Predicting resistive wall mode stability in NSTX through balanced random forests and counterfactual explanations

Piccione, A.; Berkery, J. W.; Sabbagh, S. A.; Andreopoulos, Y.

doi:10.1088/1741-4326/ac44af

Title: Predicting resistive wall mode stability in NSTX through balanced random forests and counterfactual explanations

Journal Article · 18 January 2022 · Nuclear Fusion

DOI: https://doi.org/10.1088/1741-4326/ac44af · OSTI ID:1841282

; Berkery, J. W.; Sabbagh, S. A.; Andreopoulos, Y.

Abstract Recent progress in the disruption event characterization and forecasting framework has shown that machine learning guided by physics theory can be easily implemented as a supporting tool for fast computations of ideal stability properties of spherical tokamak plasmas. In order to extend that idea, a customized random forest (RF) classifier that takes into account imbalances in the training data is hereby employed to predict resistive wall mode (RWM) stability for a set of high beta discharges from the NSTX spherical tokamak. More specifically, with this approach each tree in the forest is trained on samples that are balanced via a user-defined over/under-sampler. The proposed approach outperforms classical cost-sensitive methods for the problem at hand, in particular when used in conjunction with a random under-sampler, while also resulting in a threefold reduction in the training time. In order to further understand the model’s decisions, a diverse set of counterfactual explanations based on determinantal point processes (DPP) is generated and evaluated. Via the use of DPP, the underlying RF model infers that the presence of hypothetical magnetohydrodynamic activity would have prevented the RWM from concurrently going unstable, which is a counterfactual that is indeed expected by prior physics knowledge. Given that this result emerges from the data-driven RF classifier and the use of counterfactuals without hand-crafted embedding of prior physics intuition, it motivates the usage of counterfactuals to simulate real-time control by generating the β _N levels that would have kept the RWM stable for a set of unstable discharges.

View Journal Article

Cite

Export

Save

Research Organization:: Princeton Plasma Physics Laboratory (PPPL), Princeton, NJ (United States)

Sponsoring Organization:: Engineering and Physical Sciences Research Council (EPSRC); USDOE

Grant/Contract Number:: SC0018623

OSTI ID:: 1841282

Journal Information:: Nuclear Fusion, Journal Name: Nuclear Fusion Journal Issue: 3 Vol. 62; ISSN 0029-5515

Publisher:: IOP PublishingCopyright Statement

Country of Publication:: IAEA

Language:: English

References (49)

