# Hybrid deep learning architecture for general disruption prediction across tokamaks

arXiv: https://arxiv.org/abs/2007.01401 · HTML: https://mlsanity.com/paper/2007.01401/

## Key facts

- Cited by 17 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 2 Jul 2020.

## Built on by (most cited first, 17 of 17)

- [Fast Dynamic 1D Simulation of Divertor Plasmas with Neural PDE Surrogates](https://mlsanity.com/paper/2305.18944/) (2023)
- [Scenario adaptive disruption prediction study for next generation burning-plasma tokamaks](https://mlsanity.com/paper/2109.08956/) (2021)
- [Learning Plasma Dynamics and Robust Rampdown Trajectories with Predict-First Experiments at TCV](https://mlsanity.com/paper/2502.12327/) (2025)
- [IDP-PGFE: An Interpretable Disruption Predictor based on Physics-Guided Feature Extraction](https://mlsanity.com/paper/2208.13197/) (2022)
- [TokaMark: A Comprehensive Benchmark for MAST Tokamak Plasma Models](https://mlsanity.com/paper/2602.10132/) (2026)
- [Continuous Convolutional Neural Networks for Disruption Prediction in Nuclear Fusion Plasmas](https://mlsanity.com/paper/2312.01286/) (2023)
- [Robust Confinement State Classification with Uncertainty Quantification through Ensembled Data-Driven Methods](https://mlsanity.com/paper/2502.17397/) (2025)
- [Plasma State Monitoring and Disruption Characterization using Multimodal VAEs](https://mlsanity.com/paper/2504.17710/) (2025)
- [FusionMAE: large-scale pretrained model to optimize and simplify diagnostic and control of fusion plasma](https://mlsanity.com/paper/2509.12945/) (2025)
- [Cross-tokamak Disruption Prediction based on Physics-Guided Feature Extraction and domain adaptation](https://mlsanity.com/paper/2309.05361/) (2023)
- [EuroPED-NN: Uncertainty aware surrogate model](https://mlsanity.com/paper/2402.00760/) (2024)
- [Time Series Viewmakers for Robust Disruption Prediction](https://mlsanity.com/paper/2410.11065/) (2024)
- [Disruption Precursor Onset Time Study Based on Semi-supervised Anomaly Detection](https://mlsanity.com/paper/2303.14965/) (2023)
- [Hierarchical Multi-to-Single-Modal Knowledge Distillation for Disruption Prediction in EAST](https://mlsanity.com/paper/2607.04241/) (2026)
- [Interpretable statistical feature engineering for early disruption prediction in the short pulse ADITYA tokamak](https://mlsanity.com/paper/2608.22515/) (2026)
- [Horizon-Aware Early Event Prediction for Tokamak Disruption Alarms](https://mlsanity.com/paper/2609.24443/) (2026)
- [FusionMMT: A Unified Multimodal and Multitask Learning Framework for Nuclear Fusion](https://mlsanity.com/paper/2609.26095/) (2026)

## Abstract (arXiv)

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 disruptive 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 instantaneous plasma state data with further advantageous implications for a sequence-based predictor. Based on such important findings, we have designed a new algorithm for multi-machine disruption prediction that achieves high predictive accuracy on the C-Mod (AUC=0.801), DIII-D (AUC=0.947) and EAST (AUC=0.973) tokamaks with limited hyperparameter tuning. Through numerical experiments, we show that boosted accuracy (AUC=0.959) is achieved on EAST predictions by including in the training only 20 disruptive discharges, thousands of non-disruptive discharges from EAST, and combining this with more than a thousand discharges from DIII-D and C-Mod. The improvement of predictive ability obtained by combining disruptive data from other devices is found to be true for all permutations of the three devices. Furthermore, by comparing the predictive performance of each individual numerical experiment, we find that non-disruptive data are machine-specific while disruptive data from multiple devices contain device-independent knowledge that can be used to inform predictions for disruptions occurring on a new device.