Page 116 - Fundamentals of Magnetic Thermonuclear Reactor Design

100 Fundamentals of Magnetic Thermonuclear Reactor Design

Thus, these algorithmic approaches enable the field source modelling to be
accurate to the accuracy of the plasma model for all operation scenarios.
The total contribution of all currents cannot be estimated only from their
integrated value or the integrated value of associated loads because crucial
local mechanical and thermal loads may occur. The tokamak-type devices
typically have many electrical contacts between their components. In the
ITER machine, this is a critical issue because of the high number of in-
vessel components. A local contact area is usually small. However, the total
effect can be significant that necessitates reasonable FE meshing over these
areas.
Despite the limited usage of ferromagnetic materials, in some cases one
has to account for their perturbation effect on the magnetic field. This includes
evaluation of the toroidal field ripple, the field level at the plasma breakdown
area, anticipated error fields, and so on. The field perturbations associated with
ferromagnetic components can be estimated by solving linearised local prob-
lems, in which magnetic permeability generally features a piecewise constant
distribution in ferromagnetics.
An important issue is high-precision computations of magnetic fields due
to coil currents to assess field perturbations caused by the coil manufacture/
assembly deviations that are the main contributors to error fields in tokamaks.
For this purpose, an original computational technique is applied to provide a
precision 3D field simulation applicable for all typical current configurations.
This technique has been implemented in the KLONDIKE code and associated
procedure libraries.
Also, note that, first, a need for high-precision field computations is as-
sociated with the stiffness of the original systems of differential equations
describing EM transients. The methods for improving the accuracy are quite
standard. They include partitioning of current-carrying components into
smaller elements and reduction of interpolation errors involved in a numeri-
cal integration.

4.4.3 Global Computational Models Based
on Conducting Shells

The ITER’s complex EM behaviour resulting from a strong inductive coupling
calls for global computational models.
A global model must contain a mathematical description of two groups of
components, namely, the sources of ‘primary’ magnetic fields noted earlier, and
the conducting structures, where eddy currents are induced. Those conducting
structures include
l the vacuum vessel and in-vessel components,
l the thermal shield, and
l the cryostat.