On-line prediction and mitigation of disruptions in ASDEX Upgrade

Pautasso, G.; Tichmann, C.; Egorov, S.M.; Zehetbauer, T.; Gruber, O.; Maraschek, M.; Mast, K. F.; Mertens, V.; Perchermeier, I.; Raupp, G.; Treutterer, W.; Windsor, C. G.

doi:10.1088/0029-5515/42/1/314

Cited by 81 publications

(88 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In order to avoid or to mitigate the disruptive events a number of disruption prediction techniques have been developed. In many cases neural networks approaches are used and they seems to be the most suitable to predict the event or, more precisely, to build an impending disruption warning indicator [12][13][14][15][16].…”

Section: Disruption Prediction In Nuclear Fusionmentioning

confidence: 99%

Geometrical Kernel Machine for Prediction and Novelty Detection of Disruptive Events in TOKAMAK Machines

Cannas

Delogu

Fanni

et al. 2009

J Sign Process Syst

View full text Add to dashboard Cite

This paper presents a recently introduced Kernel\ud Machine, called Geometrical Kernel Machine, used to\ud predict disruptive events in nuclear fusion reactors. The\ud algorithm proposed to construct the Kernel Machine is able\ud to automatically determine both the number of neurons and\ud the synaptic weights of a Multilayer Perceptron neural\ud network with a single hidden layer. It has been demonstrated\ud that the resulting network is able to classify any finite set\ud of patterns defined in a real domain. The prediction\ud problem has been here modeled as a two classes classification\ud problem. The geometrical interpretation of the\ud network equations allows us both to develop the disruption\ud predictor and to manage the so called ageing of the kernel\ud machine. In fact, using the same kernel machine, a novelty\ud detection system has been integrated in the predictor,\ud increasing the overall system performance, and the reliability\ud of the predictor

show abstract

Section: Disruption Prediction In Nuclear Fusionmentioning

confidence: 99%

Geometrical Kernel Machine for Prediction and Novelty Detection of Disruptive Events in TOKAMAK Machines

Cannas

Delogu

Fanni

et al. 2009

J Sign Process Syst

View full text Add to dashboard Cite

show abstract

“…Methods for predicting disruption onset have been developed (e.g. [34][35][36][37]), but almost all are based on training algorithms with data. Generally, a significant quantity of disruptive data is required for such training, which is likely to be difficult in ITER, which can tolerate only a small number of major disruptions before in-vessel different control objectives may be heavily coupled.…”

Section: Off-normal Eventsmentioning

confidence: 99%

Open and emerging control problems in tokamak plasma control

Walker

Schuster

Mazon

et al. 2008

2008 47th IEEE Conference on Decision and Control

View full text Add to dashboard Cite

Tbe tokamak concept for magnetic confinement of fusion plasmas is now quite mature scientifically. This maturity is evidenced by tbe ongoing worldwide effort to design and construct an internationally supported multi-billion dollar experimental tokamak called ITER, wbose purpose is to demonstrate the scientific and technical feasibility of fusion energy as 8 power source. To achieve its scientific objectives, tbe ITER device will need to implement solutions to several challenging control problems. Some solutions to tbese control problems are already mature, e.g. control of the plasma boundary sbape and stabilization of the vertical stability, but many otber solutions are currently in development or do not yet have viable solution approaches. In almost all cases, control solutions developed on existing tokamaks are made more challenging on ITER by safety issues arising from its nuclear mission and control actuation margins tbat are reduced due to cost considerations. However, many of these problems must ha\'e robust solutions in place before ITER comes online in approximately 2016. In this paper, we summarize a set of the most urgently needed control solutions and describe the progress made toward solving a few of these problems.

show abstract

“…Neural networks are typically trained, in the sense that a predetermined sample of input and output data are used to determine the optimal values of the coefficients of the network. Neural networks have been used to predict various forms of disruptions on ASDEX Upgrade [84][85][86], DIII-D [87], ADITYA [88,89], TEXT [90,91], JET [85,92,93], and JT-60 [94,95].…”

Section: : Introductionmentioning

confidence: 99%