Experiment-free disruption prediction for new devices enabled by synthetic diagnostic data augmentation

A new study published on arXiv proposes a deep learning approach to predict plasma disruptions in tokamaks, a critical safety concern for fusion reactors like ITER.

The method addresses the challenge of data scarcity for new devices by augmenting limited experimental data with synthetic diagnostic signals generated from the target device.

This 'experiment-free' prediction strategy aims to ensure the safety of initial plasma operations and subsequent experiments by enabling robust disruption prediction even before real-world data is available.