Return To: Session EXP3 - MHD & Stability
Prev Page: (EXP3/16(R)) Active Feedback Control of the Wall Stabilized
Next Page: Session EXP4 - Current Drive, Heating, Fuelling, Divertors

(EXP3/17) Progress in the Prediction of Disruptions in ASDEX-Upgrade via Neural and Fuzzy-Neural Techniques

M. Versaci¹⁾, F. C. Morabito¹⁾, C. Tichmann²⁾, G. Pautasso²⁾, the ASDEX Upgrade Team^2)
 
1) Università ``Mediterranea'' degli Studi di Reggio Calabria, Associazione EURATOM-ENEA-CREATE, Via Graziella, Feo di Vito, I-89100 Reggio Calabria, Italy.
²⁾ Max-Planck-Institut für Plasmaphysik (IPP), D-85748 Garching bei München, Germany.

Abstract.  The paper addresses the problem of predicting the onset of a disruption on the basis of some known precursors possibly announcing the event. The availability in real time of a large set of diagnostic signals allows us to collectively interpret the data in order to decide whether we are near a disruption or during a normal operation scenario. As a relevant experimental example, a database of disruptive discharges in ASDEX-Upgrade has been analysed in this work. Both Neural Networks (NN's) and Fuzzy Inference Systems (FIS) have been investigated as suitable tools to cope with the prediction problem. The experimental database has been exploited aiming to gain information about the mechanisms which drive the plasma column to a disruption. The proposed processor will operate by implementing a classification of the shot type, and outputing a real number that indicates the time left before the disruption will effectively take place (ttd).

Read the full paper in PDF format.