Search

Results from two different sets of JET experiments are presented. In experiments in which toroidicity-induced Alfvén eigenmodes (TAEs) localized at different radial locations had the same frequencies and toroidal mode numbers, the occurrence of enhanced losses after the excitation of TAEs in the core of the plasma was observed. On the contrary, enhanced losses were not observed if the TAEs localized at different radial locations had different frequencies and toroidal mode numbers. Numerical modeling indicates that, in the first set of experiments, the enhanced losses were caused by a combined effect of the TAEs localized at different radial locations. The TAEs localized in the plasma core transported energetic ions from the core to outer regions of the plasma. Then, the TAEs localized in outer regions of the plasma interacted with these ions just transported by the core-localized TAEs causing a further radial displacement of the ions to the plasma edge. This process eventually ends up causing the loss of the resonant ions. In the second set of experiments, it was found that TAEs localized in the plasma core and in outer regions did not interact with the same ions and so no enhanced losses were measured. Sheared profiles of the safety factor combined with flat mass density profiles lead to larger differences on the frequencies of the TAEs localized at different radial locations, eventually avoiding loss of energetic ions through the described mechanism. [ABSTRACT FROM AUTHOR]

Fast ions in fusion plasmas often leave characteristic signatures in the neutron emission from the plasma. In this paper, we show how neutron measurements can be used to study fast ions and give examples of physics results obtained on present day tokamaks. The focus is on measurements with dedicated neutron spectrometers and with compact neutron detectors used in each channel of neutron profile monitors. A measured neutron spectrum can be analyzed in several different ways, depending on the physics scenario under consideration. Gross features of a fast ion energy distribution can be studied by applying suitably chosen thresholds to the measured spectrum, thus probing ions with different energies. With this technique it is possible to study the interaction between fast ions and MHD activity, such as toroidal Alfvén eigenmodes (TAEs) and sawtooth instabilities. Quantitative comparisons with modeling can be performed by a direct computation of the neutron emission expected from a given fast ion distribution. Within this framework it is also possible to determine physics parameters, such as the supra-thermal fraction of the neutron emission, by fitting model parameters to the data. A detailed, model-independent estimate of the fast ion distribution can be obtained by analyzing the data in terms of velocity space weight functions. Using this method, fast ion distributions can be resolved in both energy and pitch by combining neutron and gamma-ray measurements obtained along several different sightlines. Fast ion measurements of the type described in this paper will also be possible at ITER, provided that the spectrometers have the dynamic range required to resolve the fast ion spectral features in the presence of the dominating thermonuclear neutron emission. A dedicated high-resolution neutron spectrometer has been designed for this purpose. [ABSTRACT FROM AUTHOR]

A novel approach to uncertainty quantification in codes simulating fusion plasma, deterministic sampling (DS), is evaluated. This method uses a few carefully selected samples and can be used to propagate input parameter uncertainties through calculations where other sampling methods are unmanageable due to time constraints. The primary analysis is performed on the transport code TRANSP, but another faster code is also tested where a comparison with Monte Carlo sampling is made. The tests, performed with two pulses at the Joint European Torus (JET), show that even lower order DS will give a reliable estimation of the standard deviation of the calculated neutron rate. However, a higher order DS can give information about higher output moments, such as skewness an kurtosis. The TRANSP-simulated neutron rates of both examined pulses are found to have an uncertainty with an upward skewness, meaning input parameter uncertainties are can better explain an underestimation of the neutron rate than an overestimation. This information can, for example, be lead to a better benchmarking comparison between the measured and calculated neutron rates. [ABSTRACT FROM AUTHOR]

In this paper, the scattered and direct neutron fluxes in the line of sight (LOS) of the TOFOR neutron spectrometer at JET are simulated and the simulations compared with measurement results. The Monte Carlo code MCNPX is used in the simulations, with a vessel material composition obtained from the JET drawing office and neutron emission profiles calculated from TRANSP simulations of beam ion density profiles. The MCNPX simulations show that the material composition of the scattering wall has a large effect on the shape of the scattered neutron spectrum. Neutron source profile shapes as well as radial and vertical source displacements in the TOFOR LOS are shown to only marginally affect the scatter, while having a larger impact on the direct neutron flux. A matrix of simulated scatter spectra for mono-energetic source neutrons is created which is folded with an approximation of the source spectrum for each JET pulse studied to obtain a scatter component for use in the data analysis. The scatter components thus obtained are shown to describe the measured data. It is also demonstrated that the scattered flux is approximately constant relative to the total neutron yield as measured with the JET fission chambers, while there is a larger spread in the direct flux, consistent with simulations. The simulated effect on the integrated scattered/direct ratio of an increase with movements outward along the radial direction and a drop at higher values of the vertical plasma position is also reproduced in the measurements. Finally, the quantitative agreement found in scatter/direct ratios between simulations (0.185 +- 0.005) and measurements (0.187 +- 0.050) serves as a solid benchmark of the MCNPX model used. [ABSTRACT FROM AUTHOR]