Skip to main content
OSTI.GOVU.S. Department of Energy Office of Scientific and Technical Information
U.S. Department of Energy
Office of Scientific and Technical Information
 

Disruption prediction investigations using Machine Learning tools on DIII-D and Alcator C-Mod

Journal Article · · Plasma Physics and Controlled Fusion
 [1];  [2];  [2];  [2];  [3];  [4];  [3]
  1. Massachusetts Inst. of Technology (MIT), Cambridge, MA (United States). Plasma Science and Fusion Center (PSFC); General Atomics
  2. Massachusetts Inst. of Technology (MIT), Cambridge, MA (United States). Plasma Science and Fusion Center (PSFC)
  3. General Atomics, San Diego, CA (United States)
  4. Columbia Univ., New York, NY (United States)
+ Show Author Affiliations

Using data-driven methodology, we exploit the time series of relevant plasma parameters for a large set of disrupted and non-disrupted discharges to develop a classification algorithm for detecting disruptive phases in shots that eventually disrupt. Comparing the same methodology on different devices is crucial in order to have information on the portability of the developed algorithm and the possible extrapolation to ITER. Therefore, we use data from two very different tokamaks, DIII-D and Alcator C-Mod. We then focus on a subset of disruption predictors, most of which are dimensionless and/or machine-independent parameters, coming from both plasma diagnostics and equilibrium reconstructions, such as the normalized plasma internal inductance ℓ and the n = 1 mode amplitude normalized to the toroidal magnetic field. Using such dimensionless indicators facilitates a more direct comparison between DIII-D and C-Mod. We then choose a shallow Machine Learning technique, called Random Forests, to explore the databases available for the two devices. We show results from the classification task, where we introduce a time dependency through the definition of class labels on the basis of the elapsed time before the disruption (i.e. ‘far from a disruption’ and ‘close to a disruption’). The performances of the different Random Forest classifiers are discussed in terms of several metrics, by showing the number of successfully detected samples, as well as the misclassifications. The overall model accuracies are above 97% when identifying a ‘far from disruption’ and a ‘disruptive’ phase for disrupted discharges. Nevertheless, the Forests are intrinsically different in their capability of predicting a disruptive behavior, with C-Mod predictions comparable to random guesses. Indeed, we show that C-Mod recall index, i.e. the sensitivity to a disruptive behavior, is as low as 0.47, while DIII-D recall is ~0.72. The portability of the developed algorithm is also tested across the two devices, by using DIII-D data for training the forests and C-Mod for testing and vice versa.

Research Organization:
Massachusetts Inst. of Technology (MIT), Cambridge, MA (United States)
Sponsoring Organization:
USDOE Office of Science (SC), Fusion Energy Sciences (FES) (SC-24)
Grant/Contract Number:
FC02-04ER54698; SC0014264; FG02-04ER54761
OSTI ID:
1454268
Journal Information:
Plasma Physics and Controlled Fusion, Journal Name: Plasma Physics and Controlled Fusion Journal Issue: 8 Vol. 60; ISSN 0741-3335
Publisher:
IOP ScienceCopyright Statement
Country of Publication:
United States
Language:
English

References (44)