The tokamak Monte Carlo fast ion module NUBEAM in the National Transport Code Collaboration library Pankin, Alexei; McCune, Douglas; Andre, Robert Computer Physics Communications, Vol. 159, Issue 3 https://doi.org/10.1016/j.cpc.2003.11.002	journal	June 2004
Prediction of high-beta disruptions in JT-60U based on sparse modeling using exhaustive search Yokoyama, T.; Miyoshi, Y.; Hiwatari, R. Fusion Engineering and Design, Vol. 140 https://doi.org/10.1016/j.fusengdes.2019.01.128	journal	March 2019
Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations Raissi, M.; Perdikaris, P.; Karniadakis, G. E. Journal of Computational Physics, Vol. 378 https://doi.org/10.1016/j.jcp.2018.10.045	journal	February 2019
Predicting disruptive instabilities in controlled fusion plasmas through deep learning Kates-Harbeck, Julian; Svyatkovskiy, Alexey; Tang, William Nature, Vol. 568, Issue 7753 https://doi.org/10.1038/s41586-019-1116-4	journal	April 2019
Beta-limiting instabilities and global mode stabilization in the National Spherical Torus Experiment Sabbagh, S. A.; Bell, R. E.; Bell, M. G. Physics of Plasmas, Vol. 9, Issue 5 https://doi.org/10.1063/1.1468230	journal	May 2002
Application of the low-frequency energy principle to wall modes Hu, Bo; Betti, R.; Manickam, J. Physics of Plasmas, Vol. 12, Issue 5 https://doi.org/10.1063/1.1873852	journal	May 2005
Neoclassical tearing modes and their control La Haye, R. J. Physics of Plasmas, Vol. 13, Issue 5 https://doi.org/10.1063/1.2180747	journal	May 2006
Toroidal self-consistent modeling of drift kinetic effects on the resistive wall mode Liu, Yueqiang; Chu, M. S.; Chapman, I. T. Physics of Plasmas, Vol. 15, Issue 11 https://doi.org/10.1063/1.3008045	journal	November 2008
The role of kinetic effects, including plasma rotation and energetic particles, in resistive wall mode stability Berkery, J. W.; Sabbagh, S. A.; Reimerdes, H. Physics of Plasmas, Vol. 17, Issue 8 https://doi.org/10.1063/1.3474925	journal	August 2010
Benchmarking kinetic calculations of resistive wall mode stability Berkery, J. W.; Liu, Y. Q.; Wang, Z. R. Physics of Plasmas, Vol. 21, Issue 5 https://doi.org/10.1063/1.4873894	journal	May 2014
Measured improvement of global magnetohydrodynamic mode stability at high-beta, and in reduced collisionality spherical torus plasmas Berkery, J. W.; Sabbagh, S. A.; Balbaky, A. Physics of Plasmas, Vol. 21, Issue 5 https://doi.org/10.1063/1.4876610	journal	May 2014
The effect of an anisotropic pressure of thermal particles on resistive wall mode stability Berkery, J. W.; Betti, R.; Sabbagh, S. A. Physics of Plasmas, Vol. 21, Issue 11 https://doi.org/10.1063/1.4901568	journal	November 2014
The direct criterion of Newcomb for the ideal MHD stability of an axisymmetric toroidal plasma Glasser, A. H. Physics of Plasmas, Vol. 23, Issue 7 https://doi.org/10.1063/1.4958328	journal	July 2016
A reduced resistive wall mode kinetic stability model for disruption forecasting Berkery, J. W.; Sabbagh, S. A.; Bell, R. E. Physics of Plasmas, Vol. 24, Issue 5 https://doi.org/10.1063/1.4977464	journal	May 2017
Application of benchmarked kinetic resistive wall mode stability codes to ITER, including additional physics Berkery, J. W.; Wang, Z. R.; Sabbagh, S. A. Physics of Plasmas, Vol. 24, Issue 11 https://doi.org/10.1063/1.4989503	journal	November 2017
Deep convolutional neural networks for multi-scale time-series classification and application to tokamak disruption prediction using raw, high temporal resolution diagnostic data Churchill, R. M.; Tobias, B.; Zhu, Y. Physics of Plasmas, Vol. 27, Issue 6 https://doi.org/10.1063/1.5144458	journal	June 2020
Progress Toward Interpretable Machine Learning–Based Disruption Predictors Across Tokamaks Rea, C.; Montes, K. J.; Pau, A. Fusion Science and Technology, Vol. 76, Issue 8 https://doi.org/10.1080/15361055.2020.1798589	journal	September 2020
Resistive wall stabilized operation in rotating high beta NSTX plasmas Sabbagh, S. A.; Sontag, A. C.; Bialek, J. M. Nuclear Fusion, Vol. 46, Issue 5 https://doi.org/10.1088/0029-5515/46/5/014	journal	April 2006
