Artificial Neural Network (ANN) models are applied to both simulated and experimental data from the ASDEX Upgrade tokarnak. The results are compared in detail to those obtained from the real time version of the Function Parameterization (FP) technique currently in use for the feedback control of the plasma discharge and for after-shot analysis. Though the on-line evaluation algorithm for both methods is quite similar, the choice of nonlinear models and the strategies for determining the representation coefficients are very different. Both methods can be advantageously used to solve problems of plasma identification.
Get full access to this article
View all access options for this article.
References
1.
CoccoreseE., MartoneR. and MorabitoF.C., On line plasma shape identification in a tokamak reactor via neural networks, in: Proc. 4th Italian Workshop on Neural Networ L., ed. CaianielloE.R., Word Scientific, Singapore, 1992, p. 349.
2.
CoccoreseE., MartoneR. and MorabitoF.C., Nuclear Fusion34(10) (1994), 1349.
3.
ListerJ.B. and SchnurrenbergerH., Nuclear Fusion31(7) (1991), 1291.
4.
MorabitoF.C., in: Proc. of the World Congress on Neural Networ L., WCNN'94, Vol. 1, San Diego CA, 1994, Lawrence Erlbaum and Press INNS, p. 254.
5.
BishopC.M.., in: Fusion Technology, Proc. 17th Symp. SOFT-92, Vol. 2, Rome, 1992, North Holland, Amsterdam, p. 997.
6.
The ASDEX-Upgrade Team, in: Fusion Technology, Proc. 13th Symp. SOFT-84, Vol. 1, Varese, 1984, Pergamon Press, p. 203.
7.
WindH., Function parametrization, in: Proceedings of the 1972 CERN Computing and Data Processing School, CERN 72-21, 1972, p. 53.
8.
BraamsB.J., Computational Studies in Tokamak Equilibrium and Transport, Ph. Thesis D., Utrecht, 1986.
9.
BraamsB.J., JilgeW. and LacknerK., Nuclear Fusion26 (1986), 699.
10.
McCarthyP.J., An Integrated Data Interpretation System for Tokamak Discharges, Ph. Dissertation D., University College Cork, 1992.
11.
RebutP.H., Fusion Engineering and Design30 (1995), 85.
12.
CoccoreseE., McCarthyP.J. and MorabitoF.C., in: Fusion Technology, Proc. of the 18th Symp. SOFT'94, Vol. 1, Karlsruhe, Germany, 1994, eds HerschbachK., MaurerW. and VetterJ.E., North Holland, Amsterdam, p. 795.
13.
LacknerK., Comp. Phys. Commun. 12 (1976), 33.
14.
RumelhartD.E., HintonG.E. and WilliamsR.J., Nature323 (1986), 533.
15.
SmithM., Neural Networks for Statistical Processing, Van Nostrand Reinhold, New York, 1994.
16.
HaykinS., Neural Networb: Comprehensive A. Foundation, IEEE Press and MacMillan, New York, 1994.
17.
BattitiR., Neural Computation4 (1992), 141.
18.
HwangJ.-N., LayS.-R., MaechlerM., MartinR.D. and SchimertJ., IEEE Transactions on Neural Networks5(3) (1994), 342.
19.
MinskyM.L. and PapertS.A., Perceptrons: An Introduction to Computational Geometry, The Press MIT, Cambridge MA, 1988.
20.
PaoY.-H., Adaptive Pattern Recognition and Neural Networb, Addison-Wesley, 1989.
21.
KohonenT., Self-Organization and Associative Memory, Springer, Berlin, 1984.
22.
ReedR., MarksR.J. and OhS., IEEE Transactions on Neural Networks6 (1995), 529.
23.
CoccoreseE., MartoneR. and MorabitoF.C., in: Proc. 5th Italian Workshop on Neural Networb, ed. CaianielloE.R., Word Scientific, Singapore, 1993, p. 349.
24.
NetiC., SchneiderM.H. and YoungE.D., IEEE Transactions on Neural Networks3 (1992), 1.
