Using advanced camera image and data analysis to address an important hurdle for magnetic fusion energy

A magnetically confined fusion tokamak reactor does not operate in isolation: Fusion reactions generate power which, in steady state, is transferred to surrounding surfaces. In particular, 20% of the fusion power heats the core plasma sustaining the fusion reactions. That same power is slowly transported to the plasma edge, through the separatrix and flows through a thin (~mm) channel (SOL) along the magnetic field to the divertor plasma and ‘targets’; such flows can be far beyond what material surfaces can survive (GW/m2). We use the divertor plasma to remove power from the heat flow through atomic and molecular processes, and distribute that power over the entire divertor region surfaces as opposed to the narrow footprint at the targets. While that dissipation is generally successful there are large uncertainties (compared to core plasma predictions) in what dissipation can be achieved in a reactor - and thus whether a tokamak fusion reactor is viable.
Central to improved understanding of divertor plasma physics, and thus a research path to optimise fusion, is to improve the information that we can extract from measurements of divertor characteristics. While there are a number of divertor diagnostics measurements, the standard methods for their analysis focus on one diagnostic at a time, limiting our ability to infer properties of the divertor that are consistent with all the measurements.
To address the shortcomings in divertor physics understanding the York-CCFE divertor team has developed a prototype ‘Integrated Data Analysis’ (IDA) technique that uses a Bayesian framework to simultaneously ‘fit’ a complete 2D divertor solution to all diagnostic measurements. In the development test cases, it provides the most ‘probable’ 2D map of the divertor characteristics (e.g. density and temperature). The goal of this work is to continue to develop the IDA past this first stage and apply it to real divertor plasmas - with primary focus on the divertor of the MAST-U tokamak which is being brought into operation towards the end of 2019. Application to other tokamak experiments elsewhere is under consideration as well.
While the IDA method utilises multiple divertor diagnostics, CCD cameras viewing the divertor region have a unique characteristic: each camera image is of the entire divertor region as opposed to other diagnostics which provide information only at specific points across the divertor. Each camera is filtered in wavelength to select particular plasma emission lines, the intensity of which, through atomic physics, is dependent on the local plasma characteristics of density and temperature. Given the centrality of the camera system the student would join CCFE staff in the camera system operation and data analysis in addition to the development and application of the IDA method.
There are a number of proposed new divertor configurations that have been theoretically suggested to significantly improve the heat spreading action and control of the divertor plasma, thus making a tokamak fusion reactor less risky. MAST-U has been designed to provide an optimal platform for the study and comparison of such configurations. The IDA provides the best tool for making that comparison as well as generally improving our understanding of divertor physics needed for increasing our confidence in designing/building a tokamak fusion reactor.
This project will be mainly based at the Culham Centre for Fusion Energy (CCFE), the location of the MAST-U and JET tokamaks.
The student will gain experience in a number of areas. Computational skills will be used in most aspects of the research from experiments and the analysis. The student will gain experience with Bayesian analysis to a level that the student finds comfortable. Through assisting and basic analysis of the camera data the student will acquire skills associated with calibrations and operation.