Exploration of the equilibrium operating space for NSTX-Upgrade Gerhardt, S. P.; Andre, R.; Menard, J. E. Nuclear Fusion, Vol. 52, Issue 8 https://doi.org/10.1088/0029-5515/52/8/083020	journal	August 2012
Modifications to ideal stability by kinetic effects in NSTX Berkery, J. W.; Sabbagh, S. A.; Bell, R. E. Nuclear Fusion, Vol. 55, Issue 12 https://doi.org/10.1088/0029-5515/55/12/123007	journal	October 2015
Central safety factor and β _N control on NSTX-U via beam power and plasma boundary shape modification, using TRANSP for closed loop simulations Boyer, M. D.; Andre, R.; Gates, D. A. Nuclear Fusion, Vol. 55, Issue 5 https://doi.org/10.1088/0029-5515/55/5/053033	journal	April 2015
Stabilization of the external kink and the resistive wall mode Chu, M. S.; Okabayashi, M. Plasma Physics and Controlled Fusion, Vol. 52, Issue 12 https://doi.org/10.1088/0741-3335/52/12/123001	journal	October 2010
Neural network based prediction of no-wall β _N limits due to ideal external kink instabilities Liu, Yueqiang; Lao, Lang; Li, Li Plasma Physics and Controlled Fusion, Vol. 62, Issue 4 https://doi.org/10.1088/1361-6587/ab6f56	journal	February 2020
Projected global stability of high beta MAST-U spherical tokamak plasmas Berkery, J. W.; Xia, G.; Sabbagh, S. A. Plasma Physics and Controlled Fusion, Vol. 62, Issue 8 https://doi.org/10.1088/1361-6587/ab98e1	journal	July 2020
NSTX/NSTX-U theory, modeling and analysis results Kaye, S. M.; Battaglia, D. J.; Baver, D. Nuclear Fusion, Vol. 59, Issue 11 https://doi.org/10.1088/1741-4326/ab023a	journal	June 2019
Real-time capable modeling of neutral beam injection on NSTX-U using neural networks Boyer, M. D.; Kaye, S.; Erickson, K. Nuclear Fusion, Vol. 59, Issue 5 https://doi.org/10.1088/1741-4326/ab0762	journal	March 2019
Progress in disruption prevention for ITER Strait, E. J.; Barr, J. L.; Baruzzo, M. Nuclear Fusion, Vol. 59, Issue 11 https://doi.org/10.1088/1741-4326/ab15de	journal	June 2019
Machine learning for disruption warnings on Alcator C-Mod, DIII-D, and EAST Montes, K. J.; Rea, C.; Granetz, R. S. Nuclear Fusion, Vol. 59, Issue 9 https://doi.org/10.1088/1741-4326/ab1df4	journal	July 2019
A real-time machine learning-based disruption predictor in DIII-D Rea, C.; Montes, K. J.; Erickson, K. G. Nuclear Fusion, Vol. 59, Issue 9 https://doi.org/10.1088/1741-4326/ab28bf	journal	July 2019
A machine learning approach based on generative topographic mapping for disruption prevention and avoidance at JET Pau, A.; Fanni, A.; Carcangiu, S. Nuclear Fusion, Vol. 59, Issue 10 https://doi.org/10.1088/1741-4326/ab2ea9	journal	August 2019
Physics-guided machine learning approaches to predict the ideal stability properties of fusion plasmas Piccione, A.; Berkery, J. W.; Sabbagh, S. A. Nuclear Fusion, Vol. 60, Issue 4 https://doi.org/10.1088/1741-4326/ab7597	journal	March 2020
On the transfer of adaptive predictors between different devices for both mitigation and prevention of disruptions Murari, A.; Rossi, R.; Peluso, E. Nuclear Fusion, Vol. 60, Issue 5 https://doi.org/10.1088/1741-4326/ab77a6	journal	March 2020
Hybrid deep-learning architecture for general disruption prediction across multiple tokamaks Zhu, J. X.; Rea, C.; Montes, K. Nuclear Fusion, Vol. 61, Issue 2 https://doi.org/10.1088/1741-4326/abc664	journal	December 2020
A semi-supervised machine learning detector for physics events in tokamak discharges Montes, K. J.; Rea, C.; Tinguely, R. A. Nuclear Fusion, Vol. 61, Issue 2 https://doi.org/10.1088/1741-4326/abcdb9	journal	January 2021
Prediction of electron density and pressure profile shapes on NSTX-U using neural networks Boyer, M. D.; Chadwick, J. Nuclear Fusion, Vol. 61, Issue 4 https://doi.org/10.1088/1741-4326/abe08b	journal	March 2021
Real-time prediction of high-density EAST disruptions using random forest Hu, W. H.; Rea, C.; Yuan, Q. P. Nuclear Fusion, Vol. 61, Issue 6 https://doi.org/10.1088/1741-4326/abf74d	journal	May 2021
