Predicting Resistive Wall Mode Stability in NSTX through Balanced Random Forests and Counterfactual Explanations

Piccione, Andrea; Sabbagh, Steven; Andreopoulos, Yiannis

doi:10.11578/1888266

Title: Predicting Resistive Wall Mode Stability in NSTX through Balanced Random Forests and Counterfactual Explanations

Dataset
Other Related Research

Authors:

Piccione, Andrea; Sabbagh, Steven; Andreopoulos, Yiannis

Princeton Plasma Physics Laboratory

Publication Date:: Tue Dec 07 04:00:00 UTC 2021

DOE Contract Number:: SC0018623

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

Sponsoring Org.:: USDOE Office of Science (SC)

OSTI Identifier:: 1888266

DOI:: https://doi.org/10.11578/1888266

Citation Formats


                    Piccione, Andrea, Sabbagh, Steven, and Andreopoulos, Yiannis. Predicting Resistive Wall Mode Stability in NSTX through Balanced Random Forests and Counterfactual Explanations.  United States: N. p., 2021. 
        Web.  doi:10.11578/1888266.

Copy to clipboard


                    Piccione, Andrea, Sabbagh, Steven, & Andreopoulos, Yiannis. Predicting Resistive Wall Mode Stability in NSTX through Balanced Random Forests and Counterfactual Explanations.  United States.  doi:https://doi.org/10.11578/1888266

Copy to clipboard


                    Piccione, Andrea, Sabbagh, Steven, and Andreopoulos, Yiannis. 2021.  
"Predicting Resistive Wall Mode Stability in NSTX through Balanced Random Forests and Counterfactual Explanations".  United States.  doi:https://doi.org/10.11578/1888266.  https://www.osti.gov/servlets/purl/1888266. Pub date:Tue Dec 07 04:00:00 UTC 2021

Copy to clipboard


                    
@article{osti_1888266,

  title        = {Predicting Resistive Wall Mode Stability in NSTX through Balanced Random Forests and Counterfactual Explanations},

  author       = {Piccione, Andrea and Sabbagh, Steven and Andreopoulos, Yiannis},

  abstractNote = {},

  doi          = {10.11578/1888266},

  journal      = {},

  number       = ,

  volume       = ,

  place        = {United States},

  year         = {Tue Dec 07 04:00:00 UTC 2021},

  month        = {Tue Dec 07 04:00:00 UTC 2021}

}

Copy to clipboard

Dataset:

View Dataset

DOI: https://doi.org/10.11578/1888266

Save / Share:

Export Metadata

Save to My Library

Similar records in DOE Data Explorer and OSTI.GOV collections:

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

Journal Article Piccione, A. ; Berkery, J. W. ; Sabbagh, S. A. ; ... - Nuclear Fusion

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 viamore »« less
Predicting resistive wall mode stability in NSTX through balanced random forests and counterfactual explanations

Journal Article Piccione, A. ; Berkery, J. W. ; Sabbagh, S. A. ; ... - Nuclear Fusion

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 amore »« less
A Reduced Resistive Wall Mode Kinetic Stability Model for Disruption Forecasting

Dataset Berkery, J.W. ; Sabbagh, S.A. ; Bell, R.E. ; ...

Kinetic modification of ideal stability theory from stabilizing resonances of mode-particle interaction has had success in explaining resistive wall mode (RWM) stability limits in tokamaks. With the goal of real-time stability forecasting, a reduced kinetic stability model has been implemented in the new Disruption Event Characterization and Forecasting (DECAF) code, which has been written to analyze disruptions in tokamaks. The reduced model incorporates parameterized models for ideal limits on beta, a ratio of plasma pressure to magnetic pressure, which are shown to be in good agreement with DCON code calculations. Increased beta between these ideal limits causes a shift inmore »« less
Stabilizing effect of resistivity towards ELM-free H-mode discharge in lithium-conditioned NSTX

Dataset Banerjee, D. ; Zhu, P. ; Maingi, R.

Linear stability analysis of the national spherical torus experiment (NSTX) Li-conditioned ELM-free H-mode equilibria is carried out in the context of the extended magneto-hydrodynamic (MHD) model in NIMROD. The purpose is to investigate the physical cause behind edge localized mode (ELM) suppression in experiment after the Li-coating of the divertor and the first wall of the NSTX tokamak. Besides ideal MHD modeling, including finite-Larmor radius effect and two-fluid Hall and electron diamagnetic drift contributions, a non-ideal resistivity model is employed, taking into account the increase of Z eff after Li-conditioning in ELM-free H-mode. Unlike an earlier conclusion from an eigenvaluemore »« less
An application of survival analysis to disruption prediction via Random Forests

Dataset Tinguely, R. A. ; Montes, K. J. ; Rea, C. ; ...

One of the most pressing challenges facing the fusion community is adequately mitigating or, even better, avoiding disruptions of tokamak plasmas. However, before this can be done, disruptions must first be predicted with sufficient warning time to actuate a response. The established field of survival analysis provides a convenient statistical framework for time-to-event (i.e. time-to-disruption) studies. This paper demonstrates the integration of an existing disruption prediction machine learning algorithm with the Kaplan-Meier estimator of survival probability. Specifically discussed are the implied warning times from binary classification of disruption databases and the interpretation of output signals from Random Forest algorithms trainedmore »« less

Similar Records