A real-time machine learning-based disruption predictor in DIII-D
Abstract
A disruption prediction algorithm, named DPRF (Disruption Prediction using Random Forests), has run in real-time in the DIII-D plasma control system (PCS) over more than 900 discharges. DPRF naturally provides a probability mapping associated to its predictions, i.e. the disruptivity signal, now incorporated in the DIII-D PCS. Here, we report on disruption prediction accomplishments in terms of shot-by-shot performances, by simulating alarms on each discharge as in the PCS framework. Due to the optimised performance metric chosen to evaluate DPRF, we find that almost all disruptive discharges are detected in average with few hundred of milliseconds of warning time, but this comes at a high cost of false alarms produced. Performances are not satisfying ITER requirements, where the success rate has to be higher than 95%, but this is not completely unexpected. DPRF is trained on many years of major disruptions occurring during the flattop phase of the plasma current on DIII-D, but without any differentiation by cause. Furthermore, we find that DPRF is affected by a relatively high fraction of false alarms occurring during the first 500 milliseconds from the flattop onset. This subtle effect, more evident on the discharges where DPRF ran in real-time, can be marginalized bymore »
- Authors:
-
- General Atomics, San Diego, CA (United States); Massachusetts Inst. of Technology (MIT), Cambridge, MA (United States)
- Massachusetts Inst. of Technology (MIT), Cambridge, MA (United States)
- Princeton Plasma Physics Lab. (PPPL), Princeton, NJ (United States)
- Publication Date:
- Research Org.:
- General Atomics, San Diego, CA (United States); Massachusetts Inst. of Technology (MIT), Cambridge, MA (United States); Princeton Univ., NJ (United States)
- Sponsoring Org.:
- USDOE Office of Science (SC), Fusion Energy Sciences (FES)
- OSTI Identifier:
- 1562300
- Report Number(s):
- DOE-GA-54698
Journal ID: ISSN 0029-5515
- Grant/Contract Number:
- FC02-04ER54698; SC0014264; AC02-09CH11466
- 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; disruption prediction; machine learning; random forests; DIII-D; real-time
Citation Formats
Rea, C., Montes, K. J., Erickson, K. G., Granetz, R. S., and Tinguely, R. A.. A real-time machine learning-based disruption predictor in DIII-D. United States: N. p., 2019.
Web. doi:10.1088/1741-4326/ab28bf.
Rea, C., Montes, K. J., Erickson, K. G., Granetz, R. S., & Tinguely, R. A.. A real-time machine learning-based disruption predictor in DIII-D. United States. https://doi.org/10.1088/1741-4326/ab28bf
Rea, C., Montes, K. J., Erickson, K. G., Granetz, R. S., and Tinguely, R. A.. Tue .
"A real-time machine learning-based disruption predictor in DIII-D". United States. https://doi.org/10.1088/1741-4326/ab28bf. https://www.osti.gov/servlets/purl/1562300.
@article{osti_1562300,
title = {A real-time machine learning-based disruption predictor in DIII-D},
author = {Rea, C. and Montes, K. J. and Erickson, K. G. and Granetz, R. S. and Tinguely, R. A.},
abstractNote = {A disruption prediction algorithm, named DPRF (Disruption Prediction using Random Forests), has run in real-time in the DIII-D plasma control system (PCS) over more than 900 discharges. DPRF naturally provides a probability mapping associated to its predictions, i.e. the disruptivity signal, now incorporated in the DIII-D PCS. Here, we report on disruption prediction accomplishments in terms of shot-by-shot performances, by simulating alarms on each discharge as in the PCS framework. Due to the optimised performance metric chosen to evaluate DPRF, we find that almost all disruptive discharges are detected in average with few hundred of milliseconds of warning time, but this comes at a high cost of false alarms produced. Performances are not satisfying ITER requirements, where the success rate has to be higher than 95%, but this is not completely unexpected. DPRF is trained on many years of major disruptions occurring during the flattop phase of the plasma current on DIII-D, but without any differentiation by cause. Furthermore, we find that DPRF is affected by a relatively high fraction of false alarms occurring during the first 500 milliseconds from the flattop onset. This subtle effect, more evident on the discharges where DPRF ran in real-time, can be marginalized by taking specific precautions on the validity range of the predictions, and performances do improve. Even if presently burdened by some limitations, DPRF provides an incredible and novel advantage. Thanks to the feature contribution analysis (e.g., the identification of which signals contributed to triggering an alarm), it is possible to interpret and explain DPRF predictions. It is the first time that such interpretability features is exploited by a disruption predictor: by uncovering the disruption events' causes, a better understanding of disruption dynamics is achieved, and a clear path toward the design of disruption avoidance strategies can be provided.},
doi = {10.1088/1741-4326/ab28bf},
journal = {Nuclear Fusion},
number = 9,
volume = 59,
place = {United States},
year = {2019},
month = {7}
}
Web of Science
Figures / Tables:

Works referenced in this record:
Development of an efficient real-time disruption predictor from scratch on JET and implications for ITER
journal, September 2013
- Dormido-Canto, S.; Vega, J.; Ramírez, J. M.
