Search

The detailed design of the magnetic diagnostic for ITER is presented. The system consists of groups of pickup coils and flux loops on the vessel wall, the back plate, and the divertor. These sensors provide the measurements for the equilibrium reconstruction and the fluctuation analysis. The system is supplemented by Rogowski coils for halo current measurements and by a diamagnetic loop. The complete system meets the measurement requirements, matching those of contemporary large divertor tokamaks. The maximum radiation exposure is such that the sensors will survive for the lifetime of the ITER.

In real electromagnetic problems, it is often required to rapidly interpret a lot of experimental raw data provided by a set of sensors with the aim of controlling the time evolution of a system under observation. This is the case, for example, of the real time control of a plasma discharge in a tokamak device for nuclear fusion experiments. Some procedures apt to carry out principal component analysis (PCA) by using suitable artificial neural network (ANN) models are presented. The related models allow one to adaptively extract the PCs directly from the input data without estimating in advance the covariance matrix over the sample database. The proposed architectures may also cope with non-stationary problems. Two examples of application in electromagnetics are presented which concern respectively the reduction of dimensionality in a typical identification problem and the adaptive recovery of the PCs during a slow change of the statistics of the simulated experiment.

A numerical technique based on a suitable combination of artificial neural networks (ANNs) and fuzzy logic (FL) is presented. It is shown how the ANN solution of typical inverse problems can take advantage of the introduction of fuzzy information. The study case is an inverse electrostatic problem of some relevance for nondestructive testing (NDT) applications. The performance of both standard ANNs and the novel hybrid neuro-fuzzy model are compared, and it is shown that the structured approach is superior to the unstructured one, particularly in terms of speed of the learning phase.

In this paper, a short review of the work done in the framework of a nation-wide research programme on 'Models and Methods for Plasma Control in Magnetically Confined Fusion Experiments' is presented. The broad aim of the overall programme is to develop and propose a new effective and reliable approach to the on-line plasma control for future fusion experiments, starting from the today's theoretical background, validated by experimental evidence from a number of tests performed on existing experiments. The proposed formulation to approach the control problem is a linearized model in terms of suitable state variables and input/output relationships. The basic project has been subdivided into four major areas of investigation: the linearized response plasma model, the three-dimensional electromagnetic model, the identification techniques and finally the plasma control requirements. The most remarkable results, achieved so far in each area above, are presented in the paper. (C) 2001 Elsevier Science B.V. All rights reserved.

The specification of the large magnets needed in future fusion reactors based on the tokamak concept is presented as an inverse electromagnetic problem. The influence of the eddy currents flowing in the metallic structure is illustrated. Currently used computational tools and related mathematical formulations are reviewed. The behavior predicted by the numerical models is in agreement with experimental measurements, as demonstrated by a campaign of code validation. Recent progress on the techniques for the identification of the plasma parameters and for the scenario optimization is presented.

The problem of the vertical instability of the plasma column in the NET/ITER (International Thermonuclear Experimental Reactor) reactor is analyzed via a suitable low-frequency electromagnetic modeling of the structure. For this purpose a 3-D eddy current code is coupled with techniques belonging to MHD (magnetohydrodynamics) and control theory domains. In particular, the specific design concept called twin loops, which has recently been proposed for the passive control of the plasma vertical instability in NET/ITER, is analyzed. The results show good agreement among models of different natures adopted to solve this kind of coupled problem. >

The properties of finite dimensional nonlinear systems relative to the numerical formulations of a class of three-dimensional electromagnetic field problems are discussed. These structural properties increase confidence in the results of the numerical calculation, since problems arising from multiple or chaotic solutions are a priori excluded. Numerical procedures for the computation of transient and steady-state solutions are presented and applied. The error introduced by the finite-element discretization is estimated making use of the properties of the dual electric and magnetic formulations. >