A Customer Relationship Management Case Study Based on Banking Data book January 2016
Bagging predictors journal August 1996
Bagging predictors journal August 1996
Selecting and interpreting measures of thematic classification accuracy journal October 1997
The MDSplus data acquisition system, current status and future directions journal January 1999
Support vector machines for disruption prediction and novelty detection at JET journal October 2007
Results of the JET real-time disruption predictor in the ITER-like wall campaigns journal October 2013
TokSearch: A search engine for fusion experimental data journal April 2018
Selecting and interpreting measures of thematic classification accuracy journal October 1997
The MDSplus data acquisition system, current status and future directions journal January 1999
Random Forests journal January 2001
Random Forests journal January 2001
Achievement of Sustained Net Plasma Heating in a Fusion Experiment with the Optometrist Algorithm journal July 2017
miRNALoc: predicting miRNA subcellular localizations based on principal component scores of physico-chemical properties and pseudo compositions of di-nucleotides journal September 2020
Nonaxisymmetric field effects on Alcator C-Mod journal May 2005
Magnetic diagnostic system of the DIII-D tokamak journal February 2006
A reduced resistive wall mode kinetic stability model for disruption forecasting journal May 2017
Critical error fields for locked mode instability in tokamaks journal March 1992
Exploratory Machine Learning Studies for Disruption Prediction Using Large Databases on DIII-D journal February 2018
Reconstruction of current profile parameters and plasma shapes in tokamaks journal November 1985
Neural network prediction of some classes of tokamak disruptions journal August 1996
Neural network prediction of some classes of tokamak disruptions journal August 1996
Tokamak disruption alarm based on a neural network model of the high- beta limit journal June 1997
Tokamak disruption alarm based on a neural network model of the high- beta limit journal June 1997
Real time equilibrium reconstruction for tokamak discharge control journal July 1998
Chapter 3: MHD stability, operational limits and disruptions journal December 1999
Forecast of TEXT plasma disruptions using soft X rays as input signal in a neural network journal February 1999
Considerations on energy confinement time scalings using present tokamak databases and prediction for ITER size experiments journal May 2000
On-line prediction and mitigation of disruptions in ASDEX Upgrade journal January 2002
Neural-net disruption predictor in JT-60U journal December 2003
Disruption forecasting at JET using neural networks journal December 2003
A cross-tokamak neural network disruption predictor for the JET and ASDEX Upgrade tokamaks journal April 2005
Automatic disruption classification at JET: comparison of different pattern recognition techniques journal May 2006
A prediction tool for real-time application in the disruption protection system at JET journal October 2007
Survey of disruption causes at JET journal April 2011
Advances in lower hybrid current drive technology on Alcator C-Mod journal June 2013
Integrated modeling applications for tokamak experiments with OMFIT journal July 2015
Statistical analysis of m / n   =  2/1 locked and quasi-stationary modes with rotating precursors at DIII-D journal November 2016
Neoclassical tearing modes journal December 2000
Neoclassical tearing modes journal December 2000
Advanced tokamak research in DIII-D journal November 2004
Advanced tokamak research in DIII-D journal November 2004
A review of feature selection techniques in bioinformatics journal August 2007
Statistical analysis of $m/n$ = 2/1 locked and quasi-stationary modes with rotating precursors at DIII-D text January 2016

Cited By (1)

Machine learning control for disruption and tearing mode avoidance journal February 2020

Figures / Tables (14)

Table 1(p. 5)table Table 1
Table 2(p. 6)table Table 2
Figure 1(p. 7)figure Figure 1
Figure 2(p. 9)figure Figure 2
Figure 3(p. 9)figure Figure 3
Figure 4(p. 11)figure Figure 4
Figure 5(p. 12)figure Figure 5
Figure 6(p. 13)figure Figure 6
Figure 7(p. 14)figure Figure 7
Figure 8(p. 18)figure Figure 8
Table 3(p. 18)table Table 3
Figure 9(p. 19)figure Figure 9
Table 4(p. 19)table Table 4
Table 5(p. 20)table Table 5

Similar Records

Machine learning for disruption warnings on Alcator C-Mod, DIII-D, and EAST
Journal Article · Tue Apr 30 00:00:00 EDT 2019 · Nuclear Fusion · OSTI ID:1569035
Hybrid deep learning architecture for general disruption prediction across tokamaks
Dataset · Fri Jun 25 00:00:00 EDT 2021 · OSTI ID:1886671
Hybrid deep learning architecture for general disruption prediction across tokamaks
Journal Article · Fri Oct 30 00:00:00 EDT 2020 · Nuclear Fusion · OSTI ID:1735947

Related Subjects

43 PARTICLE ACCELERATORS
70 PLASMA PHYSICS AND FUSION TECHNOLOGY
cross-device study
disruptions
machine learning