Skip to content

Latest commit

 

History

History
10 lines (8 loc) · 745 Bytes

README.md

File metadata and controls

10 lines (8 loc) · 745 Bytes
Raw

DisruptCNN

DisruptCNN presents code using the Temporal Convolutional Network (a convolutional neural network with dilated convolutions, see https://github.com/locuslab/TCN) to predict tokamak disruption using raw, high temporal resolution, multi-scale diagnostic data. Specifically, the Electron Cyclotron Emission imaging (ECEi) diagnostic at the DIII-D tokamak was used. Paper here: https://arxiv.org/abs/1911.00149, please cite using:

@article{ChurchillNeurIPS2019,
	author    = {R.M. Churchill and the DIII-D team},
	title     = {Deep convolutional neural networks for multi-scale time-series classification and application to disruption prediction in fusion devices},
	journal   = {arXiv:1911.00149v1},
	year      = {2019},
}