- Nuclear Fusion, Vol. 53, Issue 11
Integrated modeling applications for tokamak experiments with OMFIT
journal, July 2015
- Meneghini, O.; Smith, S. P.; Lao, L. L.
- Nuclear Fusion, Vol. 55, Issue 8
Unbiased and non-supervised learning methods for disruption prediction at JET
journal, April 2009
- Murari, A.; Vega, J.; Rattá, G. A.
- Nuclear Fusion, Vol. 49, Issue 5
Magnetic diagnostic system of the DIII-D tokamak
journal, February 2006
- Strait, E. J.
- Review of Scientific Instruments, Vol. 77, Issue 2
A First Analysis of JET Plasma Profile-Based Indicators for Disruption Prediction and Avoidance
journal, July 2018
- Pau, A.; Fanni, A.; Cannas, B.
- IEEE Transactions on Plasma Science, Vol. 46, Issue 7
Exploratory Machine Learning Studies for Disruption Prediction Using Large Databases on DIII-D
journal, February 2018
- Rea, Cristina; Granetz, Robert S.
- Fusion Science and Technology, Vol. 74, Issue 1-2
Results of the JET real-time disruption predictor in the ITER-like wall campaigns
journal, October 2013
- Vega, Jesús; Dormido-Canto, Sebastián; López, Juan M.
- Fusion Engineering and Design, Vol. 88, Issue 6-8
Automatic disruption classification in JET with the ITER-like wall
journal, October 2015
- Cannas, B.; de Vries, P. C.; Fanni, A.
- Plasma Physics and Controlled Fusion, Vol. 57, Issue 12
Survey of disruption causes at JET
journal, April 2011
- de Vries, P. C.; Johnson, M. F.; Alper, B.
- Nuclear Fusion, Vol. 51, Issue 5
Improvements in disruption prediction at ASDEX Upgrade
journal, October 2015
- Aledda, R.; Cannas, B.; Fanni, A.
- Fusion Engineering and Design, Vol. 96-97
Tokamak disruption alarm based on a neural network model of the high- beta limit
journal, June 1997
- Wroblewski, D.; Jahns, G. L.; Leuer, J. A.
- Nuclear Fusion, Vol. 37, Issue 6
Implementing a finite-state off-normal and fault response system for disruption avoidance in tokamaks
journal, March 2018
- Eidietis, N. W.; Choi, W.; Hahn, S. H.
- Nuclear Fusion, Vol. 58, Issue 5
Advanced tokamak research in DIII-D
journal, November 2004
- Greenfield, C. M.; Murakami, M.; Ferron, J. R.
- Plasma Physics and Controlled Fusion, Vol. 46, Issue 12B
Multivariate statistical models for disruption prediction at ASDEX Upgrade
journal, October 2013
- Aledda, R.; Cannas, B.; Fanni, A.
- Fusion Engineering and Design, Vol. 88, Issue 6-8
Real time equilibrium reconstruction for tokamak discharge control
journal, July 1998
- Ferron, J. R.; Walker, M. L.; Lao, L. L.
- Nuclear Fusion, Vol. 38, Issue 7
An advanced disruption predictor for JET tested in a simulated real-time environment
journal, January 2010
- Rattá, G. A.; Vega, J.; Murari, A.
- Nuclear Fusion, Vol. 50, Issue 2
A reduced resistive wall mode kinetic stability model for disruption forecasting
journal, May 2017
- Berkery, J. W.; Sabbagh, S. A.; Bell, R. E.
- Physics of Plasmas, Vol. 24, Issue 5
Machine learning for disruption warnings on Alcator C-Mod, DIII-D, and EAST
journal, July 2019
- Montes, K. J.; Rea, C.; Granetz, R. S.
- Nuclear Fusion, Vol. 59, Issue 9
Advanced Disruption Predictor Based On The Locked Mode Signal: Application To Jet
conference, October 2016
- Vega, Jesús; Moreno, Raul; Pereira, Augusto
- Proceedings of 1st EPS conference on Plasma Diagnostics — PoS(ECPD2015)
A cross-tokamak neural network disruption predictor for the JET and ASDEX Upgrade tokamaks
journal, April 2005
- Windsor, C. G.; Pautasso, G.; Tichmann, C.
- Nuclear Fusion, Vol. 45, Issue 5
miRNALoc: predicting miRNA subcellular localizations based on principal component scores of physico-chemical properties and pseudo compositions of di-nucleotides
journal, September 2020
- Meher, Prabina Kumar; Satpathy, Subhrajit; Rao, Atmakuri Ramakrishna
- Scientific Reports, Vol. 10, Issue 1
Optimization of probiotic therapeutics using machine learning in an artificial human gastrointestinal tract
journal, January 2021
- Westfall, Susan; Carracci, Francesca; Estill, Molly
- Scientific Reports, Vol. 11, Issue 1
The Elements of Statistical Learning
book, January 2001
- Hastie, Trevor; Friedman, Jerome; Tibshirani, Robert
- Springer Series in Statistics
Figures / Tables found in this record: