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

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.

arXiv commentAccepted at the Second Workshop on Machine Learning and the Physical Sciences (NeurIPS 2019), Vancouver, Canada

References in corpus (1)

• A systematic study of the class imbalance problem in convolutional neural networks

Cited by in corpus (7)

• 2022 Review of Data-Driven Plasma Science
• Plasma steering to avoid disruptions in ITER and tokamak power plants
• Using HPC infrastructures for deep learning applications in fusion research
• A data management system for machine learning research of tokamak
• Interpreting AI for Fusion: an application to Plasma Profile Analysis for Tearing Mode Stability
• Parallelized Real-time Physics Codes for Plasma Control on DIII-D
• Opacity predictions in plasmas under stellar conditions using deep learning