skip to main content
DOE PAGES title logo U.S. Department of Energy
Office of Scientific and Technical Information

This content will become publicly available on April 30, 2020

Title: Machine learning for disruption warnings on Alcator C-Mod, DIII-D, and EAST

Abstract

In this paper, we report on disruption prediction using a shallow machinelearning method known as Random Forests, trained on large databases containingonly plasma parameters that are available in real-time on Alcator C-Mod, DIII-D, and EAST. The database for each tokamak contains parameters sampled ~10 6 times throughout ~10 4 discharges (disruptive and non-disruptive) over the last 4years of operation. It is found that a number of parameters (e.g.P rad/P input,l i,n/n G,B n=1/B 0) exhibit changes in aggregate as a disruption is approached on one ormore of these tokamaks. However, for each machine, the most useful parameters,as well as the details of their precursor behaviors, are markedly different. When the prediction problem is framed using a binary classification scheme to discriminate between time slices "close to disruption" and "far from disruption," it is found that the prediction algorithms differ substantially in performance among the three machines ona time slice-by-time slice basis, but have similar disruption detection rates (~80-90%) on a shot-by-shot basis after appropriate optimisation. This could have important implications for disruption prediction and avoidance on ITER, for which developmentof a training database of disruptions may be infeasible. The algorithm's output is interpretable using a method that identifies the mostmore » strongly contributing inputsignals, which may have implications for avoiding disruptive scenarios. Lastly, to further support its real-time capability, successful applications in inter-shot and real-time environments on EAST and DIII-D are also discussed.« less

Authors:
ORCiD logo [1]; ORCiD logo [1];  [1]; ORCiD logo [1]; ORCiD logo [2];  [2];  [3];  [3];  [3];  [4]; ORCiD logo [4]
  1. Massachusetts Inst. of Technology (MIT), Cambridge, MA (United States)
  2. General Atomics, San Diego, CA (United States)
  3. Chinese Academy of Sciences (CAS), Hefei (China)
  4. Princeton Plasma Physics Lab. (PPPL), Princeton, NJ (United States)
Publication Date:
Research Org.:
Dept. of Energy (DOE), Washington DC (United States)
Sponsoring Org.:
USDOE Office of Science (SC), Fusion Energy Sciences (FES) (SC-24)
OSTI Identifier:
1569035
Grant/Contract Number:  
FC02-04ER54698; FC02-99ER54512; SC0014264; SC0010720; SC0010492
Resource Type:
Accepted Manuscript
Journal Name:
Nuclear Fusion
Additional Journal Information:
Journal Volume: 59; Journal Issue: 9; Journal ID: ISSN 0029-5515
Publisher:
IOP Science
Country of Publication:
United States
Language:
English
Subject:
70 PLASMA PHYSICS AND FUSION TECHNOLOGY

Citation Formats

Montes, K. J., Rea, C., Granetz, R. S., Tinguely, R. A., Eidietis, N., Meneghini, O. M., Chen, D. L., Shen, B., Xiao, B. J., Erickson, K., and Boyer, M. D. Machine learning for disruption warnings on Alcator C-Mod, DIII-D, and EAST. United States: N. p., 2019. Web. doi:10.1088/1741-4326/ab1df4.
Montes, K. J., Rea, C., Granetz, R. S., Tinguely, R. A., Eidietis, N., Meneghini, O. M., Chen, D. L., Shen, B., Xiao, B. J., Erickson, K., & Boyer, M. D. Machine learning for disruption warnings on Alcator C-Mod, DIII-D, and EAST. United States. doi:10.1088/1741-4326/ab1df4.
Montes, K. J., Rea, C., Granetz, R. S., Tinguely, R. A., Eidietis, N., Meneghini, O. M., Chen, D. L., Shen, B., Xiao, B. J., Erickson, K., and Boyer, M. D. Tue . "Machine learning for disruption warnings on Alcator C-Mod, DIII-D, and EAST". United States. doi:10.1088/1741-4326/ab1df4.
@article{osti_1569035,
title = {Machine learning for disruption warnings on Alcator C-Mod, DIII-D, and EAST},
author = {Montes, K. J. and Rea, C. and Granetz, R. S. and Tinguely, R. A. and Eidietis, N. and Meneghini, O. M. and Chen, D. L. and Shen, B. and Xiao, B. J. and Erickson, K. and Boyer, M. D.},
abstractNote = {In this paper, we report on disruption prediction using a shallow machinelearning method known as Random Forests, trained on large databases containingonly plasma parameters that are available in real-time on Alcator C-Mod, DIII-D, and EAST. The database for each tokamak contains parameters sampled ~106 times throughout ~104 discharges (disruptive and non-disruptive) over the last 4years of operation. It is found that a number of parameters (e.g.Prad/Pinput,li,n/nG,Bn=1/B0) exhibit changes in aggregate as a disruption is approached on one ormore of these tokamaks. However, for each machine, the most useful parameters,as well as the details of their precursor behaviors, are markedly different. When the prediction problem is framed using a binary classification scheme to discriminate between time slices "close to disruption" and "far from disruption," it is found that the prediction algorithms differ substantially in performance among the three machines ona time slice-by-time slice basis, but have similar disruption detection rates (~80-90%) on a shot-by-shot basis after appropriate optimisation. This could have important implications for disruption prediction and avoidance on ITER, for which developmentof a training database of disruptions may be infeasible. The algorithm's output is interpretable using a method that identifies the most strongly contributing inputsignals, which may have implications for avoiding disruptive scenarios. Lastly, to further support its real-time capability, successful applications in inter-shot and real-time environments on EAST and DIII-D are also discussed.},
doi = {10.1088/1741-4326/ab1df4},
journal = {Nuclear Fusion},
number = 9,
volume = 59,
place = {United States},
year = {2019},
month = {4}
}

Journal Article:
Free Publicly Available Full Text
This content will become publicly available on April 30, 2020
Publisher's Version of Record

Citation Metrics:
Cited by: 3 works
Citation information provided by
Web of Science

Save / Share:

Works referenced in this record:

Random Forests
journal, January 2001