Deep convolutional neural networks for multi-scale time-series classification and application to disruption prediction in fusion devices

TL;DR

This paper demonstrates that deep convolutional neural networks with dilated convolutions can effectively predict disruptions in fusion plasmas by analyzing high-resolution electron temperature data, achieving high accuracy.

Contribution

It introduces a deep CNN architecture with large receptive fields for multi-scale time-series analysis in fusion disruption prediction, using only a single diagnostic data source.

Findings

Achieved approximately 91% F1-score on disruption prediction

Successfully applied CNN to raw ECEi data for plasma event forecasting

Validated the approach on real fusion device data

Abstract

The multi-scale, mutli-physics nature of fusion plasmas makes predicting plasma events challenging. Recent advances in deep convolutional neural network architectures (CNN) utilizing dilated convolutions enable accurate predictions on sequences which have long-range, multi-scale characteristics, such as the time-series generated by diagnostic instruments observing fusion plasmas. Here we apply this neural network architecture to the popular problem of disruption prediction in fusion tokamaks, utilizing raw data from a single diagnostic, the Electron Cyclotron Emission imaging (ECEi) diagnostic from the DIII-D tokamak. ECEi measures a fundamental plasma quantity (electron temperature) with high temporal resolution over the entire plasma discharge, making it sensitive to a number of potential pre-disruptions markers with different temporal and spatial scales. Promising, initial disruption prediction results are obtained training a deep CNN with large receptive field (~30k), achieving an $F_1$-score of ~91% on individual time-slices using only the ECEi data.