Novelty detection for on-line disruption prediction systems

B. Cannas, A. Fanni, P. Sonato, M.K. Zedda

Research output: Contribution to conferencePaper

2 Citations (Scopus)

Abstract

One of the main factors limiting the implementation of neural networks in industrial applications is the difficulty of detecting potentially unreliable outputs. This could be the case of the neural disruption predictor installed in JET, where new plasma configurations might present features completely different from the ones observed in the experiments used in the training set. This 'novelty' can lead to incorrect behaviour of the network. A Novelty Detection method, which determines the novelty of the output of the neural network, can be used to assess the network reliability. In this paper, two approaches to Novelty Detection are tested, i.e., Self Organising Maps and Support Vector Machines. Preliminary results are encouraging, in particular when referring to false alarms. Copyright © (2005) by the European Physical Society (EPS).
Original languageEnglish
Publication statusPublished - 2005
Externally publishedYes
Event32nd European Physical Society Conference on Plasma Physics and Controlled Fusion combined with the 8th International Workshop on Fast Ignition of Fusion Targets - , Spain
Duration: 1 Jan 2005 → …

Conference

Conference32nd European Physical Society Conference on Plasma Physics and Controlled Fusion combined with the 8th International Workshop on Fast Ignition of Fusion Targets
CountrySpain
Period1/1/05 → …

    Fingerprint

All Science Journal Classification (ASJC) codes

  • Atomic and Molecular Physics, and Optics

Cite this

Cannas, B., Fanni, A., Sonato, P., & Zedda, M. K. (2005). Novelty detection for on-line disruption prediction systems. Paper presented at 32nd European Physical Society Conference on Plasma Physics and Controlled Fusion combined with the 8th International Workshop on Fast Ignition of Fusion Targets, Spain.