Interpretation of machine-learning-based disruption models for plasma control

Parsons, Matthew

doi:10.1088/1361-6587/aa72a3

Cited by 15 publications

(5 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, these algorithms are often trained and tested on limited datasets from a single tokamak, and few cross-machine comparison studies have been implemented [11,12]. The methods typically used in previous work have also generally relied on black-box ML algorithms, which have few methods available to interpret their predictions [13].…”

Section: Introductionmentioning

confidence: 99%

Machine learning for disruption warnings on Alcator C-Mod, DIII-D, and EAST

et al. 2019

View full text Add to dashboard Cite

This paper reports on disruption prediction using a shallow machine learning method known as a random forest, trained on large databases containing only plasma parameters that are available in real-time on Alcator C-Mod, DIII-D, and EAST. The database for each tokamak contains parameters sampled ∼10 6 times throughout ∼10 4 discharges (disruptive and nondisruptive) over the last four years of operation. It is found that a number of parameters (e.g. P rad /P input , i , n/n G , B n=1 /B 0 ) exhibit changes in aggregate as a disruption is approached on one or more of these tokamaks. However, for each machine, the most useful parameters, as well as the details of their precursor behaviors, are markedly different. When the prediction problem is framed using a binary classification scheme to discriminate between time slices 'close to disruption' and 'far from disruption', it is found that the prediction algorithms differ substantially in performance among the three machines on a time slice-by-time slice basis, but have similar disruption detection rates (∼80%-90%) on a shot-by-shot basis after appropriate optimisation. This could have important implications for disruption prediction and avoidance on ITER, for which development of a training database of disruptions may be infeasible. The algorithm's output is interpretable using a method that identifies the most strongly contributing input signals, which may have implications for avoiding disruptive scenarios. To further support its real-time capability, successful applications in inter-shot and real-time environments on EAST and DIII-D are also discussed.

show abstract

Section: Introductionmentioning

confidence: 99%

Machine learning for disruption warnings on Alcator C-Mod, DIII-D, and EAST

et al. 2019

View full text Add to dashboard Cite

show abstract

“…Thus, these estimates should be considered as a preliminary measures of signal importance. It must be noted that a method has been proposed to determine how sensitive the state of a plasma is at any given time with respect to the input signals [28]. It is demonstrated that the normalized gradients function is a key factor for the disruption as it can describe the sensitivity of the parameter to the features of disruption.…”

Section: Performance Of Disruption Prediction Modelmentioning

confidence: 99%

Disruption prediction using a full convolutional neural network on EAST

Guo

Shen

Chen

et al. 2020

Plasma Phys. Control. Fusion

View full text Add to dashboard Cite

In this study, a full convolutional neural network is trained on a large database of experimental EAST data to classify disruptive discharges and distinguish them from non-disruptive discharges. The database contains 14 diagnostic parameters from the ∼104 discharges (disruptive and non-disruptive). The test set contains 417 disruptive discharges and 999 non-disruptive discharges, which are used to evaluate the performance of the model. The results reveal that the true positive (TP) rate is ∼ 0.827, while the false positive (FP) rate is ∼0.067. This indicates that 72 disruptive discharges and 67 non-disruptive discharges are misclassified in the test set. The FPs are investigated in detail and are found to emerge due to some subtle disturbances in the signals, which lead to misjudgment of the model. Therefore, hundreds of non-disruptive discharges from training set, containing time slices of small disturbances, are artificially added into the training database for retraining the model. The same test set is used to assess the performance of the improved model. The TP rate of the improved model increases up to 0.875, while its FP rate decreases to 0.061. Overall, the proposed data-driven predicted model exhibits immense potential for application in long pulse fusion devices such as ITER.

show abstract

“…(v) Finally, the linear extrapolation of disruptivity employed in this paper is quite simple (even crude), especially considering that plasma parameters can be actuated in real time to navigate in "disruptivity space." A smarter method, such as that described in [36], could be used to calculate gradients in parameter space and map possible trajectories away from the disruptive boundary (P D = 1, in this case). Furthermore, dynamical models of the the plasma state vector x(t)-which is input into the disruption predictor, i.e.…”

Section: Opportunities For Future Workmentioning

confidence: 99%

An application of survival analysis to disruption prediction via Random Forests

Tinguely

Montes

Rea

et al. 2019

Plasma Phys. Control. Fusion

View full text Add to dashboard Cite

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 trained and tested on these databases. This survival analysis approach is applied to both smooth and noisy test data to highlight important features of the survival and hazard functions. In addition, this method is applied to three Alcator C-Mod plasma discharges and compared to a threshold-based scheme for triggering alarms. In one case, both techniques successfully predict the disruption; although, in another, neither warns of the impending disruption with enough time to mitigate. For the final discharge, the survival analysis approach could avoid the false alarm triggered by the threshold method. Limitations of this analysis and opportunities for future work are also presented.

show abstract

Interpretation of machine-learning-based disruption models for plasma control

Cited by 15 publications

References 14 publications

Machine learning for disruption warnings on Alcator C-Mod, DIII-D, and EAST

Machine learning for disruption warnings on Alcator C-Mod, DIII-D, and EAST

Disruption prediction using a full convolutional neural network on EAST

An application of survival analysis to disruption prediction via Random Forests

Contact Info

Product

Resources

About