large dataset of high-beta experiments at JT-60U to develop a method for predicting the occurrence of disruptions. The method is based on sparse modeling that exploits the inherent sparseness common to all high-dimensional data, and it enables us to extract the maximum amount of information from the data efficiently. To carry out the sparse modeling, we have used exhaustive searches with a support vector machine and a neural network. In this research, we repeated the training and evaluation of the predictor while changing the combination of plasma parameters. As a result of the exhaustive search, we found |Brn=1| and d|Brn=1|/dt to be the dominant parameters for disruption predictions. This is not surprising, because MHD instabilities are considered to be the direct triggers of disruption. In addition, we have succeeded in identifying several important parameters that may also be strongly related to disruptions, i.e., βN, βP, q95, δ, fGW, and frad.


Keywords

tokamak, disruption, data-driven science, machine learning, sparse modeling, ES-K

DOI: 10.1585/pfr.13.3405021


References

  • [1] T.C. Hender et al., Nucl. Fusion 47, S128 (2007).
  • [2] A.H. Boozer, Phys. Plasmas 19, 058101 (2012).
  • [3] G.A. Rattá, J. Vega, A. Murari et al., Nucl. Fusion 50, 025005 (2010).
  • [4] J. Vega, R. Moreno, A. Pereira et al., In 1st EPS Conf. on Plasma Diagnostics, 2015.
  • [5] R. Yoshino, Nucl. Fusion 45, 1232 (2005).
  • [6] G.A. Rattá, J. Vega and A. Murari, Fusion Eng. Des. 87, 1670 (2012).
  • [7] Y. Igarashi, H. Takenaka, Y. Nakanishi-Ohno et al., IEICE Technical Report, 116 (300): 313-320, 2016.
  • [8] I. Takeuchi and M. Karasuyama, Sapoto Bekutoru Mashin (Support Vector Machine), (Kodansha, 2015).
  • [9] T. Okatani, Shinsogakushu (Deep Learning), (Kodansha, 2015).