Image of Plasma Disruption and Non-Disruption on Jet Tokamak (IMAGE)

DOE/Princeton Plasma Physics Laboratory

Image of Plasma Disruption and Non-Disruption on Jet Tokamak

Caption

Image of plasma disruption in experiment on JET, left, and disruption-free experiment on JET, right. Training the FRNN neural network to predict disruptions calls for assigning weights to the data flow along the connections between nodes. Data from new experiments is then put through the network, which predicts "disruption" or "non-disruption." The ultimate goal is at least 95 percent correct predictions of disruption events.

Credit

Image and explanation courtesy of Eliot Feibush.

Usage Restrictions

None

License

Licensed content

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.