The paper aims to give a contribution to the critical issue of the plasma disruptions in the next generation large tokamaks. In this framework the electromagnetic effects occurring in the vacuum vessel structure proposed for NET-I are predicted by means of a 3-D eddy current code. Typical disruption scenarios are considered, including plasma current decays preceeded by radial or vertical motions. The mesh used for the computations includes vacuum vessel, poloidal field coils and a toroidally segmented outboard first wall. The main results of the analysis (time history of toroidal current, forces and torques on the vessel, forces on the first wall, magnetic field on P.F. coils, etc.) are summarized. Comments and physical interpretations are of general type and therefore applicable to other devices (e.g. NET-II, ITER); in this respect a simple scaling with the plasma current is provided.

A fully three-dimensional eddy-current code is adapted to treat the problem of control of the vertical instability in an air-core tokamak. A rigid model of the plasma displacement is adopted. The definitions of stability margin and residual growth time are generalized. Examples of passive stability analyses are presented in the limit case of zero plasma mass. >

The paper deals with some electromagnetic aspects of interest for their impact on the ITER project. In particular the following topics are illustrated: • feedback control of the distance between plasma separatrix and first wall; • identification of the plasma separatrix from signals detected by magnetic sensors; • ohmic losses in the cold structures, during normal operation scenario and plasma disruptions. The numerical results, which mainly refer to the ITER TAC-8 configuration, are discussed in the paper with the aim of contributing to the still open ITER design options.

This paper introduces a simple technique based on the use of a neural network model for shape reconstruction in ITER as well as for selecting a number of sensors and locations of particular relevance for shape identification. The method has been applied to a dataset of about 300 limiter discharges.

Electromagnetic aspects in the design of ITER-like reactors call for an exrensive use of complex and advanced numerical codes. For this reason a strong attention has been paid within the NET-Team to the code development. In particular, through a cooperation with some Italian universities, during the last years a number of numerical procedures were developed and integrated. In order to assess the code reliability and to gain confidence on their predictions for next generation ITER-like reactors, the validation of the codes against experiments has to be considered as a strict requirement. Aim of this paper is to give a comprehensive presentation of this problem in the light of the results of a campaign of validation runs. The main outcone of this work is that the computational procedures, which have been developed for the NET project and then extensively used also for ITER studies, can be considered as experimentally validated in a sufficiently wide range of cases of interest. In particular, computed values are compared with experimental measurements made during some typical ASDEX-Upgrade discharges. From the electromagnetic point of view, many features of this machine are common to the ITER concept, so that the results of the validation can reasonably be extended to the ITER case.

Electromagnetic aspects play a central role in the design of ITER-like reactors and call then for an extensive use of complex and advanced numerical codes; therefore the validation of the code predictions against experiments has to be considered as a strict requirement. Aim of the paper is to give a comprehensive presentation of this problem in the light of the results of a campaign of validation runs. The main outcome of this work is that a number of computational procedures, which have been developed for the NET project and then extensively used also for ITER studies, can be considered as experimentally validated in a sufficiently wide range of cases of interest. In particular, computed values of magnetic signals are compared with experimental measurements made during some typical ASDEX-U discharges. From the electromagnetic point of view, many features of this machine are common to the ITER concept, so that the results of the validation can reasonably be extended to the ITER case.

The aim of the work is to present the results of some computer simulations of the ASDEX-Upgrade shot #942 obtained with a free boundary MHD equilibrium code The computed magnetic flux and poloidal field are compared to the measured values. Furthermore the sensitivity of these results to the variations of some physical and modeling parameters is studied.

The class of electromagnetic phenomena occurring in a fusion reactor based on the tokamak concept is so wide that a variety of models and related numerical techniques are needed to give a satisfactory solution to the main design problems.

A technique is presented which is based on the balancing of the dynamic system resulting from an eddy current problem. The algorithm employed is suitable for reducing high-order systems, as required when studying a geometrically complex current pattern. The technique has been successfully applied to the reduction of the model for the study of the vertical instability of the plasma in the ITER tokamak reactor