Feedforward beta control in the KSTAR tokamak by deep reinforcement learning Seo, Jaemin; Na, Y. -S.; Kim, B. Nuclear Fusion, Vol. 61, Issue 10 https://doi.org/10.1088/1741-4326/ac121b	journal	September 2021
Resistive Wall Mode Instability at Intermediate Plasma Rotation Berkery, J. W.; Sabbagh, S. A.; Betti, R. Physical Review Letters, Vol. 104, Issue 3 https://doi.org/10.1103/physrevlett.104.035003	journal	January 2010
Effect of Collisionality on Kinetic Stability of the Resistive Wall Mode Berkery, J. W.; Sabbagh, S. A.; Betti, R. Physical Review Letters, Vol. 106, Issue 7 https://doi.org/10.1103/physrevlett.106.075004	journal	February 2011
Stabilization of external modes in tokamaks by resistive walls and plasma rotation Bondeson, A.; Ward, D. J. Physical Review Letters, Vol. 72, Issue 17 https://doi.org/10.1103/physrevlett.72.2709	journal	April 1994
Wall Stabilization of High Beta Tokamak Discharges in DIII-D Strait, E. J.; Taylor, T. S.; Turnbull, A. D. Physical Review Letters, Vol. 74, Issue 13 https://doi.org/10.1103/physrevlett.74.2483	journal	March 1995
Observation of Plasma Toroidal-Momentum Dissipation by Neoclassical Toroidal Viscosity Zhu, W.; Sabbagh, S. A.; Bell, R. E. Physical Review Letters, Vol. 96, Issue 22 https://doi.org/10.1103/physrevlett.96.225002	journal	June 2006
Active Stabilization of the Resistive-Wall Mode in High-Beta, Low-Rotation Plasmas Sabbagh, S. A.; Bell, R. E.; Menard, J. E. Physical Review Letters, Vol. 97, Issue 4 https://doi.org/10.1103/physrevlett.97.045004	journal	July 2006
Theory-Guided Data Science: A New Paradigm for Scientific Discovery from Data Karpatne, Anuj; Atluri, Gowtham; Faghmous, James H. IEEE Transactions on Knowledge and Data Engineering, Vol. 29, Issue 10 https://doi.org/10.1109/tkde.2017.2720168	journal	October 2017
IFME: information filtering by multiple examples with under-sampling in a digital library environment Zhu, Mingzhu; Xu, Chao; Wu, Yi-Fang Brook Proceedings of the 13th ACM/IEEE-CS joint conference on Digital libraries - JCDL '13 https://doi.org/10.1145/2467696.2467736	conference	January 2013
Explaining machine learning classifiers through diverse counterfactual explanations Mothilal, Ramaravind K.; Sharma, Amit; Tan, Chenhao FAT* '20: Conference on Fairness, Accountability, and Transparency, Proceedings of the 2020 Conference on Fairness, Accountability, and Transparency https://doi.org/10.1145/3351095.3372850	conference	January 2020
Implementation of β _N Control in the National Spherical Torus Experiment Gerhardt, S. P.; Mastrovito, D.; Bell, M. G. Fusion Science and Technology, Vol. 61, Issue 1 https://doi.org/10.13182/fst12-a13336	journal	January 2012
Disruption Prediction on JET during the ILW Experimental Campaigns Moreno, R.; Vega, J.; Dormido-Canto, S. Fusion Science and Technology, Vol. 69, Issue 2 https://doi.org/10.13182/fst15-167	journal	April 2016
SMOTE: Synthetic Minority Over-sampling Technique Chawla, N. V.; Bowyer, K. W.; Hall, L. O. Journal of Artificial Intelligence Research, Vol. 16 https://doi.org/10.1613/jair.953	journal	January 2002

Similar Records

Physics-guided machine learning approaches to predict the ideal stability properties of fusion plasmas

Journal Article · 2020 · Nuclear Fusion · OSTI ID:1803428

Piccione, A.; Berkery, J. W.; Sabbagh, S. A.; +1 more

Meaning of counterfactual statements in quantum physics

Journal Article · 1998 · American Journal of Physics · OSTI ID:657798

Stapp, H P

A Reduced Resistive Wall Mode Kinetic Stability Model for Disruption Forecasting

Dataset · 2017 · OSTI ID:1367870

Berkery, J.W.; Sabbagh, S.A.; Bell, R.E.; +2 more

Related Subjects

70 PLASMA PHYSICS AND FUSION TECHNOLOGY
counterfactuals
machine learning
random forest
resistive wall mode

Title: Predicting resistive wall mode stability in NSTX through balanced random forests and counterfactual explanations

Citation Formats

References (49)

Similar Records

Related Subjects