Hybrid deep-learning architecture for general disruption prediction across multiple tokamaks

  title={Hybrid deep-learning architecture for general disruption prediction across multiple tokamaks},
  author={J.X. Zhu and Cristina Rea and K. J. Montes and R S Granetz and R M Sweeney and R. A. Tinguely},
  journal={Nuclear Fusion},
In this paper, we present a new deep-learning disruption-prediction algorithm based on important findings from explorative data analysis which effectively allows knowledge transfer from existing devices to new ones, thereby predicting disruptions using very limited disruption data from the new devices. The explorative data analysis, conducted via unsupervised clustering techniques confirms that time-sequence data are much better separators of disruptive and non-disruptive behavior than the… 

IDP-PGFE: An Interpretable Disruption Predictor based on Physics-Guided Feature Extraction

Disruption prediction has made rapid progress in recent years, especially machine learning (ML)-based methods. Understanding why a predictor makes a certain prediction can be as crucial as the

Big Data Analysis of Marketing User Intelligence Information Based on Deep Learning

It is proved that the system proposed in this paper can process massive data very effectively and make a great contribution to the development of the industry.

A Wood Quality Defect Detection System Based on Deep Learning and Multicriterion Framework

In order to solve the problems of image perception and quality decision-making of wood defects with typical bionic intelligent algorithms, the existence of multidimensional degradation factors causes

Overview of the SPARC physics basis towards the exploration of burning-plasma regimes in high-field, compact tokamaks

The SPARC tokamak project, currently in engineering design, aims to achieve breakeven and burning plasma conditions in a compact device, thanks to new developments in high-temperature superconductor

Development of robust indicators for the identification of electron temperature profile anomalies and application to JET

Recent experience with metallic devices operating in ITER relevant regions of the operational space, has shown that the disruptivity of these plasmas is unacceptably high. The main causes of the

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

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.

Disruption prediction on EAST tokamak using a deep learning algorithm

It is proved that deep learning algorithms exhibit immense application potential in the disruption prediction of long-pulse fusion devices.

Scenario adaptive disruption prediction study for next generation burning-plasma tokamaks

The results suggest data-driven disruption predictors trained on abundant LP discharges work poorly on the HP regime of the same tokamak, which is a consequence of the distinct distributions of the tightly correlated signals related to disruptions in these two regimes, and find that matching operational parameters among toKamaks strongly improves cross-machine accuracy.

Experiment data-driven modeling of tokamak discharge in EAST

A neural network model of tokamak discharge is developed based on the experimental dataset of a superconducting long-pulse tokamak (EAST) campaign 2016–2018. The purpose is to reproduce the response



WaveNet: A Generative Model for Raw Audio

WaveNet, a deep neural network for generating raw audio waveforms, is introduced; it is shown that it can be efficiently trained on data with tens of thousands of samples per second of audio, and can be employed as a discriminative model, returning promising results for phoneme recognition.

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

One of the biggest challenges to achieve the goal of producing fusion energy in tokamak devices is the necessity of avoiding disruptions of the plasma current due to instabilities. The disruption

Chapter 7: Measurement of plasma parameters

The physics issues of the measurements of the plasma properties necessary to provide both the control and science data for achieving the goals of the ITER device are discussed. The assessment of the

A machine learning approach based on generative topographic mapping for disruption prevention and avoidance at JET

The proposed approach to predict disruptions has been evaluated on a training and an independent test set and achieves very good performance with only one tardive detection and a limited number of false detections.

A real-time machine learning-based disruption predictor in DIII-D

By uncovering the causes of the disruption events, a better understanding of disruption dynamics is achieved, and a clear path toward the design of disruption avoidance strategies can be provided.

How We Learn; How We Remember: Toward an Understanding of Brain and Neural Systems : Selected Papers of Leon N. Cooper

Some Properties of a Neural Model for Memory A Possible Organization of Animal Memory and Learning A Theory for the Acquisition and Loss of Neuron Specificity in Visual Cortex On the Development of

The causes of the disruptive tearing instabilities of the ITER Baseline Scenario in DIII-D

Analysis of the evolving current density (J), pedestal and rotation profiles in a database of 200 ITER Baseline Scenario (IBS) demonstration discharges in the DIII-D tokamak identifies the current

Viability Assessment of a Cross-Tokamak AUG-JET Disruption Predictor

This work computes a genetic algorithm–optimized predictor inspired by a previous study using initially only ASDEX-Upgrade data and tested with the wide database of JET, and pursues the future extrapolation of this technique to ITER.

Adaptive predictors based on probabilistic SVM for real time disruption mitigation on JET

A new set of tools, based on probabilistic extensions of support vector machines (SVM), are introduced and applied for the first time to JET data, which give the probability of disruption and improves the interpretability of the results, provides an estimate of the predictor quality and gives new insights into the physics.

Disruption Event Characterization and Forecasting in Tokamaks

Disruption prediction and avoidance is a critical need for next-step tokamaks such as ITER. The Disruption Event Characterization and Forecasting Code (DECAF) is used to fully automate analysis of